Commit
•
8e957b4
1
Parent(s):
5314277
Upload folder using huggingface_hub (#2)
Browse files- Upload folder using huggingface_hub (8db594905b288f03a9b4377d4b0a631203fecea1)
This view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +6 -0
- README.md +34 -0
- adapter_config.json +25 -0
- adapter_model.bin +3 -0
- checkpoint-100/README.md +34 -0
- checkpoint-100/adapter_config.json +25 -0
- checkpoint-100/adapter_model.bin +3 -0
- checkpoint-100/optimizer.pt +3 -0
- checkpoint-100/rng_state.pth +3 -0
- checkpoint-100/scheduler.pt +3 -0
- checkpoint-100/special_tokens_map.json +6 -0
- checkpoint-100/tokenizer.json +3 -0
- checkpoint-100/tokenizer_config.json +10 -0
- checkpoint-100/trainer_state.json +619 -0
- checkpoint-100/training_args.bin +3 -0
- checkpoint-200/README.md +34 -0
- checkpoint-200/adapter_config.json +25 -0
- checkpoint-200/adapter_model.bin +3 -0
- checkpoint-200/optimizer.pt +3 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/special_tokens_map.json +6 -0
- checkpoint-200/tokenizer.json +3 -0
- checkpoint-200/tokenizer_config.json +10 -0
- checkpoint-200/trainer_state.json +1219 -0
- checkpoint-200/training_args.bin +3 -0
- checkpoint-300/README.md +34 -0
- checkpoint-300/adapter_config.json +25 -0
- checkpoint-300/adapter_model.bin +3 -0
- checkpoint-300/optimizer.pt +3 -0
- checkpoint-300/rng_state.pth +3 -0
- checkpoint-300/scheduler.pt +3 -0
- checkpoint-300/special_tokens_map.json +6 -0
- checkpoint-300/tokenizer.json +3 -0
- checkpoint-300/tokenizer_config.json +10 -0
- checkpoint-300/trainer_state.json +1819 -0
- checkpoint-300/training_args.bin +3 -0
- checkpoint-400/README.md +34 -0
- checkpoint-400/adapter_config.json +25 -0
- checkpoint-400/adapter_model.bin +3 -0
- checkpoint-400/optimizer.pt +3 -0
- checkpoint-400/rng_state.pth +3 -0
- checkpoint-400/scheduler.pt +3 -0
- checkpoint-400/special_tokens_map.json +6 -0
- checkpoint-400/tokenizer.json +3 -0
- checkpoint-400/tokenizer_config.json +10 -0
- checkpoint-400/trainer_state.json +2419 -0
- checkpoint-400/training_args.bin +3 -0
- checkpoint-500/README.md +34 -0
- checkpoint-500/adapter_config.json +25 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a125cc125172a0c45ae8a435cdae72b0c657f9b899e914953f9df21478011a8
|
| 3 |
+
size 39409357
|
checkpoint-100/README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
checkpoint-100/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-100/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1494d9295feb24e846e9762b1b2118d8c5592c6b44f650b463565b33581b2bd
|
| 3 |
+
size 39409357
|
checkpoint-100/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0a5a340ededaf6a479a86b5f1d38b1ad96df10d24d2e3b600016733e406c965
|
| 3 |
+
size 78844421
|
checkpoint-100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6869750f95a25c4e970298a33adf90e2d7ab52680bf3317239bff1b10103235
|
| 3 |
+
size 14575
|
checkpoint-100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a58b95c1126bab491ac28f5f742cbad6d9169c46a57b0c332b4be7e7507c82f5
|
| 3 |
+
size 627
|
checkpoint-100/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
checkpoint-100/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
|
| 3 |
+
size 14500471
|
checkpoint-100/tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"tokenizer_class": "BloomTokenizer",
|
| 9 |
+
"unk_token": "<unk>"
|
| 10 |
+
}
|
checkpoint-100/trainer_state.json
ADDED
|
@@ -0,0 +1,619 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 0.0001996,
|
| 14 |
+
"loss": 2.4683,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 0.00019920000000000002,
|
| 20 |
+
"loss": 2.2023,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 0.0001988,
|
| 26 |
+
"loss": 2.3039,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 0.0001984,
|
| 32 |
+
"loss": 2.1855,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.2,
|
| 37 |
+
"learning_rate": 0.00019800000000000002,
|
| 38 |
+
"loss": 2.328,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.24,
|
| 43 |
+
"learning_rate": 0.0001976,
|
| 44 |
+
"loss": 2.3407,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.28,
|
| 49 |
+
"learning_rate": 0.0001972,
|
| 50 |
+
"loss": 2.3432,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"learning_rate": 0.0001968,
|
| 56 |
+
"loss": 2.2333,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.36,
|
| 61 |
+
"learning_rate": 0.0001964,
|
| 62 |
+
"loss": 2.1761,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.4,
|
| 67 |
+
"learning_rate": 0.000196,
|
| 68 |
+
"loss": 2.1473,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.44,
|
| 73 |
+
"learning_rate": 0.0001956,
|
| 74 |
+
"loss": 2.2076,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.48,
|
| 79 |
+
"learning_rate": 0.0001952,
|
| 80 |
+
"loss": 2.1925,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.52,
|
| 85 |
+
"learning_rate": 0.0001948,
|
| 86 |
+
"loss": 2.1694,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.56,
|
| 91 |
+
"learning_rate": 0.0001944,
|
| 92 |
+
"loss": 2.1056,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6,
|
| 97 |
+
"learning_rate": 0.000194,
|
| 98 |
+
"loss": 2.2426,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.64,
|
| 103 |
+
"learning_rate": 0.00019360000000000002,
|
| 104 |
+
"loss": 2.2635,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.68,
|
| 109 |
+
"learning_rate": 0.0001932,
|
| 110 |
+
"loss": 2.251,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.72,
|
| 115 |
+
"learning_rate": 0.0001928,
|
| 116 |
+
"loss": 2.224,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.76,
|
| 121 |
+
"learning_rate": 0.00019240000000000001,
|
| 122 |
+
"loss": 2.1848,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.8,
|
| 127 |
+
"learning_rate": 0.000192,
|
| 128 |
+
"loss": 2.1341,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.84,
|
| 133 |
+
"learning_rate": 0.0001916,
|
| 134 |
+
"loss": 2.126,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.88,
|
| 139 |
+
"learning_rate": 0.0001912,
|
| 140 |
+
"loss": 2.0959,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.92,
|
| 145 |
+
"learning_rate": 0.0001908,
|
| 146 |
+
"loss": 2.2446,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.96,
|
| 151 |
+
"learning_rate": 0.0001904,
|
| 152 |
+
"loss": 1.9448,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.0,
|
| 157 |
+
"learning_rate": 0.00019,
|
| 158 |
+
"loss": 2.0939,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.04,
|
| 163 |
+
"learning_rate": 0.0001896,
|
| 164 |
+
"loss": 2.1247,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.08,
|
| 169 |
+
"learning_rate": 0.0001892,
|
| 170 |
+
"loss": 1.9417,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.12,
|
| 175 |
+
"learning_rate": 0.0001888,
|
| 176 |
+
"loss": 2.105,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.16,
|
| 181 |
+
"learning_rate": 0.0001884,
|
| 182 |
+
"loss": 2.1595,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.2,
|
| 187 |
+
"learning_rate": 0.000188,
|
| 188 |
+
"loss": 2.2009,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.24,
|
| 193 |
+
"learning_rate": 0.0001876,
|
| 194 |
+
"loss": 2.0784,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.28,
|
| 199 |
+
"learning_rate": 0.00018720000000000002,
|
| 200 |
+
"loss": 2.2173,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.32,
|
| 205 |
+
"learning_rate": 0.00018680000000000001,
|
| 206 |
+
"loss": 2.1185,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.36,
|
| 211 |
+
"learning_rate": 0.00018640000000000003,
|
| 212 |
+
"loss": 1.988,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4,
|
| 217 |
+
"learning_rate": 0.00018600000000000002,
|
| 218 |
+
"loss": 1.9493,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.44,
|
| 223 |
+
"learning_rate": 0.0001856,
|
| 224 |
+
"loss": 1.9947,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.48,
|
| 229 |
+
"learning_rate": 0.00018520000000000003,
|
| 230 |
+
"loss": 2.0506,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.52,
|
| 235 |
+
"learning_rate": 0.00018480000000000002,
|
| 236 |
+
"loss": 2.0829,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.56,
|
| 241 |
+
"learning_rate": 0.0001844,
|
| 242 |
+
"loss": 2.1455,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.6,
|
| 247 |
+
"learning_rate": 0.00018400000000000003,
|
| 248 |
+
"loss": 2.0304,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.64,
|
| 253 |
+
"learning_rate": 0.00018360000000000002,
|
| 254 |
+
"loss": 2.006,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.68,
|
| 259 |
+
"learning_rate": 0.0001832,
|
| 260 |
+
"loss": 2.1759,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.72,
|
| 265 |
+
"learning_rate": 0.00018280000000000003,
|
| 266 |
+
"loss": 2.0093,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.76,
|
| 271 |
+
"learning_rate": 0.00018240000000000002,
|
| 272 |
+
"loss": 2.0683,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.8,
|
| 277 |
+
"learning_rate": 0.000182,
|
| 278 |
+
"loss": 2.1352,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.84,
|
| 283 |
+
"learning_rate": 0.00018160000000000002,
|
| 284 |
+
"loss": 1.9736,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.88,
|
| 289 |
+
"learning_rate": 0.0001812,
|
| 290 |
+
"loss": 1.9631,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.92,
|
| 295 |
+
"learning_rate": 0.0001808,
|
| 296 |
+
"loss": 2.1024,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.96,
|
| 301 |
+
"learning_rate": 0.00018040000000000002,
|
| 302 |
+
"loss": 1.9895,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.0,
|
| 307 |
+
"learning_rate": 0.00018,
|
| 308 |
+
"loss": 2.0067,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.04,
|
| 313 |
+
"learning_rate": 0.0001796,
|
| 314 |
+
"loss": 2.0509,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.08,
|
| 319 |
+
"learning_rate": 0.00017920000000000002,
|
| 320 |
+
"loss": 1.8795,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.12,
|
| 325 |
+
"learning_rate": 0.0001788,
|
| 326 |
+
"loss": 2.0633,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.16,
|
| 331 |
+
"learning_rate": 0.0001784,
|
| 332 |
+
"loss": 1.9998,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.2,
|
| 337 |
+
"learning_rate": 0.00017800000000000002,
|
| 338 |
+
"loss": 2.0675,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.24,
|
| 343 |
+
"learning_rate": 0.0001776,
|
| 344 |
+
"loss": 2.0129,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.28,
|
| 349 |
+
"learning_rate": 0.0001772,
|
| 350 |
+
"loss": 1.9302,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.32,
|
| 355 |
+
"learning_rate": 0.00017680000000000001,
|
| 356 |
+
"loss": 2.0472,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.36,
|
| 361 |
+
"learning_rate": 0.0001764,
|
| 362 |
+
"loss": 2.0031,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.4,
|
| 367 |
+
"learning_rate": 0.00017600000000000002,
|
| 368 |
+
"loss": 2.0231,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.44,
|
| 373 |
+
"learning_rate": 0.0001756,
|
| 374 |
+
"loss": 2.0192,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.48,
|
| 379 |
+
"learning_rate": 0.0001752,
|
| 380 |
+
"loss": 2.1336,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.52,
|
| 385 |
+
"learning_rate": 0.00017480000000000002,
|
| 386 |
+
"loss": 1.7753,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.56,
|
| 391 |
+
"learning_rate": 0.0001744,
|
| 392 |
+
"loss": 2.0274,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.6,
|
| 397 |
+
"learning_rate": 0.000174,
|
| 398 |
+
"loss": 1.8893,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.64,
|
| 403 |
+
"learning_rate": 0.00017360000000000002,
|
| 404 |
+
"loss": 2.0341,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.68,
|
| 409 |
+
"learning_rate": 0.0001732,
|
| 410 |
+
"loss": 2.1081,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.72,
|
| 415 |
+
"learning_rate": 0.0001728,
|
| 416 |
+
"loss": 1.9463,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.76,
|
| 421 |
+
"learning_rate": 0.00017240000000000002,
|
| 422 |
+
"loss": 2.0607,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.8,
|
| 427 |
+
"learning_rate": 0.000172,
|
| 428 |
+
"loss": 1.9803,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.84,
|
| 433 |
+
"learning_rate": 0.0001716,
|
| 434 |
+
"loss": 1.7663,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.88,
|
| 439 |
+
"learning_rate": 0.00017120000000000001,
|
| 440 |
+
"loss": 2.1116,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.92,
|
| 445 |
+
"learning_rate": 0.0001708,
|
| 446 |
+
"loss": 1.9039,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.96,
|
| 451 |
+
"learning_rate": 0.0001704,
|
| 452 |
+
"loss": 2.049,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.0,
|
| 457 |
+
"learning_rate": 0.00017,
|
| 458 |
+
"loss": 1.991,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.04,
|
| 463 |
+
"learning_rate": 0.0001696,
|
| 464 |
+
"loss": 1.8755,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 3.08,
|
| 469 |
+
"learning_rate": 0.0001692,
|
| 470 |
+
"loss": 2.0367,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.12,
|
| 475 |
+
"learning_rate": 0.0001688,
|
| 476 |
+
"loss": 1.9959,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.16,
|
| 481 |
+
"learning_rate": 0.0001684,
|
| 482 |
+
"loss": 1.8051,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.2,
|
| 487 |
+
"learning_rate": 0.000168,
|
| 488 |
+
"loss": 2.0446,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.24,
|
| 493 |
+
"learning_rate": 0.0001676,
|
| 494 |
+
"loss": 1.7873,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.28,
|
| 499 |
+
"learning_rate": 0.0001672,
|
| 500 |
+
"loss": 1.9264,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 3.32,
|
| 505 |
+
"learning_rate": 0.0001668,
|
| 506 |
+
"loss": 1.8595,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.36,
|
| 511 |
+
"learning_rate": 0.0001664,
|
| 512 |
+
"loss": 1.9992,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.4,
|
| 517 |
+
"learning_rate": 0.000166,
|
| 518 |
+
"loss": 1.8783,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.44,
|
| 523 |
+
"learning_rate": 0.0001656,
|
| 524 |
+
"loss": 2.0216,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.48,
|
| 529 |
+
"learning_rate": 0.0001652,
|
| 530 |
+
"loss": 1.947,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.52,
|
| 535 |
+
"learning_rate": 0.0001648,
|
| 536 |
+
"loss": 1.9554,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.56,
|
| 541 |
+
"learning_rate": 0.0001644,
|
| 542 |
+
"loss": 1.8563,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.6,
|
| 547 |
+
"learning_rate": 0.000164,
|
| 548 |
+
"loss": 1.7795,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.64,
|
| 553 |
+
"learning_rate": 0.0001636,
|
| 554 |
+
"loss": 1.9347,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.68,
|
| 559 |
+
"learning_rate": 0.0001632,
|
| 560 |
+
"loss": 2.0078,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.72,
|
| 565 |
+
"learning_rate": 0.0001628,
|
| 566 |
+
"loss": 1.9964,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 3.76,
|
| 571 |
+
"learning_rate": 0.00016240000000000002,
|
| 572 |
+
"loss": 1.9004,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.8,
|
| 577 |
+
"learning_rate": 0.000162,
|
| 578 |
+
"loss": 1.9446,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.84,
|
| 583 |
+
"learning_rate": 0.00016160000000000002,
|
| 584 |
+
"loss": 1.8857,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.88,
|
| 589 |
+
"learning_rate": 0.00016120000000000002,
|
| 590 |
+
"loss": 1.8797,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.92,
|
| 595 |
+
"learning_rate": 0.0001608,
|
| 596 |
+
"loss": 1.7643,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.96,
|
| 601 |
+
"learning_rate": 0.00016040000000000002,
|
| 602 |
+
"loss": 1.9724,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 4.0,
|
| 607 |
+
"learning_rate": 0.00016,
|
| 608 |
+
"loss": 1.9058,
|
| 609 |
+
"step": 100
|
| 610 |
+
}
|
| 611 |
+
],
|
| 612 |
+
"logging_steps": 1,
|
| 613 |
+
"max_steps": 500,
|
| 614 |
+
"num_train_epochs": 20,
|
| 615 |
+
"save_steps": 100,
|
| 616 |
+
"total_flos": 3.406529677264896e+16,
|
| 617 |
+
"trial_name": null,
|
| 618 |
+
"trial_params": null
|
| 619 |
+
}
|
checkpoint-100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ef74b31950ae6c8955316bed48c343fb06cda0cb6a8a54ca46ca3cb681c8736
|
| 3 |
+
size 4027
|
checkpoint-200/README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
checkpoint-200/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-200/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6eb0dcfb489eef093d9f20354ad4dbd76c3cacdee8012aaf3a92dc2a4267e697
|
| 3 |
+
size 39409357
|
checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5f31f7a2996c207ab778c7745622649ab7de438c906222dc8ef5f81d8c4ed10
|
| 3 |
+
size 78844421
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36fc71bd44bd7f04f2599c5dface64c517de1a7ab7bac3600f3f6470c6c72673
|
| 3 |
+
size 14575
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cf1ea83db88b65824ac70ed44968b05e8a83b81aa97e9d6a33c16e33168c1fe
|
| 3 |
+
size 627
|
checkpoint-200/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
checkpoint-200/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
|
| 3 |
+
size 14500471
|
checkpoint-200/tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"tokenizer_class": "BloomTokenizer",
|
| 9 |
+
"unk_token": "<unk>"
|
| 10 |
+
}
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,1219 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 0.0001996,
|
| 14 |
+
"loss": 2.4683,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 0.00019920000000000002,
|
| 20 |
+
"loss": 2.2023,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 0.0001988,
|
| 26 |
+
"loss": 2.3039,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 0.0001984,
|
| 32 |
+
"loss": 2.1855,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.2,
|
| 37 |
+
"learning_rate": 0.00019800000000000002,
|
| 38 |
+
"loss": 2.328,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.24,
|
| 43 |
+
"learning_rate": 0.0001976,
|
| 44 |
+
"loss": 2.3407,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.28,
|
| 49 |
+
"learning_rate": 0.0001972,
|
| 50 |
+
"loss": 2.3432,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"learning_rate": 0.0001968,
|
| 56 |
+
"loss": 2.2333,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.36,
|
| 61 |
+
"learning_rate": 0.0001964,
|
| 62 |
+
"loss": 2.1761,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.4,
|
| 67 |
+
"learning_rate": 0.000196,
|
| 68 |
+
"loss": 2.1473,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.44,
|
| 73 |
+
"learning_rate": 0.0001956,
|
| 74 |
+
"loss": 2.2076,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.48,
|
| 79 |
+
"learning_rate": 0.0001952,
|
| 80 |
+
"loss": 2.1925,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.52,
|
| 85 |
+
"learning_rate": 0.0001948,
|
| 86 |
+
"loss": 2.1694,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.56,
|
| 91 |
+
"learning_rate": 0.0001944,
|
| 92 |
+
"loss": 2.1056,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6,
|
| 97 |
+
"learning_rate": 0.000194,
|
| 98 |
+
"loss": 2.2426,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.64,
|
| 103 |
+
"learning_rate": 0.00019360000000000002,
|
| 104 |
+
"loss": 2.2635,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.68,
|
| 109 |
+
"learning_rate": 0.0001932,
|
| 110 |
+
"loss": 2.251,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.72,
|
| 115 |
+
"learning_rate": 0.0001928,
|
| 116 |
+
"loss": 2.224,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.76,
|
| 121 |
+
"learning_rate": 0.00019240000000000001,
|
| 122 |
+
"loss": 2.1848,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.8,
|
| 127 |
+
"learning_rate": 0.000192,
|
| 128 |
+
"loss": 2.1341,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.84,
|
| 133 |
+
"learning_rate": 0.0001916,
|
| 134 |
+
"loss": 2.126,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.88,
|
| 139 |
+
"learning_rate": 0.0001912,
|
| 140 |
+
"loss": 2.0959,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.92,
|
| 145 |
+
"learning_rate": 0.0001908,
|
| 146 |
+
"loss": 2.2446,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.96,
|
| 151 |
+
"learning_rate": 0.0001904,
|
| 152 |
+
"loss": 1.9448,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.0,
|
| 157 |
+
"learning_rate": 0.00019,
|
| 158 |
+
"loss": 2.0939,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.04,
|
| 163 |
+
"learning_rate": 0.0001896,
|
| 164 |
+
"loss": 2.1247,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.08,
|
| 169 |
+
"learning_rate": 0.0001892,
|
| 170 |
+
"loss": 1.9417,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.12,
|
| 175 |
+
"learning_rate": 0.0001888,
|
| 176 |
+
"loss": 2.105,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.16,
|
| 181 |
+
"learning_rate": 0.0001884,
|
| 182 |
+
"loss": 2.1595,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.2,
|
| 187 |
+
"learning_rate": 0.000188,
|
| 188 |
+
"loss": 2.2009,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.24,
|
| 193 |
+
"learning_rate": 0.0001876,
|
| 194 |
+
"loss": 2.0784,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.28,
|
| 199 |
+
"learning_rate": 0.00018720000000000002,
|
| 200 |
+
"loss": 2.2173,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.32,
|
| 205 |
+
"learning_rate": 0.00018680000000000001,
|
| 206 |
+
"loss": 2.1185,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.36,
|
| 211 |
+
"learning_rate": 0.00018640000000000003,
|
| 212 |
+
"loss": 1.988,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4,
|
| 217 |
+
"learning_rate": 0.00018600000000000002,
|
| 218 |
+
"loss": 1.9493,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.44,
|
| 223 |
+
"learning_rate": 0.0001856,
|
| 224 |
+
"loss": 1.9947,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.48,
|
| 229 |
+
"learning_rate": 0.00018520000000000003,
|
| 230 |
+
"loss": 2.0506,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.52,
|
| 235 |
+
"learning_rate": 0.00018480000000000002,
|
| 236 |
+
"loss": 2.0829,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.56,
|
| 241 |
+
"learning_rate": 0.0001844,
|
| 242 |
+
"loss": 2.1455,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.6,
|
| 247 |
+
"learning_rate": 0.00018400000000000003,
|
| 248 |
+
"loss": 2.0304,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.64,
|
| 253 |
+
"learning_rate": 0.00018360000000000002,
|
| 254 |
+
"loss": 2.006,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.68,
|
| 259 |
+
"learning_rate": 0.0001832,
|
| 260 |
+
"loss": 2.1759,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.72,
|
| 265 |
+
"learning_rate": 0.00018280000000000003,
|
| 266 |
+
"loss": 2.0093,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.76,
|
| 271 |
+
"learning_rate": 0.00018240000000000002,
|
| 272 |
+
"loss": 2.0683,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.8,
|
| 277 |
+
"learning_rate": 0.000182,
|
| 278 |
+
"loss": 2.1352,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.84,
|
| 283 |
+
"learning_rate": 0.00018160000000000002,
|
| 284 |
+
"loss": 1.9736,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.88,
|
| 289 |
+
"learning_rate": 0.0001812,
|
| 290 |
+
"loss": 1.9631,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.92,
|
| 295 |
+
"learning_rate": 0.0001808,
|
| 296 |
+
"loss": 2.1024,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.96,
|
| 301 |
+
"learning_rate": 0.00018040000000000002,
|
| 302 |
+
"loss": 1.9895,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.0,
|
| 307 |
+
"learning_rate": 0.00018,
|
| 308 |
+
"loss": 2.0067,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.04,
|
| 313 |
+
"learning_rate": 0.0001796,
|
| 314 |
+
"loss": 2.0509,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.08,
|
| 319 |
+
"learning_rate": 0.00017920000000000002,
|
| 320 |
+
"loss": 1.8795,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.12,
|
| 325 |
+
"learning_rate": 0.0001788,
|
| 326 |
+
"loss": 2.0633,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.16,
|
| 331 |
+
"learning_rate": 0.0001784,
|
| 332 |
+
"loss": 1.9998,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.2,
|
| 337 |
+
"learning_rate": 0.00017800000000000002,
|
| 338 |
+
"loss": 2.0675,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.24,
|
| 343 |
+
"learning_rate": 0.0001776,
|
| 344 |
+
"loss": 2.0129,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.28,
|
| 349 |
+
"learning_rate": 0.0001772,
|
| 350 |
+
"loss": 1.9302,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.32,
|
| 355 |
+
"learning_rate": 0.00017680000000000001,
|
| 356 |
+
"loss": 2.0472,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.36,
|
| 361 |
+
"learning_rate": 0.0001764,
|
| 362 |
+
"loss": 2.0031,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.4,
|
| 367 |
+
"learning_rate": 0.00017600000000000002,
|
| 368 |
+
"loss": 2.0231,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.44,
|
| 373 |
+
"learning_rate": 0.0001756,
|
| 374 |
+
"loss": 2.0192,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.48,
|
| 379 |
+
"learning_rate": 0.0001752,
|
| 380 |
+
"loss": 2.1336,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.52,
|
| 385 |
+
"learning_rate": 0.00017480000000000002,
|
| 386 |
+
"loss": 1.7753,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.56,
|
| 391 |
+
"learning_rate": 0.0001744,
|
| 392 |
+
"loss": 2.0274,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.6,
|
| 397 |
+
"learning_rate": 0.000174,
|
| 398 |
+
"loss": 1.8893,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.64,
|
| 403 |
+
"learning_rate": 0.00017360000000000002,
|
| 404 |
+
"loss": 2.0341,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.68,
|
| 409 |
+
"learning_rate": 0.0001732,
|
| 410 |
+
"loss": 2.1081,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.72,
|
| 415 |
+
"learning_rate": 0.0001728,
|
| 416 |
+
"loss": 1.9463,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.76,
|
| 421 |
+
"learning_rate": 0.00017240000000000002,
|
| 422 |
+
"loss": 2.0607,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.8,
|
| 427 |
+
"learning_rate": 0.000172,
|
| 428 |
+
"loss": 1.9803,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.84,
|
| 433 |
+
"learning_rate": 0.0001716,
|
| 434 |
+
"loss": 1.7663,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.88,
|
| 439 |
+
"learning_rate": 0.00017120000000000001,
|
| 440 |
+
"loss": 2.1116,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.92,
|
| 445 |
+
"learning_rate": 0.0001708,
|
| 446 |
+
"loss": 1.9039,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.96,
|
| 451 |
+
"learning_rate": 0.0001704,
|
| 452 |
+
"loss": 2.049,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.0,
|
| 457 |
+
"learning_rate": 0.00017,
|
| 458 |
+
"loss": 1.991,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.04,
|
| 463 |
+
"learning_rate": 0.0001696,
|
| 464 |
+
"loss": 1.8755,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 3.08,
|
| 469 |
+
"learning_rate": 0.0001692,
|
| 470 |
+
"loss": 2.0367,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.12,
|
| 475 |
+
"learning_rate": 0.0001688,
|
| 476 |
+
"loss": 1.9959,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.16,
|
| 481 |
+
"learning_rate": 0.0001684,
|
| 482 |
+
"loss": 1.8051,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.2,
|
| 487 |
+
"learning_rate": 0.000168,
|
| 488 |
+
"loss": 2.0446,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.24,
|
| 493 |
+
"learning_rate": 0.0001676,
|
| 494 |
+
"loss": 1.7873,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.28,
|
| 499 |
+
"learning_rate": 0.0001672,
|
| 500 |
+
"loss": 1.9264,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 3.32,
|
| 505 |
+
"learning_rate": 0.0001668,
|
| 506 |
+
"loss": 1.8595,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.36,
|
| 511 |
+
"learning_rate": 0.0001664,
|
| 512 |
+
"loss": 1.9992,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.4,
|
| 517 |
+
"learning_rate": 0.000166,
|
| 518 |
+
"loss": 1.8783,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.44,
|
| 523 |
+
"learning_rate": 0.0001656,
|
| 524 |
+
"loss": 2.0216,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.48,
|
| 529 |
+
"learning_rate": 0.0001652,
|
| 530 |
+
"loss": 1.947,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.52,
|
| 535 |
+
"learning_rate": 0.0001648,
|
| 536 |
+
"loss": 1.9554,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.56,
|
| 541 |
+
"learning_rate": 0.0001644,
|
| 542 |
+
"loss": 1.8563,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.6,
|
| 547 |
+
"learning_rate": 0.000164,
|
| 548 |
+
"loss": 1.7795,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.64,
|
| 553 |
+
"learning_rate": 0.0001636,
|
| 554 |
+
"loss": 1.9347,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.68,
|
| 559 |
+
"learning_rate": 0.0001632,
|
| 560 |
+
"loss": 2.0078,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.72,
|
| 565 |
+
"learning_rate": 0.0001628,
|
| 566 |
+
"loss": 1.9964,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 3.76,
|
| 571 |
+
"learning_rate": 0.00016240000000000002,
|
| 572 |
+
"loss": 1.9004,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.8,
|
| 577 |
+
"learning_rate": 0.000162,
|
| 578 |
+
"loss": 1.9446,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.84,
|
| 583 |
+
"learning_rate": 0.00016160000000000002,
|
| 584 |
+
"loss": 1.8857,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.88,
|
| 589 |
+
"learning_rate": 0.00016120000000000002,
|
| 590 |
+
"loss": 1.8797,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.92,
|
| 595 |
+
"learning_rate": 0.0001608,
|
| 596 |
+
"loss": 1.7643,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.96,
|
| 601 |
+
"learning_rate": 0.00016040000000000002,
|
| 602 |
+
"loss": 1.9724,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 4.0,
|
| 607 |
+
"learning_rate": 0.00016,
|
| 608 |
+
"loss": 1.9058,
|
| 609 |
+
"step": 100
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.04,
|
| 613 |
+
"learning_rate": 0.0001596,
|
| 614 |
+
"loss": 1.8663,
|
| 615 |
+
"step": 101
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 4.08,
|
| 619 |
+
"learning_rate": 0.00015920000000000002,
|
| 620 |
+
"loss": 1.8457,
|
| 621 |
+
"step": 102
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 4.12,
|
| 625 |
+
"learning_rate": 0.0001588,
|
| 626 |
+
"loss": 1.8394,
|
| 627 |
+
"step": 103
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 4.16,
|
| 631 |
+
"learning_rate": 0.00015840000000000003,
|
| 632 |
+
"loss": 1.8941,
|
| 633 |
+
"step": 104
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 4.2,
|
| 637 |
+
"learning_rate": 0.00015800000000000002,
|
| 638 |
+
"loss": 1.8601,
|
| 639 |
+
"step": 105
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 4.24,
|
| 643 |
+
"learning_rate": 0.0001576,
|
| 644 |
+
"loss": 1.7365,
|
| 645 |
+
"step": 106
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 4.28,
|
| 649 |
+
"learning_rate": 0.00015720000000000003,
|
| 650 |
+
"loss": 1.8809,
|
| 651 |
+
"step": 107
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 4.32,
|
| 655 |
+
"learning_rate": 0.00015680000000000002,
|
| 656 |
+
"loss": 2.0206,
|
| 657 |
+
"step": 108
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 4.36,
|
| 661 |
+
"learning_rate": 0.0001564,
|
| 662 |
+
"loss": 1.7776,
|
| 663 |
+
"step": 109
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 4.4,
|
| 667 |
+
"learning_rate": 0.00015600000000000002,
|
| 668 |
+
"loss": 1.9089,
|
| 669 |
+
"step": 110
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.44,
|
| 673 |
+
"learning_rate": 0.00015560000000000001,
|
| 674 |
+
"loss": 1.8464,
|
| 675 |
+
"step": 111
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.48,
|
| 679 |
+
"learning_rate": 0.0001552,
|
| 680 |
+
"loss": 1.7943,
|
| 681 |
+
"step": 112
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.52,
|
| 685 |
+
"learning_rate": 0.00015480000000000002,
|
| 686 |
+
"loss": 1.8081,
|
| 687 |
+
"step": 113
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 4.56,
|
| 691 |
+
"learning_rate": 0.0001544,
|
| 692 |
+
"loss": 1.734,
|
| 693 |
+
"step": 114
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 4.6,
|
| 697 |
+
"learning_rate": 0.000154,
|
| 698 |
+
"loss": 1.8379,
|
| 699 |
+
"step": 115
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.64,
|
| 703 |
+
"learning_rate": 0.00015360000000000002,
|
| 704 |
+
"loss": 1.6841,
|
| 705 |
+
"step": 116
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 4.68,
|
| 709 |
+
"learning_rate": 0.0001532,
|
| 710 |
+
"loss": 1.8253,
|
| 711 |
+
"step": 117
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 4.72,
|
| 715 |
+
"learning_rate": 0.0001528,
|
| 716 |
+
"loss": 1.7607,
|
| 717 |
+
"step": 118
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 4.76,
|
| 721 |
+
"learning_rate": 0.00015240000000000002,
|
| 722 |
+
"loss": 1.8205,
|
| 723 |
+
"step": 119
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 4.8,
|
| 727 |
+
"learning_rate": 0.000152,
|
| 728 |
+
"loss": 1.8443,
|
| 729 |
+
"step": 120
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 4.84,
|
| 733 |
+
"learning_rate": 0.0001516,
|
| 734 |
+
"loss": 1.8824,
|
| 735 |
+
"step": 121
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 4.88,
|
| 739 |
+
"learning_rate": 0.00015120000000000002,
|
| 740 |
+
"loss": 1.8369,
|
| 741 |
+
"step": 122
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.92,
|
| 745 |
+
"learning_rate": 0.0001508,
|
| 746 |
+
"loss": 1.8017,
|
| 747 |
+
"step": 123
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 4.96,
|
| 751 |
+
"learning_rate": 0.0001504,
|
| 752 |
+
"loss": 1.7858,
|
| 753 |
+
"step": 124
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 5.0,
|
| 757 |
+
"learning_rate": 0.00015000000000000001,
|
| 758 |
+
"loss": 1.8366,
|
| 759 |
+
"step": 125
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 5.04,
|
| 763 |
+
"learning_rate": 0.0001496,
|
| 764 |
+
"loss": 1.7353,
|
| 765 |
+
"step": 126
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 5.08,
|
| 769 |
+
"learning_rate": 0.0001492,
|
| 770 |
+
"loss": 1.6017,
|
| 771 |
+
"step": 127
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 5.12,
|
| 775 |
+
"learning_rate": 0.0001488,
|
| 776 |
+
"loss": 1.6645,
|
| 777 |
+
"step": 128
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 5.16,
|
| 781 |
+
"learning_rate": 0.0001484,
|
| 782 |
+
"loss": 1.8663,
|
| 783 |
+
"step": 129
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 5.2,
|
| 787 |
+
"learning_rate": 0.000148,
|
| 788 |
+
"loss": 1.7009,
|
| 789 |
+
"step": 130
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 5.24,
|
| 793 |
+
"learning_rate": 0.0001476,
|
| 794 |
+
"loss": 1.7371,
|
| 795 |
+
"step": 131
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 5.28,
|
| 799 |
+
"learning_rate": 0.0001472,
|
| 800 |
+
"loss": 1.8555,
|
| 801 |
+
"step": 132
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 5.32,
|
| 805 |
+
"learning_rate": 0.00014680000000000002,
|
| 806 |
+
"loss": 1.6373,
|
| 807 |
+
"step": 133
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 5.36,
|
| 811 |
+
"learning_rate": 0.0001464,
|
| 812 |
+
"loss": 1.7211,
|
| 813 |
+
"step": 134
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 5.4,
|
| 817 |
+
"learning_rate": 0.000146,
|
| 818 |
+
"loss": 1.614,
|
| 819 |
+
"step": 135
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.44,
|
| 823 |
+
"learning_rate": 0.00014560000000000002,
|
| 824 |
+
"loss": 1.8736,
|
| 825 |
+
"step": 136
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.48,
|
| 829 |
+
"learning_rate": 0.0001452,
|
| 830 |
+
"loss": 1.7229,
|
| 831 |
+
"step": 137
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 5.52,
|
| 835 |
+
"learning_rate": 0.0001448,
|
| 836 |
+
"loss": 1.7315,
|
| 837 |
+
"step": 138
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 5.56,
|
| 841 |
+
"learning_rate": 0.0001444,
|
| 842 |
+
"loss": 1.7259,
|
| 843 |
+
"step": 139
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 5.6,
|
| 847 |
+
"learning_rate": 0.000144,
|
| 848 |
+
"loss": 1.7032,
|
| 849 |
+
"step": 140
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 5.64,
|
| 853 |
+
"learning_rate": 0.0001436,
|
| 854 |
+
"loss": 1.9237,
|
| 855 |
+
"step": 141
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 5.68,
|
| 859 |
+
"learning_rate": 0.0001432,
|
| 860 |
+
"loss": 1.7821,
|
| 861 |
+
"step": 142
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.72,
|
| 865 |
+
"learning_rate": 0.0001428,
|
| 866 |
+
"loss": 1.6081,
|
| 867 |
+
"step": 143
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 5.76,
|
| 871 |
+
"learning_rate": 0.0001424,
|
| 872 |
+
"loss": 1.7979,
|
| 873 |
+
"step": 144
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 5.8,
|
| 877 |
+
"learning_rate": 0.000142,
|
| 878 |
+
"loss": 1.6675,
|
| 879 |
+
"step": 145
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 5.84,
|
| 883 |
+
"learning_rate": 0.0001416,
|
| 884 |
+
"loss": 1.6758,
|
| 885 |
+
"step": 146
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 5.88,
|
| 889 |
+
"learning_rate": 0.0001412,
|
| 890 |
+
"loss": 1.7783,
|
| 891 |
+
"step": 147
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 5.92,
|
| 895 |
+
"learning_rate": 0.0001408,
|
| 896 |
+
"loss": 1.5935,
|
| 897 |
+
"step": 148
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 5.96,
|
| 901 |
+
"learning_rate": 0.0001404,
|
| 902 |
+
"loss": 1.7372,
|
| 903 |
+
"step": 149
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 6.0,
|
| 907 |
+
"learning_rate": 0.00014,
|
| 908 |
+
"loss": 1.6308,
|
| 909 |
+
"step": 150
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.04,
|
| 913 |
+
"learning_rate": 0.0001396,
|
| 914 |
+
"loss": 1.5829,
|
| 915 |
+
"step": 151
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 6.08,
|
| 919 |
+
"learning_rate": 0.0001392,
|
| 920 |
+
"loss": 1.6538,
|
| 921 |
+
"step": 152
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 6.12,
|
| 925 |
+
"learning_rate": 0.00013879999999999999,
|
| 926 |
+
"loss": 1.7624,
|
| 927 |
+
"step": 153
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 6.16,
|
| 931 |
+
"learning_rate": 0.0001384,
|
| 932 |
+
"loss": 1.4968,
|
| 933 |
+
"step": 154
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 6.2,
|
| 937 |
+
"learning_rate": 0.000138,
|
| 938 |
+
"loss": 1.7305,
|
| 939 |
+
"step": 155
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.24,
|
| 943 |
+
"learning_rate": 0.00013759999999999998,
|
| 944 |
+
"loss": 1.6536,
|
| 945 |
+
"step": 156
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 6.28,
|
| 949 |
+
"learning_rate": 0.00013720000000000003,
|
| 950 |
+
"loss": 1.654,
|
| 951 |
+
"step": 157
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 6.32,
|
| 955 |
+
"learning_rate": 0.00013680000000000002,
|
| 956 |
+
"loss": 1.7163,
|
| 957 |
+
"step": 158
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 6.36,
|
| 961 |
+
"learning_rate": 0.0001364,
|
| 962 |
+
"loss": 1.6599,
|
| 963 |
+
"step": 159
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 6.4,
|
| 967 |
+
"learning_rate": 0.00013600000000000003,
|
| 968 |
+
"loss": 1.5729,
|
| 969 |
+
"step": 160
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.44,
|
| 973 |
+
"learning_rate": 0.00013560000000000002,
|
| 974 |
+
"loss": 1.5291,
|
| 975 |
+
"step": 161
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 6.48,
|
| 979 |
+
"learning_rate": 0.0001352,
|
| 980 |
+
"loss": 1.5666,
|
| 981 |
+
"step": 162
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.52,
|
| 985 |
+
"learning_rate": 0.00013480000000000002,
|
| 986 |
+
"loss": 1.7576,
|
| 987 |
+
"step": 163
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 6.56,
|
| 991 |
+
"learning_rate": 0.00013440000000000001,
|
| 992 |
+
"loss": 1.4462,
|
| 993 |
+
"step": 164
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 6.6,
|
| 997 |
+
"learning_rate": 0.000134,
|
| 998 |
+
"loss": 1.5659,
|
| 999 |
+
"step": 165
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 6.64,
|
| 1003 |
+
"learning_rate": 0.00013360000000000002,
|
| 1004 |
+
"loss": 1.6234,
|
| 1005 |
+
"step": 166
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 6.68,
|
| 1009 |
+
"learning_rate": 0.0001332,
|
| 1010 |
+
"loss": 1.5176,
|
| 1011 |
+
"step": 167
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 6.72,
|
| 1015 |
+
"learning_rate": 0.0001328,
|
| 1016 |
+
"loss": 1.667,
|
| 1017 |
+
"step": 168
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 6.76,
|
| 1021 |
+
"learning_rate": 0.00013240000000000002,
|
| 1022 |
+
"loss": 1.6514,
|
| 1023 |
+
"step": 169
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.8,
|
| 1027 |
+
"learning_rate": 0.000132,
|
| 1028 |
+
"loss": 1.7179,
|
| 1029 |
+
"step": 170
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 6.84,
|
| 1033 |
+
"learning_rate": 0.0001316,
|
| 1034 |
+
"loss": 1.4373,
|
| 1035 |
+
"step": 171
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 6.88,
|
| 1039 |
+
"learning_rate": 0.00013120000000000002,
|
| 1040 |
+
"loss": 1.7044,
|
| 1041 |
+
"step": 172
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 6.92,
|
| 1045 |
+
"learning_rate": 0.0001308,
|
| 1046 |
+
"loss": 1.5673,
|
| 1047 |
+
"step": 173
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 6.96,
|
| 1051 |
+
"learning_rate": 0.0001304,
|
| 1052 |
+
"loss": 1.6303,
|
| 1053 |
+
"step": 174
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 7.0,
|
| 1057 |
+
"learning_rate": 0.00013000000000000002,
|
| 1058 |
+
"loss": 1.4828,
|
| 1059 |
+
"step": 175
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 7.04,
|
| 1063 |
+
"learning_rate": 0.0001296,
|
| 1064 |
+
"loss": 1.6204,
|
| 1065 |
+
"step": 176
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 7.08,
|
| 1069 |
+
"learning_rate": 0.00012920000000000002,
|
| 1070 |
+
"loss": 1.4955,
|
| 1071 |
+
"step": 177
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 7.12,
|
| 1075 |
+
"learning_rate": 0.00012880000000000001,
|
| 1076 |
+
"loss": 1.6109,
|
| 1077 |
+
"step": 178
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 7.16,
|
| 1081 |
+
"learning_rate": 0.0001284,
|
| 1082 |
+
"loss": 1.5318,
|
| 1083 |
+
"step": 179
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 7.2,
|
| 1087 |
+
"learning_rate": 0.00012800000000000002,
|
| 1088 |
+
"loss": 1.5702,
|
| 1089 |
+
"step": 180
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 7.24,
|
| 1093 |
+
"learning_rate": 0.0001276,
|
| 1094 |
+
"loss": 1.4598,
|
| 1095 |
+
"step": 181
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 7.28,
|
| 1099 |
+
"learning_rate": 0.0001272,
|
| 1100 |
+
"loss": 1.6145,
|
| 1101 |
+
"step": 182
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 7.32,
|
| 1105 |
+
"learning_rate": 0.00012680000000000002,
|
| 1106 |
+
"loss": 1.3628,
|
| 1107 |
+
"step": 183
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 7.36,
|
| 1111 |
+
"learning_rate": 0.0001264,
|
| 1112 |
+
"loss": 1.4578,
|
| 1113 |
+
"step": 184
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 7.4,
|
| 1117 |
+
"learning_rate": 0.000126,
|
| 1118 |
+
"loss": 1.7094,
|
| 1119 |
+
"step": 185
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 7.44,
|
| 1123 |
+
"learning_rate": 0.00012560000000000002,
|
| 1124 |
+
"loss": 1.4376,
|
| 1125 |
+
"step": 186
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 7.48,
|
| 1129 |
+
"learning_rate": 0.0001252,
|
| 1130 |
+
"loss": 1.3393,
|
| 1131 |
+
"step": 187
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 7.52,
|
| 1135 |
+
"learning_rate": 0.0001248,
|
| 1136 |
+
"loss": 1.4273,
|
| 1137 |
+
"step": 188
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 7.56,
|
| 1141 |
+
"learning_rate": 0.00012440000000000002,
|
| 1142 |
+
"loss": 1.5506,
|
| 1143 |
+
"step": 189
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.6,
|
| 1147 |
+
"learning_rate": 0.000124,
|
| 1148 |
+
"loss": 1.4774,
|
| 1149 |
+
"step": 190
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 7.64,
|
| 1153 |
+
"learning_rate": 0.0001236,
|
| 1154 |
+
"loss": 1.4632,
|
| 1155 |
+
"step": 191
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 7.68,
|
| 1159 |
+
"learning_rate": 0.0001232,
|
| 1160 |
+
"loss": 1.4568,
|
| 1161 |
+
"step": 192
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 7.72,
|
| 1165 |
+
"learning_rate": 0.0001228,
|
| 1166 |
+
"loss": 1.6106,
|
| 1167 |
+
"step": 193
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 7.76,
|
| 1171 |
+
"learning_rate": 0.0001224,
|
| 1172 |
+
"loss": 1.577,
|
| 1173 |
+
"step": 194
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 7.8,
|
| 1177 |
+
"learning_rate": 0.000122,
|
| 1178 |
+
"loss": 1.4805,
|
| 1179 |
+
"step": 195
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 7.84,
|
| 1183 |
+
"learning_rate": 0.0001216,
|
| 1184 |
+
"loss": 1.4546,
|
| 1185 |
+
"step": 196
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.88,
|
| 1189 |
+
"learning_rate": 0.0001212,
|
| 1190 |
+
"loss": 1.5164,
|
| 1191 |
+
"step": 197
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 7.92,
|
| 1195 |
+
"learning_rate": 0.0001208,
|
| 1196 |
+
"loss": 1.5061,
|
| 1197 |
+
"step": 198
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 7.96,
|
| 1201 |
+
"learning_rate": 0.0001204,
|
| 1202 |
+
"loss": 1.5312,
|
| 1203 |
+
"step": 199
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 8.0,
|
| 1207 |
+
"learning_rate": 0.00012,
|
| 1208 |
+
"loss": 1.4463,
|
| 1209 |
+
"step": 200
|
| 1210 |
+
}
|
| 1211 |
+
],
|
| 1212 |
+
"logging_steps": 1,
|
| 1213 |
+
"max_steps": 500,
|
| 1214 |
+
"num_train_epochs": 20,
|
| 1215 |
+
"save_steps": 100,
|
| 1216 |
+
"total_flos": 6.816544875119616e+16,
|
| 1217 |
+
"trial_name": null,
|
| 1218 |
+
"trial_params": null
|
| 1219 |
+
}
|
checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ef74b31950ae6c8955316bed48c343fb06cda0cb6a8a54ca46ca3cb681c8736
|
| 3 |
+
size 4027
|
checkpoint-300/README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
checkpoint-300/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-300/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fced119b995138b5b4c9d37dd51211a238534b883b617216cadd221bc8ae36d
|
| 3 |
+
size 39409357
|
checkpoint-300/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67fdec376938273ad2ba5e201ffaba7ad32265728524da3562b132b79fdf2ede
|
| 3 |
+
size 78844421
|
checkpoint-300/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:370c3a07f37a8aae6ea141b54ca992b21699546baf7407eb587b6056f787333b
|
| 3 |
+
size 14575
|
checkpoint-300/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2bc1def4a9cb9244bae2e6cfff139bec4634b7353bf4cbc7d9c4c7018fdd400
|
| 3 |
+
size 627
|
checkpoint-300/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
checkpoint-300/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
|
| 3 |
+
size 14500471
|
checkpoint-300/tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"tokenizer_class": "BloomTokenizer",
|
| 9 |
+
"unk_token": "<unk>"
|
| 10 |
+
}
|
checkpoint-300/trainer_state.json
ADDED
|
@@ -0,0 +1,1819 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 12.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 0.0001996,
|
| 14 |
+
"loss": 2.4683,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 0.00019920000000000002,
|
| 20 |
+
"loss": 2.2023,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 0.0001988,
|
| 26 |
+
"loss": 2.3039,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 0.0001984,
|
| 32 |
+
"loss": 2.1855,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.2,
|
| 37 |
+
"learning_rate": 0.00019800000000000002,
|
| 38 |
+
"loss": 2.328,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.24,
|
| 43 |
+
"learning_rate": 0.0001976,
|
| 44 |
+
"loss": 2.3407,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.28,
|
| 49 |
+
"learning_rate": 0.0001972,
|
| 50 |
+
"loss": 2.3432,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"learning_rate": 0.0001968,
|
| 56 |
+
"loss": 2.2333,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.36,
|
| 61 |
+
"learning_rate": 0.0001964,
|
| 62 |
+
"loss": 2.1761,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.4,
|
| 67 |
+
"learning_rate": 0.000196,
|
| 68 |
+
"loss": 2.1473,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.44,
|
| 73 |
+
"learning_rate": 0.0001956,
|
| 74 |
+
"loss": 2.2076,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.48,
|
| 79 |
+
"learning_rate": 0.0001952,
|
| 80 |
+
"loss": 2.1925,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.52,
|
| 85 |
+
"learning_rate": 0.0001948,
|
| 86 |
+
"loss": 2.1694,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.56,
|
| 91 |
+
"learning_rate": 0.0001944,
|
| 92 |
+
"loss": 2.1056,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6,
|
| 97 |
+
"learning_rate": 0.000194,
|
| 98 |
+
"loss": 2.2426,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.64,
|
| 103 |
+
"learning_rate": 0.00019360000000000002,
|
| 104 |
+
"loss": 2.2635,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.68,
|
| 109 |
+
"learning_rate": 0.0001932,
|
| 110 |
+
"loss": 2.251,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.72,
|
| 115 |
+
"learning_rate": 0.0001928,
|
| 116 |
+
"loss": 2.224,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.76,
|
| 121 |
+
"learning_rate": 0.00019240000000000001,
|
| 122 |
+
"loss": 2.1848,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.8,
|
| 127 |
+
"learning_rate": 0.000192,
|
| 128 |
+
"loss": 2.1341,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.84,
|
| 133 |
+
"learning_rate": 0.0001916,
|
| 134 |
+
"loss": 2.126,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.88,
|
| 139 |
+
"learning_rate": 0.0001912,
|
| 140 |
+
"loss": 2.0959,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.92,
|
| 145 |
+
"learning_rate": 0.0001908,
|
| 146 |
+
"loss": 2.2446,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.96,
|
| 151 |
+
"learning_rate": 0.0001904,
|
| 152 |
+
"loss": 1.9448,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.0,
|
| 157 |
+
"learning_rate": 0.00019,
|
| 158 |
+
"loss": 2.0939,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.04,
|
| 163 |
+
"learning_rate": 0.0001896,
|
| 164 |
+
"loss": 2.1247,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.08,
|
| 169 |
+
"learning_rate": 0.0001892,
|
| 170 |
+
"loss": 1.9417,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.12,
|
| 175 |
+
"learning_rate": 0.0001888,
|
| 176 |
+
"loss": 2.105,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.16,
|
| 181 |
+
"learning_rate": 0.0001884,
|
| 182 |
+
"loss": 2.1595,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.2,
|
| 187 |
+
"learning_rate": 0.000188,
|
| 188 |
+
"loss": 2.2009,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.24,
|
| 193 |
+
"learning_rate": 0.0001876,
|
| 194 |
+
"loss": 2.0784,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.28,
|
| 199 |
+
"learning_rate": 0.00018720000000000002,
|
| 200 |
+
"loss": 2.2173,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.32,
|
| 205 |
+
"learning_rate": 0.00018680000000000001,
|
| 206 |
+
"loss": 2.1185,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.36,
|
| 211 |
+
"learning_rate": 0.00018640000000000003,
|
| 212 |
+
"loss": 1.988,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4,
|
| 217 |
+
"learning_rate": 0.00018600000000000002,
|
| 218 |
+
"loss": 1.9493,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.44,
|
| 223 |
+
"learning_rate": 0.0001856,
|
| 224 |
+
"loss": 1.9947,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.48,
|
| 229 |
+
"learning_rate": 0.00018520000000000003,
|
| 230 |
+
"loss": 2.0506,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.52,
|
| 235 |
+
"learning_rate": 0.00018480000000000002,
|
| 236 |
+
"loss": 2.0829,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.56,
|
| 241 |
+
"learning_rate": 0.0001844,
|
| 242 |
+
"loss": 2.1455,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.6,
|
| 247 |
+
"learning_rate": 0.00018400000000000003,
|
| 248 |
+
"loss": 2.0304,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.64,
|
| 253 |
+
"learning_rate": 0.00018360000000000002,
|
| 254 |
+
"loss": 2.006,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.68,
|
| 259 |
+
"learning_rate": 0.0001832,
|
| 260 |
+
"loss": 2.1759,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.72,
|
| 265 |
+
"learning_rate": 0.00018280000000000003,
|
| 266 |
+
"loss": 2.0093,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.76,
|
| 271 |
+
"learning_rate": 0.00018240000000000002,
|
| 272 |
+
"loss": 2.0683,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.8,
|
| 277 |
+
"learning_rate": 0.000182,
|
| 278 |
+
"loss": 2.1352,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.84,
|
| 283 |
+
"learning_rate": 0.00018160000000000002,
|
| 284 |
+
"loss": 1.9736,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.88,
|
| 289 |
+
"learning_rate": 0.0001812,
|
| 290 |
+
"loss": 1.9631,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.92,
|
| 295 |
+
"learning_rate": 0.0001808,
|
| 296 |
+
"loss": 2.1024,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.96,
|
| 301 |
+
"learning_rate": 0.00018040000000000002,
|
| 302 |
+
"loss": 1.9895,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.0,
|
| 307 |
+
"learning_rate": 0.00018,
|
| 308 |
+
"loss": 2.0067,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.04,
|
| 313 |
+
"learning_rate": 0.0001796,
|
| 314 |
+
"loss": 2.0509,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.08,
|
| 319 |
+
"learning_rate": 0.00017920000000000002,
|
| 320 |
+
"loss": 1.8795,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.12,
|
| 325 |
+
"learning_rate": 0.0001788,
|
| 326 |
+
"loss": 2.0633,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.16,
|
| 331 |
+
"learning_rate": 0.0001784,
|
| 332 |
+
"loss": 1.9998,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.2,
|
| 337 |
+
"learning_rate": 0.00017800000000000002,
|
| 338 |
+
"loss": 2.0675,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.24,
|
| 343 |
+
"learning_rate": 0.0001776,
|
| 344 |
+
"loss": 2.0129,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.28,
|
| 349 |
+
"learning_rate": 0.0001772,
|
| 350 |
+
"loss": 1.9302,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.32,
|
| 355 |
+
"learning_rate": 0.00017680000000000001,
|
| 356 |
+
"loss": 2.0472,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.36,
|
| 361 |
+
"learning_rate": 0.0001764,
|
| 362 |
+
"loss": 2.0031,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.4,
|
| 367 |
+
"learning_rate": 0.00017600000000000002,
|
| 368 |
+
"loss": 2.0231,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.44,
|
| 373 |
+
"learning_rate": 0.0001756,
|
| 374 |
+
"loss": 2.0192,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.48,
|
| 379 |
+
"learning_rate": 0.0001752,
|
| 380 |
+
"loss": 2.1336,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.52,
|
| 385 |
+
"learning_rate": 0.00017480000000000002,
|
| 386 |
+
"loss": 1.7753,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.56,
|
| 391 |
+
"learning_rate": 0.0001744,
|
| 392 |
+
"loss": 2.0274,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.6,
|
| 397 |
+
"learning_rate": 0.000174,
|
| 398 |
+
"loss": 1.8893,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.64,
|
| 403 |
+
"learning_rate": 0.00017360000000000002,
|
| 404 |
+
"loss": 2.0341,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.68,
|
| 409 |
+
"learning_rate": 0.0001732,
|
| 410 |
+
"loss": 2.1081,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.72,
|
| 415 |
+
"learning_rate": 0.0001728,
|
| 416 |
+
"loss": 1.9463,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.76,
|
| 421 |
+
"learning_rate": 0.00017240000000000002,
|
| 422 |
+
"loss": 2.0607,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.8,
|
| 427 |
+
"learning_rate": 0.000172,
|
| 428 |
+
"loss": 1.9803,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.84,
|
| 433 |
+
"learning_rate": 0.0001716,
|
| 434 |
+
"loss": 1.7663,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.88,
|
| 439 |
+
"learning_rate": 0.00017120000000000001,
|
| 440 |
+
"loss": 2.1116,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.92,
|
| 445 |
+
"learning_rate": 0.0001708,
|
| 446 |
+
"loss": 1.9039,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.96,
|
| 451 |
+
"learning_rate": 0.0001704,
|
| 452 |
+
"loss": 2.049,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.0,
|
| 457 |
+
"learning_rate": 0.00017,
|
| 458 |
+
"loss": 1.991,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.04,
|
| 463 |
+
"learning_rate": 0.0001696,
|
| 464 |
+
"loss": 1.8755,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 3.08,
|
| 469 |
+
"learning_rate": 0.0001692,
|
| 470 |
+
"loss": 2.0367,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.12,
|
| 475 |
+
"learning_rate": 0.0001688,
|
| 476 |
+
"loss": 1.9959,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.16,
|
| 481 |
+
"learning_rate": 0.0001684,
|
| 482 |
+
"loss": 1.8051,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.2,
|
| 487 |
+
"learning_rate": 0.000168,
|
| 488 |
+
"loss": 2.0446,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.24,
|
| 493 |
+
"learning_rate": 0.0001676,
|
| 494 |
+
"loss": 1.7873,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.28,
|
| 499 |
+
"learning_rate": 0.0001672,
|
| 500 |
+
"loss": 1.9264,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 3.32,
|
| 505 |
+
"learning_rate": 0.0001668,
|
| 506 |
+
"loss": 1.8595,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.36,
|
| 511 |
+
"learning_rate": 0.0001664,
|
| 512 |
+
"loss": 1.9992,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.4,
|
| 517 |
+
"learning_rate": 0.000166,
|
| 518 |
+
"loss": 1.8783,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.44,
|
| 523 |
+
"learning_rate": 0.0001656,
|
| 524 |
+
"loss": 2.0216,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.48,
|
| 529 |
+
"learning_rate": 0.0001652,
|
| 530 |
+
"loss": 1.947,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.52,
|
| 535 |
+
"learning_rate": 0.0001648,
|
| 536 |
+
"loss": 1.9554,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.56,
|
| 541 |
+
"learning_rate": 0.0001644,
|
| 542 |
+
"loss": 1.8563,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.6,
|
| 547 |
+
"learning_rate": 0.000164,
|
| 548 |
+
"loss": 1.7795,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.64,
|
| 553 |
+
"learning_rate": 0.0001636,
|
| 554 |
+
"loss": 1.9347,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.68,
|
| 559 |
+
"learning_rate": 0.0001632,
|
| 560 |
+
"loss": 2.0078,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.72,
|
| 565 |
+
"learning_rate": 0.0001628,
|
| 566 |
+
"loss": 1.9964,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 3.76,
|
| 571 |
+
"learning_rate": 0.00016240000000000002,
|
| 572 |
+
"loss": 1.9004,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.8,
|
| 577 |
+
"learning_rate": 0.000162,
|
| 578 |
+
"loss": 1.9446,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.84,
|
| 583 |
+
"learning_rate": 0.00016160000000000002,
|
| 584 |
+
"loss": 1.8857,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.88,
|
| 589 |
+
"learning_rate": 0.00016120000000000002,
|
| 590 |
+
"loss": 1.8797,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.92,
|
| 595 |
+
"learning_rate": 0.0001608,
|
| 596 |
+
"loss": 1.7643,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.96,
|
| 601 |
+
"learning_rate": 0.00016040000000000002,
|
| 602 |
+
"loss": 1.9724,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 4.0,
|
| 607 |
+
"learning_rate": 0.00016,
|
| 608 |
+
"loss": 1.9058,
|
| 609 |
+
"step": 100
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.04,
|
| 613 |
+
"learning_rate": 0.0001596,
|
| 614 |
+
"loss": 1.8663,
|
| 615 |
+
"step": 101
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 4.08,
|
| 619 |
+
"learning_rate": 0.00015920000000000002,
|
| 620 |
+
"loss": 1.8457,
|
| 621 |
+
"step": 102
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 4.12,
|
| 625 |
+
"learning_rate": 0.0001588,
|
| 626 |
+
"loss": 1.8394,
|
| 627 |
+
"step": 103
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 4.16,
|
| 631 |
+
"learning_rate": 0.00015840000000000003,
|
| 632 |
+
"loss": 1.8941,
|
| 633 |
+
"step": 104
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 4.2,
|
| 637 |
+
"learning_rate": 0.00015800000000000002,
|
| 638 |
+
"loss": 1.8601,
|
| 639 |
+
"step": 105
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 4.24,
|
| 643 |
+
"learning_rate": 0.0001576,
|
| 644 |
+
"loss": 1.7365,
|
| 645 |
+
"step": 106
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 4.28,
|
| 649 |
+
"learning_rate": 0.00015720000000000003,
|
| 650 |
+
"loss": 1.8809,
|
| 651 |
+
"step": 107
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 4.32,
|
| 655 |
+
"learning_rate": 0.00015680000000000002,
|
| 656 |
+
"loss": 2.0206,
|
| 657 |
+
"step": 108
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 4.36,
|
| 661 |
+
"learning_rate": 0.0001564,
|
| 662 |
+
"loss": 1.7776,
|
| 663 |
+
"step": 109
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 4.4,
|
| 667 |
+
"learning_rate": 0.00015600000000000002,
|
| 668 |
+
"loss": 1.9089,
|
| 669 |
+
"step": 110
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.44,
|
| 673 |
+
"learning_rate": 0.00015560000000000001,
|
| 674 |
+
"loss": 1.8464,
|
| 675 |
+
"step": 111
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.48,
|
| 679 |
+
"learning_rate": 0.0001552,
|
| 680 |
+
"loss": 1.7943,
|
| 681 |
+
"step": 112
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.52,
|
| 685 |
+
"learning_rate": 0.00015480000000000002,
|
| 686 |
+
"loss": 1.8081,
|
| 687 |
+
"step": 113
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 4.56,
|
| 691 |
+
"learning_rate": 0.0001544,
|
| 692 |
+
"loss": 1.734,
|
| 693 |
+
"step": 114
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 4.6,
|
| 697 |
+
"learning_rate": 0.000154,
|
| 698 |
+
"loss": 1.8379,
|
| 699 |
+
"step": 115
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.64,
|
| 703 |
+
"learning_rate": 0.00015360000000000002,
|
| 704 |
+
"loss": 1.6841,
|
| 705 |
+
"step": 116
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 4.68,
|
| 709 |
+
"learning_rate": 0.0001532,
|
| 710 |
+
"loss": 1.8253,
|
| 711 |
+
"step": 117
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 4.72,
|
| 715 |
+
"learning_rate": 0.0001528,
|
| 716 |
+
"loss": 1.7607,
|
| 717 |
+
"step": 118
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 4.76,
|
| 721 |
+
"learning_rate": 0.00015240000000000002,
|
| 722 |
+
"loss": 1.8205,
|
| 723 |
+
"step": 119
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 4.8,
|
| 727 |
+
"learning_rate": 0.000152,
|
| 728 |
+
"loss": 1.8443,
|
| 729 |
+
"step": 120
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 4.84,
|
| 733 |
+
"learning_rate": 0.0001516,
|
| 734 |
+
"loss": 1.8824,
|
| 735 |
+
"step": 121
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 4.88,
|
| 739 |
+
"learning_rate": 0.00015120000000000002,
|
| 740 |
+
"loss": 1.8369,
|
| 741 |
+
"step": 122
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.92,
|
| 745 |
+
"learning_rate": 0.0001508,
|
| 746 |
+
"loss": 1.8017,
|
| 747 |
+
"step": 123
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 4.96,
|
| 751 |
+
"learning_rate": 0.0001504,
|
| 752 |
+
"loss": 1.7858,
|
| 753 |
+
"step": 124
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 5.0,
|
| 757 |
+
"learning_rate": 0.00015000000000000001,
|
| 758 |
+
"loss": 1.8366,
|
| 759 |
+
"step": 125
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 5.04,
|
| 763 |
+
"learning_rate": 0.0001496,
|
| 764 |
+
"loss": 1.7353,
|
| 765 |
+
"step": 126
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 5.08,
|
| 769 |
+
"learning_rate": 0.0001492,
|
| 770 |
+
"loss": 1.6017,
|
| 771 |
+
"step": 127
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 5.12,
|
| 775 |
+
"learning_rate": 0.0001488,
|
| 776 |
+
"loss": 1.6645,
|
| 777 |
+
"step": 128
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 5.16,
|
| 781 |
+
"learning_rate": 0.0001484,
|
| 782 |
+
"loss": 1.8663,
|
| 783 |
+
"step": 129
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 5.2,
|
| 787 |
+
"learning_rate": 0.000148,
|
| 788 |
+
"loss": 1.7009,
|
| 789 |
+
"step": 130
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 5.24,
|
| 793 |
+
"learning_rate": 0.0001476,
|
| 794 |
+
"loss": 1.7371,
|
| 795 |
+
"step": 131
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 5.28,
|
| 799 |
+
"learning_rate": 0.0001472,
|
| 800 |
+
"loss": 1.8555,
|
| 801 |
+
"step": 132
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 5.32,
|
| 805 |
+
"learning_rate": 0.00014680000000000002,
|
| 806 |
+
"loss": 1.6373,
|
| 807 |
+
"step": 133
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 5.36,
|
| 811 |
+
"learning_rate": 0.0001464,
|
| 812 |
+
"loss": 1.7211,
|
| 813 |
+
"step": 134
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 5.4,
|
| 817 |
+
"learning_rate": 0.000146,
|
| 818 |
+
"loss": 1.614,
|
| 819 |
+
"step": 135
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.44,
|
| 823 |
+
"learning_rate": 0.00014560000000000002,
|
| 824 |
+
"loss": 1.8736,
|
| 825 |
+
"step": 136
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.48,
|
| 829 |
+
"learning_rate": 0.0001452,
|
| 830 |
+
"loss": 1.7229,
|
| 831 |
+
"step": 137
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 5.52,
|
| 835 |
+
"learning_rate": 0.0001448,
|
| 836 |
+
"loss": 1.7315,
|
| 837 |
+
"step": 138
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 5.56,
|
| 841 |
+
"learning_rate": 0.0001444,
|
| 842 |
+
"loss": 1.7259,
|
| 843 |
+
"step": 139
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 5.6,
|
| 847 |
+
"learning_rate": 0.000144,
|
| 848 |
+
"loss": 1.7032,
|
| 849 |
+
"step": 140
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 5.64,
|
| 853 |
+
"learning_rate": 0.0001436,
|
| 854 |
+
"loss": 1.9237,
|
| 855 |
+
"step": 141
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 5.68,
|
| 859 |
+
"learning_rate": 0.0001432,
|
| 860 |
+
"loss": 1.7821,
|
| 861 |
+
"step": 142
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.72,
|
| 865 |
+
"learning_rate": 0.0001428,
|
| 866 |
+
"loss": 1.6081,
|
| 867 |
+
"step": 143
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 5.76,
|
| 871 |
+
"learning_rate": 0.0001424,
|
| 872 |
+
"loss": 1.7979,
|
| 873 |
+
"step": 144
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 5.8,
|
| 877 |
+
"learning_rate": 0.000142,
|
| 878 |
+
"loss": 1.6675,
|
| 879 |
+
"step": 145
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 5.84,
|
| 883 |
+
"learning_rate": 0.0001416,
|
| 884 |
+
"loss": 1.6758,
|
| 885 |
+
"step": 146
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 5.88,
|
| 889 |
+
"learning_rate": 0.0001412,
|
| 890 |
+
"loss": 1.7783,
|
| 891 |
+
"step": 147
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 5.92,
|
| 895 |
+
"learning_rate": 0.0001408,
|
| 896 |
+
"loss": 1.5935,
|
| 897 |
+
"step": 148
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 5.96,
|
| 901 |
+
"learning_rate": 0.0001404,
|
| 902 |
+
"loss": 1.7372,
|
| 903 |
+
"step": 149
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 6.0,
|
| 907 |
+
"learning_rate": 0.00014,
|
| 908 |
+
"loss": 1.6308,
|
| 909 |
+
"step": 150
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.04,
|
| 913 |
+
"learning_rate": 0.0001396,
|
| 914 |
+
"loss": 1.5829,
|
| 915 |
+
"step": 151
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 6.08,
|
| 919 |
+
"learning_rate": 0.0001392,
|
| 920 |
+
"loss": 1.6538,
|
| 921 |
+
"step": 152
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 6.12,
|
| 925 |
+
"learning_rate": 0.00013879999999999999,
|
| 926 |
+
"loss": 1.7624,
|
| 927 |
+
"step": 153
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 6.16,
|
| 931 |
+
"learning_rate": 0.0001384,
|
| 932 |
+
"loss": 1.4968,
|
| 933 |
+
"step": 154
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 6.2,
|
| 937 |
+
"learning_rate": 0.000138,
|
| 938 |
+
"loss": 1.7305,
|
| 939 |
+
"step": 155
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.24,
|
| 943 |
+
"learning_rate": 0.00013759999999999998,
|
| 944 |
+
"loss": 1.6536,
|
| 945 |
+
"step": 156
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 6.28,
|
| 949 |
+
"learning_rate": 0.00013720000000000003,
|
| 950 |
+
"loss": 1.654,
|
| 951 |
+
"step": 157
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 6.32,
|
| 955 |
+
"learning_rate": 0.00013680000000000002,
|
| 956 |
+
"loss": 1.7163,
|
| 957 |
+
"step": 158
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 6.36,
|
| 961 |
+
"learning_rate": 0.0001364,
|
| 962 |
+
"loss": 1.6599,
|
| 963 |
+
"step": 159
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 6.4,
|
| 967 |
+
"learning_rate": 0.00013600000000000003,
|
| 968 |
+
"loss": 1.5729,
|
| 969 |
+
"step": 160
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.44,
|
| 973 |
+
"learning_rate": 0.00013560000000000002,
|
| 974 |
+
"loss": 1.5291,
|
| 975 |
+
"step": 161
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 6.48,
|
| 979 |
+
"learning_rate": 0.0001352,
|
| 980 |
+
"loss": 1.5666,
|
| 981 |
+
"step": 162
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.52,
|
| 985 |
+
"learning_rate": 0.00013480000000000002,
|
| 986 |
+
"loss": 1.7576,
|
| 987 |
+
"step": 163
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 6.56,
|
| 991 |
+
"learning_rate": 0.00013440000000000001,
|
| 992 |
+
"loss": 1.4462,
|
| 993 |
+
"step": 164
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 6.6,
|
| 997 |
+
"learning_rate": 0.000134,
|
| 998 |
+
"loss": 1.5659,
|
| 999 |
+
"step": 165
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 6.64,
|
| 1003 |
+
"learning_rate": 0.00013360000000000002,
|
| 1004 |
+
"loss": 1.6234,
|
| 1005 |
+
"step": 166
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 6.68,
|
| 1009 |
+
"learning_rate": 0.0001332,
|
| 1010 |
+
"loss": 1.5176,
|
| 1011 |
+
"step": 167
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 6.72,
|
| 1015 |
+
"learning_rate": 0.0001328,
|
| 1016 |
+
"loss": 1.667,
|
| 1017 |
+
"step": 168
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 6.76,
|
| 1021 |
+
"learning_rate": 0.00013240000000000002,
|
| 1022 |
+
"loss": 1.6514,
|
| 1023 |
+
"step": 169
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.8,
|
| 1027 |
+
"learning_rate": 0.000132,
|
| 1028 |
+
"loss": 1.7179,
|
| 1029 |
+
"step": 170
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 6.84,
|
| 1033 |
+
"learning_rate": 0.0001316,
|
| 1034 |
+
"loss": 1.4373,
|
| 1035 |
+
"step": 171
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 6.88,
|
| 1039 |
+
"learning_rate": 0.00013120000000000002,
|
| 1040 |
+
"loss": 1.7044,
|
| 1041 |
+
"step": 172
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 6.92,
|
| 1045 |
+
"learning_rate": 0.0001308,
|
| 1046 |
+
"loss": 1.5673,
|
| 1047 |
+
"step": 173
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 6.96,
|
| 1051 |
+
"learning_rate": 0.0001304,
|
| 1052 |
+
"loss": 1.6303,
|
| 1053 |
+
"step": 174
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 7.0,
|
| 1057 |
+
"learning_rate": 0.00013000000000000002,
|
| 1058 |
+
"loss": 1.4828,
|
| 1059 |
+
"step": 175
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 7.04,
|
| 1063 |
+
"learning_rate": 0.0001296,
|
| 1064 |
+
"loss": 1.6204,
|
| 1065 |
+
"step": 176
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 7.08,
|
| 1069 |
+
"learning_rate": 0.00012920000000000002,
|
| 1070 |
+
"loss": 1.4955,
|
| 1071 |
+
"step": 177
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 7.12,
|
| 1075 |
+
"learning_rate": 0.00012880000000000001,
|
| 1076 |
+
"loss": 1.6109,
|
| 1077 |
+
"step": 178
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 7.16,
|
| 1081 |
+
"learning_rate": 0.0001284,
|
| 1082 |
+
"loss": 1.5318,
|
| 1083 |
+
"step": 179
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 7.2,
|
| 1087 |
+
"learning_rate": 0.00012800000000000002,
|
| 1088 |
+
"loss": 1.5702,
|
| 1089 |
+
"step": 180
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 7.24,
|
| 1093 |
+
"learning_rate": 0.0001276,
|
| 1094 |
+
"loss": 1.4598,
|
| 1095 |
+
"step": 181
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 7.28,
|
| 1099 |
+
"learning_rate": 0.0001272,
|
| 1100 |
+
"loss": 1.6145,
|
| 1101 |
+
"step": 182
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 7.32,
|
| 1105 |
+
"learning_rate": 0.00012680000000000002,
|
| 1106 |
+
"loss": 1.3628,
|
| 1107 |
+
"step": 183
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 7.36,
|
| 1111 |
+
"learning_rate": 0.0001264,
|
| 1112 |
+
"loss": 1.4578,
|
| 1113 |
+
"step": 184
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 7.4,
|
| 1117 |
+
"learning_rate": 0.000126,
|
| 1118 |
+
"loss": 1.7094,
|
| 1119 |
+
"step": 185
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 7.44,
|
| 1123 |
+
"learning_rate": 0.00012560000000000002,
|
| 1124 |
+
"loss": 1.4376,
|
| 1125 |
+
"step": 186
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 7.48,
|
| 1129 |
+
"learning_rate": 0.0001252,
|
| 1130 |
+
"loss": 1.3393,
|
| 1131 |
+
"step": 187
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 7.52,
|
| 1135 |
+
"learning_rate": 0.0001248,
|
| 1136 |
+
"loss": 1.4273,
|
| 1137 |
+
"step": 188
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 7.56,
|
| 1141 |
+
"learning_rate": 0.00012440000000000002,
|
| 1142 |
+
"loss": 1.5506,
|
| 1143 |
+
"step": 189
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.6,
|
| 1147 |
+
"learning_rate": 0.000124,
|
| 1148 |
+
"loss": 1.4774,
|
| 1149 |
+
"step": 190
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 7.64,
|
| 1153 |
+
"learning_rate": 0.0001236,
|
| 1154 |
+
"loss": 1.4632,
|
| 1155 |
+
"step": 191
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 7.68,
|
| 1159 |
+
"learning_rate": 0.0001232,
|
| 1160 |
+
"loss": 1.4568,
|
| 1161 |
+
"step": 192
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 7.72,
|
| 1165 |
+
"learning_rate": 0.0001228,
|
| 1166 |
+
"loss": 1.6106,
|
| 1167 |
+
"step": 193
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 7.76,
|
| 1171 |
+
"learning_rate": 0.0001224,
|
| 1172 |
+
"loss": 1.577,
|
| 1173 |
+
"step": 194
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 7.8,
|
| 1177 |
+
"learning_rate": 0.000122,
|
| 1178 |
+
"loss": 1.4805,
|
| 1179 |
+
"step": 195
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 7.84,
|
| 1183 |
+
"learning_rate": 0.0001216,
|
| 1184 |
+
"loss": 1.4546,
|
| 1185 |
+
"step": 196
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.88,
|
| 1189 |
+
"learning_rate": 0.0001212,
|
| 1190 |
+
"loss": 1.5164,
|
| 1191 |
+
"step": 197
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 7.92,
|
| 1195 |
+
"learning_rate": 0.0001208,
|
| 1196 |
+
"loss": 1.5061,
|
| 1197 |
+
"step": 198
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 7.96,
|
| 1201 |
+
"learning_rate": 0.0001204,
|
| 1202 |
+
"loss": 1.5312,
|
| 1203 |
+
"step": 199
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 8.0,
|
| 1207 |
+
"learning_rate": 0.00012,
|
| 1208 |
+
"loss": 1.4463,
|
| 1209 |
+
"step": 200
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 8.04,
|
| 1213 |
+
"learning_rate": 0.00011960000000000001,
|
| 1214 |
+
"loss": 1.4207,
|
| 1215 |
+
"step": 201
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 8.08,
|
| 1219 |
+
"learning_rate": 0.0001192,
|
| 1220 |
+
"loss": 1.4688,
|
| 1221 |
+
"step": 202
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 8.12,
|
| 1225 |
+
"learning_rate": 0.0001188,
|
| 1226 |
+
"loss": 1.4445,
|
| 1227 |
+
"step": 203
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 8.16,
|
| 1231 |
+
"learning_rate": 0.0001184,
|
| 1232 |
+
"loss": 1.4158,
|
| 1233 |
+
"step": 204
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 8.2,
|
| 1237 |
+
"learning_rate": 0.000118,
|
| 1238 |
+
"loss": 1.3483,
|
| 1239 |
+
"step": 205
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 8.24,
|
| 1243 |
+
"learning_rate": 0.0001176,
|
| 1244 |
+
"loss": 1.4117,
|
| 1245 |
+
"step": 206
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 8.28,
|
| 1249 |
+
"learning_rate": 0.0001172,
|
| 1250 |
+
"loss": 1.4267,
|
| 1251 |
+
"step": 207
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 8.32,
|
| 1255 |
+
"learning_rate": 0.00011679999999999999,
|
| 1256 |
+
"loss": 1.5547,
|
| 1257 |
+
"step": 208
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 8.36,
|
| 1261 |
+
"learning_rate": 0.0001164,
|
| 1262 |
+
"loss": 1.4883,
|
| 1263 |
+
"step": 209
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 8.4,
|
| 1267 |
+
"learning_rate": 0.000116,
|
| 1268 |
+
"loss": 1.414,
|
| 1269 |
+
"step": 210
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 8.44,
|
| 1273 |
+
"learning_rate": 0.00011559999999999999,
|
| 1274 |
+
"loss": 1.3254,
|
| 1275 |
+
"step": 211
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 8.48,
|
| 1279 |
+
"learning_rate": 0.0001152,
|
| 1280 |
+
"loss": 1.22,
|
| 1281 |
+
"step": 212
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 8.52,
|
| 1285 |
+
"learning_rate": 0.0001148,
|
| 1286 |
+
"loss": 1.3224,
|
| 1287 |
+
"step": 213
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 8.56,
|
| 1291 |
+
"learning_rate": 0.0001144,
|
| 1292 |
+
"loss": 1.6158,
|
| 1293 |
+
"step": 214
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 8.6,
|
| 1297 |
+
"learning_rate": 0.00011399999999999999,
|
| 1298 |
+
"loss": 1.2817,
|
| 1299 |
+
"step": 215
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 8.64,
|
| 1303 |
+
"learning_rate": 0.0001136,
|
| 1304 |
+
"loss": 1.4077,
|
| 1305 |
+
"step": 216
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 8.68,
|
| 1309 |
+
"learning_rate": 0.0001132,
|
| 1310 |
+
"loss": 1.5311,
|
| 1311 |
+
"step": 217
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 8.72,
|
| 1315 |
+
"learning_rate": 0.00011279999999999999,
|
| 1316 |
+
"loss": 1.5289,
|
| 1317 |
+
"step": 218
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 8.76,
|
| 1321 |
+
"learning_rate": 0.00011240000000000002,
|
| 1322 |
+
"loss": 1.4576,
|
| 1323 |
+
"step": 219
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 8.8,
|
| 1327 |
+
"learning_rate": 0.00011200000000000001,
|
| 1328 |
+
"loss": 1.2063,
|
| 1329 |
+
"step": 220
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 8.84,
|
| 1333 |
+
"learning_rate": 0.00011160000000000002,
|
| 1334 |
+
"loss": 1.3543,
|
| 1335 |
+
"step": 221
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 8.88,
|
| 1339 |
+
"learning_rate": 0.00011120000000000002,
|
| 1340 |
+
"loss": 1.4352,
|
| 1341 |
+
"step": 222
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 8.92,
|
| 1345 |
+
"learning_rate": 0.00011080000000000001,
|
| 1346 |
+
"loss": 1.3058,
|
| 1347 |
+
"step": 223
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.96,
|
| 1351 |
+
"learning_rate": 0.00011040000000000001,
|
| 1352 |
+
"loss": 1.3475,
|
| 1353 |
+
"step": 224
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 9.0,
|
| 1357 |
+
"learning_rate": 0.00011000000000000002,
|
| 1358 |
+
"loss": 1.1293,
|
| 1359 |
+
"step": 225
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 9.04,
|
| 1363 |
+
"learning_rate": 0.00010960000000000001,
|
| 1364 |
+
"loss": 1.1671,
|
| 1365 |
+
"step": 226
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 9.08,
|
| 1369 |
+
"learning_rate": 0.00010920000000000001,
|
| 1370 |
+
"loss": 1.4726,
|
| 1371 |
+
"step": 227
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 9.12,
|
| 1375 |
+
"learning_rate": 0.00010880000000000002,
|
| 1376 |
+
"loss": 1.3986,
|
| 1377 |
+
"step": 228
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 9.16,
|
| 1381 |
+
"learning_rate": 0.00010840000000000002,
|
| 1382 |
+
"loss": 1.2731,
|
| 1383 |
+
"step": 229
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 9.2,
|
| 1387 |
+
"learning_rate": 0.00010800000000000001,
|
| 1388 |
+
"loss": 1.2212,
|
| 1389 |
+
"step": 230
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 9.24,
|
| 1393 |
+
"learning_rate": 0.00010760000000000001,
|
| 1394 |
+
"loss": 1.3067,
|
| 1395 |
+
"step": 231
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 9.28,
|
| 1399 |
+
"learning_rate": 0.00010720000000000002,
|
| 1400 |
+
"loss": 1.278,
|
| 1401 |
+
"step": 232
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 9.32,
|
| 1405 |
+
"learning_rate": 0.00010680000000000001,
|
| 1406 |
+
"loss": 1.4141,
|
| 1407 |
+
"step": 233
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 9.36,
|
| 1411 |
+
"learning_rate": 0.00010640000000000001,
|
| 1412 |
+
"loss": 1.2885,
|
| 1413 |
+
"step": 234
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 9.4,
|
| 1417 |
+
"learning_rate": 0.00010600000000000002,
|
| 1418 |
+
"loss": 1.466,
|
| 1419 |
+
"step": 235
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 9.44,
|
| 1423 |
+
"learning_rate": 0.0001056,
|
| 1424 |
+
"loss": 1.2979,
|
| 1425 |
+
"step": 236
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 9.48,
|
| 1429 |
+
"learning_rate": 0.00010520000000000001,
|
| 1430 |
+
"loss": 1.1622,
|
| 1431 |
+
"step": 237
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 9.52,
|
| 1435 |
+
"learning_rate": 0.00010480000000000001,
|
| 1436 |
+
"loss": 1.2504,
|
| 1437 |
+
"step": 238
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 9.56,
|
| 1441 |
+
"learning_rate": 0.0001044,
|
| 1442 |
+
"loss": 1.228,
|
| 1443 |
+
"step": 239
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 9.6,
|
| 1447 |
+
"learning_rate": 0.00010400000000000001,
|
| 1448 |
+
"loss": 1.3607,
|
| 1449 |
+
"step": 240
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 9.64,
|
| 1453 |
+
"learning_rate": 0.00010360000000000001,
|
| 1454 |
+
"loss": 1.1921,
|
| 1455 |
+
"step": 241
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 9.68,
|
| 1459 |
+
"learning_rate": 0.0001032,
|
| 1460 |
+
"loss": 1.2114,
|
| 1461 |
+
"step": 242
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 9.72,
|
| 1465 |
+
"learning_rate": 0.0001028,
|
| 1466 |
+
"loss": 1.2385,
|
| 1467 |
+
"step": 243
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 9.76,
|
| 1471 |
+
"learning_rate": 0.00010240000000000001,
|
| 1472 |
+
"loss": 1.3227,
|
| 1473 |
+
"step": 244
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 9.8,
|
| 1477 |
+
"learning_rate": 0.00010200000000000001,
|
| 1478 |
+
"loss": 1.3935,
|
| 1479 |
+
"step": 245
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 9.84,
|
| 1483 |
+
"learning_rate": 0.0001016,
|
| 1484 |
+
"loss": 1.2209,
|
| 1485 |
+
"step": 246
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 9.88,
|
| 1489 |
+
"learning_rate": 0.00010120000000000001,
|
| 1490 |
+
"loss": 1.1541,
|
| 1491 |
+
"step": 247
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 9.92,
|
| 1495 |
+
"learning_rate": 0.00010080000000000001,
|
| 1496 |
+
"loss": 1.4049,
|
| 1497 |
+
"step": 248
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 9.96,
|
| 1501 |
+
"learning_rate": 0.0001004,
|
| 1502 |
+
"loss": 1.13,
|
| 1503 |
+
"step": 249
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 10.0,
|
| 1507 |
+
"learning_rate": 0.0001,
|
| 1508 |
+
"loss": 1.3783,
|
| 1509 |
+
"step": 250
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 10.04,
|
| 1513 |
+
"learning_rate": 9.960000000000001e-05,
|
| 1514 |
+
"loss": 1.3207,
|
| 1515 |
+
"step": 251
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 10.08,
|
| 1519 |
+
"learning_rate": 9.92e-05,
|
| 1520 |
+
"loss": 1.2697,
|
| 1521 |
+
"step": 252
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 10.12,
|
| 1525 |
+
"learning_rate": 9.88e-05,
|
| 1526 |
+
"loss": 1.091,
|
| 1527 |
+
"step": 253
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 10.16,
|
| 1531 |
+
"learning_rate": 9.84e-05,
|
| 1532 |
+
"loss": 1.0463,
|
| 1533 |
+
"step": 254
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 10.2,
|
| 1537 |
+
"learning_rate": 9.8e-05,
|
| 1538 |
+
"loss": 1.3686,
|
| 1539 |
+
"step": 255
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 10.24,
|
| 1543 |
+
"learning_rate": 9.76e-05,
|
| 1544 |
+
"loss": 1.3109,
|
| 1545 |
+
"step": 256
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 10.28,
|
| 1549 |
+
"learning_rate": 9.72e-05,
|
| 1550 |
+
"loss": 1.2338,
|
| 1551 |
+
"step": 257
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 10.32,
|
| 1555 |
+
"learning_rate": 9.680000000000001e-05,
|
| 1556 |
+
"loss": 1.12,
|
| 1557 |
+
"step": 258
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 10.36,
|
| 1561 |
+
"learning_rate": 9.64e-05,
|
| 1562 |
+
"loss": 1.1864,
|
| 1563 |
+
"step": 259
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 10.4,
|
| 1567 |
+
"learning_rate": 9.6e-05,
|
| 1568 |
+
"loss": 1.1062,
|
| 1569 |
+
"step": 260
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 10.44,
|
| 1573 |
+
"learning_rate": 9.56e-05,
|
| 1574 |
+
"loss": 1.2924,
|
| 1575 |
+
"step": 261
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 10.48,
|
| 1579 |
+
"learning_rate": 9.52e-05,
|
| 1580 |
+
"loss": 1.1931,
|
| 1581 |
+
"step": 262
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 10.52,
|
| 1585 |
+
"learning_rate": 9.48e-05,
|
| 1586 |
+
"loss": 1.2366,
|
| 1587 |
+
"step": 263
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 10.56,
|
| 1591 |
+
"learning_rate": 9.44e-05,
|
| 1592 |
+
"loss": 1.1589,
|
| 1593 |
+
"step": 264
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 10.6,
|
| 1597 |
+
"learning_rate": 9.4e-05,
|
| 1598 |
+
"loss": 1.1197,
|
| 1599 |
+
"step": 265
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 10.64,
|
| 1603 |
+
"learning_rate": 9.360000000000001e-05,
|
| 1604 |
+
"loss": 1.3072,
|
| 1605 |
+
"step": 266
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 10.68,
|
| 1609 |
+
"learning_rate": 9.320000000000002e-05,
|
| 1610 |
+
"loss": 1.1173,
|
| 1611 |
+
"step": 267
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 10.72,
|
| 1615 |
+
"learning_rate": 9.28e-05,
|
| 1616 |
+
"loss": 1.0378,
|
| 1617 |
+
"step": 268
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 10.76,
|
| 1621 |
+
"learning_rate": 9.240000000000001e-05,
|
| 1622 |
+
"loss": 1.1746,
|
| 1623 |
+
"step": 269
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 10.8,
|
| 1627 |
+
"learning_rate": 9.200000000000001e-05,
|
| 1628 |
+
"loss": 1.1144,
|
| 1629 |
+
"step": 270
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 10.84,
|
| 1633 |
+
"learning_rate": 9.16e-05,
|
| 1634 |
+
"loss": 1.1862,
|
| 1635 |
+
"step": 271
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 10.88,
|
| 1639 |
+
"learning_rate": 9.120000000000001e-05,
|
| 1640 |
+
"loss": 1.0816,
|
| 1641 |
+
"step": 272
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 10.92,
|
| 1645 |
+
"learning_rate": 9.080000000000001e-05,
|
| 1646 |
+
"loss": 1.2288,
|
| 1647 |
+
"step": 273
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 10.96,
|
| 1651 |
+
"learning_rate": 9.04e-05,
|
| 1652 |
+
"loss": 1.3031,
|
| 1653 |
+
"step": 274
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 11.0,
|
| 1657 |
+
"learning_rate": 9e-05,
|
| 1658 |
+
"loss": 1.1646,
|
| 1659 |
+
"step": 275
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 11.04,
|
| 1663 |
+
"learning_rate": 8.960000000000001e-05,
|
| 1664 |
+
"loss": 1.2088,
|
| 1665 |
+
"step": 276
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 11.08,
|
| 1669 |
+
"learning_rate": 8.92e-05,
|
| 1670 |
+
"loss": 1.1465,
|
| 1671 |
+
"step": 277
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 11.12,
|
| 1675 |
+
"learning_rate": 8.88e-05,
|
| 1676 |
+
"loss": 1.0313,
|
| 1677 |
+
"step": 278
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 11.16,
|
| 1681 |
+
"learning_rate": 8.840000000000001e-05,
|
| 1682 |
+
"loss": 0.9662,
|
| 1683 |
+
"step": 279
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 11.2,
|
| 1687 |
+
"learning_rate": 8.800000000000001e-05,
|
| 1688 |
+
"loss": 1.051,
|
| 1689 |
+
"step": 280
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 11.24,
|
| 1693 |
+
"learning_rate": 8.76e-05,
|
| 1694 |
+
"loss": 1.1708,
|
| 1695 |
+
"step": 281
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 11.28,
|
| 1699 |
+
"learning_rate": 8.72e-05,
|
| 1700 |
+
"loss": 1.0236,
|
| 1701 |
+
"step": 282
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 11.32,
|
| 1705 |
+
"learning_rate": 8.680000000000001e-05,
|
| 1706 |
+
"loss": 1.1359,
|
| 1707 |
+
"step": 283
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 11.36,
|
| 1711 |
+
"learning_rate": 8.64e-05,
|
| 1712 |
+
"loss": 1.1398,
|
| 1713 |
+
"step": 284
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 11.4,
|
| 1717 |
+
"learning_rate": 8.6e-05,
|
| 1718 |
+
"loss": 1.0213,
|
| 1719 |
+
"step": 285
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 11.44,
|
| 1723 |
+
"learning_rate": 8.560000000000001e-05,
|
| 1724 |
+
"loss": 1.1788,
|
| 1725 |
+
"step": 286
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 11.48,
|
| 1729 |
+
"learning_rate": 8.52e-05,
|
| 1730 |
+
"loss": 1.0387,
|
| 1731 |
+
"step": 287
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 11.52,
|
| 1735 |
+
"learning_rate": 8.48e-05,
|
| 1736 |
+
"loss": 0.9476,
|
| 1737 |
+
"step": 288
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 11.56,
|
| 1741 |
+
"learning_rate": 8.44e-05,
|
| 1742 |
+
"loss": 1.1954,
|
| 1743 |
+
"step": 289
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 11.6,
|
| 1747 |
+
"learning_rate": 8.4e-05,
|
| 1748 |
+
"loss": 1.0309,
|
| 1749 |
+
"step": 290
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 11.64,
|
| 1753 |
+
"learning_rate": 8.36e-05,
|
| 1754 |
+
"loss": 1.1321,
|
| 1755 |
+
"step": 291
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 11.68,
|
| 1759 |
+
"learning_rate": 8.32e-05,
|
| 1760 |
+
"loss": 1.0914,
|
| 1761 |
+
"step": 292
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 11.72,
|
| 1765 |
+
"learning_rate": 8.28e-05,
|
| 1766 |
+
"loss": 1.222,
|
| 1767 |
+
"step": 293
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 11.76,
|
| 1771 |
+
"learning_rate": 8.24e-05,
|
| 1772 |
+
"loss": 1.0127,
|
| 1773 |
+
"step": 294
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 11.8,
|
| 1777 |
+
"learning_rate": 8.2e-05,
|
| 1778 |
+
"loss": 1.1977,
|
| 1779 |
+
"step": 295
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 11.84,
|
| 1783 |
+
"learning_rate": 8.16e-05,
|
| 1784 |
+
"loss": 1.122,
|
| 1785 |
+
"step": 296
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 11.88,
|
| 1789 |
+
"learning_rate": 8.120000000000001e-05,
|
| 1790 |
+
"loss": 1.1543,
|
| 1791 |
+
"step": 297
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 11.92,
|
| 1795 |
+
"learning_rate": 8.080000000000001e-05,
|
| 1796 |
+
"loss": 1.1767,
|
| 1797 |
+
"step": 298
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 11.96,
|
| 1801 |
+
"learning_rate": 8.04e-05,
|
| 1802 |
+
"loss": 1.1341,
|
| 1803 |
+
"step": 299
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 12.0,
|
| 1807 |
+
"learning_rate": 8e-05,
|
| 1808 |
+
"loss": 1.0107,
|
| 1809 |
+
"step": 300
|
| 1810 |
+
}
|
| 1811 |
+
],
|
| 1812 |
+
"logging_steps": 1,
|
| 1813 |
+
"max_steps": 500,
|
| 1814 |
+
"num_train_epochs": 20,
|
| 1815 |
+
"save_steps": 100,
|
| 1816 |
+
"total_flos": 1.0209241970264064e+17,
|
| 1817 |
+
"trial_name": null,
|
| 1818 |
+
"trial_params": null
|
| 1819 |
+
}
|
checkpoint-300/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ef74b31950ae6c8955316bed48c343fb06cda0cb6a8a54ca46ca3cb681c8736
|
| 3 |
+
size 4027
|
checkpoint-400/README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
checkpoint-400/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-400/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2314e06a7452fc12576ddd903f95a2753dfd474755a766d263b3d938b19d66f
|
| 3 |
+
size 39409357
|
checkpoint-400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:744a5e62f698effd66997373bf91c530543a99d8828c460626ac0cd2032b7420
|
| 3 |
+
size 78844421
|
checkpoint-400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51f3c17c1e7ce128318be2166e96b6dda4d1ff79c8b44e6fb06488bf0bbfcf8d
|
| 3 |
+
size 14575
|
checkpoint-400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9aafe6849dc1b38ea86a67ee0dd398e835f1054275ee4627209b6ffeca7902f
|
| 3 |
+
size 627
|
checkpoint-400/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
checkpoint-400/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17a208233d2ee8d8c83b23bc214df737c44806a1919f444e89b31e586cd956ba
|
| 3 |
+
size 14500471
|
checkpoint-400/tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"tokenizer_class": "BloomTokenizer",
|
| 9 |
+
"unk_token": "<unk>"
|
| 10 |
+
}
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,2419 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 16.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"learning_rate": 0.0001996,
|
| 14 |
+
"loss": 2.4683,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.08,
|
| 19 |
+
"learning_rate": 0.00019920000000000002,
|
| 20 |
+
"loss": 2.2023,
|
| 21 |
+
"step": 2
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.12,
|
| 25 |
+
"learning_rate": 0.0001988,
|
| 26 |
+
"loss": 2.3039,
|
| 27 |
+
"step": 3
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.16,
|
| 31 |
+
"learning_rate": 0.0001984,
|
| 32 |
+
"loss": 2.1855,
|
| 33 |
+
"step": 4
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.2,
|
| 37 |
+
"learning_rate": 0.00019800000000000002,
|
| 38 |
+
"loss": 2.328,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.24,
|
| 43 |
+
"learning_rate": 0.0001976,
|
| 44 |
+
"loss": 2.3407,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.28,
|
| 49 |
+
"learning_rate": 0.0001972,
|
| 50 |
+
"loss": 2.3432,
|
| 51 |
+
"step": 7
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"learning_rate": 0.0001968,
|
| 56 |
+
"loss": 2.2333,
|
| 57 |
+
"step": 8
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.36,
|
| 61 |
+
"learning_rate": 0.0001964,
|
| 62 |
+
"loss": 2.1761,
|
| 63 |
+
"step": 9
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.4,
|
| 67 |
+
"learning_rate": 0.000196,
|
| 68 |
+
"loss": 2.1473,
|
| 69 |
+
"step": 10
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.44,
|
| 73 |
+
"learning_rate": 0.0001956,
|
| 74 |
+
"loss": 2.2076,
|
| 75 |
+
"step": 11
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.48,
|
| 79 |
+
"learning_rate": 0.0001952,
|
| 80 |
+
"loss": 2.1925,
|
| 81 |
+
"step": 12
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.52,
|
| 85 |
+
"learning_rate": 0.0001948,
|
| 86 |
+
"loss": 2.1694,
|
| 87 |
+
"step": 13
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.56,
|
| 91 |
+
"learning_rate": 0.0001944,
|
| 92 |
+
"loss": 2.1056,
|
| 93 |
+
"step": 14
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6,
|
| 97 |
+
"learning_rate": 0.000194,
|
| 98 |
+
"loss": 2.2426,
|
| 99 |
+
"step": 15
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.64,
|
| 103 |
+
"learning_rate": 0.00019360000000000002,
|
| 104 |
+
"loss": 2.2635,
|
| 105 |
+
"step": 16
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.68,
|
| 109 |
+
"learning_rate": 0.0001932,
|
| 110 |
+
"loss": 2.251,
|
| 111 |
+
"step": 17
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.72,
|
| 115 |
+
"learning_rate": 0.0001928,
|
| 116 |
+
"loss": 2.224,
|
| 117 |
+
"step": 18
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.76,
|
| 121 |
+
"learning_rate": 0.00019240000000000001,
|
| 122 |
+
"loss": 2.1848,
|
| 123 |
+
"step": 19
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.8,
|
| 127 |
+
"learning_rate": 0.000192,
|
| 128 |
+
"loss": 2.1341,
|
| 129 |
+
"step": 20
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.84,
|
| 133 |
+
"learning_rate": 0.0001916,
|
| 134 |
+
"loss": 2.126,
|
| 135 |
+
"step": 21
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.88,
|
| 139 |
+
"learning_rate": 0.0001912,
|
| 140 |
+
"loss": 2.0959,
|
| 141 |
+
"step": 22
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.92,
|
| 145 |
+
"learning_rate": 0.0001908,
|
| 146 |
+
"loss": 2.2446,
|
| 147 |
+
"step": 23
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.96,
|
| 151 |
+
"learning_rate": 0.0001904,
|
| 152 |
+
"loss": 1.9448,
|
| 153 |
+
"step": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.0,
|
| 157 |
+
"learning_rate": 0.00019,
|
| 158 |
+
"loss": 2.0939,
|
| 159 |
+
"step": 25
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.04,
|
| 163 |
+
"learning_rate": 0.0001896,
|
| 164 |
+
"loss": 2.1247,
|
| 165 |
+
"step": 26
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.08,
|
| 169 |
+
"learning_rate": 0.0001892,
|
| 170 |
+
"loss": 1.9417,
|
| 171 |
+
"step": 27
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.12,
|
| 175 |
+
"learning_rate": 0.0001888,
|
| 176 |
+
"loss": 2.105,
|
| 177 |
+
"step": 28
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.16,
|
| 181 |
+
"learning_rate": 0.0001884,
|
| 182 |
+
"loss": 2.1595,
|
| 183 |
+
"step": 29
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.2,
|
| 187 |
+
"learning_rate": 0.000188,
|
| 188 |
+
"loss": 2.2009,
|
| 189 |
+
"step": 30
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.24,
|
| 193 |
+
"learning_rate": 0.0001876,
|
| 194 |
+
"loss": 2.0784,
|
| 195 |
+
"step": 31
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.28,
|
| 199 |
+
"learning_rate": 0.00018720000000000002,
|
| 200 |
+
"loss": 2.2173,
|
| 201 |
+
"step": 32
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.32,
|
| 205 |
+
"learning_rate": 0.00018680000000000001,
|
| 206 |
+
"loss": 2.1185,
|
| 207 |
+
"step": 33
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.36,
|
| 211 |
+
"learning_rate": 0.00018640000000000003,
|
| 212 |
+
"loss": 1.988,
|
| 213 |
+
"step": 34
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.4,
|
| 217 |
+
"learning_rate": 0.00018600000000000002,
|
| 218 |
+
"loss": 1.9493,
|
| 219 |
+
"step": 35
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.44,
|
| 223 |
+
"learning_rate": 0.0001856,
|
| 224 |
+
"loss": 1.9947,
|
| 225 |
+
"step": 36
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.48,
|
| 229 |
+
"learning_rate": 0.00018520000000000003,
|
| 230 |
+
"loss": 2.0506,
|
| 231 |
+
"step": 37
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.52,
|
| 235 |
+
"learning_rate": 0.00018480000000000002,
|
| 236 |
+
"loss": 2.0829,
|
| 237 |
+
"step": 38
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.56,
|
| 241 |
+
"learning_rate": 0.0001844,
|
| 242 |
+
"loss": 2.1455,
|
| 243 |
+
"step": 39
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.6,
|
| 247 |
+
"learning_rate": 0.00018400000000000003,
|
| 248 |
+
"loss": 2.0304,
|
| 249 |
+
"step": 40
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.64,
|
| 253 |
+
"learning_rate": 0.00018360000000000002,
|
| 254 |
+
"loss": 2.006,
|
| 255 |
+
"step": 41
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.68,
|
| 259 |
+
"learning_rate": 0.0001832,
|
| 260 |
+
"loss": 2.1759,
|
| 261 |
+
"step": 42
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.72,
|
| 265 |
+
"learning_rate": 0.00018280000000000003,
|
| 266 |
+
"loss": 2.0093,
|
| 267 |
+
"step": 43
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.76,
|
| 271 |
+
"learning_rate": 0.00018240000000000002,
|
| 272 |
+
"loss": 2.0683,
|
| 273 |
+
"step": 44
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 1.8,
|
| 277 |
+
"learning_rate": 0.000182,
|
| 278 |
+
"loss": 2.1352,
|
| 279 |
+
"step": 45
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.84,
|
| 283 |
+
"learning_rate": 0.00018160000000000002,
|
| 284 |
+
"loss": 1.9736,
|
| 285 |
+
"step": 46
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.88,
|
| 289 |
+
"learning_rate": 0.0001812,
|
| 290 |
+
"loss": 1.9631,
|
| 291 |
+
"step": 47
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.92,
|
| 295 |
+
"learning_rate": 0.0001808,
|
| 296 |
+
"loss": 2.1024,
|
| 297 |
+
"step": 48
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.96,
|
| 301 |
+
"learning_rate": 0.00018040000000000002,
|
| 302 |
+
"loss": 1.9895,
|
| 303 |
+
"step": 49
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 2.0,
|
| 307 |
+
"learning_rate": 0.00018,
|
| 308 |
+
"loss": 2.0067,
|
| 309 |
+
"step": 50
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.04,
|
| 313 |
+
"learning_rate": 0.0001796,
|
| 314 |
+
"loss": 2.0509,
|
| 315 |
+
"step": 51
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.08,
|
| 319 |
+
"learning_rate": 0.00017920000000000002,
|
| 320 |
+
"loss": 1.8795,
|
| 321 |
+
"step": 52
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.12,
|
| 325 |
+
"learning_rate": 0.0001788,
|
| 326 |
+
"loss": 2.0633,
|
| 327 |
+
"step": 53
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.16,
|
| 331 |
+
"learning_rate": 0.0001784,
|
| 332 |
+
"loss": 1.9998,
|
| 333 |
+
"step": 54
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.2,
|
| 337 |
+
"learning_rate": 0.00017800000000000002,
|
| 338 |
+
"loss": 2.0675,
|
| 339 |
+
"step": 55
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.24,
|
| 343 |
+
"learning_rate": 0.0001776,
|
| 344 |
+
"loss": 2.0129,
|
| 345 |
+
"step": 56
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.28,
|
| 349 |
+
"learning_rate": 0.0001772,
|
| 350 |
+
"loss": 1.9302,
|
| 351 |
+
"step": 57
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.32,
|
| 355 |
+
"learning_rate": 0.00017680000000000001,
|
| 356 |
+
"loss": 2.0472,
|
| 357 |
+
"step": 58
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.36,
|
| 361 |
+
"learning_rate": 0.0001764,
|
| 362 |
+
"loss": 2.0031,
|
| 363 |
+
"step": 59
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.4,
|
| 367 |
+
"learning_rate": 0.00017600000000000002,
|
| 368 |
+
"loss": 2.0231,
|
| 369 |
+
"step": 60
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.44,
|
| 373 |
+
"learning_rate": 0.0001756,
|
| 374 |
+
"loss": 2.0192,
|
| 375 |
+
"step": 61
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.48,
|
| 379 |
+
"learning_rate": 0.0001752,
|
| 380 |
+
"loss": 2.1336,
|
| 381 |
+
"step": 62
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.52,
|
| 385 |
+
"learning_rate": 0.00017480000000000002,
|
| 386 |
+
"loss": 1.7753,
|
| 387 |
+
"step": 63
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.56,
|
| 391 |
+
"learning_rate": 0.0001744,
|
| 392 |
+
"loss": 2.0274,
|
| 393 |
+
"step": 64
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.6,
|
| 397 |
+
"learning_rate": 0.000174,
|
| 398 |
+
"loss": 1.8893,
|
| 399 |
+
"step": 65
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.64,
|
| 403 |
+
"learning_rate": 0.00017360000000000002,
|
| 404 |
+
"loss": 2.0341,
|
| 405 |
+
"step": 66
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 2.68,
|
| 409 |
+
"learning_rate": 0.0001732,
|
| 410 |
+
"loss": 2.1081,
|
| 411 |
+
"step": 67
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.72,
|
| 415 |
+
"learning_rate": 0.0001728,
|
| 416 |
+
"loss": 1.9463,
|
| 417 |
+
"step": 68
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.76,
|
| 421 |
+
"learning_rate": 0.00017240000000000002,
|
| 422 |
+
"loss": 2.0607,
|
| 423 |
+
"step": 69
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.8,
|
| 427 |
+
"learning_rate": 0.000172,
|
| 428 |
+
"loss": 1.9803,
|
| 429 |
+
"step": 70
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.84,
|
| 433 |
+
"learning_rate": 0.0001716,
|
| 434 |
+
"loss": 1.7663,
|
| 435 |
+
"step": 71
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.88,
|
| 439 |
+
"learning_rate": 0.00017120000000000001,
|
| 440 |
+
"loss": 2.1116,
|
| 441 |
+
"step": 72
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.92,
|
| 445 |
+
"learning_rate": 0.0001708,
|
| 446 |
+
"loss": 1.9039,
|
| 447 |
+
"step": 73
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.96,
|
| 451 |
+
"learning_rate": 0.0001704,
|
| 452 |
+
"loss": 2.049,
|
| 453 |
+
"step": 74
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.0,
|
| 457 |
+
"learning_rate": 0.00017,
|
| 458 |
+
"loss": 1.991,
|
| 459 |
+
"step": 75
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.04,
|
| 463 |
+
"learning_rate": 0.0001696,
|
| 464 |
+
"loss": 1.8755,
|
| 465 |
+
"step": 76
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 3.08,
|
| 469 |
+
"learning_rate": 0.0001692,
|
| 470 |
+
"loss": 2.0367,
|
| 471 |
+
"step": 77
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.12,
|
| 475 |
+
"learning_rate": 0.0001688,
|
| 476 |
+
"loss": 1.9959,
|
| 477 |
+
"step": 78
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.16,
|
| 481 |
+
"learning_rate": 0.0001684,
|
| 482 |
+
"loss": 1.8051,
|
| 483 |
+
"step": 79
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.2,
|
| 487 |
+
"learning_rate": 0.000168,
|
| 488 |
+
"loss": 2.0446,
|
| 489 |
+
"step": 80
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.24,
|
| 493 |
+
"learning_rate": 0.0001676,
|
| 494 |
+
"loss": 1.7873,
|
| 495 |
+
"step": 81
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.28,
|
| 499 |
+
"learning_rate": 0.0001672,
|
| 500 |
+
"loss": 1.9264,
|
| 501 |
+
"step": 82
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 3.32,
|
| 505 |
+
"learning_rate": 0.0001668,
|
| 506 |
+
"loss": 1.8595,
|
| 507 |
+
"step": 83
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.36,
|
| 511 |
+
"learning_rate": 0.0001664,
|
| 512 |
+
"loss": 1.9992,
|
| 513 |
+
"step": 84
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.4,
|
| 517 |
+
"learning_rate": 0.000166,
|
| 518 |
+
"loss": 1.8783,
|
| 519 |
+
"step": 85
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.44,
|
| 523 |
+
"learning_rate": 0.0001656,
|
| 524 |
+
"loss": 2.0216,
|
| 525 |
+
"step": 86
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 3.48,
|
| 529 |
+
"learning_rate": 0.0001652,
|
| 530 |
+
"loss": 1.947,
|
| 531 |
+
"step": 87
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.52,
|
| 535 |
+
"learning_rate": 0.0001648,
|
| 536 |
+
"loss": 1.9554,
|
| 537 |
+
"step": 88
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.56,
|
| 541 |
+
"learning_rate": 0.0001644,
|
| 542 |
+
"loss": 1.8563,
|
| 543 |
+
"step": 89
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.6,
|
| 547 |
+
"learning_rate": 0.000164,
|
| 548 |
+
"loss": 1.7795,
|
| 549 |
+
"step": 90
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.64,
|
| 553 |
+
"learning_rate": 0.0001636,
|
| 554 |
+
"loss": 1.9347,
|
| 555 |
+
"step": 91
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.68,
|
| 559 |
+
"learning_rate": 0.0001632,
|
| 560 |
+
"loss": 2.0078,
|
| 561 |
+
"step": 92
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.72,
|
| 565 |
+
"learning_rate": 0.0001628,
|
| 566 |
+
"loss": 1.9964,
|
| 567 |
+
"step": 93
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 3.76,
|
| 571 |
+
"learning_rate": 0.00016240000000000002,
|
| 572 |
+
"loss": 1.9004,
|
| 573 |
+
"step": 94
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.8,
|
| 577 |
+
"learning_rate": 0.000162,
|
| 578 |
+
"loss": 1.9446,
|
| 579 |
+
"step": 95
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.84,
|
| 583 |
+
"learning_rate": 0.00016160000000000002,
|
| 584 |
+
"loss": 1.8857,
|
| 585 |
+
"step": 96
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.88,
|
| 589 |
+
"learning_rate": 0.00016120000000000002,
|
| 590 |
+
"loss": 1.8797,
|
| 591 |
+
"step": 97
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.92,
|
| 595 |
+
"learning_rate": 0.0001608,
|
| 596 |
+
"loss": 1.7643,
|
| 597 |
+
"step": 98
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.96,
|
| 601 |
+
"learning_rate": 0.00016040000000000002,
|
| 602 |
+
"loss": 1.9724,
|
| 603 |
+
"step": 99
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 4.0,
|
| 607 |
+
"learning_rate": 0.00016,
|
| 608 |
+
"loss": 1.9058,
|
| 609 |
+
"step": 100
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 4.04,
|
| 613 |
+
"learning_rate": 0.0001596,
|
| 614 |
+
"loss": 1.8663,
|
| 615 |
+
"step": 101
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 4.08,
|
| 619 |
+
"learning_rate": 0.00015920000000000002,
|
| 620 |
+
"loss": 1.8457,
|
| 621 |
+
"step": 102
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 4.12,
|
| 625 |
+
"learning_rate": 0.0001588,
|
| 626 |
+
"loss": 1.8394,
|
| 627 |
+
"step": 103
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 4.16,
|
| 631 |
+
"learning_rate": 0.00015840000000000003,
|
| 632 |
+
"loss": 1.8941,
|
| 633 |
+
"step": 104
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 4.2,
|
| 637 |
+
"learning_rate": 0.00015800000000000002,
|
| 638 |
+
"loss": 1.8601,
|
| 639 |
+
"step": 105
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 4.24,
|
| 643 |
+
"learning_rate": 0.0001576,
|
| 644 |
+
"loss": 1.7365,
|
| 645 |
+
"step": 106
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 4.28,
|
| 649 |
+
"learning_rate": 0.00015720000000000003,
|
| 650 |
+
"loss": 1.8809,
|
| 651 |
+
"step": 107
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 4.32,
|
| 655 |
+
"learning_rate": 0.00015680000000000002,
|
| 656 |
+
"loss": 2.0206,
|
| 657 |
+
"step": 108
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 4.36,
|
| 661 |
+
"learning_rate": 0.0001564,
|
| 662 |
+
"loss": 1.7776,
|
| 663 |
+
"step": 109
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 4.4,
|
| 667 |
+
"learning_rate": 0.00015600000000000002,
|
| 668 |
+
"loss": 1.9089,
|
| 669 |
+
"step": 110
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.44,
|
| 673 |
+
"learning_rate": 0.00015560000000000001,
|
| 674 |
+
"loss": 1.8464,
|
| 675 |
+
"step": 111
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.48,
|
| 679 |
+
"learning_rate": 0.0001552,
|
| 680 |
+
"loss": 1.7943,
|
| 681 |
+
"step": 112
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.52,
|
| 685 |
+
"learning_rate": 0.00015480000000000002,
|
| 686 |
+
"loss": 1.8081,
|
| 687 |
+
"step": 113
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 4.56,
|
| 691 |
+
"learning_rate": 0.0001544,
|
| 692 |
+
"loss": 1.734,
|
| 693 |
+
"step": 114
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 4.6,
|
| 697 |
+
"learning_rate": 0.000154,
|
| 698 |
+
"loss": 1.8379,
|
| 699 |
+
"step": 115
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.64,
|
| 703 |
+
"learning_rate": 0.00015360000000000002,
|
| 704 |
+
"loss": 1.6841,
|
| 705 |
+
"step": 116
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 4.68,
|
| 709 |
+
"learning_rate": 0.0001532,
|
| 710 |
+
"loss": 1.8253,
|
| 711 |
+
"step": 117
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 4.72,
|
| 715 |
+
"learning_rate": 0.0001528,
|
| 716 |
+
"loss": 1.7607,
|
| 717 |
+
"step": 118
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 4.76,
|
| 721 |
+
"learning_rate": 0.00015240000000000002,
|
| 722 |
+
"loss": 1.8205,
|
| 723 |
+
"step": 119
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 4.8,
|
| 727 |
+
"learning_rate": 0.000152,
|
| 728 |
+
"loss": 1.8443,
|
| 729 |
+
"step": 120
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 4.84,
|
| 733 |
+
"learning_rate": 0.0001516,
|
| 734 |
+
"loss": 1.8824,
|
| 735 |
+
"step": 121
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 4.88,
|
| 739 |
+
"learning_rate": 0.00015120000000000002,
|
| 740 |
+
"loss": 1.8369,
|
| 741 |
+
"step": 122
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.92,
|
| 745 |
+
"learning_rate": 0.0001508,
|
| 746 |
+
"loss": 1.8017,
|
| 747 |
+
"step": 123
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 4.96,
|
| 751 |
+
"learning_rate": 0.0001504,
|
| 752 |
+
"loss": 1.7858,
|
| 753 |
+
"step": 124
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 5.0,
|
| 757 |
+
"learning_rate": 0.00015000000000000001,
|
| 758 |
+
"loss": 1.8366,
|
| 759 |
+
"step": 125
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 5.04,
|
| 763 |
+
"learning_rate": 0.0001496,
|
| 764 |
+
"loss": 1.7353,
|
| 765 |
+
"step": 126
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 5.08,
|
| 769 |
+
"learning_rate": 0.0001492,
|
| 770 |
+
"loss": 1.6017,
|
| 771 |
+
"step": 127
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 5.12,
|
| 775 |
+
"learning_rate": 0.0001488,
|
| 776 |
+
"loss": 1.6645,
|
| 777 |
+
"step": 128
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 5.16,
|
| 781 |
+
"learning_rate": 0.0001484,
|
| 782 |
+
"loss": 1.8663,
|
| 783 |
+
"step": 129
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 5.2,
|
| 787 |
+
"learning_rate": 0.000148,
|
| 788 |
+
"loss": 1.7009,
|
| 789 |
+
"step": 130
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 5.24,
|
| 793 |
+
"learning_rate": 0.0001476,
|
| 794 |
+
"loss": 1.7371,
|
| 795 |
+
"step": 131
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 5.28,
|
| 799 |
+
"learning_rate": 0.0001472,
|
| 800 |
+
"loss": 1.8555,
|
| 801 |
+
"step": 132
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 5.32,
|
| 805 |
+
"learning_rate": 0.00014680000000000002,
|
| 806 |
+
"loss": 1.6373,
|
| 807 |
+
"step": 133
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 5.36,
|
| 811 |
+
"learning_rate": 0.0001464,
|
| 812 |
+
"loss": 1.7211,
|
| 813 |
+
"step": 134
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 5.4,
|
| 817 |
+
"learning_rate": 0.000146,
|
| 818 |
+
"loss": 1.614,
|
| 819 |
+
"step": 135
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.44,
|
| 823 |
+
"learning_rate": 0.00014560000000000002,
|
| 824 |
+
"loss": 1.8736,
|
| 825 |
+
"step": 136
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.48,
|
| 829 |
+
"learning_rate": 0.0001452,
|
| 830 |
+
"loss": 1.7229,
|
| 831 |
+
"step": 137
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 5.52,
|
| 835 |
+
"learning_rate": 0.0001448,
|
| 836 |
+
"loss": 1.7315,
|
| 837 |
+
"step": 138
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 5.56,
|
| 841 |
+
"learning_rate": 0.0001444,
|
| 842 |
+
"loss": 1.7259,
|
| 843 |
+
"step": 139
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 5.6,
|
| 847 |
+
"learning_rate": 0.000144,
|
| 848 |
+
"loss": 1.7032,
|
| 849 |
+
"step": 140
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 5.64,
|
| 853 |
+
"learning_rate": 0.0001436,
|
| 854 |
+
"loss": 1.9237,
|
| 855 |
+
"step": 141
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 5.68,
|
| 859 |
+
"learning_rate": 0.0001432,
|
| 860 |
+
"loss": 1.7821,
|
| 861 |
+
"step": 142
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.72,
|
| 865 |
+
"learning_rate": 0.0001428,
|
| 866 |
+
"loss": 1.6081,
|
| 867 |
+
"step": 143
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 5.76,
|
| 871 |
+
"learning_rate": 0.0001424,
|
| 872 |
+
"loss": 1.7979,
|
| 873 |
+
"step": 144
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 5.8,
|
| 877 |
+
"learning_rate": 0.000142,
|
| 878 |
+
"loss": 1.6675,
|
| 879 |
+
"step": 145
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 5.84,
|
| 883 |
+
"learning_rate": 0.0001416,
|
| 884 |
+
"loss": 1.6758,
|
| 885 |
+
"step": 146
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 5.88,
|
| 889 |
+
"learning_rate": 0.0001412,
|
| 890 |
+
"loss": 1.7783,
|
| 891 |
+
"step": 147
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 5.92,
|
| 895 |
+
"learning_rate": 0.0001408,
|
| 896 |
+
"loss": 1.5935,
|
| 897 |
+
"step": 148
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 5.96,
|
| 901 |
+
"learning_rate": 0.0001404,
|
| 902 |
+
"loss": 1.7372,
|
| 903 |
+
"step": 149
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 6.0,
|
| 907 |
+
"learning_rate": 0.00014,
|
| 908 |
+
"loss": 1.6308,
|
| 909 |
+
"step": 150
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.04,
|
| 913 |
+
"learning_rate": 0.0001396,
|
| 914 |
+
"loss": 1.5829,
|
| 915 |
+
"step": 151
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 6.08,
|
| 919 |
+
"learning_rate": 0.0001392,
|
| 920 |
+
"loss": 1.6538,
|
| 921 |
+
"step": 152
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 6.12,
|
| 925 |
+
"learning_rate": 0.00013879999999999999,
|
| 926 |
+
"loss": 1.7624,
|
| 927 |
+
"step": 153
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 6.16,
|
| 931 |
+
"learning_rate": 0.0001384,
|
| 932 |
+
"loss": 1.4968,
|
| 933 |
+
"step": 154
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 6.2,
|
| 937 |
+
"learning_rate": 0.000138,
|
| 938 |
+
"loss": 1.7305,
|
| 939 |
+
"step": 155
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.24,
|
| 943 |
+
"learning_rate": 0.00013759999999999998,
|
| 944 |
+
"loss": 1.6536,
|
| 945 |
+
"step": 156
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 6.28,
|
| 949 |
+
"learning_rate": 0.00013720000000000003,
|
| 950 |
+
"loss": 1.654,
|
| 951 |
+
"step": 157
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 6.32,
|
| 955 |
+
"learning_rate": 0.00013680000000000002,
|
| 956 |
+
"loss": 1.7163,
|
| 957 |
+
"step": 158
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 6.36,
|
| 961 |
+
"learning_rate": 0.0001364,
|
| 962 |
+
"loss": 1.6599,
|
| 963 |
+
"step": 159
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 6.4,
|
| 967 |
+
"learning_rate": 0.00013600000000000003,
|
| 968 |
+
"loss": 1.5729,
|
| 969 |
+
"step": 160
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 6.44,
|
| 973 |
+
"learning_rate": 0.00013560000000000002,
|
| 974 |
+
"loss": 1.5291,
|
| 975 |
+
"step": 161
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 6.48,
|
| 979 |
+
"learning_rate": 0.0001352,
|
| 980 |
+
"loss": 1.5666,
|
| 981 |
+
"step": 162
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.52,
|
| 985 |
+
"learning_rate": 0.00013480000000000002,
|
| 986 |
+
"loss": 1.7576,
|
| 987 |
+
"step": 163
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 6.56,
|
| 991 |
+
"learning_rate": 0.00013440000000000001,
|
| 992 |
+
"loss": 1.4462,
|
| 993 |
+
"step": 164
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 6.6,
|
| 997 |
+
"learning_rate": 0.000134,
|
| 998 |
+
"loss": 1.5659,
|
| 999 |
+
"step": 165
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 6.64,
|
| 1003 |
+
"learning_rate": 0.00013360000000000002,
|
| 1004 |
+
"loss": 1.6234,
|
| 1005 |
+
"step": 166
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 6.68,
|
| 1009 |
+
"learning_rate": 0.0001332,
|
| 1010 |
+
"loss": 1.5176,
|
| 1011 |
+
"step": 167
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 6.72,
|
| 1015 |
+
"learning_rate": 0.0001328,
|
| 1016 |
+
"loss": 1.667,
|
| 1017 |
+
"step": 168
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 6.76,
|
| 1021 |
+
"learning_rate": 0.00013240000000000002,
|
| 1022 |
+
"loss": 1.6514,
|
| 1023 |
+
"step": 169
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.8,
|
| 1027 |
+
"learning_rate": 0.000132,
|
| 1028 |
+
"loss": 1.7179,
|
| 1029 |
+
"step": 170
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 6.84,
|
| 1033 |
+
"learning_rate": 0.0001316,
|
| 1034 |
+
"loss": 1.4373,
|
| 1035 |
+
"step": 171
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 6.88,
|
| 1039 |
+
"learning_rate": 0.00013120000000000002,
|
| 1040 |
+
"loss": 1.7044,
|
| 1041 |
+
"step": 172
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 6.92,
|
| 1045 |
+
"learning_rate": 0.0001308,
|
| 1046 |
+
"loss": 1.5673,
|
| 1047 |
+
"step": 173
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 6.96,
|
| 1051 |
+
"learning_rate": 0.0001304,
|
| 1052 |
+
"loss": 1.6303,
|
| 1053 |
+
"step": 174
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 7.0,
|
| 1057 |
+
"learning_rate": 0.00013000000000000002,
|
| 1058 |
+
"loss": 1.4828,
|
| 1059 |
+
"step": 175
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 7.04,
|
| 1063 |
+
"learning_rate": 0.0001296,
|
| 1064 |
+
"loss": 1.6204,
|
| 1065 |
+
"step": 176
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 7.08,
|
| 1069 |
+
"learning_rate": 0.00012920000000000002,
|
| 1070 |
+
"loss": 1.4955,
|
| 1071 |
+
"step": 177
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 7.12,
|
| 1075 |
+
"learning_rate": 0.00012880000000000001,
|
| 1076 |
+
"loss": 1.6109,
|
| 1077 |
+
"step": 178
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 7.16,
|
| 1081 |
+
"learning_rate": 0.0001284,
|
| 1082 |
+
"loss": 1.5318,
|
| 1083 |
+
"step": 179
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 7.2,
|
| 1087 |
+
"learning_rate": 0.00012800000000000002,
|
| 1088 |
+
"loss": 1.5702,
|
| 1089 |
+
"step": 180
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 7.24,
|
| 1093 |
+
"learning_rate": 0.0001276,
|
| 1094 |
+
"loss": 1.4598,
|
| 1095 |
+
"step": 181
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 7.28,
|
| 1099 |
+
"learning_rate": 0.0001272,
|
| 1100 |
+
"loss": 1.6145,
|
| 1101 |
+
"step": 182
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 7.32,
|
| 1105 |
+
"learning_rate": 0.00012680000000000002,
|
| 1106 |
+
"loss": 1.3628,
|
| 1107 |
+
"step": 183
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 7.36,
|
| 1111 |
+
"learning_rate": 0.0001264,
|
| 1112 |
+
"loss": 1.4578,
|
| 1113 |
+
"step": 184
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 7.4,
|
| 1117 |
+
"learning_rate": 0.000126,
|
| 1118 |
+
"loss": 1.7094,
|
| 1119 |
+
"step": 185
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 7.44,
|
| 1123 |
+
"learning_rate": 0.00012560000000000002,
|
| 1124 |
+
"loss": 1.4376,
|
| 1125 |
+
"step": 186
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 7.48,
|
| 1129 |
+
"learning_rate": 0.0001252,
|
| 1130 |
+
"loss": 1.3393,
|
| 1131 |
+
"step": 187
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 7.52,
|
| 1135 |
+
"learning_rate": 0.0001248,
|
| 1136 |
+
"loss": 1.4273,
|
| 1137 |
+
"step": 188
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 7.56,
|
| 1141 |
+
"learning_rate": 0.00012440000000000002,
|
| 1142 |
+
"loss": 1.5506,
|
| 1143 |
+
"step": 189
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.6,
|
| 1147 |
+
"learning_rate": 0.000124,
|
| 1148 |
+
"loss": 1.4774,
|
| 1149 |
+
"step": 190
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 7.64,
|
| 1153 |
+
"learning_rate": 0.0001236,
|
| 1154 |
+
"loss": 1.4632,
|
| 1155 |
+
"step": 191
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 7.68,
|
| 1159 |
+
"learning_rate": 0.0001232,
|
| 1160 |
+
"loss": 1.4568,
|
| 1161 |
+
"step": 192
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 7.72,
|
| 1165 |
+
"learning_rate": 0.0001228,
|
| 1166 |
+
"loss": 1.6106,
|
| 1167 |
+
"step": 193
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 7.76,
|
| 1171 |
+
"learning_rate": 0.0001224,
|
| 1172 |
+
"loss": 1.577,
|
| 1173 |
+
"step": 194
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 7.8,
|
| 1177 |
+
"learning_rate": 0.000122,
|
| 1178 |
+
"loss": 1.4805,
|
| 1179 |
+
"step": 195
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 7.84,
|
| 1183 |
+
"learning_rate": 0.0001216,
|
| 1184 |
+
"loss": 1.4546,
|
| 1185 |
+
"step": 196
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.88,
|
| 1189 |
+
"learning_rate": 0.0001212,
|
| 1190 |
+
"loss": 1.5164,
|
| 1191 |
+
"step": 197
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 7.92,
|
| 1195 |
+
"learning_rate": 0.0001208,
|
| 1196 |
+
"loss": 1.5061,
|
| 1197 |
+
"step": 198
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 7.96,
|
| 1201 |
+
"learning_rate": 0.0001204,
|
| 1202 |
+
"loss": 1.5312,
|
| 1203 |
+
"step": 199
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 8.0,
|
| 1207 |
+
"learning_rate": 0.00012,
|
| 1208 |
+
"loss": 1.4463,
|
| 1209 |
+
"step": 200
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 8.04,
|
| 1213 |
+
"learning_rate": 0.00011960000000000001,
|
| 1214 |
+
"loss": 1.4207,
|
| 1215 |
+
"step": 201
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 8.08,
|
| 1219 |
+
"learning_rate": 0.0001192,
|
| 1220 |
+
"loss": 1.4688,
|
| 1221 |
+
"step": 202
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 8.12,
|
| 1225 |
+
"learning_rate": 0.0001188,
|
| 1226 |
+
"loss": 1.4445,
|
| 1227 |
+
"step": 203
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 8.16,
|
| 1231 |
+
"learning_rate": 0.0001184,
|
| 1232 |
+
"loss": 1.4158,
|
| 1233 |
+
"step": 204
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 8.2,
|
| 1237 |
+
"learning_rate": 0.000118,
|
| 1238 |
+
"loss": 1.3483,
|
| 1239 |
+
"step": 205
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 8.24,
|
| 1243 |
+
"learning_rate": 0.0001176,
|
| 1244 |
+
"loss": 1.4117,
|
| 1245 |
+
"step": 206
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 8.28,
|
| 1249 |
+
"learning_rate": 0.0001172,
|
| 1250 |
+
"loss": 1.4267,
|
| 1251 |
+
"step": 207
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 8.32,
|
| 1255 |
+
"learning_rate": 0.00011679999999999999,
|
| 1256 |
+
"loss": 1.5547,
|
| 1257 |
+
"step": 208
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 8.36,
|
| 1261 |
+
"learning_rate": 0.0001164,
|
| 1262 |
+
"loss": 1.4883,
|
| 1263 |
+
"step": 209
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 8.4,
|
| 1267 |
+
"learning_rate": 0.000116,
|
| 1268 |
+
"loss": 1.414,
|
| 1269 |
+
"step": 210
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 8.44,
|
| 1273 |
+
"learning_rate": 0.00011559999999999999,
|
| 1274 |
+
"loss": 1.3254,
|
| 1275 |
+
"step": 211
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 8.48,
|
| 1279 |
+
"learning_rate": 0.0001152,
|
| 1280 |
+
"loss": 1.22,
|
| 1281 |
+
"step": 212
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 8.52,
|
| 1285 |
+
"learning_rate": 0.0001148,
|
| 1286 |
+
"loss": 1.3224,
|
| 1287 |
+
"step": 213
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 8.56,
|
| 1291 |
+
"learning_rate": 0.0001144,
|
| 1292 |
+
"loss": 1.6158,
|
| 1293 |
+
"step": 214
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 8.6,
|
| 1297 |
+
"learning_rate": 0.00011399999999999999,
|
| 1298 |
+
"loss": 1.2817,
|
| 1299 |
+
"step": 215
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 8.64,
|
| 1303 |
+
"learning_rate": 0.0001136,
|
| 1304 |
+
"loss": 1.4077,
|
| 1305 |
+
"step": 216
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 8.68,
|
| 1309 |
+
"learning_rate": 0.0001132,
|
| 1310 |
+
"loss": 1.5311,
|
| 1311 |
+
"step": 217
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 8.72,
|
| 1315 |
+
"learning_rate": 0.00011279999999999999,
|
| 1316 |
+
"loss": 1.5289,
|
| 1317 |
+
"step": 218
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 8.76,
|
| 1321 |
+
"learning_rate": 0.00011240000000000002,
|
| 1322 |
+
"loss": 1.4576,
|
| 1323 |
+
"step": 219
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 8.8,
|
| 1327 |
+
"learning_rate": 0.00011200000000000001,
|
| 1328 |
+
"loss": 1.2063,
|
| 1329 |
+
"step": 220
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 8.84,
|
| 1333 |
+
"learning_rate": 0.00011160000000000002,
|
| 1334 |
+
"loss": 1.3543,
|
| 1335 |
+
"step": 221
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 8.88,
|
| 1339 |
+
"learning_rate": 0.00011120000000000002,
|
| 1340 |
+
"loss": 1.4352,
|
| 1341 |
+
"step": 222
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 8.92,
|
| 1345 |
+
"learning_rate": 0.00011080000000000001,
|
| 1346 |
+
"loss": 1.3058,
|
| 1347 |
+
"step": 223
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.96,
|
| 1351 |
+
"learning_rate": 0.00011040000000000001,
|
| 1352 |
+
"loss": 1.3475,
|
| 1353 |
+
"step": 224
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 9.0,
|
| 1357 |
+
"learning_rate": 0.00011000000000000002,
|
| 1358 |
+
"loss": 1.1293,
|
| 1359 |
+
"step": 225
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 9.04,
|
| 1363 |
+
"learning_rate": 0.00010960000000000001,
|
| 1364 |
+
"loss": 1.1671,
|
| 1365 |
+
"step": 226
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 9.08,
|
| 1369 |
+
"learning_rate": 0.00010920000000000001,
|
| 1370 |
+
"loss": 1.4726,
|
| 1371 |
+
"step": 227
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 9.12,
|
| 1375 |
+
"learning_rate": 0.00010880000000000002,
|
| 1376 |
+
"loss": 1.3986,
|
| 1377 |
+
"step": 228
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 9.16,
|
| 1381 |
+
"learning_rate": 0.00010840000000000002,
|
| 1382 |
+
"loss": 1.2731,
|
| 1383 |
+
"step": 229
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 9.2,
|
| 1387 |
+
"learning_rate": 0.00010800000000000001,
|
| 1388 |
+
"loss": 1.2212,
|
| 1389 |
+
"step": 230
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 9.24,
|
| 1393 |
+
"learning_rate": 0.00010760000000000001,
|
| 1394 |
+
"loss": 1.3067,
|
| 1395 |
+
"step": 231
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 9.28,
|
| 1399 |
+
"learning_rate": 0.00010720000000000002,
|
| 1400 |
+
"loss": 1.278,
|
| 1401 |
+
"step": 232
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 9.32,
|
| 1405 |
+
"learning_rate": 0.00010680000000000001,
|
| 1406 |
+
"loss": 1.4141,
|
| 1407 |
+
"step": 233
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 9.36,
|
| 1411 |
+
"learning_rate": 0.00010640000000000001,
|
| 1412 |
+
"loss": 1.2885,
|
| 1413 |
+
"step": 234
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 9.4,
|
| 1417 |
+
"learning_rate": 0.00010600000000000002,
|
| 1418 |
+
"loss": 1.466,
|
| 1419 |
+
"step": 235
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 9.44,
|
| 1423 |
+
"learning_rate": 0.0001056,
|
| 1424 |
+
"loss": 1.2979,
|
| 1425 |
+
"step": 236
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 9.48,
|
| 1429 |
+
"learning_rate": 0.00010520000000000001,
|
| 1430 |
+
"loss": 1.1622,
|
| 1431 |
+
"step": 237
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 9.52,
|
| 1435 |
+
"learning_rate": 0.00010480000000000001,
|
| 1436 |
+
"loss": 1.2504,
|
| 1437 |
+
"step": 238
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 9.56,
|
| 1441 |
+
"learning_rate": 0.0001044,
|
| 1442 |
+
"loss": 1.228,
|
| 1443 |
+
"step": 239
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 9.6,
|
| 1447 |
+
"learning_rate": 0.00010400000000000001,
|
| 1448 |
+
"loss": 1.3607,
|
| 1449 |
+
"step": 240
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 9.64,
|
| 1453 |
+
"learning_rate": 0.00010360000000000001,
|
| 1454 |
+
"loss": 1.1921,
|
| 1455 |
+
"step": 241
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 9.68,
|
| 1459 |
+
"learning_rate": 0.0001032,
|
| 1460 |
+
"loss": 1.2114,
|
| 1461 |
+
"step": 242
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 9.72,
|
| 1465 |
+
"learning_rate": 0.0001028,
|
| 1466 |
+
"loss": 1.2385,
|
| 1467 |
+
"step": 243
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 9.76,
|
| 1471 |
+
"learning_rate": 0.00010240000000000001,
|
| 1472 |
+
"loss": 1.3227,
|
| 1473 |
+
"step": 244
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 9.8,
|
| 1477 |
+
"learning_rate": 0.00010200000000000001,
|
| 1478 |
+
"loss": 1.3935,
|
| 1479 |
+
"step": 245
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 9.84,
|
| 1483 |
+
"learning_rate": 0.0001016,
|
| 1484 |
+
"loss": 1.2209,
|
| 1485 |
+
"step": 246
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 9.88,
|
| 1489 |
+
"learning_rate": 0.00010120000000000001,
|
| 1490 |
+
"loss": 1.1541,
|
| 1491 |
+
"step": 247
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 9.92,
|
| 1495 |
+
"learning_rate": 0.00010080000000000001,
|
| 1496 |
+
"loss": 1.4049,
|
| 1497 |
+
"step": 248
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 9.96,
|
| 1501 |
+
"learning_rate": 0.0001004,
|
| 1502 |
+
"loss": 1.13,
|
| 1503 |
+
"step": 249
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 10.0,
|
| 1507 |
+
"learning_rate": 0.0001,
|
| 1508 |
+
"loss": 1.3783,
|
| 1509 |
+
"step": 250
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 10.04,
|
| 1513 |
+
"learning_rate": 9.960000000000001e-05,
|
| 1514 |
+
"loss": 1.3207,
|
| 1515 |
+
"step": 251
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 10.08,
|
| 1519 |
+
"learning_rate": 9.92e-05,
|
| 1520 |
+
"loss": 1.2697,
|
| 1521 |
+
"step": 252
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 10.12,
|
| 1525 |
+
"learning_rate": 9.88e-05,
|
| 1526 |
+
"loss": 1.091,
|
| 1527 |
+
"step": 253
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 10.16,
|
| 1531 |
+
"learning_rate": 9.84e-05,
|
| 1532 |
+
"loss": 1.0463,
|
| 1533 |
+
"step": 254
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 10.2,
|
| 1537 |
+
"learning_rate": 9.8e-05,
|
| 1538 |
+
"loss": 1.3686,
|
| 1539 |
+
"step": 255
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 10.24,
|
| 1543 |
+
"learning_rate": 9.76e-05,
|
| 1544 |
+
"loss": 1.3109,
|
| 1545 |
+
"step": 256
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 10.28,
|
| 1549 |
+
"learning_rate": 9.72e-05,
|
| 1550 |
+
"loss": 1.2338,
|
| 1551 |
+
"step": 257
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 10.32,
|
| 1555 |
+
"learning_rate": 9.680000000000001e-05,
|
| 1556 |
+
"loss": 1.12,
|
| 1557 |
+
"step": 258
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 10.36,
|
| 1561 |
+
"learning_rate": 9.64e-05,
|
| 1562 |
+
"loss": 1.1864,
|
| 1563 |
+
"step": 259
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 10.4,
|
| 1567 |
+
"learning_rate": 9.6e-05,
|
| 1568 |
+
"loss": 1.1062,
|
| 1569 |
+
"step": 260
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 10.44,
|
| 1573 |
+
"learning_rate": 9.56e-05,
|
| 1574 |
+
"loss": 1.2924,
|
| 1575 |
+
"step": 261
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 10.48,
|
| 1579 |
+
"learning_rate": 9.52e-05,
|
| 1580 |
+
"loss": 1.1931,
|
| 1581 |
+
"step": 262
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 10.52,
|
| 1585 |
+
"learning_rate": 9.48e-05,
|
| 1586 |
+
"loss": 1.2366,
|
| 1587 |
+
"step": 263
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 10.56,
|
| 1591 |
+
"learning_rate": 9.44e-05,
|
| 1592 |
+
"loss": 1.1589,
|
| 1593 |
+
"step": 264
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 10.6,
|
| 1597 |
+
"learning_rate": 9.4e-05,
|
| 1598 |
+
"loss": 1.1197,
|
| 1599 |
+
"step": 265
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 10.64,
|
| 1603 |
+
"learning_rate": 9.360000000000001e-05,
|
| 1604 |
+
"loss": 1.3072,
|
| 1605 |
+
"step": 266
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 10.68,
|
| 1609 |
+
"learning_rate": 9.320000000000002e-05,
|
| 1610 |
+
"loss": 1.1173,
|
| 1611 |
+
"step": 267
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 10.72,
|
| 1615 |
+
"learning_rate": 9.28e-05,
|
| 1616 |
+
"loss": 1.0378,
|
| 1617 |
+
"step": 268
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 10.76,
|
| 1621 |
+
"learning_rate": 9.240000000000001e-05,
|
| 1622 |
+
"loss": 1.1746,
|
| 1623 |
+
"step": 269
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 10.8,
|
| 1627 |
+
"learning_rate": 9.200000000000001e-05,
|
| 1628 |
+
"loss": 1.1144,
|
| 1629 |
+
"step": 270
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 10.84,
|
| 1633 |
+
"learning_rate": 9.16e-05,
|
| 1634 |
+
"loss": 1.1862,
|
| 1635 |
+
"step": 271
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 10.88,
|
| 1639 |
+
"learning_rate": 9.120000000000001e-05,
|
| 1640 |
+
"loss": 1.0816,
|
| 1641 |
+
"step": 272
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 10.92,
|
| 1645 |
+
"learning_rate": 9.080000000000001e-05,
|
| 1646 |
+
"loss": 1.2288,
|
| 1647 |
+
"step": 273
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 10.96,
|
| 1651 |
+
"learning_rate": 9.04e-05,
|
| 1652 |
+
"loss": 1.3031,
|
| 1653 |
+
"step": 274
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 11.0,
|
| 1657 |
+
"learning_rate": 9e-05,
|
| 1658 |
+
"loss": 1.1646,
|
| 1659 |
+
"step": 275
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 11.04,
|
| 1663 |
+
"learning_rate": 8.960000000000001e-05,
|
| 1664 |
+
"loss": 1.2088,
|
| 1665 |
+
"step": 276
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 11.08,
|
| 1669 |
+
"learning_rate": 8.92e-05,
|
| 1670 |
+
"loss": 1.1465,
|
| 1671 |
+
"step": 277
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 11.12,
|
| 1675 |
+
"learning_rate": 8.88e-05,
|
| 1676 |
+
"loss": 1.0313,
|
| 1677 |
+
"step": 278
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 11.16,
|
| 1681 |
+
"learning_rate": 8.840000000000001e-05,
|
| 1682 |
+
"loss": 0.9662,
|
| 1683 |
+
"step": 279
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 11.2,
|
| 1687 |
+
"learning_rate": 8.800000000000001e-05,
|
| 1688 |
+
"loss": 1.051,
|
| 1689 |
+
"step": 280
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 11.24,
|
| 1693 |
+
"learning_rate": 8.76e-05,
|
| 1694 |
+
"loss": 1.1708,
|
| 1695 |
+
"step": 281
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 11.28,
|
| 1699 |
+
"learning_rate": 8.72e-05,
|
| 1700 |
+
"loss": 1.0236,
|
| 1701 |
+
"step": 282
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 11.32,
|
| 1705 |
+
"learning_rate": 8.680000000000001e-05,
|
| 1706 |
+
"loss": 1.1359,
|
| 1707 |
+
"step": 283
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 11.36,
|
| 1711 |
+
"learning_rate": 8.64e-05,
|
| 1712 |
+
"loss": 1.1398,
|
| 1713 |
+
"step": 284
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 11.4,
|
| 1717 |
+
"learning_rate": 8.6e-05,
|
| 1718 |
+
"loss": 1.0213,
|
| 1719 |
+
"step": 285
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 11.44,
|
| 1723 |
+
"learning_rate": 8.560000000000001e-05,
|
| 1724 |
+
"loss": 1.1788,
|
| 1725 |
+
"step": 286
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 11.48,
|
| 1729 |
+
"learning_rate": 8.52e-05,
|
| 1730 |
+
"loss": 1.0387,
|
| 1731 |
+
"step": 287
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 11.52,
|
| 1735 |
+
"learning_rate": 8.48e-05,
|
| 1736 |
+
"loss": 0.9476,
|
| 1737 |
+
"step": 288
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 11.56,
|
| 1741 |
+
"learning_rate": 8.44e-05,
|
| 1742 |
+
"loss": 1.1954,
|
| 1743 |
+
"step": 289
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 11.6,
|
| 1747 |
+
"learning_rate": 8.4e-05,
|
| 1748 |
+
"loss": 1.0309,
|
| 1749 |
+
"step": 290
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 11.64,
|
| 1753 |
+
"learning_rate": 8.36e-05,
|
| 1754 |
+
"loss": 1.1321,
|
| 1755 |
+
"step": 291
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 11.68,
|
| 1759 |
+
"learning_rate": 8.32e-05,
|
| 1760 |
+
"loss": 1.0914,
|
| 1761 |
+
"step": 292
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 11.72,
|
| 1765 |
+
"learning_rate": 8.28e-05,
|
| 1766 |
+
"loss": 1.222,
|
| 1767 |
+
"step": 293
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 11.76,
|
| 1771 |
+
"learning_rate": 8.24e-05,
|
| 1772 |
+
"loss": 1.0127,
|
| 1773 |
+
"step": 294
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 11.8,
|
| 1777 |
+
"learning_rate": 8.2e-05,
|
| 1778 |
+
"loss": 1.1977,
|
| 1779 |
+
"step": 295
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 11.84,
|
| 1783 |
+
"learning_rate": 8.16e-05,
|
| 1784 |
+
"loss": 1.122,
|
| 1785 |
+
"step": 296
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 11.88,
|
| 1789 |
+
"learning_rate": 8.120000000000001e-05,
|
| 1790 |
+
"loss": 1.1543,
|
| 1791 |
+
"step": 297
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 11.92,
|
| 1795 |
+
"learning_rate": 8.080000000000001e-05,
|
| 1796 |
+
"loss": 1.1767,
|
| 1797 |
+
"step": 298
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 11.96,
|
| 1801 |
+
"learning_rate": 8.04e-05,
|
| 1802 |
+
"loss": 1.1341,
|
| 1803 |
+
"step": 299
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 12.0,
|
| 1807 |
+
"learning_rate": 8e-05,
|
| 1808 |
+
"loss": 1.0107,
|
| 1809 |
+
"step": 300
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 12.04,
|
| 1813 |
+
"learning_rate": 7.960000000000001e-05,
|
| 1814 |
+
"loss": 1.1455,
|
| 1815 |
+
"step": 301
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 12.08,
|
| 1819 |
+
"learning_rate": 7.920000000000001e-05,
|
| 1820 |
+
"loss": 1.0091,
|
| 1821 |
+
"step": 302
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 12.12,
|
| 1825 |
+
"learning_rate": 7.88e-05,
|
| 1826 |
+
"loss": 0.9888,
|
| 1827 |
+
"step": 303
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 12.16,
|
| 1831 |
+
"learning_rate": 7.840000000000001e-05,
|
| 1832 |
+
"loss": 0.9894,
|
| 1833 |
+
"step": 304
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 12.2,
|
| 1837 |
+
"learning_rate": 7.800000000000001e-05,
|
| 1838 |
+
"loss": 1.0095,
|
| 1839 |
+
"step": 305
|
| 1840 |
+
},
|
| 1841 |
+
{
|
| 1842 |
+
"epoch": 12.24,
|
| 1843 |
+
"learning_rate": 7.76e-05,
|
| 1844 |
+
"loss": 1.1163,
|
| 1845 |
+
"step": 306
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 12.28,
|
| 1849 |
+
"learning_rate": 7.72e-05,
|
| 1850 |
+
"loss": 1.0345,
|
| 1851 |
+
"step": 307
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 12.32,
|
| 1855 |
+
"learning_rate": 7.680000000000001e-05,
|
| 1856 |
+
"loss": 0.9662,
|
| 1857 |
+
"step": 308
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 12.36,
|
| 1861 |
+
"learning_rate": 7.64e-05,
|
| 1862 |
+
"loss": 0.9064,
|
| 1863 |
+
"step": 309
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 12.4,
|
| 1867 |
+
"learning_rate": 7.6e-05,
|
| 1868 |
+
"loss": 0.9448,
|
| 1869 |
+
"step": 310
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 12.44,
|
| 1873 |
+
"learning_rate": 7.560000000000001e-05,
|
| 1874 |
+
"loss": 0.9509,
|
| 1875 |
+
"step": 311
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 12.48,
|
| 1879 |
+
"learning_rate": 7.52e-05,
|
| 1880 |
+
"loss": 1.123,
|
| 1881 |
+
"step": 312
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 12.52,
|
| 1885 |
+
"learning_rate": 7.48e-05,
|
| 1886 |
+
"loss": 1.1011,
|
| 1887 |
+
"step": 313
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 12.56,
|
| 1891 |
+
"learning_rate": 7.44e-05,
|
| 1892 |
+
"loss": 0.9398,
|
| 1893 |
+
"step": 314
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 12.6,
|
| 1897 |
+
"learning_rate": 7.4e-05,
|
| 1898 |
+
"loss": 1.0498,
|
| 1899 |
+
"step": 315
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 12.64,
|
| 1903 |
+
"learning_rate": 7.36e-05,
|
| 1904 |
+
"loss": 1.0442,
|
| 1905 |
+
"step": 316
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 12.68,
|
| 1909 |
+
"learning_rate": 7.32e-05,
|
| 1910 |
+
"loss": 0.9599,
|
| 1911 |
+
"step": 317
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 12.72,
|
| 1915 |
+
"learning_rate": 7.280000000000001e-05,
|
| 1916 |
+
"loss": 0.8778,
|
| 1917 |
+
"step": 318
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 12.76,
|
| 1921 |
+
"learning_rate": 7.24e-05,
|
| 1922 |
+
"loss": 0.8727,
|
| 1923 |
+
"step": 319
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 12.8,
|
| 1927 |
+
"learning_rate": 7.2e-05,
|
| 1928 |
+
"loss": 1.0189,
|
| 1929 |
+
"step": 320
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 12.84,
|
| 1933 |
+
"learning_rate": 7.16e-05,
|
| 1934 |
+
"loss": 1.0543,
|
| 1935 |
+
"step": 321
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 12.88,
|
| 1939 |
+
"learning_rate": 7.12e-05,
|
| 1940 |
+
"loss": 1.1282,
|
| 1941 |
+
"step": 322
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 12.92,
|
| 1945 |
+
"learning_rate": 7.08e-05,
|
| 1946 |
+
"loss": 1.1018,
|
| 1947 |
+
"step": 323
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 12.96,
|
| 1951 |
+
"learning_rate": 7.04e-05,
|
| 1952 |
+
"loss": 1.1827,
|
| 1953 |
+
"step": 324
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 13.0,
|
| 1957 |
+
"learning_rate": 7e-05,
|
| 1958 |
+
"loss": 1.108,
|
| 1959 |
+
"step": 325
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 13.04,
|
| 1963 |
+
"learning_rate": 6.96e-05,
|
| 1964 |
+
"loss": 0.9618,
|
| 1965 |
+
"step": 326
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 13.08,
|
| 1969 |
+
"learning_rate": 6.92e-05,
|
| 1970 |
+
"loss": 1.0666,
|
| 1971 |
+
"step": 327
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 13.12,
|
| 1975 |
+
"learning_rate": 6.879999999999999e-05,
|
| 1976 |
+
"loss": 0.9973,
|
| 1977 |
+
"step": 328
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 13.16,
|
| 1981 |
+
"learning_rate": 6.840000000000001e-05,
|
| 1982 |
+
"loss": 1.0103,
|
| 1983 |
+
"step": 329
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 13.2,
|
| 1987 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1988 |
+
"loss": 0.9974,
|
| 1989 |
+
"step": 330
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 13.24,
|
| 1993 |
+
"learning_rate": 6.76e-05,
|
| 1994 |
+
"loss": 1.2283,
|
| 1995 |
+
"step": 331
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 13.28,
|
| 1999 |
+
"learning_rate": 6.720000000000001e-05,
|
| 2000 |
+
"loss": 0.7485,
|
| 2001 |
+
"step": 332
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 13.32,
|
| 2005 |
+
"learning_rate": 6.680000000000001e-05,
|
| 2006 |
+
"loss": 0.8072,
|
| 2007 |
+
"step": 333
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 13.36,
|
| 2011 |
+
"learning_rate": 6.64e-05,
|
| 2012 |
+
"loss": 1.1106,
|
| 2013 |
+
"step": 334
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 13.4,
|
| 2017 |
+
"learning_rate": 6.6e-05,
|
| 2018 |
+
"loss": 0.8216,
|
| 2019 |
+
"step": 335
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 13.44,
|
| 2023 |
+
"learning_rate": 6.560000000000001e-05,
|
| 2024 |
+
"loss": 0.959,
|
| 2025 |
+
"step": 336
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 13.48,
|
| 2029 |
+
"learning_rate": 6.52e-05,
|
| 2030 |
+
"loss": 1.0767,
|
| 2031 |
+
"step": 337
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 13.52,
|
| 2035 |
+
"learning_rate": 6.48e-05,
|
| 2036 |
+
"loss": 1.0395,
|
| 2037 |
+
"step": 338
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 13.56,
|
| 2041 |
+
"learning_rate": 6.440000000000001e-05,
|
| 2042 |
+
"loss": 0.9102,
|
| 2043 |
+
"step": 339
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 13.6,
|
| 2047 |
+
"learning_rate": 6.400000000000001e-05,
|
| 2048 |
+
"loss": 0.8875,
|
| 2049 |
+
"step": 340
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 13.64,
|
| 2053 |
+
"learning_rate": 6.36e-05,
|
| 2054 |
+
"loss": 1.0005,
|
| 2055 |
+
"step": 341
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 13.68,
|
| 2059 |
+
"learning_rate": 6.32e-05,
|
| 2060 |
+
"loss": 0.9208,
|
| 2061 |
+
"step": 342
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 13.72,
|
| 2065 |
+
"learning_rate": 6.280000000000001e-05,
|
| 2066 |
+
"loss": 0.8943,
|
| 2067 |
+
"step": 343
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 13.76,
|
| 2071 |
+
"learning_rate": 6.24e-05,
|
| 2072 |
+
"loss": 0.8441,
|
| 2073 |
+
"step": 344
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 13.8,
|
| 2077 |
+
"learning_rate": 6.2e-05,
|
| 2078 |
+
"loss": 0.9953,
|
| 2079 |
+
"step": 345
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 13.84,
|
| 2083 |
+
"learning_rate": 6.16e-05,
|
| 2084 |
+
"loss": 0.796,
|
| 2085 |
+
"step": 346
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 13.88,
|
| 2089 |
+
"learning_rate": 6.12e-05,
|
| 2090 |
+
"loss": 1.0745,
|
| 2091 |
+
"step": 347
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 13.92,
|
| 2095 |
+
"learning_rate": 6.08e-05,
|
| 2096 |
+
"loss": 0.9898,
|
| 2097 |
+
"step": 348
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 13.96,
|
| 2101 |
+
"learning_rate": 6.04e-05,
|
| 2102 |
+
"loss": 0.9559,
|
| 2103 |
+
"step": 349
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 14.0,
|
| 2107 |
+
"learning_rate": 6e-05,
|
| 2108 |
+
"loss": 0.8236,
|
| 2109 |
+
"step": 350
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 14.04,
|
| 2113 |
+
"learning_rate": 5.96e-05,
|
| 2114 |
+
"loss": 0.7318,
|
| 2115 |
+
"step": 351
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 14.08,
|
| 2119 |
+
"learning_rate": 5.92e-05,
|
| 2120 |
+
"loss": 0.8994,
|
| 2121 |
+
"step": 352
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 14.12,
|
| 2125 |
+
"learning_rate": 5.88e-05,
|
| 2126 |
+
"loss": 0.9707,
|
| 2127 |
+
"step": 353
|
| 2128 |
+
},
|
| 2129 |
+
{
|
| 2130 |
+
"epoch": 14.16,
|
| 2131 |
+
"learning_rate": 5.8399999999999997e-05,
|
| 2132 |
+
"loss": 0.806,
|
| 2133 |
+
"step": 354
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 14.2,
|
| 2137 |
+
"learning_rate": 5.8e-05,
|
| 2138 |
+
"loss": 1.0525,
|
| 2139 |
+
"step": 355
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 14.24,
|
| 2143 |
+
"learning_rate": 5.76e-05,
|
| 2144 |
+
"loss": 0.7589,
|
| 2145 |
+
"step": 356
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 14.28,
|
| 2149 |
+
"learning_rate": 5.72e-05,
|
| 2150 |
+
"loss": 0.7703,
|
| 2151 |
+
"step": 357
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 14.32,
|
| 2155 |
+
"learning_rate": 5.68e-05,
|
| 2156 |
+
"loss": 0.9869,
|
| 2157 |
+
"step": 358
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 14.36,
|
| 2161 |
+
"learning_rate": 5.6399999999999995e-05,
|
| 2162 |
+
"loss": 0.9673,
|
| 2163 |
+
"step": 359
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 14.4,
|
| 2167 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 2168 |
+
"loss": 0.9913,
|
| 2169 |
+
"step": 360
|
| 2170 |
+
},
|
| 2171 |
+
{
|
| 2172 |
+
"epoch": 14.44,
|
| 2173 |
+
"learning_rate": 5.560000000000001e-05,
|
| 2174 |
+
"loss": 1.0219,
|
| 2175 |
+
"step": 361
|
| 2176 |
+
},
|
| 2177 |
+
{
|
| 2178 |
+
"epoch": 14.48,
|
| 2179 |
+
"learning_rate": 5.520000000000001e-05,
|
| 2180 |
+
"loss": 0.8672,
|
| 2181 |
+
"step": 362
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 14.52,
|
| 2185 |
+
"learning_rate": 5.4800000000000004e-05,
|
| 2186 |
+
"loss": 1.0464,
|
| 2187 |
+
"step": 363
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 14.56,
|
| 2191 |
+
"learning_rate": 5.440000000000001e-05,
|
| 2192 |
+
"loss": 1.0109,
|
| 2193 |
+
"step": 364
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 14.6,
|
| 2197 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 2198 |
+
"loss": 0.9913,
|
| 2199 |
+
"step": 365
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 14.64,
|
| 2203 |
+
"learning_rate": 5.360000000000001e-05,
|
| 2204 |
+
"loss": 0.779,
|
| 2205 |
+
"step": 366
|
| 2206 |
+
},
|
| 2207 |
+
{
|
| 2208 |
+
"epoch": 14.68,
|
| 2209 |
+
"learning_rate": 5.3200000000000006e-05,
|
| 2210 |
+
"loss": 0.8675,
|
| 2211 |
+
"step": 367
|
| 2212 |
+
},
|
| 2213 |
+
{
|
| 2214 |
+
"epoch": 14.72,
|
| 2215 |
+
"learning_rate": 5.28e-05,
|
| 2216 |
+
"loss": 0.987,
|
| 2217 |
+
"step": 368
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"epoch": 14.76,
|
| 2221 |
+
"learning_rate": 5.2400000000000007e-05,
|
| 2222 |
+
"loss": 0.741,
|
| 2223 |
+
"step": 369
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 14.8,
|
| 2227 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 2228 |
+
"loss": 0.9585,
|
| 2229 |
+
"step": 370
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 14.84,
|
| 2233 |
+
"learning_rate": 5.16e-05,
|
| 2234 |
+
"loss": 0.8977,
|
| 2235 |
+
"step": 371
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 14.88,
|
| 2239 |
+
"learning_rate": 5.1200000000000004e-05,
|
| 2240 |
+
"loss": 0.7836,
|
| 2241 |
+
"step": 372
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"epoch": 14.92,
|
| 2245 |
+
"learning_rate": 5.08e-05,
|
| 2246 |
+
"loss": 0.8461,
|
| 2247 |
+
"step": 373
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"epoch": 14.96,
|
| 2251 |
+
"learning_rate": 5.0400000000000005e-05,
|
| 2252 |
+
"loss": 0.7531,
|
| 2253 |
+
"step": 374
|
| 2254 |
+
},
|
| 2255 |
+
{
|
| 2256 |
+
"epoch": 15.0,
|
| 2257 |
+
"learning_rate": 5e-05,
|
| 2258 |
+
"loss": 0.8906,
|
| 2259 |
+
"step": 375
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 15.04,
|
| 2263 |
+
"learning_rate": 4.96e-05,
|
| 2264 |
+
"loss": 0.8793,
|
| 2265 |
+
"step": 376
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 15.08,
|
| 2269 |
+
"learning_rate": 4.92e-05,
|
| 2270 |
+
"loss": 0.8596,
|
| 2271 |
+
"step": 377
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 15.12,
|
| 2275 |
+
"learning_rate": 4.88e-05,
|
| 2276 |
+
"loss": 0.7947,
|
| 2277 |
+
"step": 378
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 15.16,
|
| 2281 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 2282 |
+
"loss": 0.7976,
|
| 2283 |
+
"step": 379
|
| 2284 |
+
},
|
| 2285 |
+
{
|
| 2286 |
+
"epoch": 15.2,
|
| 2287 |
+
"learning_rate": 4.8e-05,
|
| 2288 |
+
"loss": 0.9037,
|
| 2289 |
+
"step": 380
|
| 2290 |
+
},
|
| 2291 |
+
{
|
| 2292 |
+
"epoch": 15.24,
|
| 2293 |
+
"learning_rate": 4.76e-05,
|
| 2294 |
+
"loss": 0.9863,
|
| 2295 |
+
"step": 381
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"epoch": 15.28,
|
| 2299 |
+
"learning_rate": 4.72e-05,
|
| 2300 |
+
"loss": 0.7735,
|
| 2301 |
+
"step": 382
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 15.32,
|
| 2305 |
+
"learning_rate": 4.6800000000000006e-05,
|
| 2306 |
+
"loss": 0.6723,
|
| 2307 |
+
"step": 383
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 15.36,
|
| 2311 |
+
"learning_rate": 4.64e-05,
|
| 2312 |
+
"loss": 0.7002,
|
| 2313 |
+
"step": 384
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 15.4,
|
| 2317 |
+
"learning_rate": 4.600000000000001e-05,
|
| 2318 |
+
"loss": 0.9479,
|
| 2319 |
+
"step": 385
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 15.44,
|
| 2323 |
+
"learning_rate": 4.5600000000000004e-05,
|
| 2324 |
+
"loss": 0.8159,
|
| 2325 |
+
"step": 386
|
| 2326 |
+
},
|
| 2327 |
+
{
|
| 2328 |
+
"epoch": 15.48,
|
| 2329 |
+
"learning_rate": 4.52e-05,
|
| 2330 |
+
"loss": 0.7954,
|
| 2331 |
+
"step": 387
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 15.52,
|
| 2335 |
+
"learning_rate": 4.4800000000000005e-05,
|
| 2336 |
+
"loss": 0.8016,
|
| 2337 |
+
"step": 388
|
| 2338 |
+
},
|
| 2339 |
+
{
|
| 2340 |
+
"epoch": 15.56,
|
| 2341 |
+
"learning_rate": 4.44e-05,
|
| 2342 |
+
"loss": 0.7995,
|
| 2343 |
+
"step": 389
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 15.6,
|
| 2347 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 2348 |
+
"loss": 0.8243,
|
| 2349 |
+
"step": 390
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 15.64,
|
| 2353 |
+
"learning_rate": 4.36e-05,
|
| 2354 |
+
"loss": 0.8335,
|
| 2355 |
+
"step": 391
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 15.68,
|
| 2359 |
+
"learning_rate": 4.32e-05,
|
| 2360 |
+
"loss": 1.001,
|
| 2361 |
+
"step": 392
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 15.72,
|
| 2365 |
+
"learning_rate": 4.2800000000000004e-05,
|
| 2366 |
+
"loss": 0.9385,
|
| 2367 |
+
"step": 393
|
| 2368 |
+
},
|
| 2369 |
+
{
|
| 2370 |
+
"epoch": 15.76,
|
| 2371 |
+
"learning_rate": 4.24e-05,
|
| 2372 |
+
"loss": 0.8796,
|
| 2373 |
+
"step": 394
|
| 2374 |
+
},
|
| 2375 |
+
{
|
| 2376 |
+
"epoch": 15.8,
|
| 2377 |
+
"learning_rate": 4.2e-05,
|
| 2378 |
+
"loss": 0.98,
|
| 2379 |
+
"step": 395
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 15.84,
|
| 2383 |
+
"learning_rate": 4.16e-05,
|
| 2384 |
+
"loss": 0.7547,
|
| 2385 |
+
"step": 396
|
| 2386 |
+
},
|
| 2387 |
+
{
|
| 2388 |
+
"epoch": 15.88,
|
| 2389 |
+
"learning_rate": 4.12e-05,
|
| 2390 |
+
"loss": 0.7943,
|
| 2391 |
+
"step": 397
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 15.92,
|
| 2395 |
+
"learning_rate": 4.08e-05,
|
| 2396 |
+
"loss": 0.8135,
|
| 2397 |
+
"step": 398
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 15.96,
|
| 2401 |
+
"learning_rate": 4.0400000000000006e-05,
|
| 2402 |
+
"loss": 0.87,
|
| 2403 |
+
"step": 399
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 16.0,
|
| 2407 |
+
"learning_rate": 4e-05,
|
| 2408 |
+
"loss": 0.8818,
|
| 2409 |
+
"step": 400
|
| 2410 |
+
}
|
| 2411 |
+
],
|
| 2412 |
+
"logging_steps": 1,
|
| 2413 |
+
"max_steps": 500,
|
| 2414 |
+
"num_train_epochs": 20,
|
| 2415 |
+
"save_steps": 100,
|
| 2416 |
+
"total_flos": 1.361597073887232e+17,
|
| 2417 |
+
"trial_name": null,
|
| 2418 |
+
"trial_params": null
|
| 2419 |
+
}
|
checkpoint-400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ef74b31950ae6c8955316bed48c343fb06cda0cb6a8a54ca46ca3cb681c8736
|
| 3 |
+
size 4027
|
checkpoint-500/README.md
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: True
|
| 10 |
+
- load_in_4bit: False
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: fp4
|
| 16 |
+
- bnb_4bit_use_double_quant: False
|
| 17 |
+
- bnb_4bit_compute_dtype: float32
|
| 18 |
+
|
| 19 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 20 |
+
- quant_method: bitsandbytes
|
| 21 |
+
- load_in_8bit: True
|
| 22 |
+
- load_in_4bit: False
|
| 23 |
+
- llm_int8_threshold: 6.0
|
| 24 |
+
- llm_int8_skip_modules: None
|
| 25 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 26 |
+
- llm_int8_has_fp16_weight: False
|
| 27 |
+
- bnb_4bit_quant_type: fp4
|
| 28 |
+
- bnb_4bit_use_double_quant: False
|
| 29 |
+
- bnb_4bit_compute_dtype: float32
|
| 30 |
+
### Framework versions
|
| 31 |
+
|
| 32 |
+
- PEFT 0.6.0.dev0
|
| 33 |
+
|
| 34 |
+
- PEFT 0.6.0.dev0
|
checkpoint-500/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "bigscience/bloomz-3b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
+
"lora_dropout": 0.0,
|
| 13 |
+
"modules_to_save": null,
|
| 14 |
+
"peft_type": "LORA",
|
| 15 |
+
"r": 8,
|
| 16 |
+
"rank_pattern": {},
|
| 17 |
+
"revision": null,
|
| 18 |
+
"target_modules": [
|
| 19 |
+
"dense_4h_to_h",
|
| 20 |
+
"dense",
|
| 21 |
+
"dense_h_to_4h",
|
| 22 |
+
"query_key_value"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|