Commit
•
6f3caf8
1
Parent(s):
cbb88ea
Model save
Browse files- README.md +110 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1308 -0
README.md
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Llama-31-8B_task-3_120-samples_config-4
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Llama-31-8B_task-3_120-samples_config-4
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4641
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 1e-05
|
| 41 |
+
- train_batch_size: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- gradient_accumulation_steps: 16
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 150
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 2.4707 | 0.9091 | 5 | 2.5024 |
|
| 57 |
+
| 2.3847 | 2.0 | 11 | 2.4794 |
|
| 58 |
+
| 2.5822 | 2.9091 | 16 | 2.4506 |
|
| 59 |
+
| 2.2635 | 4.0 | 22 | 2.3903 |
|
| 60 |
+
| 2.446 | 4.9091 | 27 | 2.3184 |
|
| 61 |
+
| 2.3172 | 6.0 | 33 | 2.1938 |
|
| 62 |
+
| 2.0582 | 6.9091 | 38 | 2.0560 |
|
| 63 |
+
| 1.9038 | 8.0 | 44 | 1.8385 |
|
| 64 |
+
| 1.7291 | 8.9091 | 49 | 1.6252 |
|
| 65 |
+
| 1.3996 | 10.0 | 55 | 1.3456 |
|
| 66 |
+
| 1.1127 | 10.9091 | 60 | 1.1189 |
|
| 67 |
+
| 0.8648 | 12.0 | 66 | 0.8627 |
|
| 68 |
+
| 0.8247 | 12.9091 | 71 | 0.7228 |
|
| 69 |
+
| 0.5681 | 14.0 | 77 | 0.6453 |
|
| 70 |
+
| 0.4968 | 14.9091 | 82 | 0.6020 |
|
| 71 |
+
| 0.589 | 16.0 | 88 | 0.5632 |
|
| 72 |
+
| 0.3902 | 16.9091 | 93 | 0.5394 |
|
| 73 |
+
| 0.4795 | 18.0 | 99 | 0.5253 |
|
| 74 |
+
| 0.3937 | 18.9091 | 104 | 0.5188 |
|
| 75 |
+
| 0.3482 | 20.0 | 110 | 0.5149 |
|
| 76 |
+
| 0.4633 | 20.9091 | 115 | 0.5076 |
|
| 77 |
+
| 0.4324 | 22.0 | 121 | 0.5073 |
|
| 78 |
+
| 0.5268 | 22.9091 | 126 | 0.5014 |
|
| 79 |
+
| 0.3829 | 24.0 | 132 | 0.4946 |
|
| 80 |
+
| 0.3884 | 24.9091 | 137 | 0.4875 |
|
| 81 |
+
| 0.3955 | 26.0 | 143 | 0.4859 |
|
| 82 |
+
| 0.3296 | 26.9091 | 148 | 0.4827 |
|
| 83 |
+
| 0.364 | 28.0 | 154 | 0.4805 |
|
| 84 |
+
| 0.3218 | 28.9091 | 159 | 0.4765 |
|
| 85 |
+
| 0.2995 | 30.0 | 165 | 0.4727 |
|
| 86 |
+
| 0.3728 | 30.9091 | 170 | 0.4668 |
|
| 87 |
+
| 0.2413 | 32.0 | 176 | 0.4653 |
|
| 88 |
+
| 0.4141 | 32.9091 | 181 | 0.4626 |
|
| 89 |
+
| 0.3693 | 34.0 | 187 | 0.4562 |
|
| 90 |
+
| 0.3666 | 34.9091 | 192 | 0.4526 |
|
| 91 |
+
| 0.3605 | 36.0 | 198 | 0.4506 |
|
| 92 |
+
| 0.2923 | 36.9091 | 203 | 0.4474 |
|
| 93 |
+
| 0.3422 | 38.0 | 209 | 0.4443 |
|
| 94 |
+
| 0.3489 | 38.9091 | 214 | 0.4427 |
|
| 95 |
+
| 0.3737 | 40.0 | 220 | 0.4435 |
|
| 96 |
+
| 0.2804 | 40.9091 | 225 | 0.4437 |
|
| 97 |
+
| 0.3212 | 42.0 | 231 | 0.4469 |
|
| 98 |
+
| 0.2322 | 42.9091 | 236 | 0.4480 |
|
| 99 |
+
| 0.2569 | 44.0 | 242 | 0.4507 |
|
| 100 |
+
| 0.2501 | 44.9091 | 247 | 0.4544 |
|
| 101 |
+
| 0.2247 | 46.0 | 253 | 0.4641 |
|
| 102 |
+
|
| 103 |
+
|
| 104 |
+
### Framework versions
|
| 105 |
+
|
| 106 |
+
- PEFT 0.12.0
|
| 107 |
+
- Transformers 4.44.0
|
| 108 |
+
- Pytorch 2.1.2+cu121
|
| 109 |
+
- Datasets 2.20.0
|
| 110 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 167832240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8b9ce14e1fef5d0695cef2bf8fdabeb76c7f98ec6c1ddbdd632537813e2a8b5
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"total_flos": 3.15960483767124e+17,
|
| 4 |
+
"train_loss": 0.7967259936888699,
|
| 5 |
+
"train_runtime": 11647.481,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.133,
|
| 8 |
+
"train_steps_per_second": 0.064
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 46.0,
|
| 3 |
+
"total_flos": 3.15960483767124e+17,
|
| 4 |
+
"train_loss": 0.7967259936888699,
|
| 5 |
+
"train_runtime": 11647.481,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.133,
|
| 8 |
+
"train_steps_per_second": 0.064
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1308 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.44272685050964355,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-4/checkpoint-214",
|
| 4 |
+
"epoch": 46.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 253,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.18181818181818182,
|
| 13 |
+
"grad_norm": 3.1690733432769775,
|
| 14 |
+
"learning_rate": 1.3333333333333336e-07,
|
| 15 |
+
"loss": 2.8174,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.36363636363636365,
|
| 20 |
+
"grad_norm": 2.785377025604248,
|
| 21 |
+
"learning_rate": 2.666666666666667e-07,
|
| 22 |
+
"loss": 2.5884,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7272727272727273,
|
| 27 |
+
"grad_norm": 2.4831278324127197,
|
| 28 |
+
"learning_rate": 5.333333333333335e-07,
|
| 29 |
+
"loss": 2.4707,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9090909090909091,
|
| 34 |
+
"eval_loss": 2.502408742904663,
|
| 35 |
+
"eval_runtime": 20.9991,
|
| 36 |
+
"eval_samples_per_second": 1.143,
|
| 37 |
+
"eval_steps_per_second": 1.143,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0909090909090908,
|
| 42 |
+
"grad_norm": 2.511209726333618,
|
| 43 |
+
"learning_rate": 8.000000000000001e-07,
|
| 44 |
+
"loss": 2.3921,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.4545454545454546,
|
| 49 |
+
"grad_norm": 3.1637580394744873,
|
| 50 |
+
"learning_rate": 1.066666666666667e-06,
|
| 51 |
+
"loss": 2.5814,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.8181818181818183,
|
| 56 |
+
"grad_norm": 2.612966775894165,
|
| 57 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 58 |
+
"loss": 2.3847,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_loss": 2.479367971420288,
|
| 64 |
+
"eval_runtime": 21.0052,
|
| 65 |
+
"eval_samples_per_second": 1.143,
|
| 66 |
+
"eval_steps_per_second": 1.143,
|
| 67 |
+
"step": 11
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.1818181818181817,
|
| 71 |
+
"grad_norm": 2.905508518218994,
|
| 72 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 73 |
+
"loss": 2.6936,
|
| 74 |
+
"step": 12
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.5454545454545454,
|
| 78 |
+
"grad_norm": 2.6060760021209717,
|
| 79 |
+
"learning_rate": 1.8666666666666669e-06,
|
| 80 |
+
"loss": 2.3757,
|
| 81 |
+
"step": 14
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.909090909090909,
|
| 85 |
+
"grad_norm": 3.210188150405884,
|
| 86 |
+
"learning_rate": 2.133333333333334e-06,
|
| 87 |
+
"loss": 2.5822,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.909090909090909,
|
| 92 |
+
"eval_loss": 2.4506218433380127,
|
| 93 |
+
"eval_runtime": 21.0015,
|
| 94 |
+
"eval_samples_per_second": 1.143,
|
| 95 |
+
"eval_steps_per_second": 1.143,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.2727272727272725,
|
| 100 |
+
"grad_norm": 3.861948013305664,
|
| 101 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 102 |
+
"loss": 2.6532,
|
| 103 |
+
"step": 18
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.6363636363636362,
|
| 107 |
+
"grad_norm": 2.5222151279449463,
|
| 108 |
+
"learning_rate": 2.666666666666667e-06,
|
| 109 |
+
"loss": 2.3743,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"grad_norm": 2.5924019813537598,
|
| 115 |
+
"learning_rate": 2.9333333333333338e-06,
|
| 116 |
+
"loss": 2.2635,
|
| 117 |
+
"step": 22
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 4.0,
|
| 121 |
+
"eval_loss": 2.3902969360351562,
|
| 122 |
+
"eval_runtime": 21.0176,
|
| 123 |
+
"eval_samples_per_second": 1.142,
|
| 124 |
+
"eval_steps_per_second": 1.142,
|
| 125 |
+
"step": 22
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.363636363636363,
|
| 129 |
+
"grad_norm": 2.239548921585083,
|
| 130 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 131 |
+
"loss": 2.4045,
|
| 132 |
+
"step": 24
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.7272727272727275,
|
| 136 |
+
"grad_norm": 2.6937167644500732,
|
| 137 |
+
"learning_rate": 3.4666666666666672e-06,
|
| 138 |
+
"loss": 2.446,
|
| 139 |
+
"step": 26
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 4.909090909090909,
|
| 143 |
+
"eval_loss": 2.3183953762054443,
|
| 144 |
+
"eval_runtime": 21.0027,
|
| 145 |
+
"eval_samples_per_second": 1.143,
|
| 146 |
+
"eval_steps_per_second": 1.143,
|
| 147 |
+
"step": 27
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.090909090909091,
|
| 151 |
+
"grad_norm": 2.26949143409729,
|
| 152 |
+
"learning_rate": 3.7333333333333337e-06,
|
| 153 |
+
"loss": 2.1955,
|
| 154 |
+
"step": 28
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.454545454545454,
|
| 158 |
+
"grad_norm": 3.2933804988861084,
|
| 159 |
+
"learning_rate": 4.000000000000001e-06,
|
| 160 |
+
"loss": 2.3794,
|
| 161 |
+
"step": 30
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.818181818181818,
|
| 165 |
+
"grad_norm": 3.081024169921875,
|
| 166 |
+
"learning_rate": 4.266666666666668e-06,
|
| 167 |
+
"loss": 2.3172,
|
| 168 |
+
"step": 32
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 2.19378399848938,
|
| 173 |
+
"eval_runtime": 20.9996,
|
| 174 |
+
"eval_samples_per_second": 1.143,
|
| 175 |
+
"eval_steps_per_second": 1.143,
|
| 176 |
+
"step": 33
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.181818181818182,
|
| 180 |
+
"grad_norm": 3.226388454437256,
|
| 181 |
+
"learning_rate": 4.533333333333334e-06,
|
| 182 |
+
"loss": 2.2373,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.545454545454545,
|
| 187 |
+
"grad_norm": 1.6646192073822021,
|
| 188 |
+
"learning_rate": 4.800000000000001e-06,
|
| 189 |
+
"loss": 2.0122,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.909090909090909,
|
| 194 |
+
"grad_norm": 2.573935031890869,
|
| 195 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 196 |
+
"loss": 2.0582,
|
| 197 |
+
"step": 38
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.909090909090909,
|
| 201 |
+
"eval_loss": 2.0559873580932617,
|
| 202 |
+
"eval_runtime": 21.0001,
|
| 203 |
+
"eval_samples_per_second": 1.143,
|
| 204 |
+
"eval_steps_per_second": 1.143,
|
| 205 |
+
"step": 38
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.2727272727272725,
|
| 209 |
+
"grad_norm": 2.028982639312744,
|
| 210 |
+
"learning_rate": 5.333333333333334e-06,
|
| 211 |
+
"loss": 1.9739,
|
| 212 |
+
"step": 40
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.636363636363637,
|
| 216 |
+
"grad_norm": 2.90993595123291,
|
| 217 |
+
"learning_rate": 5.600000000000001e-06,
|
| 218 |
+
"loss": 2.0373,
|
| 219 |
+
"step": 42
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"grad_norm": 2.5906777381896973,
|
| 224 |
+
"learning_rate": 5.8666666666666675e-06,
|
| 225 |
+
"loss": 1.9038,
|
| 226 |
+
"step": 44
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 1.8385127782821655,
|
| 231 |
+
"eval_runtime": 20.9958,
|
| 232 |
+
"eval_samples_per_second": 1.143,
|
| 233 |
+
"eval_steps_per_second": 1.143,
|
| 234 |
+
"step": 44
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.363636363636363,
|
| 238 |
+
"grad_norm": 2.548511266708374,
|
| 239 |
+
"learning_rate": 6.133333333333334e-06,
|
| 240 |
+
"loss": 1.7707,
|
| 241 |
+
"step": 46
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.727272727272727,
|
| 245 |
+
"grad_norm": 2.2620716094970703,
|
| 246 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 247 |
+
"loss": 1.7291,
|
| 248 |
+
"step": 48
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.909090909090908,
|
| 252 |
+
"eval_loss": 1.625160574913025,
|
| 253 |
+
"eval_runtime": 20.9978,
|
| 254 |
+
"eval_samples_per_second": 1.143,
|
| 255 |
+
"eval_steps_per_second": 1.143,
|
| 256 |
+
"step": 49
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.090909090909092,
|
| 260 |
+
"grad_norm": 2.2526345252990723,
|
| 261 |
+
"learning_rate": 6.666666666666667e-06,
|
| 262 |
+
"loss": 1.607,
|
| 263 |
+
"step": 50
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.454545454545455,
|
| 267 |
+
"grad_norm": 1.2636234760284424,
|
| 268 |
+
"learning_rate": 6.9333333333333344e-06,
|
| 269 |
+
"loss": 1.5173,
|
| 270 |
+
"step": 52
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.818181818181818,
|
| 274 |
+
"grad_norm": 2.455042600631714,
|
| 275 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 276 |
+
"loss": 1.3996,
|
| 277 |
+
"step": 54
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"eval_loss": 1.345643162727356,
|
| 282 |
+
"eval_runtime": 21.0044,
|
| 283 |
+
"eval_samples_per_second": 1.143,
|
| 284 |
+
"eval_steps_per_second": 1.143,
|
| 285 |
+
"step": 55
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 10.181818181818182,
|
| 289 |
+
"grad_norm": 2.40154767036438,
|
| 290 |
+
"learning_rate": 7.4666666666666675e-06,
|
| 291 |
+
"loss": 1.298,
|
| 292 |
+
"step": 56
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.545454545454545,
|
| 296 |
+
"grad_norm": 2.0422167778015137,
|
| 297 |
+
"learning_rate": 7.733333333333334e-06,
|
| 298 |
+
"loss": 1.2468,
|
| 299 |
+
"step": 58
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.909090909090908,
|
| 303 |
+
"grad_norm": 2.330064058303833,
|
| 304 |
+
"learning_rate": 8.000000000000001e-06,
|
| 305 |
+
"loss": 1.1127,
|
| 306 |
+
"step": 60
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.909090909090908,
|
| 310 |
+
"eval_loss": 1.1189473867416382,
|
| 311 |
+
"eval_runtime": 21.0042,
|
| 312 |
+
"eval_samples_per_second": 1.143,
|
| 313 |
+
"eval_steps_per_second": 1.143,
|
| 314 |
+
"step": 60
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.272727272727273,
|
| 318 |
+
"grad_norm": 1.6237120628356934,
|
| 319 |
+
"learning_rate": 8.266666666666667e-06,
|
| 320 |
+
"loss": 1.0346,
|
| 321 |
+
"step": 62
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.636363636363637,
|
| 325 |
+
"grad_norm": 1.9251773357391357,
|
| 326 |
+
"learning_rate": 8.533333333333335e-06,
|
| 327 |
+
"loss": 0.9213,
|
| 328 |
+
"step": 64
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 12.0,
|
| 332 |
+
"grad_norm": 1.3815380334854126,
|
| 333 |
+
"learning_rate": 8.8e-06,
|
| 334 |
+
"loss": 0.8648,
|
| 335 |
+
"step": 66
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.8626702427864075,
|
| 340 |
+
"eval_runtime": 21.003,
|
| 341 |
+
"eval_samples_per_second": 1.143,
|
| 342 |
+
"eval_steps_per_second": 1.143,
|
| 343 |
+
"step": 66
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.363636363636363,
|
| 347 |
+
"grad_norm": 1.8721394538879395,
|
| 348 |
+
"learning_rate": 9.066666666666667e-06,
|
| 349 |
+
"loss": 0.6295,
|
| 350 |
+
"step": 68
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.727272727272727,
|
| 354 |
+
"grad_norm": 1.1077913045883179,
|
| 355 |
+
"learning_rate": 9.333333333333334e-06,
|
| 356 |
+
"loss": 0.8247,
|
| 357 |
+
"step": 70
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.909090909090908,
|
| 361 |
+
"eval_loss": 0.7228104472160339,
|
| 362 |
+
"eval_runtime": 20.9976,
|
| 363 |
+
"eval_samples_per_second": 1.143,
|
| 364 |
+
"eval_steps_per_second": 1.143,
|
| 365 |
+
"step": 71
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.090909090909092,
|
| 369 |
+
"grad_norm": 0.9280027151107788,
|
| 370 |
+
"learning_rate": 9.600000000000001e-06,
|
| 371 |
+
"loss": 0.6643,
|
| 372 |
+
"step": 72
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.454545454545455,
|
| 376 |
+
"grad_norm": 0.8919649720191956,
|
| 377 |
+
"learning_rate": 9.866666666666668e-06,
|
| 378 |
+
"loss": 0.5425,
|
| 379 |
+
"step": 74
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.818181818181818,
|
| 383 |
+
"grad_norm": 0.8291063904762268,
|
| 384 |
+
"learning_rate": 9.999945845889795e-06,
|
| 385 |
+
"loss": 0.5681,
|
| 386 |
+
"step": 76
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 14.0,
|
| 390 |
+
"eval_loss": 0.6452742218971252,
|
| 391 |
+
"eval_runtime": 21.0019,
|
| 392 |
+
"eval_samples_per_second": 1.143,
|
| 393 |
+
"eval_steps_per_second": 1.143,
|
| 394 |
+
"step": 77
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 14.181818181818182,
|
| 398 |
+
"grad_norm": 0.8017318844795227,
|
| 399 |
+
"learning_rate": 9.999512620046523e-06,
|
| 400 |
+
"loss": 0.6102,
|
| 401 |
+
"step": 78
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 14.545454545454545,
|
| 405 |
+
"grad_norm": 0.48972490429878235,
|
| 406 |
+
"learning_rate": 9.99864620589731e-06,
|
| 407 |
+
"loss": 0.5325,
|
| 408 |
+
"step": 80
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.909090909090908,
|
| 412 |
+
"grad_norm": 0.6467877626419067,
|
| 413 |
+
"learning_rate": 9.99734667851357e-06,
|
| 414 |
+
"loss": 0.4968,
|
| 415 |
+
"step": 82
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.909090909090908,
|
| 419 |
+
"eval_loss": 0.6019716858863831,
|
| 420 |
+
"eval_runtime": 21.0041,
|
| 421 |
+
"eval_samples_per_second": 1.143,
|
| 422 |
+
"eval_steps_per_second": 1.143,
|
| 423 |
+
"step": 82
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 15.272727272727273,
|
| 427 |
+
"grad_norm": 0.4685284495353699,
|
| 428 |
+
"learning_rate": 9.995614150494293e-06,
|
| 429 |
+
"loss": 0.4665,
|
| 430 |
+
"step": 84
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.636363636363637,
|
| 434 |
+
"grad_norm": 0.6542117595672607,
|
| 435 |
+
"learning_rate": 9.993448771956285e-06,
|
| 436 |
+
"loss": 0.4592,
|
| 437 |
+
"step": 86
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.0,
|
| 441 |
+
"grad_norm": 0.7055411338806152,
|
| 442 |
+
"learning_rate": 9.99085073052117e-06,
|
| 443 |
+
"loss": 0.589,
|
| 444 |
+
"step": 88
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 16.0,
|
| 448 |
+
"eval_loss": 0.5632085800170898,
|
| 449 |
+
"eval_runtime": 21.0058,
|
| 450 |
+
"eval_samples_per_second": 1.143,
|
| 451 |
+
"eval_steps_per_second": 1.143,
|
| 452 |
+
"step": 88
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 16.363636363636363,
|
| 456 |
+
"grad_norm": 0.5362458825111389,
|
| 457 |
+
"learning_rate": 9.987820251299121e-06,
|
| 458 |
+
"loss": 0.5109,
|
| 459 |
+
"step": 90
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.727272727272727,
|
| 463 |
+
"grad_norm": 0.3406112492084503,
|
| 464 |
+
"learning_rate": 9.984357596869369e-06,
|
| 465 |
+
"loss": 0.3902,
|
| 466 |
+
"step": 92
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.90909090909091,
|
| 470 |
+
"eval_loss": 0.5393884778022766,
|
| 471 |
+
"eval_runtime": 21.0005,
|
| 472 |
+
"eval_samples_per_second": 1.143,
|
| 473 |
+
"eval_steps_per_second": 1.143,
|
| 474 |
+
"step": 93
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 17.09090909090909,
|
| 478 |
+
"grad_norm": 0.5570920705795288,
|
| 479 |
+
"learning_rate": 9.980463067257437e-06,
|
| 480 |
+
"loss": 0.5325,
|
| 481 |
+
"step": 94
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 17.454545454545453,
|
| 485 |
+
"grad_norm": 0.40238556265830994,
|
| 486 |
+
"learning_rate": 9.976136999909156e-06,
|
| 487 |
+
"loss": 0.4039,
|
| 488 |
+
"step": 96
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.818181818181817,
|
| 492 |
+
"grad_norm": 0.6511011719703674,
|
| 493 |
+
"learning_rate": 9.971379769661422e-06,
|
| 494 |
+
"loss": 0.4795,
|
| 495 |
+
"step": 98
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 18.0,
|
| 499 |
+
"eval_loss": 0.5253307819366455,
|
| 500 |
+
"eval_runtime": 21.0047,
|
| 501 |
+
"eval_samples_per_second": 1.143,
|
| 502 |
+
"eval_steps_per_second": 1.143,
|
| 503 |
+
"step": 99
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 18.181818181818183,
|
| 507 |
+
"grad_norm": 0.4080711305141449,
|
| 508 |
+
"learning_rate": 9.966191788709716e-06,
|
| 509 |
+
"loss": 0.4767,
|
| 510 |
+
"step": 100
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 18.545454545454547,
|
| 514 |
+
"grad_norm": 0.35794946551322937,
|
| 515 |
+
"learning_rate": 9.960573506572391e-06,
|
| 516 |
+
"loss": 0.4024,
|
| 517 |
+
"step": 102
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 18.90909090909091,
|
| 521 |
+
"grad_norm": 0.2140028029680252,
|
| 522 |
+
"learning_rate": 9.95452541005172e-06,
|
| 523 |
+
"loss": 0.3937,
|
| 524 |
+
"step": 104
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.90909090909091,
|
| 528 |
+
"eval_loss": 0.5187709927558899,
|
| 529 |
+
"eval_runtime": 20.9985,
|
| 530 |
+
"eval_samples_per_second": 1.143,
|
| 531 |
+
"eval_steps_per_second": 1.143,
|
| 532 |
+
"step": 104
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 19.272727272727273,
|
| 536 |
+
"grad_norm": 0.26322704553604126,
|
| 537 |
+
"learning_rate": 9.948048023191728e-06,
|
| 538 |
+
"loss": 0.3845,
|
| 539 |
+
"step": 106
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.636363636363637,
|
| 543 |
+
"grad_norm": 0.29932188987731934,
|
| 544 |
+
"learning_rate": 9.941141907232766e-06,
|
| 545 |
+
"loss": 0.5589,
|
| 546 |
+
"step": 108
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 20.0,
|
| 550 |
+
"grad_norm": 0.3077118396759033,
|
| 551 |
+
"learning_rate": 9.933807660562898e-06,
|
| 552 |
+
"loss": 0.3482,
|
| 553 |
+
"step": 110
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 20.0,
|
| 557 |
+
"eval_loss": 0.5148636698722839,
|
| 558 |
+
"eval_runtime": 21.0007,
|
| 559 |
+
"eval_samples_per_second": 1.143,
|
| 560 |
+
"eval_steps_per_second": 1.143,
|
| 561 |
+
"step": 110
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 20.363636363636363,
|
| 565 |
+
"grad_norm": 0.2408452033996582,
|
| 566 |
+
"learning_rate": 9.926045918666045e-06,
|
| 567 |
+
"loss": 0.3168,
|
| 568 |
+
"step": 112
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 20.727272727272727,
|
| 572 |
+
"grad_norm": 0.24809201061725616,
|
| 573 |
+
"learning_rate": 9.91785735406693e-06,
|
| 574 |
+
"loss": 0.4633,
|
| 575 |
+
"step": 114
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 20.90909090909091,
|
| 579 |
+
"eval_loss": 0.5076439380645752,
|
| 580 |
+
"eval_runtime": 21.008,
|
| 581 |
+
"eval_samples_per_second": 1.142,
|
| 582 |
+
"eval_steps_per_second": 1.142,
|
| 583 |
+
"step": 115
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 21.09090909090909,
|
| 587 |
+
"grad_norm": 0.4201977550983429,
|
| 588 |
+
"learning_rate": 9.909242676272797e-06,
|
| 589 |
+
"loss": 0.5529,
|
| 590 |
+
"step": 116
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 21.454545454545453,
|
| 594 |
+
"grad_norm": 0.30744773149490356,
|
| 595 |
+
"learning_rate": 9.90020263171194e-06,
|
| 596 |
+
"loss": 0.3469,
|
| 597 |
+
"step": 118
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 21.818181818181817,
|
| 601 |
+
"grad_norm": 0.2413516491651535,
|
| 602 |
+
"learning_rate": 9.890738003669029e-06,
|
| 603 |
+
"loss": 0.4324,
|
| 604 |
+
"step": 120
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 22.0,
|
| 608 |
+
"eval_loss": 0.507264256477356,
|
| 609 |
+
"eval_runtime": 20.9993,
|
| 610 |
+
"eval_samples_per_second": 1.143,
|
| 611 |
+
"eval_steps_per_second": 1.143,
|
| 612 |
+
"step": 121
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 22.181818181818183,
|
| 616 |
+
"grad_norm": 0.291746586561203,
|
| 617 |
+
"learning_rate": 9.880849612217238e-06,
|
| 618 |
+
"loss": 0.325,
|
| 619 |
+
"step": 122
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 22.545454545454547,
|
| 623 |
+
"grad_norm": 0.22690318524837494,
|
| 624 |
+
"learning_rate": 9.870538314147194e-06,
|
| 625 |
+
"loss": 0.3262,
|
| 626 |
+
"step": 124
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.90909090909091,
|
| 630 |
+
"grad_norm": 0.28505831956863403,
|
| 631 |
+
"learning_rate": 9.859805002892733e-06,
|
| 632 |
+
"loss": 0.5268,
|
| 633 |
+
"step": 126
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 22.90909090909091,
|
| 637 |
+
"eval_loss": 0.5014194846153259,
|
| 638 |
+
"eval_runtime": 21.0084,
|
| 639 |
+
"eval_samples_per_second": 1.142,
|
| 640 |
+
"eval_steps_per_second": 1.142,
|
| 641 |
+
"step": 126
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 23.272727272727273,
|
| 645 |
+
"grad_norm": 0.33074063062667847,
|
| 646 |
+
"learning_rate": 9.84865060845349e-06,
|
| 647 |
+
"loss": 0.4705,
|
| 648 |
+
"step": 128
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 23.636363636363637,
|
| 652 |
+
"grad_norm": 0.24479380249977112,
|
| 653 |
+
"learning_rate": 9.83707609731432e-06,
|
| 654 |
+
"loss": 0.3475,
|
| 655 |
+
"step": 130
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 24.0,
|
| 659 |
+
"grad_norm": 0.2692837715148926,
|
| 660 |
+
"learning_rate": 9.825082472361558e-06,
|
| 661 |
+
"loss": 0.3829,
|
| 662 |
+
"step": 132
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 24.0,
|
| 666 |
+
"eval_loss": 0.49462565779685974,
|
| 667 |
+
"eval_runtime": 21.0006,
|
| 668 |
+
"eval_samples_per_second": 1.143,
|
| 669 |
+
"eval_steps_per_second": 1.143,
|
| 670 |
+
"step": 132
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 24.363636363636363,
|
| 674 |
+
"grad_norm": 0.2481246441602707,
|
| 675 |
+
"learning_rate": 9.812670772796113e-06,
|
| 676 |
+
"loss": 0.3245,
|
| 677 |
+
"step": 134
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 24.727272727272727,
|
| 681 |
+
"grad_norm": 0.2520200312137604,
|
| 682 |
+
"learning_rate": 9.799842074043438e-06,
|
| 683 |
+
"loss": 0.3884,
|
| 684 |
+
"step": 136
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 24.90909090909091,
|
| 688 |
+
"eval_loss": 0.4874545633792877,
|
| 689 |
+
"eval_runtime": 21.0022,
|
| 690 |
+
"eval_samples_per_second": 1.143,
|
| 691 |
+
"eval_steps_per_second": 1.143,
|
| 692 |
+
"step": 137
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 25.09090909090909,
|
| 696 |
+
"grad_norm": 0.34048882126808167,
|
| 697 |
+
"learning_rate": 9.786597487660336e-06,
|
| 698 |
+
"loss": 0.5025,
|
| 699 |
+
"step": 138
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 25.454545454545453,
|
| 703 |
+
"grad_norm": 0.2133476287126541,
|
| 704 |
+
"learning_rate": 9.77293816123866e-06,
|
| 705 |
+
"loss": 0.3012,
|
| 706 |
+
"step": 140
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 25.818181818181817,
|
| 710 |
+
"grad_norm": 0.3172047734260559,
|
| 711 |
+
"learning_rate": 9.75886527830587e-06,
|
| 712 |
+
"loss": 0.3955,
|
| 713 |
+
"step": 142
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 26.0,
|
| 717 |
+
"eval_loss": 0.4859234094619751,
|
| 718 |
+
"eval_runtime": 21.0019,
|
| 719 |
+
"eval_samples_per_second": 1.143,
|
| 720 |
+
"eval_steps_per_second": 1.143,
|
| 721 |
+
"step": 143
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 26.181818181818183,
|
| 725 |
+
"grad_norm": 0.299968957901001,
|
| 726 |
+
"learning_rate": 9.744380058222483e-06,
|
| 727 |
+
"loss": 0.471,
|
| 728 |
+
"step": 144
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 26.545454545454547,
|
| 732 |
+
"grad_norm": 0.22688376903533936,
|
| 733 |
+
"learning_rate": 9.729483756076436e-06,
|
| 734 |
+
"loss": 0.361,
|
| 735 |
+
"step": 146
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 26.90909090909091,
|
| 739 |
+
"grad_norm": 0.20205911993980408,
|
| 740 |
+
"learning_rate": 9.714177662574316e-06,
|
| 741 |
+
"loss": 0.3296,
|
| 742 |
+
"step": 148
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 26.90909090909091,
|
| 746 |
+
"eval_loss": 0.482715368270874,
|
| 747 |
+
"eval_runtime": 21.0014,
|
| 748 |
+
"eval_samples_per_second": 1.143,
|
| 749 |
+
"eval_steps_per_second": 1.143,
|
| 750 |
+
"step": 148
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 27.272727272727273,
|
| 754 |
+
"grad_norm": 0.2501104772090912,
|
| 755 |
+
"learning_rate": 9.698463103929542e-06,
|
| 756 |
+
"loss": 0.3371,
|
| 757 |
+
"step": 150
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 27.636363636363637,
|
| 761 |
+
"grad_norm": 0.2598528265953064,
|
| 762 |
+
"learning_rate": 9.682341441747446e-06,
|
| 763 |
+
"loss": 0.439,
|
| 764 |
+
"step": 152
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 28.0,
|
| 768 |
+
"grad_norm": 0.25387951731681824,
|
| 769 |
+
"learning_rate": 9.665814072907293e-06,
|
| 770 |
+
"loss": 0.364,
|
| 771 |
+
"step": 154
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 28.0,
|
| 775 |
+
"eval_loss": 0.4805121123790741,
|
| 776 |
+
"eval_runtime": 21.002,
|
| 777 |
+
"eval_samples_per_second": 1.143,
|
| 778 |
+
"eval_steps_per_second": 1.143,
|
| 779 |
+
"step": 154
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 28.363636363636363,
|
| 783 |
+
"grad_norm": 0.2787465751171112,
|
| 784 |
+
"learning_rate": 9.648882429441258e-06,
|
| 785 |
+
"loss": 0.4797,
|
| 786 |
+
"step": 156
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 28.727272727272727,
|
| 790 |
+
"grad_norm": 0.25384321808815,
|
| 791 |
+
"learning_rate": 9.63154797841033e-06,
|
| 792 |
+
"loss": 0.3218,
|
| 793 |
+
"step": 158
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 28.90909090909091,
|
| 797 |
+
"eval_loss": 0.4764772951602936,
|
| 798 |
+
"eval_runtime": 21.0046,
|
| 799 |
+
"eval_samples_per_second": 1.143,
|
| 800 |
+
"eval_steps_per_second": 1.143,
|
| 801 |
+
"step": 159
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 29.09090909090909,
|
| 805 |
+
"grad_norm": 0.2499808818101883,
|
| 806 |
+
"learning_rate": 9.613812221777212e-06,
|
| 807 |
+
"loss": 0.302,
|
| 808 |
+
"step": 160
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 29.454545454545453,
|
| 812 |
+
"grad_norm": 0.24313043057918549,
|
| 813 |
+
"learning_rate": 9.595676696276173e-06,
|
| 814 |
+
"loss": 0.4349,
|
| 815 |
+
"step": 162
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 29.818181818181817,
|
| 819 |
+
"grad_norm": 0.26249364018440247,
|
| 820 |
+
"learning_rate": 9.577142973279896e-06,
|
| 821 |
+
"loss": 0.2995,
|
| 822 |
+
"step": 164
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 30.0,
|
| 826 |
+
"eval_loss": 0.4727340638637543,
|
| 827 |
+
"eval_runtime": 21.0056,
|
| 828 |
+
"eval_samples_per_second": 1.143,
|
| 829 |
+
"eval_steps_per_second": 1.143,
|
| 830 |
+
"step": 165
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 30.181818181818183,
|
| 834 |
+
"grad_norm": 0.3164134621620178,
|
| 835 |
+
"learning_rate": 9.55821265866333e-06,
|
| 836 |
+
"loss": 0.3373,
|
| 837 |
+
"step": 166
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 30.545454545454547,
|
| 841 |
+
"grad_norm": 0.23215709626674652,
|
| 842 |
+
"learning_rate": 9.538887392664544e-06,
|
| 843 |
+
"loss": 0.3434,
|
| 844 |
+
"step": 168
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 30.90909090909091,
|
| 848 |
+
"grad_norm": 0.24047435820102692,
|
| 849 |
+
"learning_rate": 9.519168849742603e-06,
|
| 850 |
+
"loss": 0.3728,
|
| 851 |
+
"step": 170
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 30.90909090909091,
|
| 855 |
+
"eval_loss": 0.4668484032154083,
|
| 856 |
+
"eval_runtime": 21.0083,
|
| 857 |
+
"eval_samples_per_second": 1.142,
|
| 858 |
+
"eval_steps_per_second": 1.142,
|
| 859 |
+
"step": 170
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 31.272727272727273,
|
| 863 |
+
"grad_norm": 0.38881421089172363,
|
| 864 |
+
"learning_rate": 9.499058738432492e-06,
|
| 865 |
+
"loss": 0.4329,
|
| 866 |
+
"step": 172
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 31.636363636363637,
|
| 870 |
+
"grad_norm": 0.38587087392807007,
|
| 871 |
+
"learning_rate": 9.478558801197065e-06,
|
| 872 |
+
"loss": 0.3596,
|
| 873 |
+
"step": 174
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 32.0,
|
| 877 |
+
"grad_norm": 0.2352934628725052,
|
| 878 |
+
"learning_rate": 9.457670814276083e-06,
|
| 879 |
+
"loss": 0.2413,
|
| 880 |
+
"step": 176
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 32.0,
|
| 884 |
+
"eval_loss": 0.46528127789497375,
|
| 885 |
+
"eval_runtime": 20.9994,
|
| 886 |
+
"eval_samples_per_second": 1.143,
|
| 887 |
+
"eval_steps_per_second": 1.143,
|
| 888 |
+
"step": 176
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 32.36363636363637,
|
| 892 |
+
"grad_norm": 0.32384243607521057,
|
| 893 |
+
"learning_rate": 9.436396587532297e-06,
|
| 894 |
+
"loss": 0.3098,
|
| 895 |
+
"step": 178
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 32.72727272727273,
|
| 899 |
+
"grad_norm": 0.3551454544067383,
|
| 900 |
+
"learning_rate": 9.414737964294636e-06,
|
| 901 |
+
"loss": 0.4141,
|
| 902 |
+
"step": 180
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 32.90909090909091,
|
| 906 |
+
"eval_loss": 0.46264728903770447,
|
| 907 |
+
"eval_runtime": 21.0065,
|
| 908 |
+
"eval_samples_per_second": 1.143,
|
| 909 |
+
"eval_steps_per_second": 1.143,
|
| 910 |
+
"step": 181
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 33.09090909090909,
|
| 914 |
+
"grad_norm": 0.365192174911499,
|
| 915 |
+
"learning_rate": 9.392696821198488e-06,
|
| 916 |
+
"loss": 0.2879,
|
| 917 |
+
"step": 182
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 33.45454545454545,
|
| 921 |
+
"grad_norm": 0.38976219296455383,
|
| 922 |
+
"learning_rate": 9.370275068023097e-06,
|
| 923 |
+
"loss": 0.3467,
|
| 924 |
+
"step": 184
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 33.81818181818182,
|
| 928 |
+
"grad_norm": 0.3077632486820221,
|
| 929 |
+
"learning_rate": 9.347474647526095e-06,
|
| 930 |
+
"loss": 0.3693,
|
| 931 |
+
"step": 186
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 34.0,
|
| 935 |
+
"eval_loss": 0.4562326967716217,
|
| 936 |
+
"eval_runtime": 21.0019,
|
| 937 |
+
"eval_samples_per_second": 1.143,
|
| 938 |
+
"eval_steps_per_second": 1.143,
|
| 939 |
+
"step": 187
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 34.18181818181818,
|
| 943 |
+
"grad_norm": 0.209530308842659,
|
| 944 |
+
"learning_rate": 9.324297535275156e-06,
|
| 945 |
+
"loss": 0.1788,
|
| 946 |
+
"step": 188
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 34.54545454545455,
|
| 950 |
+
"grad_norm": 0.36394768953323364,
|
| 951 |
+
"learning_rate": 9.30074573947683e-06,
|
| 952 |
+
"loss": 0.3143,
|
| 953 |
+
"step": 190
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 34.90909090909091,
|
| 957 |
+
"grad_norm": 0.41356202960014343,
|
| 958 |
+
"learning_rate": 9.276821300802535e-06,
|
| 959 |
+
"loss": 0.3666,
|
| 960 |
+
"step": 192
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 34.90909090909091,
|
| 964 |
+
"eval_loss": 0.45260027050971985,
|
| 965 |
+
"eval_runtime": 21.0113,
|
| 966 |
+
"eval_samples_per_second": 1.142,
|
| 967 |
+
"eval_steps_per_second": 1.142,
|
| 968 |
+
"step": 192
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 35.27272727272727,
|
| 972 |
+
"grad_norm": 0.42078325152397156,
|
| 973 |
+
"learning_rate": 9.25252629221175e-06,
|
| 974 |
+
"loss": 0.468,
|
| 975 |
+
"step": 194
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 35.63636363636363,
|
| 979 |
+
"grad_norm": 0.30819612741470337,
|
| 980 |
+
"learning_rate": 9.227862818772392e-06,
|
| 981 |
+
"loss": 0.1892,
|
| 982 |
+
"step": 196
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 36.0,
|
| 986 |
+
"grad_norm": 0.3855637013912201,
|
| 987 |
+
"learning_rate": 9.202833017478421e-06,
|
| 988 |
+
"loss": 0.3605,
|
| 989 |
+
"step": 198
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 36.0,
|
| 993 |
+
"eval_loss": 0.4506441354751587,
|
| 994 |
+
"eval_runtime": 20.9992,
|
| 995 |
+
"eval_samples_per_second": 1.143,
|
| 996 |
+
"eval_steps_per_second": 1.143,
|
| 997 |
+
"step": 198
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 36.36363636363637,
|
| 1001 |
+
"grad_norm": 0.39254796504974365,
|
| 1002 |
+
"learning_rate": 9.177439057064684e-06,
|
| 1003 |
+
"loss": 0.3432,
|
| 1004 |
+
"step": 200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 36.72727272727273,
|
| 1008 |
+
"grad_norm": 0.3540774881839752,
|
| 1009 |
+
"learning_rate": 9.151683137818989e-06,
|
| 1010 |
+
"loss": 0.2923,
|
| 1011 |
+
"step": 202
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 36.90909090909091,
|
| 1015 |
+
"eval_loss": 0.44742465019226074,
|
| 1016 |
+
"eval_runtime": 21.0039,
|
| 1017 |
+
"eval_samples_per_second": 1.143,
|
| 1018 |
+
"eval_steps_per_second": 1.143,
|
| 1019 |
+
"step": 203
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 37.09090909090909,
|
| 1023 |
+
"grad_norm": 0.4445912539958954,
|
| 1024 |
+
"learning_rate": 9.125567491391476e-06,
|
| 1025 |
+
"loss": 0.3135,
|
| 1026 |
+
"step": 204
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 37.45454545454545,
|
| 1030 |
+
"grad_norm": 0.39482685923576355,
|
| 1031 |
+
"learning_rate": 9.099094380601244e-06,
|
| 1032 |
+
"loss": 0.2407,
|
| 1033 |
+
"step": 206
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 37.81818181818182,
|
| 1037 |
+
"grad_norm": 0.3445771634578705,
|
| 1038 |
+
"learning_rate": 9.072266099240286e-06,
|
| 1039 |
+
"loss": 0.3422,
|
| 1040 |
+
"step": 208
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 38.0,
|
| 1044 |
+
"eval_loss": 0.4443068504333496,
|
| 1045 |
+
"eval_runtime": 21.0039,
|
| 1046 |
+
"eval_samples_per_second": 1.143,
|
| 1047 |
+
"eval_steps_per_second": 1.143,
|
| 1048 |
+
"step": 209
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 38.18181818181818,
|
| 1052 |
+
"grad_norm": 0.42600157856941223,
|
| 1053 |
+
"learning_rate": 9.045084971874738e-06,
|
| 1054 |
+
"loss": 0.2673,
|
| 1055 |
+
"step": 210
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 38.54545454545455,
|
| 1059 |
+
"grad_norm": 0.4967671036720276,
|
| 1060 |
+
"learning_rate": 9.017553353643479e-06,
|
| 1061 |
+
"loss": 0.3177,
|
| 1062 |
+
"step": 212
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 38.90909090909091,
|
| 1066 |
+
"grad_norm": 0.39350730180740356,
|
| 1067 |
+
"learning_rate": 8.989673630054044e-06,
|
| 1068 |
+
"loss": 0.3489,
|
| 1069 |
+
"step": 214
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 38.90909090909091,
|
| 1073 |
+
"eval_loss": 0.44272685050964355,
|
| 1074 |
+
"eval_runtime": 21.0007,
|
| 1075 |
+
"eval_samples_per_second": 1.143,
|
| 1076 |
+
"eval_steps_per_second": 1.143,
|
| 1077 |
+
"step": 214
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 39.27272727272727,
|
| 1081 |
+
"grad_norm": 0.27645283937454224,
|
| 1082 |
+
"learning_rate": 8.961448216775955e-06,
|
| 1083 |
+
"loss": 0.161,
|
| 1084 |
+
"step": 216
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 39.63636363636363,
|
| 1088 |
+
"grad_norm": 0.615763783454895,
|
| 1089 |
+
"learning_rate": 8.932879559431392e-06,
|
| 1090 |
+
"loss": 0.2562,
|
| 1091 |
+
"step": 218
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 40.0,
|
| 1095 |
+
"grad_norm": 0.491910845041275,
|
| 1096 |
+
"learning_rate": 8.903970133383297e-06,
|
| 1097 |
+
"loss": 0.3737,
|
| 1098 |
+
"step": 220
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 40.0,
|
| 1102 |
+
"eval_loss": 0.443479984998703,
|
| 1103 |
+
"eval_runtime": 21.0049,
|
| 1104 |
+
"eval_samples_per_second": 1.143,
|
| 1105 |
+
"eval_steps_per_second": 1.143,
|
| 1106 |
+
"step": 220
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 40.36363636363637,
|
| 1110 |
+
"grad_norm": 0.47909313440322876,
|
| 1111 |
+
"learning_rate": 8.874722443520898e-06,
|
| 1112 |
+
"loss": 0.2508,
|
| 1113 |
+
"step": 222
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 40.72727272727273,
|
| 1117 |
+
"grad_norm": 0.5516977906227112,
|
| 1118 |
+
"learning_rate": 8.845139024042664e-06,
|
| 1119 |
+
"loss": 0.2804,
|
| 1120 |
+
"step": 224
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 40.90909090909091,
|
| 1124 |
+
"eval_loss": 0.443651407957077,
|
| 1125 |
+
"eval_runtime": 21.007,
|
| 1126 |
+
"eval_samples_per_second": 1.142,
|
| 1127 |
+
"eval_steps_per_second": 1.142,
|
| 1128 |
+
"step": 225
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 41.09090909090909,
|
| 1132 |
+
"grad_norm": 0.5438214540481567,
|
| 1133 |
+
"learning_rate": 8.815222438236726e-06,
|
| 1134 |
+
"loss": 0.3221,
|
| 1135 |
+
"step": 226
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 41.45454545454545,
|
| 1139 |
+
"grad_norm": 0.49092897772789,
|
| 1140 |
+
"learning_rate": 8.784975278258783e-06,
|
| 1141 |
+
"loss": 0.2601,
|
| 1142 |
+
"step": 228
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 41.81818181818182,
|
| 1146 |
+
"grad_norm": 0.5809181332588196,
|
| 1147 |
+
"learning_rate": 8.754400164907496e-06,
|
| 1148 |
+
"loss": 0.3212,
|
| 1149 |
+
"step": 230
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 42.0,
|
| 1153 |
+
"eval_loss": 0.44689202308654785,
|
| 1154 |
+
"eval_runtime": 21.001,
|
| 1155 |
+
"eval_samples_per_second": 1.143,
|
| 1156 |
+
"eval_steps_per_second": 1.143,
|
| 1157 |
+
"step": 231
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 42.18181818181818,
|
| 1161 |
+
"grad_norm": 0.5147470831871033,
|
| 1162 |
+
"learning_rate": 8.723499747397415e-06,
|
| 1163 |
+
"loss": 0.1686,
|
| 1164 |
+
"step": 232
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 42.54545454545455,
|
| 1168 |
+
"grad_norm": 0.7372314929962158,
|
| 1169 |
+
"learning_rate": 8.692276703129421e-06,
|
| 1170 |
+
"loss": 0.2733,
|
| 1171 |
+
"step": 234
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 42.90909090909091,
|
| 1175 |
+
"grad_norm": 0.4814988374710083,
|
| 1176 |
+
"learning_rate": 8.660733737458751e-06,
|
| 1177 |
+
"loss": 0.2322,
|
| 1178 |
+
"step": 236
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 42.90909090909091,
|
| 1182 |
+
"eval_loss": 0.4480229318141937,
|
| 1183 |
+
"eval_runtime": 21.0124,
|
| 1184 |
+
"eval_samples_per_second": 1.142,
|
| 1185 |
+
"eval_steps_per_second": 1.142,
|
| 1186 |
+
"step": 236
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 43.27272727272727,
|
| 1190 |
+
"grad_norm": 0.7270358800888062,
|
| 1191 |
+
"learning_rate": 8.628873583460593e-06,
|
| 1192 |
+
"loss": 0.2312,
|
| 1193 |
+
"step": 238
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 43.63636363636363,
|
| 1197 |
+
"grad_norm": 0.6802980303764343,
|
| 1198 |
+
"learning_rate": 8.596699001693257e-06,
|
| 1199 |
+
"loss": 0.2551,
|
| 1200 |
+
"step": 240
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 44.0,
|
| 1204 |
+
"grad_norm": 0.6301150321960449,
|
| 1205 |
+
"learning_rate": 8.564212779959003e-06,
|
| 1206 |
+
"loss": 0.2569,
|
| 1207 |
+
"step": 242
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 44.0,
|
| 1211 |
+
"eval_loss": 0.4507286548614502,
|
| 1212 |
+
"eval_runtime": 21.0008,
|
| 1213 |
+
"eval_samples_per_second": 1.143,
|
| 1214 |
+
"eval_steps_per_second": 1.143,
|
| 1215 |
+
"step": 242
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 44.36363636363637,
|
| 1219 |
+
"grad_norm": 0.6666437983512878,
|
| 1220 |
+
"learning_rate": 8.531417733062476e-06,
|
| 1221 |
+
"loss": 0.2299,
|
| 1222 |
+
"step": 244
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 44.72727272727273,
|
| 1226 |
+
"grad_norm": 0.6867101788520813,
|
| 1227 |
+
"learning_rate": 8.498316702566828e-06,
|
| 1228 |
+
"loss": 0.2501,
|
| 1229 |
+
"step": 246
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 44.90909090909091,
|
| 1233 |
+
"eval_loss": 0.4543609619140625,
|
| 1234 |
+
"eval_runtime": 21.005,
|
| 1235 |
+
"eval_samples_per_second": 1.143,
|
| 1236 |
+
"eval_steps_per_second": 1.143,
|
| 1237 |
+
"step": 247
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 45.09090909090909,
|
| 1241 |
+
"grad_norm": 0.571156919002533,
|
| 1242 |
+
"learning_rate": 8.464912556547486e-06,
|
| 1243 |
+
"loss": 0.2035,
|
| 1244 |
+
"step": 248
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 45.45454545454545,
|
| 1248 |
+
"grad_norm": 0.761677086353302,
|
| 1249 |
+
"learning_rate": 8.43120818934367e-06,
|
| 1250 |
+
"loss": 0.1717,
|
| 1251 |
+
"step": 250
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 45.81818181818182,
|
| 1255 |
+
"grad_norm": 0.6928288340568542,
|
| 1256 |
+
"learning_rate": 8.397206521307584e-06,
|
| 1257 |
+
"loss": 0.2247,
|
| 1258 |
+
"step": 252
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 46.0,
|
| 1262 |
+
"eval_loss": 0.4641251266002655,
|
| 1263 |
+
"eval_runtime": 20.9989,
|
| 1264 |
+
"eval_samples_per_second": 1.143,
|
| 1265 |
+
"eval_steps_per_second": 1.143,
|
| 1266 |
+
"step": 253
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 46.0,
|
| 1270 |
+
"step": 253,
|
| 1271 |
+
"total_flos": 3.15960483767124e+17,
|
| 1272 |
+
"train_loss": 0.7967259936888699,
|
| 1273 |
+
"train_runtime": 11647.481,
|
| 1274 |
+
"train_samples_per_second": 1.133,
|
| 1275 |
+
"train_steps_per_second": 0.064
|
| 1276 |
+
}
|
| 1277 |
+
],
|
| 1278 |
+
"logging_steps": 2,
|
| 1279 |
+
"max_steps": 750,
|
| 1280 |
+
"num_input_tokens_seen": 0,
|
| 1281 |
+
"num_train_epochs": 150,
|
| 1282 |
+
"save_steps": 25,
|
| 1283 |
+
"stateful_callbacks": {
|
| 1284 |
+
"EarlyStoppingCallback": {
|
| 1285 |
+
"args": {
|
| 1286 |
+
"early_stopping_patience": 7,
|
| 1287 |
+
"early_stopping_threshold": 0.0
|
| 1288 |
+
},
|
| 1289 |
+
"attributes": {
|
| 1290 |
+
"early_stopping_patience_counter": 0
|
| 1291 |
+
}
|
| 1292 |
+
},
|
| 1293 |
+
"TrainerControl": {
|
| 1294 |
+
"args": {
|
| 1295 |
+
"should_epoch_stop": false,
|
| 1296 |
+
"should_evaluate": false,
|
| 1297 |
+
"should_log": false,
|
| 1298 |
+
"should_save": true,
|
| 1299 |
+
"should_training_stop": true
|
| 1300 |
+
},
|
| 1301 |
+
"attributes": {}
|
| 1302 |
+
}
|
| 1303 |
+
},
|
| 1304 |
+
"total_flos": 3.15960483767124e+17,
|
| 1305 |
+
"train_batch_size": 1,
|
| 1306 |
+
"trial_name": null,
|
| 1307 |
+
"trial_params": null
|
| 1308 |
+
}
|