Model save
Browse files- README.md +65 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- model.safetensors +1 -1
- train_results.json +9 -0
- trainer_state.json +1067 -0
README.md
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: tinyllama-sft-vicuna-full-rrr100-gaussian
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# tinyllama-sft-vicuna-full-rrr100-gaussian
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.7274
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 2e-05
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 4
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 128
|
| 47 |
+
- total_eval_batch_size: 32
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
+
| 0.7115 | 1.0 | 732 | 0.7274 |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- Transformers 4.40.2
|
| 63 |
+
- Pytorch 2.3.0+cu121
|
| 64 |
+
- Datasets 2.19.1
|
| 65 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 106034689474560.0,
|
| 4 |
+
"train_loss": 0.7575202535410397,
|
| 5 |
+
"train_runtime": 8959.9133,
|
| 6 |
+
"train_samples": 93645,
|
| 7 |
+
"train_samples_per_second": 10.452,
|
| 8 |
+
"train_steps_per_second": 0.082
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.40.2"
|
| 7 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2200128056
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:546897f2c7f6823f87e1d46b1abf0dfb2903a79da47370e84e3f05542f5fe0ac
|
| 3 |
size 2200128056
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 106034689474560.0,
|
| 4 |
+
"train_loss": 0.7575202535410397,
|
| 5 |
+
"train_runtime": 8959.9133,
|
| 6 |
+
"train_samples": 93645,
|
| 7 |
+
"train_samples_per_second": 10.452,
|
| 8 |
+
"train_steps_per_second": 0.082
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1067 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 732,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001366120218579235,
|
| 13 |
+
"grad_norm": 8.843155743885935,
|
| 14 |
+
"learning_rate": 2.702702702702703e-07,
|
| 15 |
+
"loss": 1.199,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.006830601092896175,
|
| 20 |
+
"grad_norm": 8.540631742450893,
|
| 21 |
+
"learning_rate": 1.3513513513513515e-06,
|
| 22 |
+
"loss": 1.2044,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01366120218579235,
|
| 27 |
+
"grad_norm": 8.767200579620665,
|
| 28 |
+
"learning_rate": 2.702702702702703e-06,
|
| 29 |
+
"loss": 1.186,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.020491803278688523,
|
| 34 |
+
"grad_norm": 2.6098644670988818,
|
| 35 |
+
"learning_rate": 4.0540540540540545e-06,
|
| 36 |
+
"loss": 1.0193,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0273224043715847,
|
| 41 |
+
"grad_norm": 0.9214100024259756,
|
| 42 |
+
"learning_rate": 5.405405405405406e-06,
|
| 43 |
+
"loss": 0.9217,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03415300546448088,
|
| 48 |
+
"grad_norm": 0.7472855963085931,
|
| 49 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 50 |
+
"loss": 0.859,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.040983606557377046,
|
| 55 |
+
"grad_norm": 0.7027313339104094,
|
| 56 |
+
"learning_rate": 8.108108108108109e-06,
|
| 57 |
+
"loss": 0.8567,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.04781420765027322,
|
| 62 |
+
"grad_norm": 0.606492697546335,
|
| 63 |
+
"learning_rate": 9.45945945945946e-06,
|
| 64 |
+
"loss": 0.8564,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0546448087431694,
|
| 69 |
+
"grad_norm": 0.5940228911261327,
|
| 70 |
+
"learning_rate": 1.0810810810810812e-05,
|
| 71 |
+
"loss": 0.8281,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.06147540983606557,
|
| 76 |
+
"grad_norm": 0.5848291669538546,
|
| 77 |
+
"learning_rate": 1.2162162162162164e-05,
|
| 78 |
+
"loss": 0.8319,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06830601092896176,
|
| 83 |
+
"grad_norm": 0.5535457560314438,
|
| 84 |
+
"learning_rate": 1.3513513513513515e-05,
|
| 85 |
+
"loss": 0.8194,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07513661202185792,
|
| 90 |
+
"grad_norm": 0.5893686332325634,
|
| 91 |
+
"learning_rate": 1.4864864864864865e-05,
|
| 92 |
+
"loss": 0.828,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08196721311475409,
|
| 97 |
+
"grad_norm": 0.5719534274373087,
|
| 98 |
+
"learning_rate": 1.6216216216216218e-05,
|
| 99 |
+
"loss": 0.8255,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08879781420765027,
|
| 104 |
+
"grad_norm": 0.5782518990484358,
|
| 105 |
+
"learning_rate": 1.756756756756757e-05,
|
| 106 |
+
"loss": 0.8091,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.09562841530054644,
|
| 111 |
+
"grad_norm": 0.6001941236834596,
|
| 112 |
+
"learning_rate": 1.891891891891892e-05,
|
| 113 |
+
"loss": 0.8509,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.10245901639344263,
|
| 118 |
+
"grad_norm": 0.5756334358383789,
|
| 119 |
+
"learning_rate": 1.999988602302209e-05,
|
| 120 |
+
"loss": 0.7947,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.1092896174863388,
|
| 125 |
+
"grad_norm": 0.5589039932124558,
|
| 126 |
+
"learning_rate": 1.9995897101594454e-05,
|
| 127 |
+
"loss": 0.81,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.11612021857923498,
|
| 132 |
+
"grad_norm": 0.5938570135428684,
|
| 133 |
+
"learning_rate": 1.99862119291555e-05,
|
| 134 |
+
"loss": 0.8088,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.12295081967213115,
|
| 139 |
+
"grad_norm": 0.5924615537351454,
|
| 140 |
+
"learning_rate": 1.997083602488702e-05,
|
| 141 |
+
"loss": 0.8071,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.12978142076502733,
|
| 146 |
+
"grad_norm": 0.5988590272254397,
|
| 147 |
+
"learning_rate": 1.994977815088504e-05,
|
| 148 |
+
"loss": 0.8057,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1366120218579235,
|
| 153 |
+
"grad_norm": 0.5820148883591653,
|
| 154 |
+
"learning_rate": 1.9923050307166655e-05,
|
| 155 |
+
"loss": 0.8043,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.14344262295081966,
|
| 160 |
+
"grad_norm": 0.6312826892627479,
|
| 161 |
+
"learning_rate": 1.989066772483171e-05,
|
| 162 |
+
"loss": 0.8125,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.15027322404371585,
|
| 167 |
+
"grad_norm": 0.5457727775136069,
|
| 168 |
+
"learning_rate": 1.9852648857383224e-05,
|
| 169 |
+
"loss": 0.795,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.15710382513661203,
|
| 174 |
+
"grad_norm": 0.5931361564309547,
|
| 175 |
+
"learning_rate": 1.9809015370211505e-05,
|
| 176 |
+
"loss": 0.8023,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.16393442622950818,
|
| 181 |
+
"grad_norm": 0.5429570919204767,
|
| 182 |
+
"learning_rate": 1.9759792128247922e-05,
|
| 183 |
+
"loss": 0.818,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.17076502732240437,
|
| 188 |
+
"grad_norm": 0.5941429790495816,
|
| 189 |
+
"learning_rate": 1.9705007181795416e-05,
|
| 190 |
+
"loss": 0.8076,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.17759562841530055,
|
| 195 |
+
"grad_norm": 0.586524294465549,
|
| 196 |
+
"learning_rate": 1.964469175054377e-05,
|
| 197 |
+
"loss": 0.7948,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.18442622950819673,
|
| 202 |
+
"grad_norm": 0.5701970901963691,
|
| 203 |
+
"learning_rate": 1.9578880205778793e-05,
|
| 204 |
+
"loss": 0.7901,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.1912568306010929,
|
| 209 |
+
"grad_norm": 0.5531109526566766,
|
| 210 |
+
"learning_rate": 1.950761005079556e-05,
|
| 211 |
+
"loss": 0.768,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.19808743169398907,
|
| 216 |
+
"grad_norm": 0.5905328760710993,
|
| 217 |
+
"learning_rate": 1.9430921899526786e-05,
|
| 218 |
+
"loss": 0.7892,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.20491803278688525,
|
| 223 |
+
"grad_norm": 0.5887273338658354,
|
| 224 |
+
"learning_rate": 1.934885945339865e-05,
|
| 225 |
+
"loss": 0.7814,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.21174863387978143,
|
| 230 |
+
"grad_norm": 0.6013798893637619,
|
| 231 |
+
"learning_rate": 1.9261469476427122e-05,
|
| 232 |
+
"loss": 0.7957,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.2185792349726776,
|
| 237 |
+
"grad_norm": 0.5617177815620502,
|
| 238 |
+
"learning_rate": 1.916880176856909e-05,
|
| 239 |
+
"loss": 0.7591,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.22540983606557377,
|
| 244 |
+
"grad_norm": 0.6140562235904167,
|
| 245 |
+
"learning_rate": 1.907090913734341e-05,
|
| 246 |
+
"loss": 0.7787,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.23224043715846995,
|
| 251 |
+
"grad_norm": 0.586045608127345,
|
| 252 |
+
"learning_rate": 1.896784736773805e-05,
|
| 253 |
+
"loss": 0.768,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.2390710382513661,
|
| 258 |
+
"grad_norm": 0.5524101190099344,
|
| 259 |
+
"learning_rate": 1.885967519042054e-05,
|
| 260 |
+
"loss": 0.7744,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.2459016393442623,
|
| 265 |
+
"grad_norm": 0.5523610942140554,
|
| 266 |
+
"learning_rate": 1.8746454248269777e-05,
|
| 267 |
+
"loss": 0.7602,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.2527322404371585,
|
| 272 |
+
"grad_norm": 0.546789444435313,
|
| 273 |
+
"learning_rate": 1.862824906124826e-05,
|
| 274 |
+
"loss": 0.784,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.25956284153005466,
|
| 279 |
+
"grad_norm": 0.5348363807692132,
|
| 280 |
+
"learning_rate": 1.850512698963485e-05,
|
| 281 |
+
"loss": 0.7579,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.26639344262295084,
|
| 286 |
+
"grad_norm": 0.5681852755270315,
|
| 287 |
+
"learning_rate": 1.8377158195638877e-05,
|
| 288 |
+
"loss": 0.7722,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.273224043715847,
|
| 293 |
+
"grad_norm": 0.5731618955355812,
|
| 294 |
+
"learning_rate": 1.8244415603417603e-05,
|
| 295 |
+
"loss": 0.7939,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.28005464480874315,
|
| 300 |
+
"grad_norm": 0.5778873895260302,
|
| 301 |
+
"learning_rate": 1.8106974857519737e-05,
|
| 302 |
+
"loss": 0.7762,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.28688524590163933,
|
| 307 |
+
"grad_norm": 0.614463428308708,
|
| 308 |
+
"learning_rate": 1.7964914279778716e-05,
|
| 309 |
+
"loss": 0.7522,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2937158469945355,
|
| 314 |
+
"grad_norm": 0.5703542529393416,
|
| 315 |
+
"learning_rate": 1.78183148246803e-05,
|
| 316 |
+
"loss": 0.7641,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.3005464480874317,
|
| 321 |
+
"grad_norm": 0.6153806412407196,
|
| 322 |
+
"learning_rate": 1.7667260033229953e-05,
|
| 323 |
+
"loss": 0.7798,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.3073770491803279,
|
| 328 |
+
"grad_norm": 0.5851057407860113,
|
| 329 |
+
"learning_rate": 1.751183598534625e-05,
|
| 330 |
+
"loss": 0.7647,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.31420765027322406,
|
| 335 |
+
"grad_norm": 0.6316856459688894,
|
| 336 |
+
"learning_rate": 1.7352131250807466e-05,
|
| 337 |
+
"loss": 0.769,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.32103825136612024,
|
| 342 |
+
"grad_norm": 0.5516002849247819,
|
| 343 |
+
"learning_rate": 1.7188236838779297e-05,
|
| 344 |
+
"loss": 0.7518,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.32786885245901637,
|
| 349 |
+
"grad_norm": 0.5604362168665755,
|
| 350 |
+
"learning_rate": 1.702024614595248e-05,
|
| 351 |
+
"loss": 0.7736,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.33469945355191255,
|
| 356 |
+
"grad_norm": 0.6287905939345914,
|
| 357 |
+
"learning_rate": 1.6848254903319866e-05,
|
| 358 |
+
"loss": 0.7487,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.34153005464480873,
|
| 363 |
+
"grad_norm": 0.5691120534757109,
|
| 364 |
+
"learning_rate": 1.6672361121623238e-05,
|
| 365 |
+
"loss": 0.7614,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3483606557377049,
|
| 370 |
+
"grad_norm": 0.5460125409110708,
|
| 371 |
+
"learning_rate": 1.6492665035501048e-05,
|
| 372 |
+
"loss": 0.7587,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.3551912568306011,
|
| 377 |
+
"grad_norm": 0.57597601156693,
|
| 378 |
+
"learning_rate": 1.6309269046368777e-05,
|
| 379 |
+
"loss": 0.7673,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3620218579234973,
|
| 384 |
+
"grad_norm": 0.5966304605065128,
|
| 385 |
+
"learning_rate": 1.612227766406461e-05,
|
| 386 |
+
"loss": 0.7816,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.36885245901639346,
|
| 391 |
+
"grad_norm": 0.5734953007972787,
|
| 392 |
+
"learning_rate": 1.5931797447293553e-05,
|
| 393 |
+
"loss": 0.7665,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3756830601092896,
|
| 398 |
+
"grad_norm": 0.5811602692803652,
|
| 399 |
+
"learning_rate": 1.5737936942904025e-05,
|
| 400 |
+
"loss": 0.7514,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3825136612021858,
|
| 405 |
+
"grad_norm": 0.5490754114059545,
|
| 406 |
+
"learning_rate": 1.554080662403144e-05,
|
| 407 |
+
"loss": 0.7588,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.38934426229508196,
|
| 412 |
+
"grad_norm": 0.5730964675833344,
|
| 413 |
+
"learning_rate": 1.5340518827144145e-05,
|
| 414 |
+
"loss": 0.7559,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.39617486338797814,
|
| 419 |
+
"grad_norm": 0.5589786160518924,
|
| 420 |
+
"learning_rate": 1.5137187688027437e-05,
|
| 421 |
+
"loss": 0.7535,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.4030054644808743,
|
| 426 |
+
"grad_norm": 0.5482919352886477,
|
| 427 |
+
"learning_rate": 1.4930929076742317e-05,
|
| 428 |
+
"loss": 0.7408,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.4098360655737705,
|
| 433 |
+
"grad_norm": 0.5546311270276107,
|
| 434 |
+
"learning_rate": 1.4721860531595868e-05,
|
| 435 |
+
"loss": 0.771,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.4166666666666667,
|
| 440 |
+
"grad_norm": 0.5655833815736454,
|
| 441 |
+
"learning_rate": 1.451010119216102e-05,
|
| 442 |
+
"loss": 0.7367,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.42349726775956287,
|
| 447 |
+
"grad_norm": 0.6197227214321215,
|
| 448 |
+
"learning_rate": 1.4295771731383799e-05,
|
| 449 |
+
"loss": 0.7527,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.430327868852459,
|
| 454 |
+
"grad_norm": 0.5476311997058135,
|
| 455 |
+
"learning_rate": 1.4078994286816768e-05,
|
| 456 |
+
"loss": 0.7295,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.4371584699453552,
|
| 461 |
+
"grad_norm": 0.5984718358075056,
|
| 462 |
+
"learning_rate": 1.3859892391017867e-05,
|
| 463 |
+
"loss": 0.747,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.44398907103825136,
|
| 468 |
+
"grad_norm": 0.5609136301558824,
|
| 469 |
+
"learning_rate": 1.3638590901154276e-05,
|
| 470 |
+
"loss": 0.7314,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.45081967213114754,
|
| 475 |
+
"grad_norm": 0.5678828594627737,
|
| 476 |
+
"learning_rate": 1.341521592785145e-05,
|
| 477 |
+
"loss": 0.7416,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.4576502732240437,
|
| 482 |
+
"grad_norm": 0.6045615515916248,
|
| 483 |
+
"learning_rate": 1.3189894763327851e-05,
|
| 484 |
+
"loss": 0.7575,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.4644808743169399,
|
| 489 |
+
"grad_norm": 0.6102125885340797,
|
| 490 |
+
"learning_rate": 1.2962755808856341e-05,
|
| 491 |
+
"loss": 0.7233,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.4713114754098361,
|
| 496 |
+
"grad_norm": 0.5569824101385544,
|
| 497 |
+
"learning_rate": 1.2733928501593587e-05,
|
| 498 |
+
"loss": 0.7513,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.4781420765027322,
|
| 503 |
+
"grad_norm": 0.5900237407977923,
|
| 504 |
+
"learning_rate": 1.2503543240819127e-05,
|
| 505 |
+
"loss": 0.7247,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4849726775956284,
|
| 510 |
+
"grad_norm": 0.5680127868499019,
|
| 511 |
+
"learning_rate": 1.227173131362619e-05,
|
| 512 |
+
"loss": 0.7311,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.4918032786885246,
|
| 517 |
+
"grad_norm": 0.5625573721414876,
|
| 518 |
+
"learning_rate": 1.2038624820106572e-05,
|
| 519 |
+
"loss": 0.723,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.49863387978142076,
|
| 524 |
+
"grad_norm": 0.5756982415323313,
|
| 525 |
+
"learning_rate": 1.1804356598072223e-05,
|
| 526 |
+
"loss": 0.7392,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.505464480874317,
|
| 531 |
+
"grad_norm": 0.5946932612905129,
|
| 532 |
+
"learning_rate": 1.1569060147356441e-05,
|
| 533 |
+
"loss": 0.7287,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.5122950819672131,
|
| 538 |
+
"grad_norm": 0.6113580526324344,
|
| 539 |
+
"learning_rate": 1.133286955373779e-05,
|
| 540 |
+
"loss": 0.7568,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.5191256830601093,
|
| 545 |
+
"grad_norm": 0.6964655258277292,
|
| 546 |
+
"learning_rate": 1.1095919412530136e-05,
|
| 547 |
+
"loss": 0.7257,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.5259562841530054,
|
| 552 |
+
"grad_norm": 0.558955112992284,
|
| 553 |
+
"learning_rate": 1.0858344751882304e-05,
|
| 554 |
+
"loss": 0.7468,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.5327868852459017,
|
| 559 |
+
"grad_norm": 0.6215470265457846,
|
| 560 |
+
"learning_rate": 1.0620280955831088e-05,
|
| 561 |
+
"loss": 0.7537,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.5396174863387978,
|
| 566 |
+
"grad_norm": 0.5750613153322487,
|
| 567 |
+
"learning_rate": 1.038186368715145e-05,
|
| 568 |
+
"loss": 0.7343,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.546448087431694,
|
| 573 |
+
"grad_norm": 0.5701696384507413,
|
| 574 |
+
"learning_rate": 1.0143228810047877e-05,
|
| 575 |
+
"loss": 0.7358,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.5532786885245902,
|
| 580 |
+
"grad_norm": 0.5635496322641829,
|
| 581 |
+
"learning_rate": 9.904512312730948e-06,
|
| 582 |
+
"loss": 0.723,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.5601092896174863,
|
| 587 |
+
"grad_norm": 0.5331086289619262,
|
| 588 |
+
"learning_rate": 9.665850229923258e-06,
|
| 589 |
+
"loss": 0.7359,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5669398907103825,
|
| 594 |
+
"grad_norm": 0.5696605773767074,
|
| 595 |
+
"learning_rate": 9.4273785653388e-06,
|
| 596 |
+
"loss": 0.7266,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5737704918032787,
|
| 601 |
+
"grad_norm": 0.5530919852682804,
|
| 602 |
+
"learning_rate": 9.189233214180057e-06,
|
| 603 |
+
"loss": 0.7245,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5806010928961749,
|
| 608 |
+
"grad_norm": 0.6180570767094125,
|
| 609 |
+
"learning_rate": 8.951549885696889e-06,
|
| 610 |
+
"loss": 0.7152,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.587431693989071,
|
| 615 |
+
"grad_norm": 0.5435168155874842,
|
| 616 |
+
"learning_rate": 8.714464025851428e-06,
|
| 617 |
+
"loss": 0.7329,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5942622950819673,
|
| 622 |
+
"grad_norm": 0.5918089418612534,
|
| 623 |
+
"learning_rate": 8.478110740132971e-06,
|
| 624 |
+
"loss": 0.7394,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.6010928961748634,
|
| 629 |
+
"grad_norm": 0.5775872171354361,
|
| 630 |
+
"learning_rate": 8.242624716566928e-06,
|
| 631 |
+
"loss": 0.7386,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.6079234972677595,
|
| 636 |
+
"grad_norm": 0.560805508558742,
|
| 637 |
+
"learning_rate": 8.008140148961642e-06,
|
| 638 |
+
"loss": 0.7286,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.6147540983606558,
|
| 643 |
+
"grad_norm": 0.5659035847143471,
|
| 644 |
+
"learning_rate": 7.774790660436857e-06,
|
| 645 |
+
"loss": 0.7395,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.6215846994535519,
|
| 650 |
+
"grad_norm": 0.5934950402129012,
|
| 651 |
+
"learning_rate": 7.542709227277396e-06,
|
| 652 |
+
"loss": 0.7256,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.6284153005464481,
|
| 657 |
+
"grad_norm": 0.5440391404753762,
|
| 658 |
+
"learning_rate": 7.312028103155426e-06,
|
| 659 |
+
"loss": 0.7115,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.6352459016393442,
|
| 664 |
+
"grad_norm": 0.525973972202757,
|
| 665 |
+
"learning_rate": 7.0828787437645455e-06,
|
| 666 |
+
"loss": 0.7253,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.6420765027322405,
|
| 671 |
+
"grad_norm": 0.5381242803100431,
|
| 672 |
+
"learning_rate": 6.8553917319085676e-06,
|
| 673 |
+
"loss": 0.715,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.6489071038251366,
|
| 678 |
+
"grad_norm": 0.5407688729054287,
|
| 679 |
+
"learning_rate": 6.629696703087755e-06,
|
| 680 |
+
"loss": 0.7145,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.6557377049180327,
|
| 685 |
+
"grad_norm": 0.5510841812349385,
|
| 686 |
+
"learning_rate": 6.405922271624874e-06,
|
| 687 |
+
"loss": 0.6676,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.662568306010929,
|
| 692 |
+
"grad_norm": 0.5574387613507912,
|
| 693 |
+
"learning_rate": 6.184195957373176e-06,
|
| 694 |
+
"loss": 0.7353,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.6693989071038251,
|
| 699 |
+
"grad_norm": 0.6253673572306452,
|
| 700 |
+
"learning_rate": 5.964644113048079e-06,
|
| 701 |
+
"loss": 0.7065,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.6762295081967213,
|
| 706 |
+
"grad_norm": 0.5452906786267165,
|
| 707 |
+
"learning_rate": 5.74739185222394e-06,
|
| 708 |
+
"loss": 0.7336,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.6830601092896175,
|
| 713 |
+
"grad_norm": 0.5560796717868584,
|
| 714 |
+
"learning_rate": 5.532562978036964e-06,
|
| 715 |
+
"loss": 0.7207,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.6898907103825137,
|
| 720 |
+
"grad_norm": 0.6056533738818196,
|
| 721 |
+
"learning_rate": 5.320279912634907e-06,
|
| 722 |
+
"loss": 0.7186,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6967213114754098,
|
| 727 |
+
"grad_norm": 0.569520229311181,
|
| 728 |
+
"learning_rate": 5.110663627413695e-06,
|
| 729 |
+
"loss": 0.7377,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.703551912568306,
|
| 734 |
+
"grad_norm": 0.5563155283253095,
|
| 735 |
+
"learning_rate": 4.903833574080825e-06,
|
| 736 |
+
"loss": 0.7065,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.7103825136612022,
|
| 741 |
+
"grad_norm": 0.6065562069673966,
|
| 742 |
+
"learning_rate": 4.6999076165847214e-06,
|
| 743 |
+
"loss": 0.7436,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.7172131147540983,
|
| 748 |
+
"grad_norm": 0.5561597356740903,
|
| 749 |
+
"learning_rate": 4.499001963948929e-06,
|
| 750 |
+
"loss": 0.7279,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.7240437158469946,
|
| 755 |
+
"grad_norm": 0.540260398768991,
|
| 756 |
+
"learning_rate": 4.301231104049359e-06,
|
| 757 |
+
"loss": 0.7102,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.7308743169398907,
|
| 762 |
+
"grad_norm": 0.5312643673085101,
|
| 763 |
+
"learning_rate": 4.106707738372357e-06,
|
| 764 |
+
"loss": 0.696,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.7377049180327869,
|
| 769 |
+
"grad_norm": 0.5886875453265892,
|
| 770 |
+
"learning_rate": 3.915542717790759e-06,
|
| 771 |
+
"loss": 0.7225,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.744535519125683,
|
| 776 |
+
"grad_norm": 0.5771137703202747,
|
| 777 |
+
"learning_rate": 3.727844979394526e-06,
|
| 778 |
+
"loss": 0.7369,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.7513661202185792,
|
| 783 |
+
"grad_norm": 0.5513216324725466,
|
| 784 |
+
"learning_rate": 3.543721484411976e-06,
|
| 785 |
+
"loss": 0.7381,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.7581967213114754,
|
| 790 |
+
"grad_norm": 0.5821599254967618,
|
| 791 |
+
"learning_rate": 3.3632771572569878e-06,
|
| 792 |
+
"loss": 0.7346,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7650273224043715,
|
| 797 |
+
"grad_norm": 0.5434619637496027,
|
| 798 |
+
"learning_rate": 3.1866148257368666e-06,
|
| 799 |
+
"loss": 0.6955,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.7718579234972678,
|
| 804 |
+
"grad_norm": 0.5508768490641304,
|
| 805 |
+
"learning_rate": 3.0138351624550165e-06,
|
| 806 |
+
"loss": 0.7088,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.7786885245901639,
|
| 811 |
+
"grad_norm": 0.5522627907801693,
|
| 812 |
+
"learning_rate": 2.845036627441755e-06,
|
| 813 |
+
"loss": 0.7029,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.7855191256830601,
|
| 818 |
+
"grad_norm": 0.5484073193031532,
|
| 819 |
+
"learning_rate": 2.6803154120460007e-06,
|
| 820 |
+
"loss": 0.6981,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.7923497267759563,
|
| 825 |
+
"grad_norm": 0.5460127035608994,
|
| 826 |
+
"learning_rate": 2.5197653841197546e-06,
|
| 827 |
+
"loss": 0.7359,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.7991803278688525,
|
| 832 |
+
"grad_norm": 0.5649471360604418,
|
| 833 |
+
"learning_rate": 2.3634780345266805e-06,
|
| 834 |
+
"loss": 0.7273,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.8060109289617486,
|
| 839 |
+
"grad_norm": 0.5701788726901212,
|
| 840 |
+
"learning_rate": 2.211542425005223e-06,
|
| 841 |
+
"loss": 0.7093,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.8128415300546448,
|
| 846 |
+
"grad_norm": 0.5541668671991618,
|
| 847 |
+
"learning_rate": 2.064045137415982e-06,
|
| 848 |
+
"loss": 0.7464,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.819672131147541,
|
| 853 |
+
"grad_norm": 0.5330522987281802,
|
| 854 |
+
"learning_rate": 1.9210702244022616e-06,
|
| 855 |
+
"loss": 0.715,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.8265027322404371,
|
| 860 |
+
"grad_norm": 0.5298294565996464,
|
| 861 |
+
"learning_rate": 1.7826991614919264e-06,
|
| 862 |
+
"loss": 0.7223,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.8333333333333334,
|
| 867 |
+
"grad_norm": 0.5610592085239171,
|
| 868 |
+
"learning_rate": 1.6490108006678495e-06,
|
| 869 |
+
"loss": 0.7247,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.8401639344262295,
|
| 874 |
+
"grad_norm": 0.5682288942071054,
|
| 875 |
+
"learning_rate": 1.5200813254334013e-06,
|
| 876 |
+
"loss": 0.7054,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.8469945355191257,
|
| 881 |
+
"grad_norm": 0.5693548112007653,
|
| 882 |
+
"learning_rate": 1.3959842073986085e-06,
|
| 883 |
+
"loss": 0.7266,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.8538251366120219,
|
| 888 |
+
"grad_norm": 0.5448533977882911,
|
| 889 |
+
"learning_rate": 1.2767901644116943e-06,
|
| 890 |
+
"loss": 0.7154,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.860655737704918,
|
| 895 |
+
"grad_norm": 0.567356173427535,
|
| 896 |
+
"learning_rate": 1.1625671202598875e-06,
|
| 897 |
+
"loss": 0.7042,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.8674863387978142,
|
| 902 |
+
"grad_norm": 0.5923179681072539,
|
| 903 |
+
"learning_rate": 1.0533801659624531e-06,
|
| 904 |
+
"loss": 0.696,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.8743169398907104,
|
| 909 |
+
"grad_norm": 0.5471124778528522,
|
| 910 |
+
"learning_rate": 9.492915226779809e-07,
|
| 911 |
+
"loss": 0.7037,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.8811475409836066,
|
| 916 |
+
"grad_norm": 0.7367442278445117,
|
| 917 |
+
"learning_rate": 8.503605062471187e-07,
|
| 918 |
+
"loss": 0.7285,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.8879781420765027,
|
| 923 |
+
"grad_norm": 0.5411387873270271,
|
| 924 |
+
"learning_rate": 7.566434933909006e-07,
|
| 925 |
+
"loss": 0.7167,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.894808743169399,
|
| 930 |
+
"grad_norm": 0.5402678284774802,
|
| 931 |
+
"learning_rate": 6.681938895839746e-07,
|
| 932 |
+
"loss": 0.6901,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.9016393442622951,
|
| 937 |
+
"grad_norm": 0.5526115125845628,
|
| 938 |
+
"learning_rate": 5.850620986210198e-07,
|
| 939 |
+
"loss": 0.7028,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.9084699453551912,
|
| 944 |
+
"grad_norm": 0.557503740636623,
|
| 945 |
+
"learning_rate": 5.072954938936925e-07,
|
| 946 |
+
"loss": 0.7059,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.9153005464480874,
|
| 951 |
+
"grad_norm": 0.6069774844130886,
|
| 952 |
+
"learning_rate": 4.3493839139447716e-07,
|
| 953 |
+
"loss": 0.709,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.9221311475409836,
|
| 958 |
+
"grad_norm": 0.5659313165055196,
|
| 959 |
+
"learning_rate": 3.6803202446282217e-07,
|
| 960 |
+
"loss": 0.7173,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.9289617486338798,
|
| 965 |
+
"grad_norm": 0.546500310034168,
|
| 966 |
+
"learning_rate": 3.0661452028795335e-07,
|
| 967 |
+
"loss": 0.6963,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.9357923497267759,
|
| 972 |
+
"grad_norm": 0.5278636280522998,
|
| 973 |
+
"learning_rate": 2.507208781817638e-07,
|
| 974 |
+
"loss": 0.7062,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.9426229508196722,
|
| 979 |
+
"grad_norm": 0.5368853646294492,
|
| 980 |
+
"learning_rate": 2.0038294963413251e-07,
|
| 981 |
+
"loss": 0.7303,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.9494535519125683,
|
| 986 |
+
"grad_norm": 0.5295882031597404,
|
| 987 |
+
"learning_rate": 1.556294201620734e-07,
|
| 988 |
+
"loss": 0.7077,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.9562841530054644,
|
| 993 |
+
"grad_norm": 0.5590243905068004,
|
| 994 |
+
"learning_rate": 1.1648579296304252e-07,
|
| 995 |
+
"loss": 0.71,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.9631147540983607,
|
| 1000 |
+
"grad_norm": 0.5607413731294414,
|
| 1001 |
+
"learning_rate": 8.297437438170797e-08,
|
| 1002 |
+
"loss": 0.716,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.9699453551912568,
|
| 1007 |
+
"grad_norm": 0.5580234118888867,
|
| 1008 |
+
"learning_rate": 5.51142611984834e-08,
|
| 1009 |
+
"loss": 0.7147,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.976775956284153,
|
| 1014 |
+
"grad_norm": 0.5435574659414255,
|
| 1015 |
+
"learning_rate": 3.2921329747056527e-08,
|
| 1016 |
+
"loss": 0.7131,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.9836065573770492,
|
| 1021 |
+
"grad_norm": 0.6153543980271695,
|
| 1022 |
+
"learning_rate": 1.6408226867118404e-08,
|
| 1023 |
+
"loss": 0.714,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.9904371584699454,
|
| 1028 |
+
"grad_norm": 0.5542963142607653,
|
| 1029 |
+
"learning_rate": 5.584362697453882e-09,
|
| 1030 |
+
"loss": 0.7282,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.9972677595628415,
|
| 1035 |
+
"grad_norm": 1.0209900402406562,
|
| 1036 |
+
"learning_rate": 4.5590531348227443e-10,
|
| 1037 |
+
"loss": 0.7115,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.0,
|
| 1042 |
+
"eval_loss": 0.7274145483970642,
|
| 1043 |
+
"eval_runtime": 9.4843,
|
| 1044 |
+
"eval_samples_per_second": 52.719,
|
| 1045 |
+
"eval_steps_per_second": 1.687,
|
| 1046 |
+
"step": 732
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.0,
|
| 1050 |
+
"step": 732,
|
| 1051 |
+
"total_flos": 106034689474560.0,
|
| 1052 |
+
"train_loss": 0.7575202535410397,
|
| 1053 |
+
"train_runtime": 8959.9133,
|
| 1054 |
+
"train_samples_per_second": 10.452,
|
| 1055 |
+
"train_steps_per_second": 0.082
|
| 1056 |
+
}
|
| 1057 |
+
],
|
| 1058 |
+
"logging_steps": 5,
|
| 1059 |
+
"max_steps": 732,
|
| 1060 |
+
"num_input_tokens_seen": 0,
|
| 1061 |
+
"num_train_epochs": 1,
|
| 1062 |
+
"save_steps": 100,
|
| 1063 |
+
"total_flos": 106034689474560.0,
|
| 1064 |
+
"train_batch_size": 16,
|
| 1065 |
+
"trial_name": null,
|
| 1066 |
+
"trial_params": null
|
| 1067 |
+
}
|