Model save
Browse files- README.md +64 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1681 -0
README.md
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: eeve-10.8B_C
|
| 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 |
+
# eeve-10.8B_C
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [yanolja/EEVE-Korean-Instruct-10.8B-v1.0](https://huggingface.co/yanolja/EEVE-Korean-Instruct-10.8B-v1.0) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3697
|
| 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: 1e-05
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 3
|
| 45 |
+
- total_train_batch_size: 48
|
| 46 |
+
- total_eval_batch_size: 24
|
| 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: 1
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 0.2153 | 1.0 | 233 | 0.3697 |
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
### Framework versions
|
| 60 |
+
|
| 61 |
+
- Transformers 4.41.2
|
| 62 |
+
- Pytorch 2.1.2
|
| 63 |
+
- Datasets 2.20.0
|
| 64 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 8.156803263102976e+16,
|
| 4 |
+
"train_loss": 0.28537011958754627,
|
| 5 |
+
"train_runtime": 757.3361,
|
| 6 |
+
"train_samples": 11140,
|
| 7 |
+
"train_samples_per_second": 14.709,
|
| 8 |
+
"train_steps_per_second": 0.308
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 8.156803263102976e+16,
|
| 4 |
+
"train_loss": 0.28537011958754627,
|
| 5 |
+
"train_runtime": 757.3361,
|
| 6 |
+
"train_samples": 11140,
|
| 7 |
+
"train_samples_per_second": 14.709,
|
| 8 |
+
"train_steps_per_second": 0.308
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1681 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 233,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004291845493562232,
|
| 13 |
+
"grad_norm": 64.1849594116211,
|
| 14 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 15 |
+
"loss": 1.9257,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.008583690987124463,
|
| 20 |
+
"grad_norm": 67.15699768066406,
|
| 21 |
+
"learning_rate": 8.333333333333333e-07,
|
| 22 |
+
"loss": 1.9675,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.012875536480686695,
|
| 27 |
+
"grad_norm": 63.228580474853516,
|
| 28 |
+
"learning_rate": 1.25e-06,
|
| 29 |
+
"loss": 1.9059,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.017167381974248927,
|
| 34 |
+
"grad_norm": 50.28240966796875,
|
| 35 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 36 |
+
"loss": 1.6307,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02145922746781116,
|
| 41 |
+
"grad_norm": 56.101253509521484,
|
| 42 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 43 |
+
"loss": 1.174,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02575107296137339,
|
| 48 |
+
"grad_norm": 34.66034698486328,
|
| 49 |
+
"learning_rate": 2.5e-06,
|
| 50 |
+
"loss": 0.5821,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.030042918454935622,
|
| 55 |
+
"grad_norm": 79.90634155273438,
|
| 56 |
+
"learning_rate": 2.916666666666667e-06,
|
| 57 |
+
"loss": 0.4952,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.034334763948497854,
|
| 62 |
+
"grad_norm": 31.585365295410156,
|
| 63 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 64 |
+
"loss": 0.4005,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03862660944206009,
|
| 69 |
+
"grad_norm": 20.49032974243164,
|
| 70 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 71 |
+
"loss": 0.3842,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04291845493562232,
|
| 76 |
+
"grad_norm": 12.812472343444824,
|
| 77 |
+
"learning_rate": 4.166666666666667e-06,
|
| 78 |
+
"loss": 0.3652,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04721030042918455,
|
| 83 |
+
"grad_norm": 12.473087310791016,
|
| 84 |
+
"learning_rate": 4.583333333333333e-06,
|
| 85 |
+
"loss": 0.3918,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05150214592274678,
|
| 90 |
+
"grad_norm": 12.261868476867676,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.3106,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.055793991416309016,
|
| 97 |
+
"grad_norm": 10.985730171203613,
|
| 98 |
+
"learning_rate": 5.416666666666667e-06,
|
| 99 |
+
"loss": 0.2936,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.060085836909871244,
|
| 104 |
+
"grad_norm": 6.1359944343566895,
|
| 105 |
+
"learning_rate": 5.833333333333334e-06,
|
| 106 |
+
"loss": 0.3058,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06437768240343347,
|
| 111 |
+
"grad_norm": 19.714662551879883,
|
| 112 |
+
"learning_rate": 6.25e-06,
|
| 113 |
+
"loss": 0.3188,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06866952789699571,
|
| 118 |
+
"grad_norm": 50.216400146484375,
|
| 119 |
+
"learning_rate": 6.666666666666667e-06,
|
| 120 |
+
"loss": 0.2829,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07296137339055794,
|
| 125 |
+
"grad_norm": 36.995113372802734,
|
| 126 |
+
"learning_rate": 7.083333333333335e-06,
|
| 127 |
+
"loss": 0.2931,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07725321888412018,
|
| 132 |
+
"grad_norm": 4.576523780822754,
|
| 133 |
+
"learning_rate": 7.500000000000001e-06,
|
| 134 |
+
"loss": 0.3061,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0815450643776824,
|
| 139 |
+
"grad_norm": 4.885584831237793,
|
| 140 |
+
"learning_rate": 7.916666666666667e-06,
|
| 141 |
+
"loss": 0.3042,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.08583690987124463,
|
| 146 |
+
"grad_norm": 4.763634204864502,
|
| 147 |
+
"learning_rate": 8.333333333333334e-06,
|
| 148 |
+
"loss": 0.2747,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.09012875536480687,
|
| 153 |
+
"grad_norm": 3.823352575302124,
|
| 154 |
+
"learning_rate": 8.750000000000001e-06,
|
| 155 |
+
"loss": 0.2904,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0944206008583691,
|
| 160 |
+
"grad_norm": 3.134524345397949,
|
| 161 |
+
"learning_rate": 9.166666666666666e-06,
|
| 162 |
+
"loss": 0.2695,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.09871244635193133,
|
| 167 |
+
"grad_norm": 3.5232937335968018,
|
| 168 |
+
"learning_rate": 9.583333333333335e-06,
|
| 169 |
+
"loss": 0.2811,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.10300429184549356,
|
| 174 |
+
"grad_norm": 3.035285711288452,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 0.2878,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.1072961373390558,
|
| 181 |
+
"grad_norm": 5.4150614738464355,
|
| 182 |
+
"learning_rate": 9.999435142363484e-06,
|
| 183 |
+
"loss": 0.283,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.11158798283261803,
|
| 188 |
+
"grad_norm": 34.44544982910156,
|
| 189 |
+
"learning_rate": 9.997740697079595e-06,
|
| 190 |
+
"loss": 0.3557,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.11587982832618025,
|
| 195 |
+
"grad_norm": 17.197986602783203,
|
| 196 |
+
"learning_rate": 9.994917046996472e-06,
|
| 197 |
+
"loss": 0.4027,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.12017167381974249,
|
| 202 |
+
"grad_norm": 14.15243911743164,
|
| 203 |
+
"learning_rate": 9.990964830098246e-06,
|
| 204 |
+
"loss": 0.3703,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.12446351931330472,
|
| 209 |
+
"grad_norm": 14.733880043029785,
|
| 210 |
+
"learning_rate": 9.985884939360873e-06,
|
| 211 |
+
"loss": 0.3651,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.12875536480686695,
|
| 216 |
+
"grad_norm": 5.188531398773193,
|
| 217 |
+
"learning_rate": 9.979678522550382e-06,
|
| 218 |
+
"loss": 0.3329,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.13304721030042918,
|
| 223 |
+
"grad_norm": 10.757170677185059,
|
| 224 |
+
"learning_rate": 9.972346981963546e-06,
|
| 225 |
+
"loss": 0.3351,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.13733905579399142,
|
| 230 |
+
"grad_norm": 5.147034168243408,
|
| 231 |
+
"learning_rate": 9.963891974111042e-06,
|
| 232 |
+
"loss": 0.3193,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.14163090128755365,
|
| 237 |
+
"grad_norm": 8.824125289916992,
|
| 238 |
+
"learning_rate": 9.95431540934317e-06,
|
| 239 |
+
"loss": 0.3167,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1459227467811159,
|
| 244 |
+
"grad_norm": 3.809666872024536,
|
| 245 |
+
"learning_rate": 9.943619451418225e-06,
|
| 246 |
+
"loss": 0.3238,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.15021459227467812,
|
| 251 |
+
"grad_norm": 6.210383415222168,
|
| 252 |
+
"learning_rate": 9.931806517013612e-06,
|
| 253 |
+
"loss": 0.2888,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.15450643776824036,
|
| 258 |
+
"grad_norm": 2.9546399116516113,
|
| 259 |
+
"learning_rate": 9.918879275179819e-06,
|
| 260 |
+
"loss": 0.2781,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.15879828326180256,
|
| 265 |
+
"grad_norm": 2.7553744316101074,
|
| 266 |
+
"learning_rate": 9.904840646737346e-06,
|
| 267 |
+
"loss": 0.2936,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1630901287553648,
|
| 272 |
+
"grad_norm": 2.9384710788726807,
|
| 273 |
+
"learning_rate": 9.889693803616793e-06,
|
| 274 |
+
"loss": 0.3017,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.16738197424892703,
|
| 279 |
+
"grad_norm": 3.0700764656066895,
|
| 280 |
+
"learning_rate": 9.873442168142158e-06,
|
| 281 |
+
"loss": 0.2803,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.17167381974248927,
|
| 286 |
+
"grad_norm": 3.089503049850464,
|
| 287 |
+
"learning_rate": 9.856089412257605e-06,
|
| 288 |
+
"loss": 0.2679,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1759656652360515,
|
| 293 |
+
"grad_norm": 3.7180137634277344,
|
| 294 |
+
"learning_rate": 9.837639456697802e-06,
|
| 295 |
+
"loss": 0.2983,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.18025751072961374,
|
| 300 |
+
"grad_norm": 2.5908358097076416,
|
| 301 |
+
"learning_rate": 9.818096470102067e-06,
|
| 302 |
+
"loss": 0.2669,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.18454935622317598,
|
| 307 |
+
"grad_norm": 3.1205332279205322,
|
| 308 |
+
"learning_rate": 9.797464868072489e-06,
|
| 309 |
+
"loss": 0.2712,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1888412017167382,
|
| 314 |
+
"grad_norm": 2.6999526023864746,
|
| 315 |
+
"learning_rate": 9.775749312176249e-06,
|
| 316 |
+
"loss": 0.2755,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.19313304721030042,
|
| 321 |
+
"grad_norm": 3.2999775409698486,
|
| 322 |
+
"learning_rate": 9.752954708892379e-06,
|
| 323 |
+
"loss": 0.2785,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.19742489270386265,
|
| 328 |
+
"grad_norm": 3.793800115585327,
|
| 329 |
+
"learning_rate": 9.729086208503174e-06,
|
| 330 |
+
"loss": 0.2803,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2017167381974249,
|
| 335 |
+
"grad_norm": 2.8122568130493164,
|
| 336 |
+
"learning_rate": 9.704149203930522e-06,
|
| 337 |
+
"loss": 0.2819,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.20600858369098712,
|
| 342 |
+
"grad_norm": 3.148585557937622,
|
| 343 |
+
"learning_rate": 9.67814932951741e-06,
|
| 344 |
+
"loss": 0.3123,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.21030042918454936,
|
| 349 |
+
"grad_norm": 3.0467166900634766,
|
| 350 |
+
"learning_rate": 9.651092459754879e-06,
|
| 351 |
+
"loss": 0.3074,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2145922746781116,
|
| 356 |
+
"grad_norm": 2.846820831298828,
|
| 357 |
+
"learning_rate": 9.622984707954732e-06,
|
| 358 |
+
"loss": 0.2867,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.21888412017167383,
|
| 363 |
+
"grad_norm": 3.011904239654541,
|
| 364 |
+
"learning_rate": 9.593832424868271e-06,
|
| 365 |
+
"loss": 0.2857,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.22317596566523606,
|
| 370 |
+
"grad_norm": 3.032160758972168,
|
| 371 |
+
"learning_rate": 9.563642197251382e-06,
|
| 372 |
+
"loss": 0.312,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.22746781115879827,
|
| 377 |
+
"grad_norm": 3.2738237380981445,
|
| 378 |
+
"learning_rate": 9.532420846376316e-06,
|
| 379 |
+
"loss": 0.273,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2317596566523605,
|
| 384 |
+
"grad_norm": 2.755352735519409,
|
| 385 |
+
"learning_rate": 9.500175426490455e-06,
|
| 386 |
+
"loss": 0.2749,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.23605150214592274,
|
| 391 |
+
"grad_norm": 2.5805509090423584,
|
| 392 |
+
"learning_rate": 9.466913223222467e-06,
|
| 393 |
+
"loss": 0.2828,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.24034334763948498,
|
| 398 |
+
"grad_norm": 2.1449105739593506,
|
| 399 |
+
"learning_rate": 9.432641751936162e-06,
|
| 400 |
+
"loss": 0.2776,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2446351931330472,
|
| 405 |
+
"grad_norm": 2.747648000717163,
|
| 406 |
+
"learning_rate": 9.397368756032445e-06,
|
| 407 |
+
"loss": 0.2859,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.24892703862660945,
|
| 412 |
+
"grad_norm": 2.4565420150756836,
|
| 413 |
+
"learning_rate": 9.361102205199762e-06,
|
| 414 |
+
"loss": 0.2987,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2532188841201717,
|
| 419 |
+
"grad_norm": 2.3403310775756836,
|
| 420 |
+
"learning_rate": 9.32385029361338e-06,
|
| 421 |
+
"loss": 0.2766,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2575107296137339,
|
| 426 |
+
"grad_norm": 2.400320053100586,
|
| 427 |
+
"learning_rate": 9.285621438083997e-06,
|
| 428 |
+
"loss": 0.2477,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.26180257510729615,
|
| 433 |
+
"grad_norm": 2.79807710647583,
|
| 434 |
+
"learning_rate": 9.246424276156008e-06,
|
| 435 |
+
"loss": 0.2699,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.26609442060085836,
|
| 440 |
+
"grad_norm": 2.786733627319336,
|
| 441 |
+
"learning_rate": 9.206267664155906e-06,
|
| 442 |
+
"loss": 0.2822,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2703862660944206,
|
| 447 |
+
"grad_norm": 2.586775302886963,
|
| 448 |
+
"learning_rate": 9.165160675191272e-06,
|
| 449 |
+
"loss": 0.2739,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.27467811158798283,
|
| 454 |
+
"grad_norm": 2.561934471130371,
|
| 455 |
+
"learning_rate": 9.123112597100759e-06,
|
| 456 |
+
"loss": 0.2809,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.27896995708154504,
|
| 461 |
+
"grad_norm": 2.8930823802948,
|
| 462 |
+
"learning_rate": 9.080132930355567e-06,
|
| 463 |
+
"loss": 0.2733,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2832618025751073,
|
| 468 |
+
"grad_norm": 2.4183900356292725,
|
| 469 |
+
"learning_rate": 9.03623138591289e-06,
|
| 470 |
+
"loss": 0.2667,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2875536480686695,
|
| 475 |
+
"grad_norm": 2.3127546310424805,
|
| 476 |
+
"learning_rate": 8.99141788302178e-06,
|
| 477 |
+
"loss": 0.2586,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2918454935622318,
|
| 482 |
+
"grad_norm": 1.8758937120437622,
|
| 483 |
+
"learning_rate": 8.94570254698197e-06,
|
| 484 |
+
"loss": 0.2411,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.296137339055794,
|
| 489 |
+
"grad_norm": 2.9779796600341797,
|
| 490 |
+
"learning_rate": 8.899095706856122e-06,
|
| 491 |
+
"loss": 0.2645,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.30042918454935624,
|
| 496 |
+
"grad_norm": 2.948882579803467,
|
| 497 |
+
"learning_rate": 8.851607893136065e-06,
|
| 498 |
+
"loss": 0.2872,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.30472103004291845,
|
| 503 |
+
"grad_norm": 2.4292919635772705,
|
| 504 |
+
"learning_rate": 8.803249835363486e-06,
|
| 505 |
+
"loss": 0.247,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3090128755364807,
|
| 510 |
+
"grad_norm": 2.62158465385437,
|
| 511 |
+
"learning_rate": 8.754032459705672e-06,
|
| 512 |
+
"loss": 0.2903,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3133047210300429,
|
| 517 |
+
"grad_norm": 2.522172451019287,
|
| 518 |
+
"learning_rate": 8.703966886486819e-06,
|
| 519 |
+
"loss": 0.2734,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.31759656652360513,
|
| 524 |
+
"grad_norm": 2.3345656394958496,
|
| 525 |
+
"learning_rate": 8.65306442767547e-06,
|
| 526 |
+
"loss": 0.2551,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3218884120171674,
|
| 531 |
+
"grad_norm": 2.487061023712158,
|
| 532 |
+
"learning_rate": 8.601336584328659e-06,
|
| 533 |
+
"loss": 0.2848,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3261802575107296,
|
| 538 |
+
"grad_norm": 2.012143611907959,
|
| 539 |
+
"learning_rate": 8.548795043993316e-06,
|
| 540 |
+
"loss": 0.2493,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.33047210300429186,
|
| 545 |
+
"grad_norm": 2.668299436569214,
|
| 546 |
+
"learning_rate": 8.495451678065563e-06,
|
| 547 |
+
"loss": 0.284,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.33476394849785407,
|
| 552 |
+
"grad_norm": 2.1952147483825684,
|
| 553 |
+
"learning_rate": 8.441318539108433e-06,
|
| 554 |
+
"loss": 0.2868,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.33905579399141633,
|
| 559 |
+
"grad_norm": 2.1798348426818848,
|
| 560 |
+
"learning_rate": 8.386407858128707e-06,
|
| 561 |
+
"loss": 0.2732,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.34334763948497854,
|
| 566 |
+
"grad_norm": 2.0891451835632324,
|
| 567 |
+
"learning_rate": 8.330732041813367e-06,
|
| 568 |
+
"loss": 0.2509,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.34763948497854075,
|
| 573 |
+
"grad_norm": 2.444913148880005,
|
| 574 |
+
"learning_rate": 8.274303669726427e-06,
|
| 575 |
+
"loss": 0.2836,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.351931330472103,
|
| 580 |
+
"grad_norm": 2.4908130168914795,
|
| 581 |
+
"learning_rate": 8.217135491466636e-06,
|
| 582 |
+
"loss": 0.2891,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3562231759656652,
|
| 587 |
+
"grad_norm": 2.058657169342041,
|
| 588 |
+
"learning_rate": 8.15924042378682e-06,
|
| 589 |
+
"loss": 0.2482,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3605150214592275,
|
| 594 |
+
"grad_norm": 1.9691925048828125,
|
| 595 |
+
"learning_rate": 8.100631547675417e-06,
|
| 596 |
+
"loss": 0.2429,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3648068669527897,
|
| 601 |
+
"grad_norm": 2.1856772899627686,
|
| 602 |
+
"learning_rate": 8.041322105400923e-06,
|
| 603 |
+
"loss": 0.2823,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.36909871244635195,
|
| 608 |
+
"grad_norm": 2.004932165145874,
|
| 609 |
+
"learning_rate": 7.981325497519892e-06,
|
| 610 |
+
"loss": 0.2552,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.37339055793991416,
|
| 615 |
+
"grad_norm": 2.2658703327178955,
|
| 616 |
+
"learning_rate": 7.920655279849173e-06,
|
| 617 |
+
"loss": 0.264,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3776824034334764,
|
| 622 |
+
"grad_norm": 2.124732255935669,
|
| 623 |
+
"learning_rate": 7.859325160403073e-06,
|
| 624 |
+
"loss": 0.2842,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.38197424892703863,
|
| 629 |
+
"grad_norm": 2.223417043685913,
|
| 630 |
+
"learning_rate": 7.797348996296116e-06,
|
| 631 |
+
"loss": 0.2552,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.38626609442060084,
|
| 636 |
+
"grad_norm": 2.6624255180358887,
|
| 637 |
+
"learning_rate": 7.734740790612137e-06,
|
| 638 |
+
"loss": 0.2935,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3905579399141631,
|
| 643 |
+
"grad_norm": 3.1369097232818604,
|
| 644 |
+
"learning_rate": 7.671514689240366e-06,
|
| 645 |
+
"loss": 0.2568,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3948497854077253,
|
| 650 |
+
"grad_norm": 2.0869433879852295,
|
| 651 |
+
"learning_rate": 7.607684977679284e-06,
|
| 652 |
+
"loss": 0.2414,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.39914163090128757,
|
| 657 |
+
"grad_norm": 2.3208043575286865,
|
| 658 |
+
"learning_rate": 7.543266077808893e-06,
|
| 659 |
+
"loss": 0.2272,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4034334763948498,
|
| 664 |
+
"grad_norm": 2.5530765056610107,
|
| 665 |
+
"learning_rate": 7.478272544632204e-06,
|
| 666 |
+
"loss": 0.2636,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.40772532188841204,
|
| 671 |
+
"grad_norm": 2.2211928367614746,
|
| 672 |
+
"learning_rate": 7.412719062986632e-06,
|
| 673 |
+
"loss": 0.282,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.41201716738197425,
|
| 678 |
+
"grad_norm": 1.9540337324142456,
|
| 679 |
+
"learning_rate": 7.3466204442260605e-06,
|
| 680 |
+
"loss": 0.2303,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.41630901287553645,
|
| 685 |
+
"grad_norm": 2.4863698482513428,
|
| 686 |
+
"learning_rate": 7.279991622874319e-06,
|
| 687 |
+
"loss": 0.2969,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4206008583690987,
|
| 692 |
+
"grad_norm": 1.811273217201233,
|
| 693 |
+
"learning_rate": 7.212847653250828e-06,
|
| 694 |
+
"loss": 0.2439,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4248927038626609,
|
| 699 |
+
"grad_norm": 1.7187399864196777,
|
| 700 |
+
"learning_rate": 7.145203706069183e-06,
|
| 701 |
+
"loss": 0.2353,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4291845493562232,
|
| 706 |
+
"grad_norm": 2.329214572906494,
|
| 707 |
+
"learning_rate": 7.0770750650094335e-06,
|
| 708 |
+
"loss": 0.2718,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4334763948497854,
|
| 713 |
+
"grad_norm": 2.300833225250244,
|
| 714 |
+
"learning_rate": 7.008477123264849e-06,
|
| 715 |
+
"loss": 0.2815,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.43776824034334766,
|
| 720 |
+
"grad_norm": 2.1259753704071045,
|
| 721 |
+
"learning_rate": 6.939425380063924e-06,
|
| 722 |
+
"loss": 0.2544,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.44206008583690987,
|
| 727 |
+
"grad_norm": 2.159365177154541,
|
| 728 |
+
"learning_rate": 6.869935437168449e-06,
|
| 729 |
+
"loss": 0.2632,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.44635193133047213,
|
| 734 |
+
"grad_norm": 2.122368812561035,
|
| 735 |
+
"learning_rate": 6.800022995348381e-06,
|
| 736 |
+
"loss": 0.2562,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.45064377682403434,
|
| 741 |
+
"grad_norm": 1.8842718601226807,
|
| 742 |
+
"learning_rate": 6.729703850834381e-06,
|
| 743 |
+
"loss": 0.2687,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.45493562231759654,
|
| 748 |
+
"grad_norm": 1.6606192588806152,
|
| 749 |
+
"learning_rate": 6.65899389174876e-06,
|
| 750 |
+
"loss": 0.2289,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4592274678111588,
|
| 755 |
+
"grad_norm": 2.395519256591797,
|
| 756 |
+
"learning_rate": 6.587909094515663e-06,
|
| 757 |
+
"loss": 0.2487,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.463519313304721,
|
| 762 |
+
"grad_norm": 1.7725951671600342,
|
| 763 |
+
"learning_rate": 6.5164655202513135e-06,
|
| 764 |
+
"loss": 0.2341,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4678111587982833,
|
| 769 |
+
"grad_norm": 2.764538288116455,
|
| 770 |
+
"learning_rate": 6.444679311135112e-06,
|
| 771 |
+
"loss": 0.2704,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4721030042918455,
|
| 776 |
+
"grad_norm": 2.5796592235565186,
|
| 777 |
+
"learning_rate": 6.372566686762427e-06,
|
| 778 |
+
"loss": 0.2878,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.47639484978540775,
|
| 783 |
+
"grad_norm": 2.3967843055725098,
|
| 784 |
+
"learning_rate": 6.300143940479881e-06,
|
| 785 |
+
"loss": 0.2606,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.48068669527896996,
|
| 790 |
+
"grad_norm": 2.054126739501953,
|
| 791 |
+
"learning_rate": 6.227427435703997e-06,
|
| 792 |
+
"loss": 0.2278,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.48497854077253216,
|
| 797 |
+
"grad_norm": 2.0056793689727783,
|
| 798 |
+
"learning_rate": 6.154433602223979e-06,
|
| 799 |
+
"loss": 0.2131,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4892703862660944,
|
| 804 |
+
"grad_norm": 2.2301571369171143,
|
| 805 |
+
"learning_rate": 6.0811789324895365e-06,
|
| 806 |
+
"loss": 0.2437,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.49356223175965663,
|
| 811 |
+
"grad_norm": 2.192481517791748,
|
| 812 |
+
"learning_rate": 6.0076799778845105e-06,
|
| 813 |
+
"loss": 0.2484,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4978540772532189,
|
| 818 |
+
"grad_norm": 1.9841991662979126,
|
| 819 |
+
"learning_rate": 5.933953344987215e-06,
|
| 820 |
+
"loss": 0.2561,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.5021459227467812,
|
| 825 |
+
"grad_norm": 2.0496819019317627,
|
| 826 |
+
"learning_rate": 5.860015691818292e-06,
|
| 827 |
+
"loss": 0.2586,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.5064377682403434,
|
| 832 |
+
"grad_norm": 1.8923070430755615,
|
| 833 |
+
"learning_rate": 5.78588372407695e-06,
|
| 834 |
+
"loss": 0.2329,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.5107296137339056,
|
| 839 |
+
"grad_norm": 1.734017252922058,
|
| 840 |
+
"learning_rate": 5.711574191366427e-06,
|
| 841 |
+
"loss": 0.2484,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.5150214592274678,
|
| 846 |
+
"grad_norm": 1.7440507411956787,
|
| 847 |
+
"learning_rate": 5.637103883409525e-06,
|
| 848 |
+
"loss": 0.231,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.51931330472103,
|
| 853 |
+
"grad_norm": 2.1365606784820557,
|
| 854 |
+
"learning_rate": 5.562489626255104e-06,
|
| 855 |
+
"loss": 0.2572,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5236051502145923,
|
| 860 |
+
"grad_norm": 2.0460870265960693,
|
| 861 |
+
"learning_rate": 5.487748278476342e-06,
|
| 862 |
+
"loss": 0.2535,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5278969957081545,
|
| 867 |
+
"grad_norm": 1.7586556673049927,
|
| 868 |
+
"learning_rate": 5.412896727361663e-06,
|
| 869 |
+
"loss": 0.2112,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5321888412017167,
|
| 874 |
+
"grad_norm": 2.070552110671997,
|
| 875 |
+
"learning_rate": 5.337951885099167e-06,
|
| 876 |
+
"loss": 0.2179,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5364806866952789,
|
| 881 |
+
"grad_norm": 1.604686975479126,
|
| 882 |
+
"learning_rate": 5.262930684955439e-06,
|
| 883 |
+
"loss": 0.2321,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5407725321888412,
|
| 888 |
+
"grad_norm": 2.326280117034912,
|
| 889 |
+
"learning_rate": 5.187850077449604e-06,
|
| 890 |
+
"loss": 0.2568,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5450643776824035,
|
| 895 |
+
"grad_norm": 2.1093199253082275,
|
| 896 |
+
"learning_rate": 5.112727026523461e-06,
|
| 897 |
+
"loss": 0.22,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5493562231759657,
|
| 902 |
+
"grad_norm": 1.8710286617279053,
|
| 903 |
+
"learning_rate": 5.03757850570861e-06,
|
| 904 |
+
"loss": 0.242,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5536480686695279,
|
| 909 |
+
"grad_norm": 1.5934395790100098,
|
| 910 |
+
"learning_rate": 4.9624214942913916e-06,
|
| 911 |
+
"loss": 0.2369,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5579399141630901,
|
| 916 |
+
"grad_norm": 1.6365737915039062,
|
| 917 |
+
"learning_rate": 4.88727297347654e-06,
|
| 918 |
+
"loss": 0.2463,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5622317596566524,
|
| 923 |
+
"grad_norm": 1.8989615440368652,
|
| 924 |
+
"learning_rate": 4.8121499225503974e-06,
|
| 925 |
+
"loss": 0.2265,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5665236051502146,
|
| 930 |
+
"grad_norm": 1.747003436088562,
|
| 931 |
+
"learning_rate": 4.737069315044562e-06,
|
| 932 |
+
"loss": 0.2292,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5708154506437768,
|
| 937 |
+
"grad_norm": 2.0357205867767334,
|
| 938 |
+
"learning_rate": 4.662048114900837e-06,
|
| 939 |
+
"loss": 0.2605,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.575107296137339,
|
| 944 |
+
"grad_norm": 2.0515100955963135,
|
| 945 |
+
"learning_rate": 4.587103272638339e-06,
|
| 946 |
+
"loss": 0.2586,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5793991416309013,
|
| 951 |
+
"grad_norm": 1.8800376653671265,
|
| 952 |
+
"learning_rate": 4.512251721523659e-06,
|
| 953 |
+
"loss": 0.2161,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5836909871244635,
|
| 958 |
+
"grad_norm": 2.1146693229675293,
|
| 959 |
+
"learning_rate": 4.437510373744897e-06,
|
| 960 |
+
"loss": 0.2449,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5879828326180258,
|
| 965 |
+
"grad_norm": 1.944589376449585,
|
| 966 |
+
"learning_rate": 4.362896116590475e-06,
|
| 967 |
+
"loss": 0.2438,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.592274678111588,
|
| 972 |
+
"grad_norm": 2.022193431854248,
|
| 973 |
+
"learning_rate": 4.2884258086335755e-06,
|
| 974 |
+
"loss": 0.2528,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5965665236051502,
|
| 979 |
+
"grad_norm": 1.5787066221237183,
|
| 980 |
+
"learning_rate": 4.214116275923051e-06,
|
| 981 |
+
"loss": 0.2198,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.6008583690987125,
|
| 986 |
+
"grad_norm": 2.127326726913452,
|
| 987 |
+
"learning_rate": 4.1399843081817085e-06,
|
| 988 |
+
"loss": 0.2428,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.6051502145922747,
|
| 993 |
+
"grad_norm": 1.6171480417251587,
|
| 994 |
+
"learning_rate": 4.066046655012786e-06,
|
| 995 |
+
"loss": 0.2361,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.6094420600858369,
|
| 1000 |
+
"grad_norm": 1.6328740119934082,
|
| 1001 |
+
"learning_rate": 3.992320022115492e-06,
|
| 1002 |
+
"loss": 0.2333,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.6137339055793991,
|
| 1007 |
+
"grad_norm": 1.5752973556518555,
|
| 1008 |
+
"learning_rate": 3.918821067510464e-06,
|
| 1009 |
+
"loss": 0.2159,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.6180257510729614,
|
| 1014 |
+
"grad_norm": 2.487964630126953,
|
| 1015 |
+
"learning_rate": 3.845566397776022e-06,
|
| 1016 |
+
"loss": 0.2619,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6223175965665236,
|
| 1021 |
+
"grad_norm": 2.5639259815216064,
|
| 1022 |
+
"learning_rate": 3.7725725642960047e-06,
|
| 1023 |
+
"loss": 0.2361,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6266094420600858,
|
| 1028 |
+
"grad_norm": 1.714277982711792,
|
| 1029 |
+
"learning_rate": 3.6998560595201188e-06,
|
| 1030 |
+
"loss": 0.2218,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.630901287553648,
|
| 1035 |
+
"grad_norm": 1.72953462600708,
|
| 1036 |
+
"learning_rate": 3.627433313237576e-06,
|
| 1037 |
+
"loss": 0.237,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6351931330472103,
|
| 1042 |
+
"grad_norm": 1.7776756286621094,
|
| 1043 |
+
"learning_rate": 3.555320688864889e-06,
|
| 1044 |
+
"loss": 0.2291,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6394849785407726,
|
| 1049 |
+
"grad_norm": 1.651568055152893,
|
| 1050 |
+
"learning_rate": 3.483534479748688e-06,
|
| 1051 |
+
"loss": 0.2326,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6437768240343348,
|
| 1056 |
+
"grad_norm": 1.5333776473999023,
|
| 1057 |
+
"learning_rate": 3.4120909054843375e-06,
|
| 1058 |
+
"loss": 0.1999,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.648068669527897,
|
| 1063 |
+
"grad_norm": 1.4129211902618408,
|
| 1064 |
+
"learning_rate": 3.3410061082512422e-06,
|
| 1065 |
+
"loss": 0.2014,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6523605150214592,
|
| 1070 |
+
"grad_norm": 1.9737615585327148,
|
| 1071 |
+
"learning_rate": 3.2702961491656197e-06,
|
| 1072 |
+
"loss": 0.2184,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6566523605150214,
|
| 1077 |
+
"grad_norm": 1.5220732688903809,
|
| 1078 |
+
"learning_rate": 3.1999770046516198e-06,
|
| 1079 |
+
"loss": 0.2179,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6609442060085837,
|
| 1084 |
+
"grad_norm": 1.649389386177063,
|
| 1085 |
+
"learning_rate": 3.130064562831553e-06,
|
| 1086 |
+
"loss": 0.2097,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6652360515021459,
|
| 1091 |
+
"grad_norm": 1.6441725492477417,
|
| 1092 |
+
"learning_rate": 3.0605746199360755e-06,
|
| 1093 |
+
"loss": 0.2267,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6695278969957081,
|
| 1098 |
+
"grad_norm": 1.4296993017196655,
|
| 1099 |
+
"learning_rate": 2.991522876735154e-06,
|
| 1100 |
+
"loss": 0.2149,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6738197424892703,
|
| 1105 |
+
"grad_norm": 2.1046576499938965,
|
| 1106 |
+
"learning_rate": 2.9229249349905686e-06,
|
| 1107 |
+
"loss": 0.2447,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6781115879828327,
|
| 1112 |
+
"grad_norm": 1.5854638814926147,
|
| 1113 |
+
"learning_rate": 2.8547962939308187e-06,
|
| 1114 |
+
"loss": 0.2071,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6824034334763949,
|
| 1119 |
+
"grad_norm": 1.6980681419372559,
|
| 1120 |
+
"learning_rate": 2.787152346749173e-06,
|
| 1121 |
+
"loss": 0.2211,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6866952789699571,
|
| 1126 |
+
"grad_norm": 1.6970704793930054,
|
| 1127 |
+
"learning_rate": 2.720008377125682e-06,
|
| 1128 |
+
"loss": 0.2156,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6909871244635193,
|
| 1133 |
+
"grad_norm": 1.894504427909851,
|
| 1134 |
+
"learning_rate": 2.6533795557739407e-06,
|
| 1135 |
+
"loss": 0.2499,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6952789699570815,
|
| 1140 |
+
"grad_norm": 1.8777060508728027,
|
| 1141 |
+
"learning_rate": 2.5872809370133704e-06,
|
| 1142 |
+
"loss": 0.2158,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6995708154506438,
|
| 1147 |
+
"grad_norm": 3.1802759170532227,
|
| 1148 |
+
"learning_rate": 2.5217274553677975e-06,
|
| 1149 |
+
"loss": 0.2248,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.703862660944206,
|
| 1154 |
+
"grad_norm": 1.714854121208191,
|
| 1155 |
+
"learning_rate": 2.4567339221911086e-06,
|
| 1156 |
+
"loss": 0.2338,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.7081545064377682,
|
| 1161 |
+
"grad_norm": 1.6652601957321167,
|
| 1162 |
+
"learning_rate": 2.3923150223207176e-06,
|
| 1163 |
+
"loss": 0.2411,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.7124463519313304,
|
| 1168 |
+
"grad_norm": 1.497910976409912,
|
| 1169 |
+
"learning_rate": 2.328485310759635e-06,
|
| 1170 |
+
"loss": 0.2089,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.7167381974248928,
|
| 1175 |
+
"grad_norm": 1.593087911605835,
|
| 1176 |
+
"learning_rate": 2.265259209387867e-06,
|
| 1177 |
+
"loss": 0.2284,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.721030042918455,
|
| 1182 |
+
"grad_norm": 1.6965844631195068,
|
| 1183 |
+
"learning_rate": 2.202651003703885e-06,
|
| 1184 |
+
"loss": 0.2139,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.7253218884120172,
|
| 1189 |
+
"grad_norm": 1.6895867586135864,
|
| 1190 |
+
"learning_rate": 2.140674839596931e-06,
|
| 1191 |
+
"loss": 0.209,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7296137339055794,
|
| 1196 |
+
"grad_norm": 1.9336439371109009,
|
| 1197 |
+
"learning_rate": 2.0793447201508288e-06,
|
| 1198 |
+
"loss": 0.2235,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7339055793991416,
|
| 1203 |
+
"grad_norm": 1.726787805557251,
|
| 1204 |
+
"learning_rate": 2.01867450248011e-06,
|
| 1205 |
+
"loss": 0.2363,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7381974248927039,
|
| 1210 |
+
"grad_norm": 1.6296881437301636,
|
| 1211 |
+
"learning_rate": 1.9586778945990785e-06,
|
| 1212 |
+
"loss": 0.2111,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7424892703862661,
|
| 1217 |
+
"grad_norm": 1.867713212966919,
|
| 1218 |
+
"learning_rate": 1.8993684523245842e-06,
|
| 1219 |
+
"loss": 0.2041,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7467811158798283,
|
| 1224 |
+
"grad_norm": 1.8173270225524902,
|
| 1225 |
+
"learning_rate": 1.8407595762131814e-06,
|
| 1226 |
+
"loss": 0.254,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7510729613733905,
|
| 1231 |
+
"grad_norm": 1.9783226251602173,
|
| 1232 |
+
"learning_rate": 1.7828645085333645e-06,
|
| 1233 |
+
"loss": 0.2478,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7553648068669528,
|
| 1238 |
+
"grad_norm": 1.5777461528778076,
|
| 1239 |
+
"learning_rate": 1.7256963302735752e-06,
|
| 1240 |
+
"loss": 0.2382,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.759656652360515,
|
| 1245 |
+
"grad_norm": 1.6963618993759155,
|
| 1246 |
+
"learning_rate": 1.6692679581866334e-06,
|
| 1247 |
+
"loss": 0.2268,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.7639484978540773,
|
| 1252 |
+
"grad_norm": 1.4376635551452637,
|
| 1253 |
+
"learning_rate": 1.6135921418712959e-06,
|
| 1254 |
+
"loss": 0.1918,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7682403433476395,
|
| 1259 |
+
"grad_norm": 1.6802223920822144,
|
| 1260 |
+
"learning_rate": 1.5586814608915673e-06,
|
| 1261 |
+
"loss": 0.2403,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7725321888412017,
|
| 1266 |
+
"grad_norm": 1.6209694147109985,
|
| 1267 |
+
"learning_rate": 1.5045483219344387e-06,
|
| 1268 |
+
"loss": 0.2145,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.776824034334764,
|
| 1273 |
+
"grad_norm": 1.8394389152526855,
|
| 1274 |
+
"learning_rate": 1.4512049560066837e-06,
|
| 1275 |
+
"loss": 0.2147,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7811158798283262,
|
| 1280 |
+
"grad_norm": 1.6125578880310059,
|
| 1281 |
+
"learning_rate": 1.3986634156713418e-06,
|
| 1282 |
+
"loss": 0.2106,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7854077253218884,
|
| 1287 |
+
"grad_norm": 1.5914397239685059,
|
| 1288 |
+
"learning_rate": 1.3469355723245303e-06,
|
| 1289 |
+
"loss": 0.2064,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7896995708154506,
|
| 1294 |
+
"grad_norm": 2.1572067737579346,
|
| 1295 |
+
"learning_rate": 1.2960331135131826e-06,
|
| 1296 |
+
"loss": 0.2461,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7939914163090128,
|
| 1301 |
+
"grad_norm": 1.6950761079788208,
|
| 1302 |
+
"learning_rate": 1.245967540294329e-06,
|
| 1303 |
+
"loss": 0.2076,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7982832618025751,
|
| 1308 |
+
"grad_norm": 1.7178866863250732,
|
| 1309 |
+
"learning_rate": 1.1967501646365147e-06,
|
| 1310 |
+
"loss": 0.2116,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.8025751072961373,
|
| 1315 |
+
"grad_norm": 1.9565534591674805,
|
| 1316 |
+
"learning_rate": 1.1483921068639353e-06,
|
| 1317 |
+
"loss": 0.1959,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.8068669527896996,
|
| 1322 |
+
"grad_norm": 2.3265628814697266,
|
| 1323 |
+
"learning_rate": 1.1009042931438784e-06,
|
| 1324 |
+
"loss": 0.222,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.8111587982832618,
|
| 1329 |
+
"grad_norm": 1.572929859161377,
|
| 1330 |
+
"learning_rate": 1.0542974530180327e-06,
|
| 1331 |
+
"loss": 0.216,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.8154506437768241,
|
| 1336 |
+
"grad_norm": 1.3902196884155273,
|
| 1337 |
+
"learning_rate": 1.00858211697822e-06,
|
| 1338 |
+
"loss": 0.1998,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.8197424892703863,
|
| 1343 |
+
"grad_norm": 1.702994465827942,
|
| 1344 |
+
"learning_rate": 9.637686140871121e-07,
|
| 1345 |
+
"loss": 0.2223,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.8240343347639485,
|
| 1350 |
+
"grad_norm": 1.6051976680755615,
|
| 1351 |
+
"learning_rate": 9.198670696444339e-07,
|
| 1352 |
+
"loss": 0.2146,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.8283261802575107,
|
| 1357 |
+
"grad_norm": 1.641396403312683,
|
| 1358 |
+
"learning_rate": 8.768874028992431e-07,
|
| 1359 |
+
"loss": 0.2132,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.8326180257510729,
|
| 1364 |
+
"grad_norm": 1.9438247680664062,
|
| 1365 |
+
"learning_rate": 8.348393248087289e-07,
|
| 1366 |
+
"loss": 0.2347,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.8369098712446352,
|
| 1371 |
+
"grad_norm": 1.9501413106918335,
|
| 1372 |
+
"learning_rate": 7.937323358440935e-07,
|
| 1373 |
+
"loss": 0.2492,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8412017167381974,
|
| 1378 |
+
"grad_norm": 1.5952423810958862,
|
| 1379 |
+
"learning_rate": 7.535757238439939e-07,
|
| 1380 |
+
"loss": 0.2213,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.8454935622317596,
|
| 1385 |
+
"grad_norm": 1.4453738927841187,
|
| 1386 |
+
"learning_rate": 7.143785619160026e-07,
|
| 1387 |
+
"loss": 0.1981,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8497854077253219,
|
| 1392 |
+
"grad_norm": 1.3863543272018433,
|
| 1393 |
+
"learning_rate": 6.761497063866207e-07,
|
| 1394 |
+
"loss": 0.2005,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8540772532188842,
|
| 1399 |
+
"grad_norm": 1.553649663925171,
|
| 1400 |
+
"learning_rate": 6.388977948002406e-07,
|
| 1401 |
+
"loss": 0.2026,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.8583690987124464,
|
| 1406 |
+
"grad_norm": 1.4505783319473267,
|
| 1407 |
+
"learning_rate": 6.026312439675553e-07,
|
| 1408 |
+
"loss": 0.1954,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8626609442060086,
|
| 1413 |
+
"grad_norm": 2.247232675552368,
|
| 1414 |
+
"learning_rate": 5.673582480638395e-07,
|
| 1415 |
+
"loss": 0.2119,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8669527896995708,
|
| 1420 |
+
"grad_norm": 1.8728052377700806,
|
| 1421 |
+
"learning_rate": 5.330867767775333e-07,
|
| 1422 |
+
"loss": 0.2398,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.871244635193133,
|
| 1427 |
+
"grad_norm": 1.493640422821045,
|
| 1428 |
+
"learning_rate": 4.998245735095459e-07,
|
| 1429 |
+
"loss": 0.2228,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8755364806866953,
|
| 1434 |
+
"grad_norm": 1.5458440780639648,
|
| 1435 |
+
"learning_rate": 4.6757915362368567e-07,
|
| 1436 |
+
"loss": 0.211,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8798283261802575,
|
| 1441 |
+
"grad_norm": 1.659397840499878,
|
| 1442 |
+
"learning_rate": 4.363578027486187e-07,
|
| 1443 |
+
"loss": 0.2086,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8841201716738197,
|
| 1448 |
+
"grad_norm": 1.587699055671692,
|
| 1449 |
+
"learning_rate": 4.0616757513173123e-07,
|
| 1450 |
+
"loss": 0.2016,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8884120171673819,
|
| 1455 |
+
"grad_norm": 1.5061396360397339,
|
| 1456 |
+
"learning_rate": 3.7701529204526856e-07,
|
| 1457 |
+
"loss": 0.2068,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8927038626609443,
|
| 1462 |
+
"grad_norm": 1.7324118614196777,
|
| 1463 |
+
"learning_rate": 3.4890754024512254e-07,
|
| 1464 |
+
"loss": 0.2208,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8969957081545065,
|
| 1469 |
+
"grad_norm": 1.465354084968567,
|
| 1470 |
+
"learning_rate": 3.2185067048259245e-07,
|
| 1471 |
+
"loss": 0.2151,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.9012875536480687,
|
| 1476 |
+
"grad_norm": 1.370495080947876,
|
| 1477 |
+
"learning_rate": 2.9585079606947843e-07,
|
| 1478 |
+
"loss": 0.1964,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.9055793991416309,
|
| 1483 |
+
"grad_norm": 1.866836428642273,
|
| 1484 |
+
"learning_rate": 2.7091379149682683e-07,
|
| 1485 |
+
"loss": 0.1929,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.9098712446351931,
|
| 1490 |
+
"grad_norm": 1.388830304145813,
|
| 1491 |
+
"learning_rate": 2.470452911076227e-07,
|
| 1492 |
+
"loss": 0.2049,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.9141630901287554,
|
| 1497 |
+
"grad_norm": 1.3401647806167603,
|
| 1498 |
+
"learning_rate": 2.242506878237538e-07,
|
| 1499 |
+
"loss": 0.1975,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.9184549356223176,
|
| 1504 |
+
"grad_norm": 1.3348737955093384,
|
| 1505 |
+
"learning_rate": 2.0253513192751374e-07,
|
| 1506 |
+
"loss": 0.2076,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.9227467811158798,
|
| 1511 |
+
"grad_norm": 1.2677830457687378,
|
| 1512 |
+
"learning_rate": 1.8190352989793325e-07,
|
| 1513 |
+
"loss": 0.1966,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.927038626609442,
|
| 1518 |
+
"grad_norm": 1.590033769607544,
|
| 1519 |
+
"learning_rate": 1.6236054330219853e-07,
|
| 1520 |
+
"loss": 0.2217,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.9313304721030042,
|
| 1525 |
+
"grad_norm": 1.4973030090332031,
|
| 1526 |
+
"learning_rate": 1.439105877423963e-07,
|
| 1527 |
+
"loss": 0.2107,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.9356223175965666,
|
| 1532 |
+
"grad_norm": 1.597288727760315,
|
| 1533 |
+
"learning_rate": 1.2655783185784253e-07,
|
| 1534 |
+
"loss": 0.2267,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.9399141630901288,
|
| 1539 |
+
"grad_norm": 1.720969319343567,
|
| 1540 |
+
"learning_rate": 1.1030619638320805e-07,
|
| 1541 |
+
"loss": 0.2075,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.944206008583691,
|
| 1546 |
+
"grad_norm": 1.4152257442474365,
|
| 1547 |
+
"learning_rate": 9.51593532626538e-08,
|
| 1548 |
+
"loss": 0.2003,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.9484978540772532,
|
| 1553 |
+
"grad_norm": 1.540600299835205,
|
| 1554 |
+
"learning_rate": 8.11207248201834e-08,
|
| 1555 |
+
"loss": 0.2211,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.9527896995708155,
|
| 1560 |
+
"grad_norm": 1.7658554315567017,
|
| 1561 |
+
"learning_rate": 6.819348298638839e-08,
|
| 1562 |
+
"loss": 0.2296,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9570815450643777,
|
| 1567 |
+
"grad_norm": 1.7942489385604858,
|
| 1568 |
+
"learning_rate": 5.638054858177644e-08,
|
| 1569 |
+
"loss": 0.2077,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.9613733905579399,
|
| 1574 |
+
"grad_norm": 1.565927505493164,
|
| 1575 |
+
"learning_rate": 4.568459065683206e-08,
|
| 1576 |
+
"loss": 0.1982,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.9656652360515021,
|
| 1581 |
+
"grad_norm": 1.7370845079421997,
|
| 1582 |
+
"learning_rate": 3.610802588895845e-08,
|
| 1583 |
+
"loss": 0.2303,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9699570815450643,
|
| 1588 |
+
"grad_norm": 1.3400768041610718,
|
| 1589 |
+
"learning_rate": 2.765301803645426e-08,
|
| 1590 |
+
"loss": 0.1816,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.9742489270386266,
|
| 1595 |
+
"grad_norm": 1.6028944253921509,
|
| 1596 |
+
"learning_rate": 2.0321477449619098e-08,
|
| 1597 |
+
"loss": 0.2329,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9785407725321889,
|
| 1602 |
+
"grad_norm": 1.635040044784546,
|
| 1603 |
+
"learning_rate": 1.411506063912882e-08,
|
| 1604 |
+
"loss": 0.2365,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.9828326180257511,
|
| 1609 |
+
"grad_norm": 1.5071089267730713,
|
| 1610 |
+
"learning_rate": 9.035169901754902e-09,
|
| 1611 |
+
"loss": 0.1913,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9871244635193133,
|
| 1616 |
+
"grad_norm": 1.8444362878799438,
|
| 1617 |
+
"learning_rate": 5.082953003528457e-09,
|
| 1618 |
+
"loss": 0.2008,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9914163090128756,
|
| 1623 |
+
"grad_norm": 1.3740102052688599,
|
| 1624 |
+
"learning_rate": 2.2593029204076578e-09,
|
| 1625 |
+
"loss": 0.1962,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9957081545064378,
|
| 1630 |
+
"grad_norm": 1.8837482929229736,
|
| 1631 |
+
"learning_rate": 5.648576365169245e-10,
|
| 1632 |
+
"loss": 0.2168,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 1.0,
|
| 1637 |
+
"grad_norm": 1.6341578960418701,
|
| 1638 |
+
"learning_rate": 0.0,
|
| 1639 |
+
"loss": 0.2153,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.0,
|
| 1644 |
+
"eval_loss": 0.36973223090171814,
|
| 1645 |
+
"eval_runtime": 3.4823,
|
| 1646 |
+
"eval_samples_per_second": 66.049,
|
| 1647 |
+
"eval_steps_per_second": 2.872,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.0,
|
| 1652 |
+
"step": 233,
|
| 1653 |
+
"total_flos": 8.156803263102976e+16,
|
| 1654 |
+
"train_loss": 0.28537011958754627,
|
| 1655 |
+
"train_runtime": 757.3361,
|
| 1656 |
+
"train_samples_per_second": 14.709,
|
| 1657 |
+
"train_steps_per_second": 0.308
|
| 1658 |
+
}
|
| 1659 |
+
],
|
| 1660 |
+
"logging_steps": 1,
|
| 1661 |
+
"max_steps": 233,
|
| 1662 |
+
"num_input_tokens_seen": 0,
|
| 1663 |
+
"num_train_epochs": 1,
|
| 1664 |
+
"save_steps": 800,
|
| 1665 |
+
"stateful_callbacks": {
|
| 1666 |
+
"TrainerControl": {
|
| 1667 |
+
"args": {
|
| 1668 |
+
"should_epoch_stop": false,
|
| 1669 |
+
"should_evaluate": false,
|
| 1670 |
+
"should_log": false,
|
| 1671 |
+
"should_save": false,
|
| 1672 |
+
"should_training_stop": false
|
| 1673 |
+
},
|
| 1674 |
+
"attributes": {}
|
| 1675 |
+
}
|
| 1676 |
+
},
|
| 1677 |
+
"total_flos": 8.156803263102976e+16,
|
| 1678 |
+
"train_batch_size": 16,
|
| 1679 |
+
"trial_name": null,
|
| 1680 |
+
"trial_params": null
|
| 1681 |
+
}
|