Model save
Browse files- README.md +64 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2759 -0
README.md
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: yanolja/EEVE-Korean-Instruct-2.8B-v1.0
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: eeve-2.8B_A
|
| 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-2.8B_A
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [yanolja/EEVE-Korean-Instruct-2.8B-v1.0](https://huggingface.co/yanolja/EEVE-Korean-Instruct-2.8B-v1.0) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3930
|
| 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 |
+
| 0.2752 | 1.0 | 387 | 0.3930 |
|
| 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": 3.228374416542925e+16,
|
| 4 |
+
"train_loss": 0.37563710938898476,
|
| 5 |
+
"train_runtime": 503.7567,
|
| 6 |
+
"train_samples": 18568,
|
| 7 |
+
"train_samples_per_second": 36.859,
|
| 8 |
+
"train_steps_per_second": 0.768
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 3.228374416542925e+16,
|
| 4 |
+
"train_loss": 0.37563710938898476,
|
| 5 |
+
"train_runtime": 503.7567,
|
| 6 |
+
"train_samples": 18568,
|
| 7 |
+
"train_samples_per_second": 36.859,
|
| 8 |
+
"train_steps_per_second": 0.768
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2759 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 387,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002583979328165375,
|
| 13 |
+
"grad_norm": 6.678133964538574,
|
| 14 |
+
"learning_rate": 2.564102564102564e-07,
|
| 15 |
+
"loss": 2.2997,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00516795865633075,
|
| 20 |
+
"grad_norm": 6.834577560424805,
|
| 21 |
+
"learning_rate": 5.128205128205128e-07,
|
| 22 |
+
"loss": 2.3158,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.007751937984496124,
|
| 27 |
+
"grad_norm": 7.110385417938232,
|
| 28 |
+
"learning_rate": 7.692307692307694e-07,
|
| 29 |
+
"loss": 2.3468,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0103359173126615,
|
| 34 |
+
"grad_norm": 6.803109169006348,
|
| 35 |
+
"learning_rate": 1.0256410256410257e-06,
|
| 36 |
+
"loss": 2.2928,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012919896640826873,
|
| 41 |
+
"grad_norm": 6.826089382171631,
|
| 42 |
+
"learning_rate": 1.282051282051282e-06,
|
| 43 |
+
"loss": 2.3315,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.015503875968992248,
|
| 48 |
+
"grad_norm": 6.920953750610352,
|
| 49 |
+
"learning_rate": 1.5384615384615387e-06,
|
| 50 |
+
"loss": 2.3069,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01808785529715762,
|
| 55 |
+
"grad_norm": 6.446414947509766,
|
| 56 |
+
"learning_rate": 1.794871794871795e-06,
|
| 57 |
+
"loss": 2.2685,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.020671834625323,
|
| 62 |
+
"grad_norm": 6.7022786140441895,
|
| 63 |
+
"learning_rate": 2.0512820512820513e-06,
|
| 64 |
+
"loss": 2.2895,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.023255813953488372,
|
| 69 |
+
"grad_norm": 6.589105606079102,
|
| 70 |
+
"learning_rate": 2.307692307692308e-06,
|
| 71 |
+
"loss": 2.2653,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.025839793281653745,
|
| 76 |
+
"grad_norm": 5.955119609832764,
|
| 77 |
+
"learning_rate": 2.564102564102564e-06,
|
| 78 |
+
"loss": 2.1308,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.028423772609819122,
|
| 83 |
+
"grad_norm": 5.912731647491455,
|
| 84 |
+
"learning_rate": 2.8205128205128207e-06,
|
| 85 |
+
"loss": 2.1179,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.031007751937984496,
|
| 90 |
+
"grad_norm": 5.7995123863220215,
|
| 91 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 92 |
+
"loss": 2.0886,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03359173126614987,
|
| 97 |
+
"grad_norm": 4.567681312561035,
|
| 98 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 99 |
+
"loss": 1.7614,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03617571059431524,
|
| 104 |
+
"grad_norm": 4.454052925109863,
|
| 105 |
+
"learning_rate": 3.58974358974359e-06,
|
| 106 |
+
"loss": 1.7257,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03875968992248062,
|
| 111 |
+
"grad_norm": 4.280542850494385,
|
| 112 |
+
"learning_rate": 3.846153846153847e-06,
|
| 113 |
+
"loss": 1.6546,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.041343669250646,
|
| 118 |
+
"grad_norm": 4.197136878967285,
|
| 119 |
+
"learning_rate": 4.102564102564103e-06,
|
| 120 |
+
"loss": 1.5889,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04392764857881137,
|
| 125 |
+
"grad_norm": 3.324878692626953,
|
| 126 |
+
"learning_rate": 4.358974358974359e-06,
|
| 127 |
+
"loss": 1.1931,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.046511627906976744,
|
| 132 |
+
"grad_norm": 3.594575881958008,
|
| 133 |
+
"learning_rate": 4.615384615384616e-06,
|
| 134 |
+
"loss": 1.0317,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04909560723514212,
|
| 139 |
+
"grad_norm": 3.397719144821167,
|
| 140 |
+
"learning_rate": 4.871794871794872e-06,
|
| 141 |
+
"loss": 0.9836,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05167958656330749,
|
| 146 |
+
"grad_norm": 3.1062400341033936,
|
| 147 |
+
"learning_rate": 5.128205128205128e-06,
|
| 148 |
+
"loss": 0.9742,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05426356589147287,
|
| 153 |
+
"grad_norm": 2.650336742401123,
|
| 154 |
+
"learning_rate": 5.384615384615385e-06,
|
| 155 |
+
"loss": 0.8934,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.056847545219638244,
|
| 160 |
+
"grad_norm": 2.466906785964966,
|
| 161 |
+
"learning_rate": 5.641025641025641e-06,
|
| 162 |
+
"loss": 0.8145,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.059431524547803614,
|
| 167 |
+
"grad_norm": 1.7531450986862183,
|
| 168 |
+
"learning_rate": 5.897435897435898e-06,
|
| 169 |
+
"loss": 0.606,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06201550387596899,
|
| 174 |
+
"grad_norm": 1.4234894514083862,
|
| 175 |
+
"learning_rate": 6.153846153846155e-06,
|
| 176 |
+
"loss": 0.4778,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06459948320413436,
|
| 181 |
+
"grad_norm": 1.0901433229446411,
|
| 182 |
+
"learning_rate": 6.410256410256412e-06,
|
| 183 |
+
"loss": 0.4324,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06718346253229975,
|
| 188 |
+
"grad_norm": 0.9714887142181396,
|
| 189 |
+
"learning_rate": 6.666666666666667e-06,
|
| 190 |
+
"loss": 0.4577,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06976744186046512,
|
| 195 |
+
"grad_norm": 0.801122784614563,
|
| 196 |
+
"learning_rate": 6.923076923076923e-06,
|
| 197 |
+
"loss": 0.4528,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07235142118863049,
|
| 202 |
+
"grad_norm": 0.7464026808738708,
|
| 203 |
+
"learning_rate": 7.17948717948718e-06,
|
| 204 |
+
"loss": 0.4109,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07493540051679587,
|
| 209 |
+
"grad_norm": 0.7229853272438049,
|
| 210 |
+
"learning_rate": 7.435897435897437e-06,
|
| 211 |
+
"loss": 0.4129,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07751937984496124,
|
| 216 |
+
"grad_norm": 0.6986636519432068,
|
| 217 |
+
"learning_rate": 7.692307692307694e-06,
|
| 218 |
+
"loss": 0.4276,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08010335917312661,
|
| 223 |
+
"grad_norm": 0.7045499086380005,
|
| 224 |
+
"learning_rate": 7.948717948717949e-06,
|
| 225 |
+
"loss": 0.4173,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.082687338501292,
|
| 230 |
+
"grad_norm": 0.6709201335906982,
|
| 231 |
+
"learning_rate": 8.205128205128205e-06,
|
| 232 |
+
"loss": 0.4033,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08527131782945736,
|
| 237 |
+
"grad_norm": 0.5888183116912842,
|
| 238 |
+
"learning_rate": 8.461538461538462e-06,
|
| 239 |
+
"loss": 0.3339,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08785529715762273,
|
| 244 |
+
"grad_norm": 0.621361255645752,
|
| 245 |
+
"learning_rate": 8.717948717948719e-06,
|
| 246 |
+
"loss": 0.3976,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09043927648578812,
|
| 251 |
+
"grad_norm": 0.5428451299667358,
|
| 252 |
+
"learning_rate": 8.974358974358976e-06,
|
| 253 |
+
"loss": 0.3621,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09302325581395349,
|
| 258 |
+
"grad_norm": 0.5530751347541809,
|
| 259 |
+
"learning_rate": 9.230769230769232e-06,
|
| 260 |
+
"loss": 0.3263,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09560723514211886,
|
| 265 |
+
"grad_norm": 0.6014180779457092,
|
| 266 |
+
"learning_rate": 9.487179487179487e-06,
|
| 267 |
+
"loss": 0.3204,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09819121447028424,
|
| 272 |
+
"grad_norm": 0.617842435836792,
|
| 273 |
+
"learning_rate": 9.743589743589744e-06,
|
| 274 |
+
"loss": 0.3432,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10077519379844961,
|
| 279 |
+
"grad_norm": 0.5932274460792542,
|
| 280 |
+
"learning_rate": 1e-05,
|
| 281 |
+
"loss": 0.3807,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10335917312661498,
|
| 286 |
+
"grad_norm": 0.531089723110199,
|
| 287 |
+
"learning_rate": 9.999796259054765e-06,
|
| 288 |
+
"loss": 0.3466,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10594315245478036,
|
| 293 |
+
"grad_norm": 0.5526779890060425,
|
| 294 |
+
"learning_rate": 9.999185052823207e-06,
|
| 295 |
+
"loss": 0.3495,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10852713178294573,
|
| 300 |
+
"grad_norm": 0.48364171385765076,
|
| 301 |
+
"learning_rate": 9.998166431116421e-06,
|
| 302 |
+
"loss": 0.3392,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1111111111111111,
|
| 307 |
+
"grad_norm": 0.5286124348640442,
|
| 308 |
+
"learning_rate": 9.996740476948386e-06,
|
| 309 |
+
"loss": 0.3237,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11369509043927649,
|
| 314 |
+
"grad_norm": 0.5805378556251526,
|
| 315 |
+
"learning_rate": 9.994907306529203e-06,
|
| 316 |
+
"loss": 0.3871,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11627906976744186,
|
| 321 |
+
"grad_norm": 0.48061248660087585,
|
| 322 |
+
"learning_rate": 9.99266706925562e-06,
|
| 323 |
+
"loss": 0.3396,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11886304909560723,
|
| 328 |
+
"grad_norm": 0.7416213750839233,
|
| 329 |
+
"learning_rate": 9.990019947698864e-06,
|
| 330 |
+
"loss": 0.3259,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12144702842377261,
|
| 335 |
+
"grad_norm": 0.47168344259262085,
|
| 336 |
+
"learning_rate": 9.986966157589751e-06,
|
| 337 |
+
"loss": 0.3197,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12403100775193798,
|
| 342 |
+
"grad_norm": 0.4842292070388794,
|
| 343 |
+
"learning_rate": 9.983505947801115e-06,
|
| 344 |
+
"loss": 0.3641,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12661498708010335,
|
| 349 |
+
"grad_norm": 0.5371739268302917,
|
| 350 |
+
"learning_rate": 9.979639600327522e-06,
|
| 351 |
+
"loss": 0.3614,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12919896640826872,
|
| 356 |
+
"grad_norm": 0.5148357152938843,
|
| 357 |
+
"learning_rate": 9.975367430262289e-06,
|
| 358 |
+
"loss": 0.3728,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13178294573643412,
|
| 363 |
+
"grad_norm": 0.4844192564487457,
|
| 364 |
+
"learning_rate": 9.970689785771798e-06,
|
| 365 |
+
"loss": 0.3208,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1343669250645995,
|
| 370 |
+
"grad_norm": 0.6069826483726501,
|
| 371 |
+
"learning_rate": 9.965607048067138e-06,
|
| 372 |
+
"loss": 0.3513,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13695090439276486,
|
| 377 |
+
"grad_norm": 0.47529327869415283,
|
| 378 |
+
"learning_rate": 9.960119631373023e-06,
|
| 379 |
+
"loss": 0.3379,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.13953488372093023,
|
| 384 |
+
"grad_norm": 0.49472829699516296,
|
| 385 |
+
"learning_rate": 9.954227982894034e-06,
|
| 386 |
+
"loss": 0.3013,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1421188630490956,
|
| 391 |
+
"grad_norm": 0.6334584355354309,
|
| 392 |
+
"learning_rate": 9.947932582778188e-06,
|
| 393 |
+
"loss": 0.3138,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14470284237726097,
|
| 398 |
+
"grad_norm": 0.4577198624610901,
|
| 399 |
+
"learning_rate": 9.941233944077789e-06,
|
| 400 |
+
"loss": 0.3153,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14728682170542637,
|
| 405 |
+
"grad_norm": 0.47782599925994873,
|
| 406 |
+
"learning_rate": 9.934132612707631e-06,
|
| 407 |
+
"loss": 0.329,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.14987080103359174,
|
| 412 |
+
"grad_norm": 0.514714777469635,
|
| 413 |
+
"learning_rate": 9.9266291674005e-06,
|
| 414 |
+
"loss": 0.359,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1524547803617571,
|
| 419 |
+
"grad_norm": 0.4538106322288513,
|
| 420 |
+
"learning_rate": 9.918724219660013e-06,
|
| 421 |
+
"loss": 0.2988,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15503875968992248,
|
| 426 |
+
"grad_norm": 0.46882936358451843,
|
| 427 |
+
"learning_rate": 9.91041841371078e-06,
|
| 428 |
+
"loss": 0.3246,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15762273901808785,
|
| 433 |
+
"grad_norm": 0.429165244102478,
|
| 434 |
+
"learning_rate": 9.901712426445901e-06,
|
| 435 |
+
"loss": 0.2959,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16020671834625322,
|
| 440 |
+
"grad_norm": 0.4619984030723572,
|
| 441 |
+
"learning_rate": 9.892606967371808e-06,
|
| 442 |
+
"loss": 0.3291,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16279069767441862,
|
| 447 |
+
"grad_norm": 0.43021416664123535,
|
| 448 |
+
"learning_rate": 9.883102778550434e-06,
|
| 449 |
+
"loss": 0.3331,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.165374677002584,
|
| 454 |
+
"grad_norm": 0.5037028193473816,
|
| 455 |
+
"learning_rate": 9.873200634538746e-06,
|
| 456 |
+
"loss": 0.3265,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16795865633074936,
|
| 461 |
+
"grad_norm": 0.4535587728023529,
|
| 462 |
+
"learning_rate": 9.862901342325617e-06,
|
| 463 |
+
"loss": 0.3312,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17054263565891473,
|
| 468 |
+
"grad_norm": 0.42938175797462463,
|
| 469 |
+
"learning_rate": 9.852205741266058e-06,
|
| 470 |
+
"loss": 0.2971,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.1731266149870801,
|
| 475 |
+
"grad_norm": 0.4964745342731476,
|
| 476 |
+
"learning_rate": 9.841114703012817e-06,
|
| 477 |
+
"loss": 0.3306,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17571059431524547,
|
| 482 |
+
"grad_norm": 0.4658762216567993,
|
| 483 |
+
"learning_rate": 9.829629131445342e-06,
|
| 484 |
+
"loss": 0.3055,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17829457364341086,
|
| 489 |
+
"grad_norm": 0.4396679699420929,
|
| 490 |
+
"learning_rate": 9.817749962596115e-06,
|
| 491 |
+
"loss": 0.3204,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18087855297157623,
|
| 496 |
+
"grad_norm": 0.44120684266090393,
|
| 497 |
+
"learning_rate": 9.805478164574374e-06,
|
| 498 |
+
"loss": 0.3204,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1834625322997416,
|
| 503 |
+
"grad_norm": 0.4362391531467438,
|
| 504 |
+
"learning_rate": 9.792814737487207e-06,
|
| 505 |
+
"loss": 0.3001,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18604651162790697,
|
| 510 |
+
"grad_norm": 0.45392006635665894,
|
| 511 |
+
"learning_rate": 9.77976071335806e-06,
|
| 512 |
+
"loss": 0.3204,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18863049095607234,
|
| 517 |
+
"grad_norm": 0.4473075270652771,
|
| 518 |
+
"learning_rate": 9.766317156042615e-06,
|
| 519 |
+
"loss": 0.3342,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19121447028423771,
|
| 524 |
+
"grad_norm": 0.45184287428855896,
|
| 525 |
+
"learning_rate": 9.752485161142103e-06,
|
| 526 |
+
"loss": 0.3168,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1937984496124031,
|
| 531 |
+
"grad_norm": 0.4665965735912323,
|
| 532 |
+
"learning_rate": 9.738265855914014e-06,
|
| 533 |
+
"loss": 0.3347,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19638242894056848,
|
| 538 |
+
"grad_norm": 0.44520583748817444,
|
| 539 |
+
"learning_rate": 9.723660399180216e-06,
|
| 540 |
+
"loss": 0.3012,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.19896640826873385,
|
| 545 |
+
"grad_norm": 0.4683307111263275,
|
| 546 |
+
"learning_rate": 9.708669981232542e-06,
|
| 547 |
+
"loss": 0.2947,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.20155038759689922,
|
| 552 |
+
"grad_norm": 0.5088106989860535,
|
| 553 |
+
"learning_rate": 9.693295823735754e-06,
|
| 554 |
+
"loss": 0.3519,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2041343669250646,
|
| 559 |
+
"grad_norm": 0.4688245952129364,
|
| 560 |
+
"learning_rate": 9.677539179628005e-06,
|
| 561 |
+
"loss": 0.3178,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20671834625322996,
|
| 566 |
+
"grad_norm": 0.4424334168434143,
|
| 567 |
+
"learning_rate": 9.661401333018725e-06,
|
| 568 |
+
"loss": 0.3098,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.20930232558139536,
|
| 573 |
+
"grad_norm": 0.4164547324180603,
|
| 574 |
+
"learning_rate": 9.644883599083959e-06,
|
| 575 |
+
"loss": 0.2869,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21188630490956073,
|
| 580 |
+
"grad_norm": 0.500719428062439,
|
| 581 |
+
"learning_rate": 9.627987323959195e-06,
|
| 582 |
+
"loss": 0.3038,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2144702842377261,
|
| 587 |
+
"grad_norm": 0.42712247371673584,
|
| 588 |
+
"learning_rate": 9.610713884629667e-06,
|
| 589 |
+
"loss": 0.2932,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21705426356589147,
|
| 594 |
+
"grad_norm": 0.4611964821815491,
|
| 595 |
+
"learning_rate": 9.59306468881811e-06,
|
| 596 |
+
"loss": 0.3099,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.21963824289405684,
|
| 601 |
+
"grad_norm": 0.4781152307987213,
|
| 602 |
+
"learning_rate": 9.575041174870062e-06,
|
| 603 |
+
"loss": 0.3124,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2222222222222222,
|
| 608 |
+
"grad_norm": 0.4377368986606598,
|
| 609 |
+
"learning_rate": 9.556644811636628e-06,
|
| 610 |
+
"loss": 0.3002,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2248062015503876,
|
| 615 |
+
"grad_norm": 0.49812576174736023,
|
| 616 |
+
"learning_rate": 9.537877098354787e-06,
|
| 617 |
+
"loss": 0.348,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.22739018087855298,
|
| 622 |
+
"grad_norm": 0.47140446305274963,
|
| 623 |
+
"learning_rate": 9.51873956452519e-06,
|
| 624 |
+
"loss": 0.3233,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.22997416020671835,
|
| 629 |
+
"grad_norm": 0.45923298597335815,
|
| 630 |
+
"learning_rate": 9.499233769787534e-06,
|
| 631 |
+
"loss": 0.3187,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23255813953488372,
|
| 636 |
+
"grad_norm": 0.45452624559402466,
|
| 637 |
+
"learning_rate": 9.479361303793441e-06,
|
| 638 |
+
"loss": 0.2953,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2351421188630491,
|
| 643 |
+
"grad_norm": 0.5140699744224548,
|
| 644 |
+
"learning_rate": 9.459123786076911e-06,
|
| 645 |
+
"loss": 0.3142,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.23772609819121446,
|
| 650 |
+
"grad_norm": 0.4350361227989197,
|
| 651 |
+
"learning_rate": 9.438522865922344e-06,
|
| 652 |
+
"loss": 0.3193,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24031007751937986,
|
| 657 |
+
"grad_norm": 0.4551020860671997,
|
| 658 |
+
"learning_rate": 9.417560222230115e-06,
|
| 659 |
+
"loss": 0.3084,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24289405684754523,
|
| 664 |
+
"grad_norm": 0.42140695452690125,
|
| 665 |
+
"learning_rate": 9.396237563379761e-06,
|
| 666 |
+
"loss": 0.2798,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2454780361757106,
|
| 671 |
+
"grad_norm": 0.4576823115348816,
|
| 672 |
+
"learning_rate": 9.374556627090749e-06,
|
| 673 |
+
"loss": 0.2936,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.24806201550387597,
|
| 678 |
+
"grad_norm": 0.4687109887599945,
|
| 679 |
+
"learning_rate": 9.352519180280862e-06,
|
| 680 |
+
"loss": 0.3223,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.25064599483204136,
|
| 685 |
+
"grad_norm": 0.42448845505714417,
|
| 686 |
+
"learning_rate": 9.330127018922195e-06,
|
| 687 |
+
"loss": 0.3152,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2532299741602067,
|
| 692 |
+
"grad_norm": 0.46770456433296204,
|
| 693 |
+
"learning_rate": 9.307381967894798e-06,
|
| 694 |
+
"loss": 0.3358,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2558139534883721,
|
| 699 |
+
"grad_norm": 0.4652756154537201,
|
| 700 |
+
"learning_rate": 9.284285880837947e-06,
|
| 701 |
+
"loss": 0.2824,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.25839793281653745,
|
| 706 |
+
"grad_norm": 0.4634130597114563,
|
| 707 |
+
"learning_rate": 9.26084063999909e-06,
|
| 708 |
+
"loss": 0.3005,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.26098191214470284,
|
| 713 |
+
"grad_norm": 0.4184122681617737,
|
| 714 |
+
"learning_rate": 9.237048156080433e-06,
|
| 715 |
+
"loss": 0.2782,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.26356589147286824,
|
| 720 |
+
"grad_norm": 0.4704434275627136,
|
| 721 |
+
"learning_rate": 9.212910368083246e-06,
|
| 722 |
+
"loss": 0.2759,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2661498708010336,
|
| 727 |
+
"grad_norm": 0.4219190776348114,
|
| 728 |
+
"learning_rate": 9.188429243149824e-06,
|
| 729 |
+
"loss": 0.2694,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.268733850129199,
|
| 734 |
+
"grad_norm": 0.4400491416454315,
|
| 735 |
+
"learning_rate": 9.163606776403182e-06,
|
| 736 |
+
"loss": 0.313,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2713178294573643,
|
| 741 |
+
"grad_norm": 0.46378496289253235,
|
| 742 |
+
"learning_rate": 9.138444990784455e-06,
|
| 743 |
+
"loss": 0.3117,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.2739018087855297,
|
| 748 |
+
"grad_norm": 0.42383599281311035,
|
| 749 |
+
"learning_rate": 9.112945936888034e-06,
|
| 750 |
+
"loss": 0.2952,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.27648578811369506,
|
| 755 |
+
"grad_norm": 0.4268609881401062,
|
| 756 |
+
"learning_rate": 9.08711169279446e-06,
|
| 757 |
+
"loss": 0.2607,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.27906976744186046,
|
| 762 |
+
"grad_norm": 0.43639639019966125,
|
| 763 |
+
"learning_rate": 9.060944363901057e-06,
|
| 764 |
+
"loss": 0.2745,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.28165374677002586,
|
| 769 |
+
"grad_norm": 0.4547524154186249,
|
| 770 |
+
"learning_rate": 9.034446082750352e-06,
|
| 771 |
+
"loss": 0.3008,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2842377260981912,
|
| 776 |
+
"grad_norm": 0.4684068560600281,
|
| 777 |
+
"learning_rate": 9.007619008856287e-06,
|
| 778 |
+
"loss": 0.2787,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.2868217054263566,
|
| 783 |
+
"grad_norm": 0.44395485520362854,
|
| 784 |
+
"learning_rate": 8.98046532852822e-06,
|
| 785 |
+
"loss": 0.2683,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.28940568475452194,
|
| 790 |
+
"grad_norm": 0.446803480386734,
|
| 791 |
+
"learning_rate": 8.952987254692746e-06,
|
| 792 |
+
"loss": 0.3084,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.29198966408268734,
|
| 797 |
+
"grad_norm": 0.4414445161819458,
|
| 798 |
+
"learning_rate": 8.925187026713363e-06,
|
| 799 |
+
"loss": 0.2841,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.29457364341085274,
|
| 804 |
+
"grad_norm": 0.43872445821762085,
|
| 805 |
+
"learning_rate": 8.897066910207958e-06,
|
| 806 |
+
"loss": 0.2739,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2971576227390181,
|
| 811 |
+
"grad_norm": 0.4418787956237793,
|
| 812 |
+
"learning_rate": 8.868629196864182e-06,
|
| 813 |
+
"loss": 0.2842,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2997416020671835,
|
| 818 |
+
"grad_norm": 0.4769684374332428,
|
| 819 |
+
"learning_rate": 8.83987620425267e-06,
|
| 820 |
+
"loss": 0.3137,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3023255813953488,
|
| 825 |
+
"grad_norm": 0.43923094868659973,
|
| 826 |
+
"learning_rate": 8.810810275638183e-06,
|
| 827 |
+
"loss": 0.3191,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3049095607235142,
|
| 832 |
+
"grad_norm": 0.4351806044578552,
|
| 833 |
+
"learning_rate": 8.781433779788627e-06,
|
| 834 |
+
"loss": 0.2799,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.30749354005167956,
|
| 839 |
+
"grad_norm": 0.4320261478424072,
|
| 840 |
+
"learning_rate": 8.751749110782013e-06,
|
| 841 |
+
"loss": 0.2849,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.31007751937984496,
|
| 846 |
+
"grad_norm": 0.4509352147579193,
|
| 847 |
+
"learning_rate": 8.721758687811353e-06,
|
| 848 |
+
"loss": 0.3009,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.31266149870801035,
|
| 853 |
+
"grad_norm": 0.4201003313064575,
|
| 854 |
+
"learning_rate": 8.691464954987494e-06,
|
| 855 |
+
"loss": 0.2675,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3152454780361757,
|
| 860 |
+
"grad_norm": 0.42218485474586487,
|
| 861 |
+
"learning_rate": 8.660870381139944e-06,
|
| 862 |
+
"loss": 0.2872,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3178294573643411,
|
| 867 |
+
"grad_norm": 0.41878223419189453,
|
| 868 |
+
"learning_rate": 8.629977459615655e-06,
|
| 869 |
+
"loss": 0.2667,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.32041343669250644,
|
| 874 |
+
"grad_norm": 0.4328330159187317,
|
| 875 |
+
"learning_rate": 8.598788708075844e-06,
|
| 876 |
+
"loss": 0.2916,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32299741602067183,
|
| 881 |
+
"grad_norm": 0.4555003046989441,
|
| 882 |
+
"learning_rate": 8.567306668290801e-06,
|
| 883 |
+
"loss": 0.289,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32558139534883723,
|
| 888 |
+
"grad_norm": 0.4696672558784485,
|
| 889 |
+
"learning_rate": 8.535533905932739e-06,
|
| 890 |
+
"loss": 0.2982,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3281653746770026,
|
| 895 |
+
"grad_norm": 0.4428997337818146,
|
| 896 |
+
"learning_rate": 8.503473010366713e-06,
|
| 897 |
+
"loss": 0.2989,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.330749354005168,
|
| 902 |
+
"grad_norm": 0.40596020221710205,
|
| 903 |
+
"learning_rate": 8.471126594439591e-06,
|
| 904 |
+
"loss": 0.2931,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3333333333333333,
|
| 909 |
+
"grad_norm": 0.4140114486217499,
|
| 910 |
+
"learning_rate": 8.438497294267117e-06,
|
| 911 |
+
"loss": 0.2715,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3359173126614987,
|
| 916 |
+
"grad_norm": 0.47890666127204895,
|
| 917 |
+
"learning_rate": 8.405587769019072e-06,
|
| 918 |
+
"loss": 0.3099,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3385012919896641,
|
| 923 |
+
"grad_norm": 0.6070849299430847,
|
| 924 |
+
"learning_rate": 8.372400700702569e-06,
|
| 925 |
+
"loss": 0.3061,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.34108527131782945,
|
| 930 |
+
"grad_norm": 0.39838454127311707,
|
| 931 |
+
"learning_rate": 8.338938793943478e-06,
|
| 932 |
+
"loss": 0.2751,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.34366925064599485,
|
| 937 |
+
"grad_norm": 0.3867545425891876,
|
| 938 |
+
"learning_rate": 8.305204775766003e-06,
|
| 939 |
+
"loss": 0.2835,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3462532299741602,
|
| 944 |
+
"grad_norm": 0.41791021823883057,
|
| 945 |
+
"learning_rate": 8.27120139537044e-06,
|
| 946 |
+
"loss": 0.3058,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3488372093023256,
|
| 951 |
+
"grad_norm": 0.45444855093955994,
|
| 952 |
+
"learning_rate": 8.23693142390914e-06,
|
| 953 |
+
"loss": 0.2913,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.35142118863049093,
|
| 958 |
+
"grad_norm": 0.44107621908187866,
|
| 959 |
+
"learning_rate": 8.202397654260649e-06,
|
| 960 |
+
"loss": 0.2696,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.35400516795865633,
|
| 965 |
+
"grad_norm": 0.4542611539363861,
|
| 966 |
+
"learning_rate": 8.167602900802121e-06,
|
| 967 |
+
"loss": 0.299,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.35658914728682173,
|
| 972 |
+
"grad_norm": 0.4426042139530182,
|
| 973 |
+
"learning_rate": 8.132549999179934e-06,
|
| 974 |
+
"loss": 0.2841,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.35917312661498707,
|
| 979 |
+
"grad_norm": 0.44145461916923523,
|
| 980 |
+
"learning_rate": 8.097241806078616e-06,
|
| 981 |
+
"loss": 0.2936,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.36175710594315247,
|
| 986 |
+
"grad_norm": 0.42730993032455444,
|
| 987 |
+
"learning_rate": 8.06168119898802e-06,
|
| 988 |
+
"loss": 0.3022,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3643410852713178,
|
| 993 |
+
"grad_norm": 0.4857991635799408,
|
| 994 |
+
"learning_rate": 8.025871075968828e-06,
|
| 995 |
+
"loss": 0.2884,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3669250645994832,
|
| 1000 |
+
"grad_norm": 0.5083222985267639,
|
| 1001 |
+
"learning_rate": 7.989814355416362e-06,
|
| 1002 |
+
"loss": 0.2998,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3695090439276486,
|
| 1007 |
+
"grad_norm": 0.4802829325199127,
|
| 1008 |
+
"learning_rate": 7.953513975822755e-06,
|
| 1009 |
+
"loss": 0.2652,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.37209302325581395,
|
| 1014 |
+
"grad_norm": 0.5173447132110596,
|
| 1015 |
+
"learning_rate": 7.916972895537471e-06,
|
| 1016 |
+
"loss": 0.3088,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.37467700258397935,
|
| 1021 |
+
"grad_norm": 0.4066041111946106,
|
| 1022 |
+
"learning_rate": 7.8801940925262e-06,
|
| 1023 |
+
"loss": 0.2892,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3772609819121447,
|
| 1028 |
+
"grad_norm": 0.43255695700645447,
|
| 1029 |
+
"learning_rate": 7.843180564128178e-06,
|
| 1030 |
+
"loss": 0.2852,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3798449612403101,
|
| 1035 |
+
"grad_norm": 0.5515561103820801,
|
| 1036 |
+
"learning_rate": 7.805935326811913e-06,
|
| 1037 |
+
"loss": 0.2967,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.38242894056847543,
|
| 1042 |
+
"grad_norm": 0.5547642707824707,
|
| 1043 |
+
"learning_rate": 7.768461415929344e-06,
|
| 1044 |
+
"loss": 0.3027,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3850129198966408,
|
| 1049 |
+
"grad_norm": 0.4299023449420929,
|
| 1050 |
+
"learning_rate": 7.730761885468486e-06,
|
| 1051 |
+
"loss": 0.273,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.3875968992248062,
|
| 1056 |
+
"grad_norm": 0.45513665676116943,
|
| 1057 |
+
"learning_rate": 7.692839807804522e-06,
|
| 1058 |
+
"loss": 0.2964,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.39018087855297157,
|
| 1063 |
+
"grad_norm": 0.40135088562965393,
|
| 1064 |
+
"learning_rate": 7.654698273449435e-06,
|
| 1065 |
+
"loss": 0.2625,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.39276485788113696,
|
| 1070 |
+
"grad_norm": 0.4702034592628479,
|
| 1071 |
+
"learning_rate": 7.616340390800127e-06,
|
| 1072 |
+
"loss": 0.2956,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.3953488372093023,
|
| 1077 |
+
"grad_norm": 0.42673245072364807,
|
| 1078 |
+
"learning_rate": 7.57776928588511e-06,
|
| 1079 |
+
"loss": 0.2632,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.3979328165374677,
|
| 1084 |
+
"grad_norm": 0.4496805965900421,
|
| 1085 |
+
"learning_rate": 7.538988102109728e-06,
|
| 1086 |
+
"loss": 0.2834,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4005167958656331,
|
| 1091 |
+
"grad_norm": 0.4896727204322815,
|
| 1092 |
+
"learning_rate": 7.500000000000001e-06,
|
| 1093 |
+
"loss": 0.3104,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.40310077519379844,
|
| 1098 |
+
"grad_norm": 0.44338440895080566,
|
| 1099 |
+
"learning_rate": 7.4608081569450365e-06,
|
| 1100 |
+
"loss": 0.301,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.40568475452196384,
|
| 1105 |
+
"grad_norm": 0.47325599193573,
|
| 1106 |
+
"learning_rate": 7.421415766938098e-06,
|
| 1107 |
+
"loss": 0.2898,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4082687338501292,
|
| 1112 |
+
"grad_norm": 0.45277851819992065,
|
| 1113 |
+
"learning_rate": 7.381826040316294e-06,
|
| 1114 |
+
"loss": 0.2618,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4108527131782946,
|
| 1119 |
+
"grad_norm": 0.47705507278442383,
|
| 1120 |
+
"learning_rate": 7.342042203498952e-06,
|
| 1121 |
+
"loss": 0.3078,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4134366925064599,
|
| 1126 |
+
"grad_norm": 0.4861289858818054,
|
| 1127 |
+
"learning_rate": 7.302067498724681e-06,
|
| 1128 |
+
"loss": 0.2826,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4160206718346253,
|
| 1133 |
+
"grad_norm": 0.48274198174476624,
|
| 1134 |
+
"learning_rate": 7.261905183787136e-06,
|
| 1135 |
+
"loss": 0.255,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4186046511627907,
|
| 1140 |
+
"grad_norm": 0.48785096406936646,
|
| 1141 |
+
"learning_rate": 7.221558531769519e-06,
|
| 1142 |
+
"loss": 0.312,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.42118863049095606,
|
| 1147 |
+
"grad_norm": 0.4825511872768402,
|
| 1148 |
+
"learning_rate": 7.181030830777838e-06,
|
| 1149 |
+
"loss": 0.2869,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.42377260981912146,
|
| 1154 |
+
"grad_norm": 0.4649406969547272,
|
| 1155 |
+
"learning_rate": 7.140325383672938e-06,
|
| 1156 |
+
"loss": 0.2925,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4263565891472868,
|
| 1161 |
+
"grad_norm": 0.3625810146331787,
|
| 1162 |
+
"learning_rate": 7.099445507801324e-06,
|
| 1163 |
+
"loss": 0.2364,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4289405684754522,
|
| 1168 |
+
"grad_norm": 0.5031020641326904,
|
| 1169 |
+
"learning_rate": 7.058394534724819e-06,
|
| 1170 |
+
"loss": 0.2971,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4315245478036176,
|
| 1175 |
+
"grad_norm": 0.4177357256412506,
|
| 1176 |
+
"learning_rate": 7.017175809949044e-06,
|
| 1177 |
+
"loss": 0.2574,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.43410852713178294,
|
| 1182 |
+
"grad_norm": 0.5013376474380493,
|
| 1183 |
+
"learning_rate": 6.975792692650778e-06,
|
| 1184 |
+
"loss": 0.3145,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.43669250645994834,
|
| 1189 |
+
"grad_norm": 0.45426902174949646,
|
| 1190 |
+
"learning_rate": 6.934248555404197e-06,
|
| 1191 |
+
"loss": 0.2921,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4392764857881137,
|
| 1196 |
+
"grad_norm": 0.43907979130744934,
|
| 1197 |
+
"learning_rate": 6.892546783906016e-06,
|
| 1198 |
+
"loss": 0.2883,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4418604651162791,
|
| 1203 |
+
"grad_norm": 0.42910337448120117,
|
| 1204 |
+
"learning_rate": 6.850690776699574e-06,
|
| 1205 |
+
"loss": 0.3015,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4444444444444444,
|
| 1210 |
+
"grad_norm": 0.4053723216056824,
|
| 1211 |
+
"learning_rate": 6.808683944897856e-06,
|
| 1212 |
+
"loss": 0.2627,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4470284237726098,
|
| 1217 |
+
"grad_norm": 0.4231443703174591,
|
| 1218 |
+
"learning_rate": 6.766529711905513e-06,
|
| 1219 |
+
"loss": 0.2475,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4496124031007752,
|
| 1224 |
+
"grad_norm": 0.40346193313598633,
|
| 1225 |
+
"learning_rate": 6.724231513139853e-06,
|
| 1226 |
+
"loss": 0.2872,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.45219638242894056,
|
| 1231 |
+
"grad_norm": 0.4439121186733246,
|
| 1232 |
+
"learning_rate": 6.681792795750876e-06,
|
| 1233 |
+
"loss": 0.3125,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.45478036175710596,
|
| 1238 |
+
"grad_norm": 0.4078086316585541,
|
| 1239 |
+
"learning_rate": 6.639217018340342e-06,
|
| 1240 |
+
"loss": 0.2685,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4573643410852713,
|
| 1245 |
+
"grad_norm": 0.4373514652252197,
|
| 1246 |
+
"learning_rate": 6.5965076506799e-06,
|
| 1247 |
+
"loss": 0.2877,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4599483204134367,
|
| 1252 |
+
"grad_norm": 0.40580716729164124,
|
| 1253 |
+
"learning_rate": 6.553668173428329e-06,
|
| 1254 |
+
"loss": 0.2677,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4625322997416021,
|
| 1259 |
+
"grad_norm": 0.4066397547721863,
|
| 1260 |
+
"learning_rate": 6.510702077847864e-06,
|
| 1261 |
+
"loss": 0.2735,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.46511627906976744,
|
| 1266 |
+
"grad_norm": 0.44113850593566895,
|
| 1267 |
+
"learning_rate": 6.467612865519674e-06,
|
| 1268 |
+
"loss": 0.2666,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.46770025839793283,
|
| 1273 |
+
"grad_norm": 0.38186588883399963,
|
| 1274 |
+
"learning_rate": 6.424404048058501e-06,
|
| 1275 |
+
"loss": 0.2588,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4702842377260982,
|
| 1280 |
+
"grad_norm": 0.4320223033428192,
|
| 1281 |
+
"learning_rate": 6.3810791468264654e-06,
|
| 1282 |
+
"loss": 0.2829,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4728682170542636,
|
| 1287 |
+
"grad_norm": 0.4612186551094055,
|
| 1288 |
+
"learning_rate": 6.337641692646106e-06,
|
| 1289 |
+
"loss": 0.2942,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4754521963824289,
|
| 1294 |
+
"grad_norm": 0.45380473136901855,
|
| 1295 |
+
"learning_rate": 6.294095225512604e-06,
|
| 1296 |
+
"loss": 0.2863,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4780361757105943,
|
| 1301 |
+
"grad_norm": 0.4170936346054077,
|
| 1302 |
+
"learning_rate": 6.250443294305315e-06,
|
| 1303 |
+
"loss": 0.2685,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.4806201550387597,
|
| 1308 |
+
"grad_norm": 0.4243358373641968,
|
| 1309 |
+
"learning_rate": 6.206689456498529e-06,
|
| 1310 |
+
"loss": 0.275,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.48320413436692505,
|
| 1315 |
+
"grad_norm": 0.40185001492500305,
|
| 1316 |
+
"learning_rate": 6.162837277871553e-06,
|
| 1317 |
+
"loss": 0.2707,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.48578811369509045,
|
| 1322 |
+
"grad_norm": 0.43646830320358276,
|
| 1323 |
+
"learning_rate": 6.118890332218117e-06,
|
| 1324 |
+
"loss": 0.3082,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4883720930232558,
|
| 1329 |
+
"grad_norm": 0.4524317681789398,
|
| 1330 |
+
"learning_rate": 6.074852201055121e-06,
|
| 1331 |
+
"loss": 0.2862,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.4909560723514212,
|
| 1336 |
+
"grad_norm": 0.42992088198661804,
|
| 1337 |
+
"learning_rate": 6.0307264733307515e-06,
|
| 1338 |
+
"loss": 0.2772,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.4935400516795866,
|
| 1343 |
+
"grad_norm": 0.3978720009326935,
|
| 1344 |
+
"learning_rate": 5.986516745132e-06,
|
| 1345 |
+
"loss": 0.2825,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.49612403100775193,
|
| 1350 |
+
"grad_norm": 0.399278849363327,
|
| 1351 |
+
"learning_rate": 5.942226619391592e-06,
|
| 1352 |
+
"loss": 0.266,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.49870801033591733,
|
| 1357 |
+
"grad_norm": 0.41332167387008667,
|
| 1358 |
+
"learning_rate": 5.8978597055943585e-06,
|
| 1359 |
+
"loss": 0.2676,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5012919896640827,
|
| 1364 |
+
"grad_norm": 0.41946110129356384,
|
| 1365 |
+
"learning_rate": 5.853419619483083e-06,
|
| 1366 |
+
"loss": 0.2972,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5038759689922481,
|
| 1371 |
+
"grad_norm": 0.3924684524536133,
|
| 1372 |
+
"learning_rate": 5.808909982763825e-06,
|
| 1373 |
+
"loss": 0.2779,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5064599483204134,
|
| 1378 |
+
"grad_norm": 0.4718524217605591,
|
| 1379 |
+
"learning_rate": 5.764334422810767e-06,
|
| 1380 |
+
"loss": 0.3037,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5090439276485789,
|
| 1385 |
+
"grad_norm": 0.4339485168457031,
|
| 1386 |
+
"learning_rate": 5.719696572370596e-06,
|
| 1387 |
+
"loss": 0.32,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5116279069767442,
|
| 1392 |
+
"grad_norm": 0.40892696380615234,
|
| 1393 |
+
"learning_rate": 5.675000069266451e-06,
|
| 1394 |
+
"loss": 0.297,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5142118863049095,
|
| 1399 |
+
"grad_norm": 0.4312550723552704,
|
| 1400 |
+
"learning_rate": 5.630248556101448e-06,
|
| 1401 |
+
"loss": 0.3062,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5167958656330749,
|
| 1406 |
+
"grad_norm": 0.42749255895614624,
|
| 1407 |
+
"learning_rate": 5.585445679961823e-06,
|
| 1408 |
+
"loss": 0.2777,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5193798449612403,
|
| 1413 |
+
"grad_norm": 0.428129106760025,
|
| 1414 |
+
"learning_rate": 5.540595092119709e-06,
|
| 1415 |
+
"loss": 0.2914,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5219638242894057,
|
| 1420 |
+
"grad_norm": 0.4062337279319763,
|
| 1421 |
+
"learning_rate": 5.495700447735572e-06,
|
| 1422 |
+
"loss": 0.2821,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.524547803617571,
|
| 1427 |
+
"grad_norm": 0.421667218208313,
|
| 1428 |
+
"learning_rate": 5.450765405560328e-06,
|
| 1429 |
+
"loss": 0.2984,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5271317829457365,
|
| 1434 |
+
"grad_norm": 0.3966899514198303,
|
| 1435 |
+
"learning_rate": 5.405793627637157e-06,
|
| 1436 |
+
"loss": 0.259,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5297157622739018,
|
| 1441 |
+
"grad_norm": 0.4205838441848755,
|
| 1442 |
+
"learning_rate": 5.360788779003082e-06,
|
| 1443 |
+
"loss": 0.2888,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5322997416020672,
|
| 1448 |
+
"grad_norm": 0.4325535297393799,
|
| 1449 |
+
"learning_rate": 5.3157545273902585e-06,
|
| 1450 |
+
"loss": 0.302,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5348837209302325,
|
| 1455 |
+
"grad_norm": 0.4171897768974304,
|
| 1456 |
+
"learning_rate": 5.270694542927089e-06,
|
| 1457 |
+
"loss": 0.2685,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.537467700258398,
|
| 1462 |
+
"grad_norm": 0.4270216226577759,
|
| 1463 |
+
"learning_rate": 5.225612497839099e-06,
|
| 1464 |
+
"loss": 0.2785,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5400516795865633,
|
| 1469 |
+
"grad_norm": 0.439961314201355,
|
| 1470 |
+
"learning_rate": 5.180512066149682e-06,
|
| 1471 |
+
"loss": 0.2916,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5426356589147286,
|
| 1476 |
+
"grad_norm": 0.42138805985450745,
|
| 1477 |
+
"learning_rate": 5.1353969233806735e-06,
|
| 1478 |
+
"loss": 0.2906,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5452196382428941,
|
| 1483 |
+
"grad_norm": 0.39296743273735046,
|
| 1484 |
+
"learning_rate": 5.090270746252803e-06,
|
| 1485 |
+
"loss": 0.2638,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5478036175710594,
|
| 1490 |
+
"grad_norm": 0.43339210748672485,
|
| 1491 |
+
"learning_rate": 5.045137212386065e-06,
|
| 1492 |
+
"loss": 0.2845,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5503875968992248,
|
| 1497 |
+
"grad_norm": 0.41793346405029297,
|
| 1498 |
+
"learning_rate": 5e-06,
|
| 1499 |
+
"loss": 0.2794,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5529715762273901,
|
| 1504 |
+
"grad_norm": 0.4067362844944,
|
| 1505 |
+
"learning_rate": 4.954862787613937e-06,
|
| 1506 |
+
"loss": 0.2834,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5555555555555556,
|
| 1511 |
+
"grad_norm": 0.4313544034957886,
|
| 1512 |
+
"learning_rate": 4.909729253747197e-06,
|
| 1513 |
+
"loss": 0.2786,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5581395348837209,
|
| 1518 |
+
"grad_norm": 0.422354131937027,
|
| 1519 |
+
"learning_rate": 4.864603076619329e-06,
|
| 1520 |
+
"loss": 0.284,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5607235142118863,
|
| 1525 |
+
"grad_norm": 0.4361294209957123,
|
| 1526 |
+
"learning_rate": 4.819487933850319e-06,
|
| 1527 |
+
"loss": 0.3265,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5633074935400517,
|
| 1532 |
+
"grad_norm": 0.43613091111183167,
|
| 1533 |
+
"learning_rate": 4.774387502160902e-06,
|
| 1534 |
+
"loss": 0.2969,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5658914728682171,
|
| 1539 |
+
"grad_norm": 0.446242094039917,
|
| 1540 |
+
"learning_rate": 4.729305457072913e-06,
|
| 1541 |
+
"loss": 0.3051,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5684754521963824,
|
| 1546 |
+
"grad_norm": 0.42839357256889343,
|
| 1547 |
+
"learning_rate": 4.684245472609743e-06,
|
| 1548 |
+
"loss": 0.2851,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5710594315245479,
|
| 1553 |
+
"grad_norm": 0.41208556294441223,
|
| 1554 |
+
"learning_rate": 4.6392112209969205e-06,
|
| 1555 |
+
"loss": 0.2823,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5736434108527132,
|
| 1560 |
+
"grad_norm": 0.41363725066185,
|
| 1561 |
+
"learning_rate": 4.594206372362845e-06,
|
| 1562 |
+
"loss": 0.2668,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5762273901808785,
|
| 1567 |
+
"grad_norm": 0.40598398447036743,
|
| 1568 |
+
"learning_rate": 4.549234594439674e-06,
|
| 1569 |
+
"loss": 0.2757,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5788113695090439,
|
| 1574 |
+
"grad_norm": 0.39398393034935,
|
| 1575 |
+
"learning_rate": 4.504299552264428e-06,
|
| 1576 |
+
"loss": 0.2401,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5813953488372093,
|
| 1581 |
+
"grad_norm": 0.4146032929420471,
|
| 1582 |
+
"learning_rate": 4.459404907880293e-06,
|
| 1583 |
+
"loss": 0.2904,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5839793281653747,
|
| 1588 |
+
"grad_norm": 0.42441511154174805,
|
| 1589 |
+
"learning_rate": 4.414554320038179e-06,
|
| 1590 |
+
"loss": 0.2836,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.58656330749354,
|
| 1595 |
+
"grad_norm": 0.3873770833015442,
|
| 1596 |
+
"learning_rate": 4.369751443898554e-06,
|
| 1597 |
+
"loss": 0.2625,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5891472868217055,
|
| 1602 |
+
"grad_norm": 0.3869360387325287,
|
| 1603 |
+
"learning_rate": 4.32499993073355e-06,
|
| 1604 |
+
"loss": 0.2325,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5917312661498708,
|
| 1609 |
+
"grad_norm": 0.4107780456542969,
|
| 1610 |
+
"learning_rate": 4.280303427629404e-06,
|
| 1611 |
+
"loss": 0.2747,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5943152454780362,
|
| 1616 |
+
"grad_norm": 0.5055181384086609,
|
| 1617 |
+
"learning_rate": 4.2356655771892355e-06,
|
| 1618 |
+
"loss": 0.2893,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5968992248062015,
|
| 1623 |
+
"grad_norm": 0.4263874590396881,
|
| 1624 |
+
"learning_rate": 4.191090017236177e-06,
|
| 1625 |
+
"loss": 0.2839,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.599483204134367,
|
| 1630 |
+
"grad_norm": 0.42507773637771606,
|
| 1631 |
+
"learning_rate": 4.146580380516918e-06,
|
| 1632 |
+
"loss": 0.2793,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6020671834625323,
|
| 1637 |
+
"grad_norm": 0.3988296091556549,
|
| 1638 |
+
"learning_rate": 4.1021402944056415e-06,
|
| 1639 |
+
"loss": 0.2674,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6046511627906976,
|
| 1644 |
+
"grad_norm": 0.407911092042923,
|
| 1645 |
+
"learning_rate": 4.057773380608411e-06,
|
| 1646 |
+
"loss": 0.2509,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6072351421188631,
|
| 1651 |
+
"grad_norm": 0.41657814383506775,
|
| 1652 |
+
"learning_rate": 4.013483254868001e-06,
|
| 1653 |
+
"loss": 0.2782,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6098191214470284,
|
| 1658 |
+
"grad_norm": 0.405274897813797,
|
| 1659 |
+
"learning_rate": 3.969273526669249e-06,
|
| 1660 |
+
"loss": 0.2565,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6124031007751938,
|
| 1665 |
+
"grad_norm": 0.4784335196018219,
|
| 1666 |
+
"learning_rate": 3.92514779894488e-06,
|
| 1667 |
+
"loss": 0.2989,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6149870801033591,
|
| 1672 |
+
"grad_norm": 0.3939613997936249,
|
| 1673 |
+
"learning_rate": 3.881109667781884e-06,
|
| 1674 |
+
"loss": 0.2695,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6175710594315246,
|
| 1679 |
+
"grad_norm": 0.4349983036518097,
|
| 1680 |
+
"learning_rate": 3.8371627221284495e-06,
|
| 1681 |
+
"loss": 0.2846,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6201550387596899,
|
| 1686 |
+
"grad_norm": 0.3985432982444763,
|
| 1687 |
+
"learning_rate": 3.7933105435014727e-06,
|
| 1688 |
+
"loss": 0.2551,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6227390180878553,
|
| 1693 |
+
"grad_norm": 0.40800800919532776,
|
| 1694 |
+
"learning_rate": 3.7495567056946856e-06,
|
| 1695 |
+
"loss": 0.2758,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6253229974160207,
|
| 1700 |
+
"grad_norm": 0.4621322453022003,
|
| 1701 |
+
"learning_rate": 3.705904774487396e-06,
|
| 1702 |
+
"loss": 0.2921,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.627906976744186,
|
| 1707 |
+
"grad_norm": 0.4144759476184845,
|
| 1708 |
+
"learning_rate": 3.662358307353897e-06,
|
| 1709 |
+
"loss": 0.3026,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6304909560723514,
|
| 1714 |
+
"grad_norm": 0.41925767064094543,
|
| 1715 |
+
"learning_rate": 3.6189208531735354e-06,
|
| 1716 |
+
"loss": 0.2799,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6330749354005168,
|
| 1721 |
+
"grad_norm": 0.4139888882637024,
|
| 1722 |
+
"learning_rate": 3.5755959519415008e-06,
|
| 1723 |
+
"loss": 0.2614,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6356589147286822,
|
| 1728 |
+
"grad_norm": 0.4141432046890259,
|
| 1729 |
+
"learning_rate": 3.532387134480327e-06,
|
| 1730 |
+
"loss": 0.2688,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6382428940568475,
|
| 1735 |
+
"grad_norm": 0.4194762110710144,
|
| 1736 |
+
"learning_rate": 3.489297922152136e-06,
|
| 1737 |
+
"loss": 0.2691,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6408268733850129,
|
| 1742 |
+
"grad_norm": 0.40425029397010803,
|
| 1743 |
+
"learning_rate": 3.446331826571672e-06,
|
| 1744 |
+
"loss": 0.2789,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6434108527131783,
|
| 1749 |
+
"grad_norm": 0.4381207227706909,
|
| 1750 |
+
"learning_rate": 3.403492349320101e-06,
|
| 1751 |
+
"loss": 0.3129,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6459948320413437,
|
| 1756 |
+
"grad_norm": 0.43057993054389954,
|
| 1757 |
+
"learning_rate": 3.360782981659659e-06,
|
| 1758 |
+
"loss": 0.2628,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.648578811369509,
|
| 1763 |
+
"grad_norm": 0.4394291043281555,
|
| 1764 |
+
"learning_rate": 3.3182072042491244e-06,
|
| 1765 |
+
"loss": 0.3008,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6511627906976745,
|
| 1770 |
+
"grad_norm": 0.4196249544620514,
|
| 1771 |
+
"learning_rate": 3.275768486860149e-06,
|
| 1772 |
+
"loss": 0.2747,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6537467700258398,
|
| 1777 |
+
"grad_norm": 0.4204075336456299,
|
| 1778 |
+
"learning_rate": 3.233470288094489e-06,
|
| 1779 |
+
"loss": 0.2829,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6563307493540051,
|
| 1784 |
+
"grad_norm": 0.42610615491867065,
|
| 1785 |
+
"learning_rate": 3.191316055102146e-06,
|
| 1786 |
+
"loss": 0.2727,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6589147286821705,
|
| 1791 |
+
"grad_norm": 0.3784283697605133,
|
| 1792 |
+
"learning_rate": 3.149309223300428e-06,
|
| 1793 |
+
"loss": 0.2468,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.661498708010336,
|
| 1798 |
+
"grad_norm": 0.4189954698085785,
|
| 1799 |
+
"learning_rate": 3.107453216093985e-06,
|
| 1800 |
+
"loss": 0.2679,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6640826873385013,
|
| 1805 |
+
"grad_norm": 0.41831427812576294,
|
| 1806 |
+
"learning_rate": 3.0657514445958055e-06,
|
| 1807 |
+
"loss": 0.2754,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6666666666666666,
|
| 1812 |
+
"grad_norm": 0.4036039113998413,
|
| 1813 |
+
"learning_rate": 3.0242073073492238e-06,
|
| 1814 |
+
"loss": 0.2767,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6692506459948321,
|
| 1819 |
+
"grad_norm": 0.4252059757709503,
|
| 1820 |
+
"learning_rate": 2.982824190050958e-06,
|
| 1821 |
+
"loss": 0.292,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6718346253229974,
|
| 1826 |
+
"grad_norm": 0.3725378215312958,
|
| 1827 |
+
"learning_rate": 2.9416054652751834e-06,
|
| 1828 |
+
"loss": 0.2427,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6744186046511628,
|
| 1833 |
+
"grad_norm": 0.38230738043785095,
|
| 1834 |
+
"learning_rate": 2.9005544921986774e-06,
|
| 1835 |
+
"loss": 0.2686,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6770025839793282,
|
| 1840 |
+
"grad_norm": 0.42745351791381836,
|
| 1841 |
+
"learning_rate": 2.8596746163270646e-06,
|
| 1842 |
+
"loss": 0.2448,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6795865633074936,
|
| 1847 |
+
"grad_norm": 0.46798622608184814,
|
| 1848 |
+
"learning_rate": 2.8189691692221627e-06,
|
| 1849 |
+
"loss": 0.2981,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6821705426356589,
|
| 1854 |
+
"grad_norm": 0.4152715504169464,
|
| 1855 |
+
"learning_rate": 2.778441468230483e-06,
|
| 1856 |
+
"loss": 0.2807,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6847545219638242,
|
| 1861 |
+
"grad_norm": 0.4589405953884125,
|
| 1862 |
+
"learning_rate": 2.738094816212866e-06,
|
| 1863 |
+
"loss": 0.2782,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6873385012919897,
|
| 1868 |
+
"grad_norm": 0.42450419068336487,
|
| 1869 |
+
"learning_rate": 2.6979325012753214e-06,
|
| 1870 |
+
"loss": 0.2996,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.689922480620155,
|
| 1875 |
+
"grad_norm": 0.40908730030059814,
|
| 1876 |
+
"learning_rate": 2.65795779650105e-06,
|
| 1877 |
+
"loss": 0.261,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6925064599483204,
|
| 1882 |
+
"grad_norm": 0.4043743908405304,
|
| 1883 |
+
"learning_rate": 2.6181739596837075e-06,
|
| 1884 |
+
"loss": 0.2554,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6950904392764858,
|
| 1889 |
+
"grad_norm": 0.43334946036338806,
|
| 1890 |
+
"learning_rate": 2.5785842330619038e-06,
|
| 1891 |
+
"loss": 0.2795,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6976744186046512,
|
| 1896 |
+
"grad_norm": 0.4369620382785797,
|
| 1897 |
+
"learning_rate": 2.5391918430549635e-06,
|
| 1898 |
+
"loss": 0.2758,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7002583979328165,
|
| 1903 |
+
"grad_norm": 0.35578539967536926,
|
| 1904 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1905 |
+
"loss": 0.2338,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7028423772609819,
|
| 1910 |
+
"grad_norm": 0.42097681760787964,
|
| 1911 |
+
"learning_rate": 2.4610118978902732e-06,
|
| 1912 |
+
"loss": 0.269,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7054263565891473,
|
| 1917 |
+
"grad_norm": 0.42444154620170593,
|
| 1918 |
+
"learning_rate": 2.422230714114891e-06,
|
| 1919 |
+
"loss": 0.2574,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7080103359173127,
|
| 1924 |
+
"grad_norm": 0.38183045387268066,
|
| 1925 |
+
"learning_rate": 2.383659609199873e-06,
|
| 1926 |
+
"loss": 0.2469,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.710594315245478,
|
| 1931 |
+
"grad_norm": 0.4213600754737854,
|
| 1932 |
+
"learning_rate": 2.345301726550567e-06,
|
| 1933 |
+
"loss": 0.2707,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7131782945736435,
|
| 1938 |
+
"grad_norm": 0.3678889870643616,
|
| 1939 |
+
"learning_rate": 2.3071601921954797e-06,
|
| 1940 |
+
"loss": 0.2418,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7157622739018088,
|
| 1945 |
+
"grad_norm": 0.42480871081352234,
|
| 1946 |
+
"learning_rate": 2.269238114531515e-06,
|
| 1947 |
+
"loss": 0.2828,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7183462532299741,
|
| 1952 |
+
"grad_norm": 0.4046178162097931,
|
| 1953 |
+
"learning_rate": 2.2315385840706548e-06,
|
| 1954 |
+
"loss": 0.2779,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7209302325581395,
|
| 1959 |
+
"grad_norm": 0.3712913393974304,
|
| 1960 |
+
"learning_rate": 2.1940646731880887e-06,
|
| 1961 |
+
"loss": 0.2454,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7235142118863049,
|
| 1966 |
+
"grad_norm": 0.40826621651649475,
|
| 1967 |
+
"learning_rate": 2.156819435871824e-06,
|
| 1968 |
+
"loss": 0.2714,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7260981912144703,
|
| 1973 |
+
"grad_norm": 0.40158697962760925,
|
| 1974 |
+
"learning_rate": 2.1198059074738027e-06,
|
| 1975 |
+
"loss": 0.2769,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7286821705426356,
|
| 1980 |
+
"grad_norm": 0.41186589002609253,
|
| 1981 |
+
"learning_rate": 2.08302710446253e-06,
|
| 1982 |
+
"loss": 0.2681,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7312661498708011,
|
| 1987 |
+
"grad_norm": 0.39115414023399353,
|
| 1988 |
+
"learning_rate": 2.0464860241772454e-06,
|
| 1989 |
+
"loss": 0.264,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7338501291989664,
|
| 1994 |
+
"grad_norm": 0.410695880651474,
|
| 1995 |
+
"learning_rate": 2.010185644583641e-06,
|
| 1996 |
+
"loss": 0.2725,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7364341085271318,
|
| 2001 |
+
"grad_norm": 0.43781405687332153,
|
| 2002 |
+
"learning_rate": 1.9741289240311757e-06,
|
| 2003 |
+
"loss": 0.2886,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7390180878552972,
|
| 2008 |
+
"grad_norm": 0.4308369755744934,
|
| 2009 |
+
"learning_rate": 1.9383188010119818e-06,
|
| 2010 |
+
"loss": 0.3076,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7416020671834626,
|
| 2015 |
+
"grad_norm": 0.45121103525161743,
|
| 2016 |
+
"learning_rate": 1.9027581939213852e-06,
|
| 2017 |
+
"loss": 0.307,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7441860465116279,
|
| 2022 |
+
"grad_norm": 0.41774290800094604,
|
| 2023 |
+
"learning_rate": 1.8674500008200675e-06,
|
| 2024 |
+
"loss": 0.2747,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7467700258397932,
|
| 2029 |
+
"grad_norm": 0.41470867395401,
|
| 2030 |
+
"learning_rate": 1.8323970991978823e-06,
|
| 2031 |
+
"loss": 0.2719,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7493540051679587,
|
| 2036 |
+
"grad_norm": 0.40178781747817993,
|
| 2037 |
+
"learning_rate": 1.797602345739352e-06,
|
| 2038 |
+
"loss": 0.2608,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.751937984496124,
|
| 2043 |
+
"grad_norm": 0.39378994703292847,
|
| 2044 |
+
"learning_rate": 1.7630685760908623e-06,
|
| 2045 |
+
"loss": 0.2766,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7545219638242894,
|
| 2050 |
+
"grad_norm": 0.41441816091537476,
|
| 2051 |
+
"learning_rate": 1.7287986046295597e-06,
|
| 2052 |
+
"loss": 0.2732,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7571059431524548,
|
| 2057 |
+
"grad_norm": 0.3627479672431946,
|
| 2058 |
+
"learning_rate": 1.6947952242339993e-06,
|
| 2059 |
+
"loss": 0.2413,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7596899224806202,
|
| 2064 |
+
"grad_norm": 0.4463461935520172,
|
| 2065 |
+
"learning_rate": 1.6610612060565235e-06,
|
| 2066 |
+
"loss": 0.3035,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7622739018087855,
|
| 2071 |
+
"grad_norm": 0.440087229013443,
|
| 2072 |
+
"learning_rate": 1.627599299297431e-06,
|
| 2073 |
+
"loss": 0.2831,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7648578811369509,
|
| 2078 |
+
"grad_norm": 0.4249372184276581,
|
| 2079 |
+
"learning_rate": 1.594412230980928e-06,
|
| 2080 |
+
"loss": 0.2747,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7674418604651163,
|
| 2085 |
+
"grad_norm": 0.391125351190567,
|
| 2086 |
+
"learning_rate": 1.561502705732883e-06,
|
| 2087 |
+
"loss": 0.2726,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7700258397932817,
|
| 2092 |
+
"grad_norm": 0.4350535571575165,
|
| 2093 |
+
"learning_rate": 1.5288734055604087e-06,
|
| 2094 |
+
"loss": 0.2881,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.772609819121447,
|
| 2099 |
+
"grad_norm": 0.43538013100624084,
|
| 2100 |
+
"learning_rate": 1.4965269896332884e-06,
|
| 2101 |
+
"loss": 0.2733,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7751937984496124,
|
| 2106 |
+
"grad_norm": 0.4050742983818054,
|
| 2107 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 2108 |
+
"loss": 0.2633,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7777777777777778,
|
| 2113 |
+
"grad_norm": 0.40997040271759033,
|
| 2114 |
+
"learning_rate": 1.4326933317092e-06,
|
| 2115 |
+
"loss": 0.2671,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7803617571059431,
|
| 2120 |
+
"grad_norm": 0.3840159475803375,
|
| 2121 |
+
"learning_rate": 1.4012112919241562e-06,
|
| 2122 |
+
"loss": 0.2583,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7829457364341085,
|
| 2127 |
+
"grad_norm": 0.3878195285797119,
|
| 2128 |
+
"learning_rate": 1.370022540384347e-06,
|
| 2129 |
+
"loss": 0.26,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7855297157622739,
|
| 2134 |
+
"grad_norm": 0.4388813376426697,
|
| 2135 |
+
"learning_rate": 1.3391296188600594e-06,
|
| 2136 |
+
"loss": 0.2583,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7881136950904393,
|
| 2141 |
+
"grad_norm": 0.37653830647468567,
|
| 2142 |
+
"learning_rate": 1.3085350450125073e-06,
|
| 2143 |
+
"loss": 0.2679,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7906976744186046,
|
| 2148 |
+
"grad_norm": 0.3878933787345886,
|
| 2149 |
+
"learning_rate": 1.2782413121886483e-06,
|
| 2150 |
+
"loss": 0.2369,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7932816537467701,
|
| 2155 |
+
"grad_norm": 0.38015350699424744,
|
| 2156 |
+
"learning_rate": 1.2482508892179884e-06,
|
| 2157 |
+
"loss": 0.2476,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7958656330749354,
|
| 2162 |
+
"grad_norm": 0.4102688133716583,
|
| 2163 |
+
"learning_rate": 1.2185662202113764e-06,
|
| 2164 |
+
"loss": 0.259,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7984496124031008,
|
| 2169 |
+
"grad_norm": 0.3736209273338318,
|
| 2170 |
+
"learning_rate": 1.1891897243618184e-06,
|
| 2171 |
+
"loss": 0.244,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8010335917312662,
|
| 2176 |
+
"grad_norm": 0.4116411805152893,
|
| 2177 |
+
"learning_rate": 1.16012379574733e-06,
|
| 2178 |
+
"loss": 0.2687,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8036175710594315,
|
| 2183 |
+
"grad_norm": 0.47300201654434204,
|
| 2184 |
+
"learning_rate": 1.1313708031358183e-06,
|
| 2185 |
+
"loss": 0.2728,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8062015503875969,
|
| 2190 |
+
"grad_norm": 0.4103212058544159,
|
| 2191 |
+
"learning_rate": 1.102933089792042e-06,
|
| 2192 |
+
"loss": 0.2764,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8087855297157622,
|
| 2197 |
+
"grad_norm": 0.37040975689888,
|
| 2198 |
+
"learning_rate": 1.0748129732866391e-06,
|
| 2199 |
+
"loss": 0.2318,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8113695090439277,
|
| 2204 |
+
"grad_norm": 0.3986849784851074,
|
| 2205 |
+
"learning_rate": 1.047012745307255e-06,
|
| 2206 |
+
"loss": 0.2741,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.813953488372093,
|
| 2211 |
+
"grad_norm": 0.3630805015563965,
|
| 2212 |
+
"learning_rate": 1.0195346714717813e-06,
|
| 2213 |
+
"loss": 0.2413,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8165374677002584,
|
| 2218 |
+
"grad_norm": 0.38989007472991943,
|
| 2219 |
+
"learning_rate": 9.92380991143712e-07,
|
| 2220 |
+
"loss": 0.2628,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8191214470284238,
|
| 2225 |
+
"grad_norm": 0.4112791121006012,
|
| 2226 |
+
"learning_rate": 9.65553917249648e-07,
|
| 2227 |
+
"loss": 0.2768,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8217054263565892,
|
| 2232 |
+
"grad_norm": 0.40445297956466675,
|
| 2233 |
+
"learning_rate": 9.39055636098945e-07,
|
| 2234 |
+
"loss": 0.2775,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8242894056847545,
|
| 2239 |
+
"grad_norm": 0.3824197053909302,
|
| 2240 |
+
"learning_rate": 9.128883072055411e-07,
|
| 2241 |
+
"loss": 0.2357,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8268733850129198,
|
| 2246 |
+
"grad_norm": 0.4150390028953552,
|
| 2247 |
+
"learning_rate": 8.870540631119667e-07,
|
| 2248 |
+
"loss": 0.2779,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8294573643410853,
|
| 2253 |
+
"grad_norm": 0.38705989718437195,
|
| 2254 |
+
"learning_rate": 8.615550092155478e-07,
|
| 2255 |
+
"loss": 0.2584,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8320413436692506,
|
| 2260 |
+
"grad_norm": 0.41040337085723877,
|
| 2261 |
+
"learning_rate": 8.363932235968198e-07,
|
| 2262 |
+
"loss": 0.29,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.834625322997416,
|
| 2267 |
+
"grad_norm": 0.4016845226287842,
|
| 2268 |
+
"learning_rate": 8.115707568501768e-07,
|
| 2269 |
+
"loss": 0.2502,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8372093023255814,
|
| 2274 |
+
"grad_norm": 0.3633503317832947,
|
| 2275 |
+
"learning_rate": 7.870896319167548e-07,
|
| 2276 |
+
"loss": 0.2431,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8397932816537468,
|
| 2281 |
+
"grad_norm": 0.43767327070236206,
|
| 2282 |
+
"learning_rate": 7.629518439195671e-07,
|
| 2283 |
+
"loss": 0.3077,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8423772609819121,
|
| 2288 |
+
"grad_norm": 0.4164217412471771,
|
| 2289 |
+
"learning_rate": 7.391593600009123e-07,
|
| 2290 |
+
"loss": 0.2841,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8449612403100775,
|
| 2295 |
+
"grad_norm": 0.381540983915329,
|
| 2296 |
+
"learning_rate": 7.157141191620548e-07,
|
| 2297 |
+
"loss": 0.2473,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8475452196382429,
|
| 2302 |
+
"grad_norm": 0.45407700538635254,
|
| 2303 |
+
"learning_rate": 6.926180321052045e-07,
|
| 2304 |
+
"loss": 0.2833,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8501291989664083,
|
| 2309 |
+
"grad_norm": 0.38916081190109253,
|
| 2310 |
+
"learning_rate": 6.698729810778065e-07,
|
| 2311 |
+
"loss": 0.2719,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8527131782945736,
|
| 2316 |
+
"grad_norm": 0.43010246753692627,
|
| 2317 |
+
"learning_rate": 6.474808197191401e-07,
|
| 2318 |
+
"loss": 0.2714,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8552971576227391,
|
| 2323 |
+
"grad_norm": 0.38088157773017883,
|
| 2324 |
+
"learning_rate": 6.254433729092518e-07,
|
| 2325 |
+
"loss": 0.242,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8578811369509044,
|
| 2330 |
+
"grad_norm": 0.42933011054992676,
|
| 2331 |
+
"learning_rate": 6.037624366202405e-07,
|
| 2332 |
+
"loss": 0.2735,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8604651162790697,
|
| 2337 |
+
"grad_norm": 0.42603251338005066,
|
| 2338 |
+
"learning_rate": 5.824397777698859e-07,
|
| 2339 |
+
"loss": 0.2743,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8630490956072352,
|
| 2344 |
+
"grad_norm": 0.39108431339263916,
|
| 2345 |
+
"learning_rate": 5.614771340776559e-07,
|
| 2346 |
+
"loss": 0.2728,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8656330749354005,
|
| 2351 |
+
"grad_norm": 0.3757094442844391,
|
| 2352 |
+
"learning_rate": 5.408762139230889e-07,
|
| 2353 |
+
"loss": 0.2446,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8682170542635659,
|
| 2358 |
+
"grad_norm": 0.4124755561351776,
|
| 2359 |
+
"learning_rate": 5.206386962065601e-07,
|
| 2360 |
+
"loss": 0.2682,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8708010335917312,
|
| 2365 |
+
"grad_norm": 0.42066821455955505,
|
| 2366 |
+
"learning_rate": 5.007662302124671e-07,
|
| 2367 |
+
"loss": 0.2878,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8733850129198967,
|
| 2372 |
+
"grad_norm": 0.4132254421710968,
|
| 2373 |
+
"learning_rate": 4.812604354748107e-07,
|
| 2374 |
+
"loss": 0.2693,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.875968992248062,
|
| 2379 |
+
"grad_norm": 0.4205566346645355,
|
| 2380 |
+
"learning_rate": 4.6212290164521554e-07,
|
| 2381 |
+
"loss": 0.2695,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8785529715762274,
|
| 2386 |
+
"grad_norm": 0.4251742660999298,
|
| 2387 |
+
"learning_rate": 4.433551883633719e-07,
|
| 2388 |
+
"loss": 0.2784,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8811369509043928,
|
| 2393 |
+
"grad_norm": 0.4145876169204712,
|
| 2394 |
+
"learning_rate": 4.2495882512993913e-07,
|
| 2395 |
+
"loss": 0.2778,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8837209302325582,
|
| 2400 |
+
"grad_norm": 0.3994285762310028,
|
| 2401 |
+
"learning_rate": 4.069353111818913e-07,
|
| 2402 |
+
"loss": 0.2647,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.8863049095607235,
|
| 2407 |
+
"grad_norm": 0.42055121064186096,
|
| 2408 |
+
"learning_rate": 3.8928611537033424e-07,
|
| 2409 |
+
"loss": 0.256,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8888888888888888,
|
| 2414 |
+
"grad_norm": 0.4389072358608246,
|
| 2415 |
+
"learning_rate": 3.7201267604080436e-07,
|
| 2416 |
+
"loss": 0.2858,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8914728682170543,
|
| 2421 |
+
"grad_norm": 0.4421055018901825,
|
| 2422 |
+
"learning_rate": 3.5511640091604293e-07,
|
| 2423 |
+
"loss": 0.2671,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8940568475452196,
|
| 2428 |
+
"grad_norm": 0.42173531651496887,
|
| 2429 |
+
"learning_rate": 3.385986669812769e-07,
|
| 2430 |
+
"loss": 0.2771,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.896640826873385,
|
| 2435 |
+
"grad_norm": 0.4103642702102661,
|
| 2436 |
+
"learning_rate": 3.224608203719953e-07,
|
| 2437 |
+
"loss": 0.2517,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8992248062015504,
|
| 2442 |
+
"grad_norm": 0.39951303601264954,
|
| 2443 |
+
"learning_rate": 3.067041762642475e-07,
|
| 2444 |
+
"loss": 0.2616,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9018087855297158,
|
| 2449 |
+
"grad_norm": 0.39229103922843933,
|
| 2450 |
+
"learning_rate": 2.9133001876746004e-07,
|
| 2451 |
+
"loss": 0.2629,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9043927648578811,
|
| 2456 |
+
"grad_norm": 0.41716843843460083,
|
| 2457 |
+
"learning_rate": 2.763396008197833e-07,
|
| 2458 |
+
"loss": 0.2856,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9069767441860465,
|
| 2463 |
+
"grad_norm": 0.38193899393081665,
|
| 2464 |
+
"learning_rate": 2.617341440859883e-07,
|
| 2465 |
+
"loss": 0.2436,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9095607235142119,
|
| 2470 |
+
"grad_norm": 0.3895181715488434,
|
| 2471 |
+
"learning_rate": 2.475148388578974e-07,
|
| 2472 |
+
"loss": 0.2363,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9121447028423773,
|
| 2477 |
+
"grad_norm": 0.3759653866291046,
|
| 2478 |
+
"learning_rate": 2.3368284395738684e-07,
|
| 2479 |
+
"loss": 0.2544,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9147286821705426,
|
| 2484 |
+
"grad_norm": 0.40008825063705444,
|
| 2485 |
+
"learning_rate": 2.2023928664194229e-07,
|
| 2486 |
+
"loss": 0.2724,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.917312661498708,
|
| 2491 |
+
"grad_norm": 0.39400628209114075,
|
| 2492 |
+
"learning_rate": 2.0718526251279346e-07,
|
| 2493 |
+
"loss": 0.2397,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9198966408268734,
|
| 2498 |
+
"grad_norm": 0.3940667510032654,
|
| 2499 |
+
"learning_rate": 1.9452183542562785e-07,
|
| 2500 |
+
"loss": 0.2546,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9224806201550387,
|
| 2505 |
+
"grad_norm": 0.393248587846756,
|
| 2506 |
+
"learning_rate": 1.8225003740388546e-07,
|
| 2507 |
+
"loss": 0.2657,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9250645994832042,
|
| 2512 |
+
"grad_norm": 0.3825235962867737,
|
| 2513 |
+
"learning_rate": 1.7037086855465902e-07,
|
| 2514 |
+
"loss": 0.2429,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9276485788113695,
|
| 2519 |
+
"grad_norm": 0.38378632068634033,
|
| 2520 |
+
"learning_rate": 1.5888529698718347e-07,
|
| 2521 |
+
"loss": 0.2416,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9302325581395349,
|
| 2526 |
+
"grad_norm": 0.3894941806793213,
|
| 2527 |
+
"learning_rate": 1.477942587339426e-07,
|
| 2528 |
+
"loss": 0.2568,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9328165374677002,
|
| 2533 |
+
"grad_norm": 0.44816529750823975,
|
| 2534 |
+
"learning_rate": 1.3709865767438434e-07,
|
| 2535 |
+
"loss": 0.2892,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9354005167958657,
|
| 2540 |
+
"grad_norm": 0.3683631718158722,
|
| 2541 |
+
"learning_rate": 1.2679936546125483e-07,
|
| 2542 |
+
"loss": 0.2477,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.937984496124031,
|
| 2547 |
+
"grad_norm": 0.4054432809352875,
|
| 2548 |
+
"learning_rate": 1.1689722144956672e-07,
|
| 2549 |
+
"loss": 0.263,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9405684754521964,
|
| 2554 |
+
"grad_norm": 0.45787227153778076,
|
| 2555 |
+
"learning_rate": 1.0739303262819301e-07,
|
| 2556 |
+
"loss": 0.2685,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9431524547803618,
|
| 2561 |
+
"grad_norm": 0.4092801809310913,
|
| 2562 |
+
"learning_rate": 9.82875735540989e-08,
|
| 2563 |
+
"loss": 0.2616,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9457364341085271,
|
| 2568 |
+
"grad_norm": 0.3954561948776245,
|
| 2569 |
+
"learning_rate": 8.95815862892202e-08,
|
| 2570 |
+
"loss": 0.2778,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9483204134366925,
|
| 2575 |
+
"grad_norm": 0.4244314134120941,
|
| 2576 |
+
"learning_rate": 8.127578033998663e-08,
|
| 2577 |
+
"loss": 0.2873,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9509043927648578,
|
| 2582 |
+
"grad_norm": 0.40788134932518005,
|
| 2583 |
+
"learning_rate": 7.337083259949918e-08,
|
| 2584 |
+
"loss": 0.2767,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9534883720930233,
|
| 2589 |
+
"grad_norm": 0.40445661544799805,
|
| 2590 |
+
"learning_rate": 6.58673872923693e-08,
|
| 2591 |
+
"loss": 0.2561,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9560723514211886,
|
| 2596 |
+
"grad_norm": 0.38080865144729614,
|
| 2597 |
+
"learning_rate": 5.8766055922211965e-08,
|
| 2598 |
+
"loss": 0.2562,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.958656330749354,
|
| 2603 |
+
"grad_norm": 0.40840834379196167,
|
| 2604 |
+
"learning_rate": 5.206741722181385e-08,
|
| 2605 |
+
"loss": 0.2576,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9612403100775194,
|
| 2610 |
+
"grad_norm": 0.426565945148468,
|
| 2611 |
+
"learning_rate": 4.577201710596613e-08,
|
| 2612 |
+
"loss": 0.2936,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9638242894056848,
|
| 2617 |
+
"grad_norm": 0.39672496914863586,
|
| 2618 |
+
"learning_rate": 3.98803686269783e-08,
|
| 2619 |
+
"loss": 0.2397,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9664082687338501,
|
| 2624 |
+
"grad_norm": 0.4055837094783783,
|
| 2625 |
+
"learning_rate": 3.439295193286174e-08,
|
| 2626 |
+
"loss": 0.2532,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9689922480620154,
|
| 2631 |
+
"grad_norm": 0.40761086344718933,
|
| 2632 |
+
"learning_rate": 2.9310214228202016e-08,
|
| 2633 |
+
"loss": 0.2708,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9715762273901809,
|
| 2638 |
+
"grad_norm": 0.446036696434021,
|
| 2639 |
+
"learning_rate": 2.463256973771311e-08,
|
| 2640 |
+
"loss": 0.277,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9741602067183462,
|
| 2645 |
+
"grad_norm": 0.40387487411499023,
|
| 2646 |
+
"learning_rate": 2.0360399672478826e-08,
|
| 2647 |
+
"loss": 0.2539,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9767441860465116,
|
| 2652 |
+
"grad_norm": 0.4223725497722626,
|
| 2653 |
+
"learning_rate": 1.6494052198886557e-08,
|
| 2654 |
+
"loss": 0.2714,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.979328165374677,
|
| 2659 |
+
"grad_norm": 0.373526006937027,
|
| 2660 |
+
"learning_rate": 1.3033842410251074e-08,
|
| 2661 |
+
"loss": 0.2407,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9819121447028424,
|
| 2666 |
+
"grad_norm": 0.3940108120441437,
|
| 2667 |
+
"learning_rate": 9.980052301137854e-09,
|
| 2668 |
+
"loss": 0.2554,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.9844961240310077,
|
| 2673 |
+
"grad_norm": 0.3823840320110321,
|
| 2674 |
+
"learning_rate": 7.332930744380906e-09,
|
| 2675 |
+
"loss": 0.2605,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9870801033591732,
|
| 2680 |
+
"grad_norm": 0.41570064425468445,
|
| 2681 |
+
"learning_rate": 5.092693470798438e-09,
|
| 2682 |
+
"loss": 0.2538,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9896640826873385,
|
| 2687 |
+
"grad_norm": 0.4328196346759796,
|
| 2688 |
+
"learning_rate": 3.2595230516152543e-09,
|
| 2689 |
+
"loss": 0.258,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.9922480620155039,
|
| 2694 |
+
"grad_norm": 0.3797440826892853,
|
| 2695 |
+
"learning_rate": 1.8335688835802169e-09,
|
| 2696 |
+
"loss": 0.2528,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9948320413436692,
|
| 2701 |
+
"grad_norm": 0.3894307315349579,
|
| 2702 |
+
"learning_rate": 8.149471767937567e-10,
|
| 2703 |
+
"loss": 0.2572,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.9974160206718347,
|
| 2708 |
+
"grad_norm": 0.40216878056526184,
|
| 2709 |
+
"learning_rate": 2.0374094523600841e-10,
|
| 2710 |
+
"loss": 0.2686,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 1.0,
|
| 2715 |
+
"grad_norm": 0.3972252607345581,
|
| 2716 |
+
"learning_rate": 0.0,
|
| 2717 |
+
"loss": 0.2752,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 1.0,
|
| 2722 |
+
"eval_loss": 0.39300352334976196,
|
| 2723 |
+
"eval_runtime": 1.5284,
|
| 2724 |
+
"eval_samples_per_second": 150.481,
|
| 2725 |
+
"eval_steps_per_second": 6.543,
|
| 2726 |
+
"step": 387
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 1.0,
|
| 2730 |
+
"step": 387,
|
| 2731 |
+
"total_flos": 3.228374416542925e+16,
|
| 2732 |
+
"train_loss": 0.37563710938898476,
|
| 2733 |
+
"train_runtime": 503.7567,
|
| 2734 |
+
"train_samples_per_second": 36.859,
|
| 2735 |
+
"train_steps_per_second": 0.768
|
| 2736 |
+
}
|
| 2737 |
+
],
|
| 2738 |
+
"logging_steps": 1,
|
| 2739 |
+
"max_steps": 387,
|
| 2740 |
+
"num_input_tokens_seen": 0,
|
| 2741 |
+
"num_train_epochs": 1,
|
| 2742 |
+
"save_steps": 800,
|
| 2743 |
+
"stateful_callbacks": {
|
| 2744 |
+
"TrainerControl": {
|
| 2745 |
+
"args": {
|
| 2746 |
+
"should_epoch_stop": false,
|
| 2747 |
+
"should_evaluate": false,
|
| 2748 |
+
"should_log": false,
|
| 2749 |
+
"should_save": false,
|
| 2750 |
+
"should_training_stop": false
|
| 2751 |
+
},
|
| 2752 |
+
"attributes": {}
|
| 2753 |
+
}
|
| 2754 |
+
},
|
| 2755 |
+
"total_flos": 3.228374416542925e+16,
|
| 2756 |
+
"train_batch_size": 16,
|
| 2757 |
+
"trial_name": null,
|
| 2758 |
+
"trial_params": null
|
| 2759 |
+
}
|