Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +1744 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: preference_p0.1_seed42_level2_syntax
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# preference_p0.1_seed42_level2_syntax
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.3595
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 2
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- gradient_accumulation_steps: 4
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- total_eval_batch_size: 4
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 1
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 59 |
+
| 0.345 | 1.0 | 1206 | 0.3595 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.43.4
|
| 65 |
+
- Pytorch 2.3.1+cu121
|
| 66 |
+
- Datasets 2.19.1
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 252459519836160.0,
|
| 4 |
+
"train_loss": 0.5262548677858033,
|
| 5 |
+
"train_runtime": 29272.1392,
|
| 6 |
+
"train_samples": 98793,
|
| 7 |
+
"train_samples_per_second": 1.318,
|
| 8 |
+
"train_steps_per_second": 0.041
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 252459519836160.0,
|
| 4 |
+
"train_loss": 0.5262548677858033,
|
| 5 |
+
"train_runtime": 29272.1392,
|
| 6 |
+
"train_samples": 98793,
|
| 7 |
+
"train_samples_per_second": 1.318,
|
| 8 |
+
"train_steps_per_second": 0.041
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1744 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1206,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0008291873963515755,
|
| 13 |
+
"grad_norm": 24.363085939314125,
|
| 14 |
+
"learning_rate": 8.264462809917357e-08,
|
| 15 |
+
"loss": 1.4091,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0041459369817578775,
|
| 20 |
+
"grad_norm": 23.543346063299293,
|
| 21 |
+
"learning_rate": 4.132231404958678e-07,
|
| 22 |
+
"loss": 1.4056,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008291873963515755,
|
| 27 |
+
"grad_norm": 8.313966376086048,
|
| 28 |
+
"learning_rate": 8.264462809917356e-07,
|
| 29 |
+
"loss": 1.3082,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.012437810945273632,
|
| 34 |
+
"grad_norm": 10.10778485014052,
|
| 35 |
+
"learning_rate": 1.2396694214876035e-06,
|
| 36 |
+
"loss": 1.1527,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01658374792703151,
|
| 41 |
+
"grad_norm": 3.1507455192980616,
|
| 42 |
+
"learning_rate": 1.6528925619834712e-06,
|
| 43 |
+
"loss": 1.0115,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.020729684908789386,
|
| 48 |
+
"grad_norm": 2.949214823531174,
|
| 49 |
+
"learning_rate": 2.066115702479339e-06,
|
| 50 |
+
"loss": 0.9662,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.024875621890547265,
|
| 55 |
+
"grad_norm": 2.3865704411346944,
|
| 56 |
+
"learning_rate": 2.479338842975207e-06,
|
| 57 |
+
"loss": 0.9301,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02902155887230514,
|
| 62 |
+
"grad_norm": 2.367353362561768,
|
| 63 |
+
"learning_rate": 2.8925619834710743e-06,
|
| 64 |
+
"loss": 0.9066,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03316749585406302,
|
| 69 |
+
"grad_norm": 2.168359757453863,
|
| 70 |
+
"learning_rate": 3.3057851239669424e-06,
|
| 71 |
+
"loss": 0.8864,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03731343283582089,
|
| 76 |
+
"grad_norm": 2.352027725705367,
|
| 77 |
+
"learning_rate": 3.71900826446281e-06,
|
| 78 |
+
"loss": 0.8776,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04145936981757877,
|
| 83 |
+
"grad_norm": 2.221157238277382,
|
| 84 |
+
"learning_rate": 4.132231404958678e-06,
|
| 85 |
+
"loss": 0.8738,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04560530679933665,
|
| 90 |
+
"grad_norm": 2.254296003816909,
|
| 91 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 92 |
+
"loss": 0.8654,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04975124378109453,
|
| 97 |
+
"grad_norm": 2.2549051162675258,
|
| 98 |
+
"learning_rate": 4.958677685950414e-06,
|
| 99 |
+
"loss": 0.8603,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0538971807628524,
|
| 104 |
+
"grad_norm": 2.421942208846336,
|
| 105 |
+
"learning_rate": 5.371900826446281e-06,
|
| 106 |
+
"loss": 0.8514,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05804311774461028,
|
| 111 |
+
"grad_norm": 2.2852906820767833,
|
| 112 |
+
"learning_rate": 5.785123966942149e-06,
|
| 113 |
+
"loss": 0.8448,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06218905472636816,
|
| 118 |
+
"grad_norm": 2.5429700413304293,
|
| 119 |
+
"learning_rate": 6.198347107438017e-06,
|
| 120 |
+
"loss": 0.8342,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06633499170812604,
|
| 125 |
+
"grad_norm": 2.41053818480909,
|
| 126 |
+
"learning_rate": 6.611570247933885e-06,
|
| 127 |
+
"loss": 0.8233,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07048092868988391,
|
| 132 |
+
"grad_norm": 2.467306342713205,
|
| 133 |
+
"learning_rate": 7.0247933884297525e-06,
|
| 134 |
+
"loss": 0.8191,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07462686567164178,
|
| 139 |
+
"grad_norm": 2.4370364722694924,
|
| 140 |
+
"learning_rate": 7.43801652892562e-06,
|
| 141 |
+
"loss": 0.8283,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07877280265339967,
|
| 146 |
+
"grad_norm": 2.3448779174579175,
|
| 147 |
+
"learning_rate": 7.851239669421489e-06,
|
| 148 |
+
"loss": 0.8109,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08291873963515754,
|
| 153 |
+
"grad_norm": 2.366708369360123,
|
| 154 |
+
"learning_rate": 8.264462809917356e-06,
|
| 155 |
+
"loss": 0.8175,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08706467661691543,
|
| 160 |
+
"grad_norm": 2.7803642859629445,
|
| 161 |
+
"learning_rate": 8.677685950413224e-06,
|
| 162 |
+
"loss": 0.8078,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0912106135986733,
|
| 167 |
+
"grad_norm": 2.402987189273672,
|
| 168 |
+
"learning_rate": 9.090909090909091e-06,
|
| 169 |
+
"loss": 0.8036,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09535655058043117,
|
| 174 |
+
"grad_norm": 2.485685260898161,
|
| 175 |
+
"learning_rate": 9.50413223140496e-06,
|
| 176 |
+
"loss": 0.801,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09950248756218906,
|
| 181 |
+
"grad_norm": 3.0171204848637294,
|
| 182 |
+
"learning_rate": 9.917355371900828e-06,
|
| 183 |
+
"loss": 0.7926,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.10364842454394693,
|
| 188 |
+
"grad_norm": 2.3850376443520602,
|
| 189 |
+
"learning_rate": 9.999664652243188e-06,
|
| 190 |
+
"loss": 0.7988,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1077943615257048,
|
| 195 |
+
"grad_norm": 2.489350579580748,
|
| 196 |
+
"learning_rate": 9.99830237907608e-06,
|
| 197 |
+
"loss": 0.7869,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.11194029850746269,
|
| 202 |
+
"grad_norm": 2.4001253138099843,
|
| 203 |
+
"learning_rate": 9.995892506564461e-06,
|
| 204 |
+
"loss": 0.7817,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11608623548922056,
|
| 209 |
+
"grad_norm": 2.4718753541913037,
|
| 210 |
+
"learning_rate": 9.992435539796e-06,
|
| 211 |
+
"loss": 0.7853,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.12023217247097844,
|
| 216 |
+
"grad_norm": 2.4377853310772966,
|
| 217 |
+
"learning_rate": 9.987932203319917e-06,
|
| 218 |
+
"loss": 0.7634,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12437810945273632,
|
| 223 |
+
"grad_norm": 2.8232961214963024,
|
| 224 |
+
"learning_rate": 9.982383440995146e-06,
|
| 225 |
+
"loss": 0.7688,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1285240464344942,
|
| 230 |
+
"grad_norm": 2.25234786421426,
|
| 231 |
+
"learning_rate": 9.975790415792497e-06,
|
| 232 |
+
"loss": 0.7707,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.13266998341625208,
|
| 237 |
+
"grad_norm": 2.653089890533132,
|
| 238 |
+
"learning_rate": 9.968154509550914e-06,
|
| 239 |
+
"loss": 0.759,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13681592039800994,
|
| 244 |
+
"grad_norm": 2.4692695376219946,
|
| 245 |
+
"learning_rate": 9.959477322687852e-06,
|
| 246 |
+
"loss": 0.7678,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.14096185737976782,
|
| 251 |
+
"grad_norm": 2.465268528338133,
|
| 252 |
+
"learning_rate": 9.949760673863846e-06,
|
| 253 |
+
"loss": 0.7512,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1451077943615257,
|
| 258 |
+
"grad_norm": 2.487803804934397,
|
| 259 |
+
"learning_rate": 9.93900659960133e-06,
|
| 260 |
+
"loss": 0.7366,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14925373134328357,
|
| 265 |
+
"grad_norm": 2.55421871649192,
|
| 266 |
+
"learning_rate": 9.927217353857809e-06,
|
| 267 |
+
"loss": 0.7463,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.15339966832504145,
|
| 272 |
+
"grad_norm": 2.639039784633591,
|
| 273 |
+
"learning_rate": 9.914395407553444e-06,
|
| 274 |
+
"loss": 0.7443,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.15754560530679934,
|
| 279 |
+
"grad_norm": 2.501978124475521,
|
| 280 |
+
"learning_rate": 9.900543448053164e-06,
|
| 281 |
+
"loss": 0.7197,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.16169154228855723,
|
| 286 |
+
"grad_norm": 2.3337568323918214,
|
| 287 |
+
"learning_rate": 9.885664378603432e-06,
|
| 288 |
+
"loss": 0.7332,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.16583747927031509,
|
| 293 |
+
"grad_norm": 2.378783071529505,
|
| 294 |
+
"learning_rate": 9.869761317723744e-06,
|
| 295 |
+
"loss": 0.7136,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.16998341625207297,
|
| 300 |
+
"grad_norm": 2.277527337184093,
|
| 301 |
+
"learning_rate": 9.85283759855301e-06,
|
| 302 |
+
"loss": 0.7201,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.17412935323383086,
|
| 307 |
+
"grad_norm": 2.2079190034247365,
|
| 308 |
+
"learning_rate": 9.834896768150963e-06,
|
| 309 |
+
"loss": 0.7174,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.17827529021558872,
|
| 314 |
+
"grad_norm": 2.4371753499407434,
|
| 315 |
+
"learning_rate": 9.81594258675473e-06,
|
| 316 |
+
"loss": 0.7341,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1824212271973466,
|
| 321 |
+
"grad_norm": 2.2055041342497255,
|
| 322 |
+
"learning_rate": 9.795979026990717e-06,
|
| 323 |
+
"loss": 0.7024,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1865671641791045,
|
| 328 |
+
"grad_norm": 2.406773347160676,
|
| 329 |
+
"learning_rate": 9.775010273041975e-06,
|
| 330 |
+
"loss": 0.7173,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.19071310116086235,
|
| 335 |
+
"grad_norm": 2.2838878169624173,
|
| 336 |
+
"learning_rate": 9.753040719771249e-06,
|
| 337 |
+
"loss": 0.6933,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19485903814262023,
|
| 342 |
+
"grad_norm": 2.3948306779315756,
|
| 343 |
+
"learning_rate": 9.730074971799837e-06,
|
| 344 |
+
"loss": 0.7093,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.19900497512437812,
|
| 349 |
+
"grad_norm": 2.3419375521446595,
|
| 350 |
+
"learning_rate": 9.706117842542517e-06,
|
| 351 |
+
"loss": 0.7026,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.20315091210613598,
|
| 356 |
+
"grad_norm": 2.386424037340903,
|
| 357 |
+
"learning_rate": 9.681174353198687e-06,
|
| 358 |
+
"loss": 0.6928,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.20729684908789386,
|
| 363 |
+
"grad_norm": 2.352824469415509,
|
| 364 |
+
"learning_rate": 9.655249731699973e-06,
|
| 365 |
+
"loss": 0.6758,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.21144278606965175,
|
| 370 |
+
"grad_norm": 2.448238996649246,
|
| 371 |
+
"learning_rate": 9.628349411614503e-06,
|
| 372 |
+
"loss": 0.6846,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2155887230514096,
|
| 377 |
+
"grad_norm": 2.18778612059073,
|
| 378 |
+
"learning_rate": 9.600479031008072e-06,
|
| 379 |
+
"loss": 0.6783,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2197346600331675,
|
| 384 |
+
"grad_norm": 2.4786933731659384,
|
| 385 |
+
"learning_rate": 9.571644431262463e-06,
|
| 386 |
+
"loss": 0.6862,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.22388059701492538,
|
| 391 |
+
"grad_norm": 2.1664093593847165,
|
| 392 |
+
"learning_rate": 9.54185165585114e-06,
|
| 393 |
+
"loss": 0.6598,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.22802653399668324,
|
| 398 |
+
"grad_norm": 2.3439695927983024,
|
| 399 |
+
"learning_rate": 9.511106949072588e-06,
|
| 400 |
+
"loss": 0.6538,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.23217247097844113,
|
| 405 |
+
"grad_norm": 2.573211098478694,
|
| 406 |
+
"learning_rate": 9.479416754741577e-06,
|
| 407 |
+
"loss": 0.6619,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.236318407960199,
|
| 412 |
+
"grad_norm": 2.7234227642496616,
|
| 413 |
+
"learning_rate": 9.446787714838579e-06,
|
| 414 |
+
"loss": 0.6562,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.24046434494195687,
|
| 419 |
+
"grad_norm": 2.5205068601141525,
|
| 420 |
+
"learning_rate": 9.413226668117679e-06,
|
| 421 |
+
"loss": 0.6461,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.24461028192371476,
|
| 426 |
+
"grad_norm": 2.229496276338444,
|
| 427 |
+
"learning_rate": 9.37874064867323e-06,
|
| 428 |
+
"loss": 0.6469,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.24875621890547264,
|
| 433 |
+
"grad_norm": 2.3470411543105265,
|
| 434 |
+
"learning_rate": 9.343336884465577e-06,
|
| 435 |
+
"loss": 0.6426,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.25290215588723053,
|
| 440 |
+
"grad_norm": 2.3905607702569704,
|
| 441 |
+
"learning_rate": 9.307022795806125e-06,
|
| 442 |
+
"loss": 0.6278,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2570480928689884,
|
| 447 |
+
"grad_norm": 2.159892061781749,
|
| 448 |
+
"learning_rate": 9.26980599380213e-06,
|
| 449 |
+
"loss": 0.6404,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.26119402985074625,
|
| 454 |
+
"grad_norm": 2.573673941571869,
|
| 455 |
+
"learning_rate": 9.231694278761455e-06,
|
| 456 |
+
"loss": 0.629,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.26533996683250416,
|
| 461 |
+
"grad_norm": 2.1819450147556494,
|
| 462 |
+
"learning_rate": 9.192695638557723e-06,
|
| 463 |
+
"loss": 0.6297,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.269485903814262,
|
| 468 |
+
"grad_norm": 2.307987631760188,
|
| 469 |
+
"learning_rate": 9.1528182469561e-06,
|
| 470 |
+
"loss": 0.6254,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2736318407960199,
|
| 475 |
+
"grad_norm": 2.4579207582542906,
|
| 476 |
+
"learning_rate": 9.112070461900178e-06,
|
| 477 |
+
"loss": 0.6322,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2777777777777778,
|
| 482 |
+
"grad_norm": 2.3690336709142064,
|
| 483 |
+
"learning_rate": 9.070460823760197e-06,
|
| 484 |
+
"loss": 0.6164,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.28192371475953565,
|
| 489 |
+
"grad_norm": 2.7180849820218067,
|
| 490 |
+
"learning_rate": 9.027998053543079e-06,
|
| 491 |
+
"loss": 0.6298,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2860696517412935,
|
| 496 |
+
"grad_norm": 2.4829451807056118,
|
| 497 |
+
"learning_rate": 8.984691051064576e-06,
|
| 498 |
+
"loss": 0.6162,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2902155887230514,
|
| 503 |
+
"grad_norm": 2.459002185250392,
|
| 504 |
+
"learning_rate": 8.94054889308395e-06,
|
| 505 |
+
"loss": 0.6045,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2943615257048093,
|
| 510 |
+
"grad_norm": 2.2572208925258597,
|
| 511 |
+
"learning_rate": 8.895580831401563e-06,
|
| 512 |
+
"loss": 0.6112,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.29850746268656714,
|
| 517 |
+
"grad_norm": 2.276783038981904,
|
| 518 |
+
"learning_rate": 8.849796290919787e-06,
|
| 519 |
+
"loss": 0.6115,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.30265339966832505,
|
| 524 |
+
"grad_norm": 2.415622228387365,
|
| 525 |
+
"learning_rate": 8.803204867667624e-06,
|
| 526 |
+
"loss": 0.5991,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3067993366500829,
|
| 531 |
+
"grad_norm": 2.045217334174227,
|
| 532 |
+
"learning_rate": 8.755816326789469e-06,
|
| 533 |
+
"loss": 0.5909,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.31094527363184077,
|
| 538 |
+
"grad_norm": 2.3646147099166526,
|
| 539 |
+
"learning_rate": 8.70764060049842e-06,
|
| 540 |
+
"loss": 0.5929,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3150912106135987,
|
| 545 |
+
"grad_norm": 2.179575894376864,
|
| 546 |
+
"learning_rate": 8.658687785994579e-06,
|
| 547 |
+
"loss": 0.602,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.31923714759535654,
|
| 552 |
+
"grad_norm": 2.303527412372746,
|
| 553 |
+
"learning_rate": 8.608968143348765e-06,
|
| 554 |
+
"loss": 0.5911,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.32338308457711445,
|
| 559 |
+
"grad_norm": 2.184412340951817,
|
| 560 |
+
"learning_rate": 8.558492093352098e-06,
|
| 561 |
+
"loss": 0.5834,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3275290215588723,
|
| 566 |
+
"grad_norm": 2.2982310068640963,
|
| 567 |
+
"learning_rate": 8.50727021533189e-06,
|
| 568 |
+
"loss": 0.573,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.33167495854063017,
|
| 573 |
+
"grad_norm": 2.3823718371379288,
|
| 574 |
+
"learning_rate": 8.455313244934324e-06,
|
| 575 |
+
"loss": 0.5704,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3358208955223881,
|
| 580 |
+
"grad_norm": 2.308301112102417,
|
| 581 |
+
"learning_rate": 8.402632071874348e-06,
|
| 582 |
+
"loss": 0.5676,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.33996683250414594,
|
| 587 |
+
"grad_norm": 2.1617090999565782,
|
| 588 |
+
"learning_rate": 8.349237737653288e-06,
|
| 589 |
+
"loss": 0.5709,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3441127694859038,
|
| 594 |
+
"grad_norm": 2.280161952890472,
|
| 595 |
+
"learning_rate": 8.29514143324466e-06,
|
| 596 |
+
"loss": 0.5779,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3482587064676617,
|
| 601 |
+
"grad_norm": 2.2490286708106577,
|
| 602 |
+
"learning_rate": 8.24035449674863e-06,
|
| 603 |
+
"loss": 0.5564,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3524046434494196,
|
| 608 |
+
"grad_norm": 2.2807466234509777,
|
| 609 |
+
"learning_rate": 8.184888411015655e-06,
|
| 610 |
+
"loss": 0.5554,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.35655058043117743,
|
| 615 |
+
"grad_norm": 2.180528318028485,
|
| 616 |
+
"learning_rate": 8.128754801239781e-06,
|
| 617 |
+
"loss": 0.5606,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.36069651741293535,
|
| 622 |
+
"grad_norm": 2.3314447428823,
|
| 623 |
+
"learning_rate": 8.071965432522107e-06,
|
| 624 |
+
"loss": 0.5593,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3648424543946932,
|
| 629 |
+
"grad_norm": 2.2300833962237974,
|
| 630 |
+
"learning_rate": 8.01453220740492e-06,
|
| 631 |
+
"loss": 0.5513,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.36898839137645106,
|
| 636 |
+
"grad_norm": 2.4375150414786044,
|
| 637 |
+
"learning_rate": 7.956467163377037e-06,
|
| 638 |
+
"loss": 0.5451,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.373134328358209,
|
| 643 |
+
"grad_norm": 2.345552450223764,
|
| 644 |
+
"learning_rate": 7.89778247035085e-06,
|
| 645 |
+
"loss": 0.5513,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.37728026533996684,
|
| 650 |
+
"grad_norm": 2.302225232727345,
|
| 651 |
+
"learning_rate": 7.838490428111625e-06,
|
| 652 |
+
"loss": 0.5454,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3814262023217247,
|
| 657 |
+
"grad_norm": 2.3901207498612274,
|
| 658 |
+
"learning_rate": 7.77860346373957e-06,
|
| 659 |
+
"loss": 0.5354,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3855721393034826,
|
| 664 |
+
"grad_norm": 2.1013639490757527,
|
| 665 |
+
"learning_rate": 7.718134129005238e-06,
|
| 666 |
+
"loss": 0.5338,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.38971807628524047,
|
| 671 |
+
"grad_norm": 2.1095521859807604,
|
| 672 |
+
"learning_rate": 7.657095097738793e-06,
|
| 673 |
+
"loss": 0.5281,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3938640132669983,
|
| 678 |
+
"grad_norm": 2.1090684046109778,
|
| 679 |
+
"learning_rate": 7.59549916317367e-06,
|
| 680 |
+
"loss": 0.5269,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.39800995024875624,
|
| 685 |
+
"grad_norm": 2.1756383908491466,
|
| 686 |
+
"learning_rate": 7.533359235265248e-06,
|
| 687 |
+
"loss": 0.5404,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4021558872305141,
|
| 692 |
+
"grad_norm": 2.067392945377543,
|
| 693 |
+
"learning_rate": 7.470688337985029e-06,
|
| 694 |
+
"loss": 0.5257,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.40630182421227196,
|
| 699 |
+
"grad_norm": 2.178779204316918,
|
| 700 |
+
"learning_rate": 7.407499606590934e-06,
|
| 701 |
+
"loss": 0.5108,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.41044776119402987,
|
| 706 |
+
"grad_norm": 2.1482896888352303,
|
| 707 |
+
"learning_rate": 7.343806284874268e-06,
|
| 708 |
+
"loss": 0.5195,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.41459369817578773,
|
| 713 |
+
"grad_norm": 2.122877082306712,
|
| 714 |
+
"learning_rate": 7.279621722383939e-06,
|
| 715 |
+
"loss": 0.5238,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.4187396351575456,
|
| 720 |
+
"grad_norm": 2.513475103077436,
|
| 721 |
+
"learning_rate": 7.214959371628522e-06,
|
| 722 |
+
"loss": 0.5239,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.4228855721393035,
|
| 727 |
+
"grad_norm": 2.256666838754752,
|
| 728 |
+
"learning_rate": 7.149832785256718e-06,
|
| 729 |
+
"loss": 0.5111,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.42703150912106136,
|
| 734 |
+
"grad_norm": 2.126071871777049,
|
| 735 |
+
"learning_rate": 7.084255613216855e-06,
|
| 736 |
+
"loss": 0.5109,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4311774461028192,
|
| 741 |
+
"grad_norm": 2.2958892839783003,
|
| 742 |
+
"learning_rate": 7.018241599895974e-06,
|
| 743 |
+
"loss": 0.5075,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.43532338308457713,
|
| 748 |
+
"grad_norm": 2.1249877107238166,
|
| 749 |
+
"learning_rate": 6.95180458123913e-06,
|
| 750 |
+
"loss": 0.5021,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.439469320066335,
|
| 755 |
+
"grad_norm": 2.268170350249536,
|
| 756 |
+
"learning_rate": 6.8849584818494984e-06,
|
| 757 |
+
"loss": 0.5075,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.44361525704809285,
|
| 762 |
+
"grad_norm": 2.1889766909888304,
|
| 763 |
+
"learning_rate": 6.817717312069913e-06,
|
| 764 |
+
"loss": 0.4951,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.44776119402985076,
|
| 769 |
+
"grad_norm": 2.116309228498843,
|
| 770 |
+
"learning_rate": 6.750095165046415e-06,
|
| 771 |
+
"loss": 0.4993,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4519071310116086,
|
| 776 |
+
"grad_norm": 2.2014038727267735,
|
| 777 |
+
"learning_rate": 6.682106213774459e-06,
|
| 778 |
+
"loss": 0.5014,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4560530679933665,
|
| 783 |
+
"grad_norm": 2.1791180363289833,
|
| 784 |
+
"learning_rate": 6.6137647081283776e-06,
|
| 785 |
+
"loss": 0.5004,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4601990049751244,
|
| 790 |
+
"grad_norm": 2.181037493892715,
|
| 791 |
+
"learning_rate": 6.545084971874738e-06,
|
| 792 |
+
"loss": 0.4869,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.46434494195688225,
|
| 797 |
+
"grad_norm": 2.1234780810442317,
|
| 798 |
+
"learning_rate": 6.476081399670212e-06,
|
| 799 |
+
"loss": 0.487,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4684908789386401,
|
| 804 |
+
"grad_norm": 2.1577583891835137,
|
| 805 |
+
"learning_rate": 6.406768454044581e-06,
|
| 806 |
+
"loss": 0.4925,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.472636815920398,
|
| 811 |
+
"grad_norm": 2.1435505057127053,
|
| 812 |
+
"learning_rate": 6.337160662369519e-06,
|
| 813 |
+
"loss": 0.4852,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4767827529021559,
|
| 818 |
+
"grad_norm": 2.1004015267155056,
|
| 819 |
+
"learning_rate": 6.267272613813789e-06,
|
| 820 |
+
"loss": 0.4894,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.48092868988391374,
|
| 825 |
+
"grad_norm": 2.275508822904309,
|
| 826 |
+
"learning_rate": 6.19711895628548e-06,
|
| 827 |
+
"loss": 0.4783,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.48507462686567165,
|
| 832 |
+
"grad_norm": 2.0957567591112456,
|
| 833 |
+
"learning_rate": 6.126714393361939e-06,
|
| 834 |
+
"loss": 0.4811,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4892205638474295,
|
| 839 |
+
"grad_norm": 2.2453086852427138,
|
| 840 |
+
"learning_rate": 6.056073681208038e-06,
|
| 841 |
+
"loss": 0.4778,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.49336650082918737,
|
| 846 |
+
"grad_norm": 2.086146491253835,
|
| 847 |
+
"learning_rate": 5.985211625483403e-06,
|
| 848 |
+
"loss": 0.468,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4975124378109453,
|
| 853 |
+
"grad_norm": 2.08746099925638,
|
| 854 |
+
"learning_rate": 5.914143078239293e-06,
|
| 855 |
+
"loss": 0.4702,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.5016583747927031,
|
| 860 |
+
"grad_norm": 2.183555664254409,
|
| 861 |
+
"learning_rate": 5.842882934805731e-06,
|
| 862 |
+
"loss": 0.4683,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5058043117744611,
|
| 867 |
+
"grad_norm": 2.141477284770924,
|
| 868 |
+
"learning_rate": 5.771446130669589e-06,
|
| 869 |
+
"loss": 0.4692,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5099502487562189,
|
| 874 |
+
"grad_norm": 1.999530939511801,
|
| 875 |
+
"learning_rate": 5.6998476383442345e-06,
|
| 876 |
+
"loss": 0.4672,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5140961857379768,
|
| 881 |
+
"grad_norm": 2.3861168673535476,
|
| 882 |
+
"learning_rate": 5.628102464231429e-06,
|
| 883 |
+
"loss": 0.4647,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5182421227197347,
|
| 888 |
+
"grad_norm": 2.112621163685018,
|
| 889 |
+
"learning_rate": 5.556225645476119e-06,
|
| 890 |
+
"loss": 0.4569,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5223880597014925,
|
| 895 |
+
"grad_norm": 2.2162792041456023,
|
| 896 |
+
"learning_rate": 5.4842322468147926e-06,
|
| 897 |
+
"loss": 0.4636,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5265339966832504,
|
| 902 |
+
"grad_norm": 2.1411819631797444,
|
| 903 |
+
"learning_rate": 5.412137357418037e-06,
|
| 904 |
+
"loss": 0.4609,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5306799336650083,
|
| 909 |
+
"grad_norm": 2.0506863143987277,
|
| 910 |
+
"learning_rate": 5.339956087727985e-06,
|
| 911 |
+
"loss": 0.4501,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5348258706467661,
|
| 916 |
+
"grad_norm": 2.0133156015781193,
|
| 917 |
+
"learning_rate": 5.2677035662913116e-06,
|
| 918 |
+
"loss": 0.4507,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.538971807628524,
|
| 923 |
+
"grad_norm": 2.0603049501006336,
|
| 924 |
+
"learning_rate": 5.195394936588409e-06,
|
| 925 |
+
"loss": 0.441,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.543117744610282,
|
| 930 |
+
"grad_norm": 2.012154333900926,
|
| 931 |
+
"learning_rate": 5.123045353859465e-06,
|
| 932 |
+
"loss": 0.4577,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5472636815920398,
|
| 937 |
+
"grad_norm": 2.0263592161269184,
|
| 938 |
+
"learning_rate": 5.050669981928056e-06,
|
| 939 |
+
"loss": 0.4568,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5514096185737977,
|
| 944 |
+
"grad_norm": 2.0917399842428157,
|
| 945 |
+
"learning_rate": 4.978283990022936e-06,
|
| 946 |
+
"loss": 0.4532,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5555555555555556,
|
| 951 |
+
"grad_norm": 2.188577844097009,
|
| 952 |
+
"learning_rate": 4.905902549598719e-06,
|
| 953 |
+
"loss": 0.4382,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5597014925373134,
|
| 958 |
+
"grad_norm": 2.199659099615686,
|
| 959 |
+
"learning_rate": 4.833540831156062e-06,
|
| 960 |
+
"loss": 0.4437,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5638474295190713,
|
| 965 |
+
"grad_norm": 2.0640167310940027,
|
| 966 |
+
"learning_rate": 4.761214001062079e-06,
|
| 967 |
+
"loss": 0.4379,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5679933665008292,
|
| 972 |
+
"grad_norm": 2.179786737229522,
|
| 973 |
+
"learning_rate": 4.688937218371592e-06,
|
| 974 |
+
"loss": 0.4409,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.572139303482587,
|
| 979 |
+
"grad_norm": 2.074519322035315,
|
| 980 |
+
"learning_rate": 4.616725631649938e-06,
|
| 981 |
+
"loss": 0.4383,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5762852404643449,
|
| 986 |
+
"grad_norm": 2.1399873216394036,
|
| 987 |
+
"learning_rate": 4.544594375797969e-06,
|
| 988 |
+
"loss": 0.4288,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5804311774461028,
|
| 993 |
+
"grad_norm": 2.1123595173850935,
|
| 994 |
+
"learning_rate": 4.472558568879901e-06,
|
| 995 |
+
"loss": 0.4351,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5845771144278606,
|
| 1000 |
+
"grad_norm": 2.031607056740364,
|
| 1001 |
+
"learning_rate": 4.400633308954713e-06,
|
| 1002 |
+
"loss": 0.4244,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5887230514096186,
|
| 1007 |
+
"grad_norm": 1.9626506911231658,
|
| 1008 |
+
"learning_rate": 4.3288336709117246e-06,
|
| 1009 |
+
"loss": 0.4192,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5928689883913765,
|
| 1014 |
+
"grad_norm": 2.10360277006623,
|
| 1015 |
+
"learning_rate": 4.257174703311032e-06,
|
| 1016 |
+
"loss": 0.4191,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5970149253731343,
|
| 1021 |
+
"grad_norm": 2.008208706865011,
|
| 1022 |
+
"learning_rate": 4.185671425229477e-06,
|
| 1023 |
+
"loss": 0.4191,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6011608623548922,
|
| 1028 |
+
"grad_norm": 2.0304909176887977,
|
| 1029 |
+
"learning_rate": 4.11433882311277e-06,
|
| 1030 |
+
"loss": 0.4193,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6053067993366501,
|
| 1035 |
+
"grad_norm": 2.0261313089901303,
|
| 1036 |
+
"learning_rate": 4.043191847634469e-06,
|
| 1037 |
+
"loss": 0.4164,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6094527363184079,
|
| 1042 |
+
"grad_norm": 2.0660700135752945,
|
| 1043 |
+
"learning_rate": 3.9722454105624545e-06,
|
| 1044 |
+
"loss": 0.4121,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6135986733001658,
|
| 1049 |
+
"grad_norm": 1.9536775589165198,
|
| 1050 |
+
"learning_rate": 3.901514381633555e-06,
|
| 1051 |
+
"loss": 0.4184,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6177446102819237,
|
| 1056 |
+
"grad_norm": 2.096984004473796,
|
| 1057 |
+
"learning_rate": 3.831013585436985e-06,
|
| 1058 |
+
"loss": 0.4209,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.6218905472636815,
|
| 1063 |
+
"grad_norm": 1.9585159850477776,
|
| 1064 |
+
"learning_rate": 3.7607577983072486e-06,
|
| 1065 |
+
"loss": 0.4076,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6260364842454395,
|
| 1070 |
+
"grad_norm": 2.021551204900084,
|
| 1071 |
+
"learning_rate": 3.6907617452271394e-06,
|
| 1072 |
+
"loss": 0.4054,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6301824212271974,
|
| 1077 |
+
"grad_norm": 2.2587727736565095,
|
| 1078 |
+
"learning_rate": 3.621040096741526e-06,
|
| 1079 |
+
"loss": 0.4024,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6343283582089553,
|
| 1084 |
+
"grad_norm": 2.1288856262867517,
|
| 1085 |
+
"learning_rate": 3.55160746588254e-06,
|
| 1086 |
+
"loss": 0.4071,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6384742951907131,
|
| 1091 |
+
"grad_norm": 2.0991322193698485,
|
| 1092 |
+
"learning_rate": 3.482478405106803e-06,
|
| 1093 |
+
"loss": 0.4119,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.642620232172471,
|
| 1098 |
+
"grad_norm": 2.1933844049539153,
|
| 1099 |
+
"learning_rate": 3.4136674032453787e-06,
|
| 1100 |
+
"loss": 0.4057,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6467661691542289,
|
| 1105 |
+
"grad_norm": 1.9800816023792092,
|
| 1106 |
+
"learning_rate": 3.34518888246703e-06,
|
| 1107 |
+
"loss": 0.4081,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6509121061359867,
|
| 1112 |
+
"grad_norm": 2.134679203262353,
|
| 1113 |
+
"learning_rate": 3.2770571952554674e-06,
|
| 1114 |
+
"loss": 0.4055,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6550580431177446,
|
| 1119 |
+
"grad_norm": 2.0810403529065824,
|
| 1120 |
+
"learning_rate": 3.2092866214011984e-06,
|
| 1121 |
+
"loss": 0.3953,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6592039800995025,
|
| 1126 |
+
"grad_norm": 1.9681607355510164,
|
| 1127 |
+
"learning_rate": 3.141891365008609e-06,
|
| 1128 |
+
"loss": 0.4043,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6633499170812603,
|
| 1133 |
+
"grad_norm": 2.0452177198825683,
|
| 1134 |
+
"learning_rate": 3.0748855515189104e-06,
|
| 1135 |
+
"loss": 0.4,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6674958540630183,
|
| 1140 |
+
"grad_norm": 2.0817931631950914,
|
| 1141 |
+
"learning_rate": 3.00828322474958e-06,
|
| 1142 |
+
"loss": 0.3938,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6716417910447762,
|
| 1147 |
+
"grad_norm": 2.0579281745324876,
|
| 1148 |
+
"learning_rate": 2.942098343950891e-06,
|
| 1149 |
+
"loss": 0.3848,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.675787728026534,
|
| 1154 |
+
"grad_norm": 2.0378046645796446,
|
| 1155 |
+
"learning_rate": 2.8763447808801914e-06,
|
| 1156 |
+
"loss": 0.3895,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6799336650082919,
|
| 1161 |
+
"grad_norm": 2.089276342993108,
|
| 1162 |
+
"learning_rate": 2.8110363168944976e-06,
|
| 1163 |
+
"loss": 0.3947,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6840796019900498,
|
| 1168 |
+
"grad_norm": 1.9512325130647332,
|
| 1169 |
+
"learning_rate": 2.7461866400620506e-06,
|
| 1170 |
+
"loss": 0.3981,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6882255389718076,
|
| 1175 |
+
"grad_norm": 2.006701131264969,
|
| 1176 |
+
"learning_rate": 2.6818093422934254e-06,
|
| 1177 |
+
"loss": 0.3908,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6923714759535655,
|
| 1182 |
+
"grad_norm": 2.134162418048178,
|
| 1183 |
+
"learning_rate": 2.617917916492776e-06,
|
| 1184 |
+
"loss": 0.3922,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6965174129353234,
|
| 1189 |
+
"grad_norm": 2.1121757305097226,
|
| 1190 |
+
"learning_rate": 2.5545257537298497e-06,
|
| 1191 |
+
"loss": 0.3903,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7006633499170812,
|
| 1196 |
+
"grad_norm": 2.06027998760199,
|
| 1197 |
+
"learning_rate": 2.491646140433346e-06,
|
| 1198 |
+
"loss": 0.3795,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7048092868988391,
|
| 1203 |
+
"grad_norm": 2.1221824714111035,
|
| 1204 |
+
"learning_rate": 2.4292922556061877e-06,
|
| 1205 |
+
"loss": 0.3891,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7089552238805971,
|
| 1210 |
+
"grad_norm": 2.0687201549733905,
|
| 1211 |
+
"learning_rate": 2.367477168063326e-06,
|
| 1212 |
+
"loss": 0.3804,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7131011608623549,
|
| 1217 |
+
"grad_norm": 2.1096088311491323,
|
| 1218 |
+
"learning_rate": 2.3062138336926406e-06,
|
| 1219 |
+
"loss": 0.3861,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7172470978441128,
|
| 1224 |
+
"grad_norm": 1.9369468070487719,
|
| 1225 |
+
"learning_rate": 2.245515092739488e-06,
|
| 1226 |
+
"loss": 0.38,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7213930348258707,
|
| 1231 |
+
"grad_norm": 2.0347831655952198,
|
| 1232 |
+
"learning_rate": 2.185393667115513e-06,
|
| 1233 |
+
"loss": 0.3697,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7255389718076285,
|
| 1238 |
+
"grad_norm": 2.0345030208735735,
|
| 1239 |
+
"learning_rate": 2.125862157732245e-06,
|
| 1240 |
+
"loss": 0.3686,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7296849087893864,
|
| 1245 |
+
"grad_norm": 1.9521605057602363,
|
| 1246 |
+
"learning_rate": 2.066933041860059e-06,
|
| 1247 |
+
"loss": 0.3804,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.7338308457711443,
|
| 1252 |
+
"grad_norm": 2.0778756164223364,
|
| 1253 |
+
"learning_rate": 2.0086186705130545e-06,
|
| 1254 |
+
"loss": 0.3804,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7379767827529021,
|
| 1259 |
+
"grad_norm": 2.1591475619282847,
|
| 1260 |
+
"learning_rate": 1.9509312658603954e-06,
|
| 1261 |
+
"loss": 0.3752,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.74212271973466,
|
| 1266 |
+
"grad_norm": 2.0870382174991993,
|
| 1267 |
+
"learning_rate": 1.8938829186646484e-06,
|
| 1268 |
+
"loss": 0.3705,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.746268656716418,
|
| 1273 |
+
"grad_norm": 1.9907262732293516,
|
| 1274 |
+
"learning_rate": 1.8374855857476687e-06,
|
| 1275 |
+
"loss": 0.3646,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7504145936981758,
|
| 1280 |
+
"grad_norm": 2.112014948477697,
|
| 1281 |
+
"learning_rate": 1.7817510874845585e-06,
|
| 1282 |
+
"loss": 0.378,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7545605306799337,
|
| 1287 |
+
"grad_norm": 2.00612301158268,
|
| 1288 |
+
"learning_rate": 1.7266911053262196e-06,
|
| 1289 |
+
"loss": 0.3722,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7587064676616916,
|
| 1294 |
+
"grad_norm": 1.9805581686834959,
|
| 1295 |
+
"learning_rate": 1.6723171793510363e-06,
|
| 1296 |
+
"loss": 0.3571,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7628524046434494,
|
| 1301 |
+
"grad_norm": 2.1705390220024876,
|
| 1302 |
+
"learning_rate": 1.6186407058461622e-06,
|
| 1303 |
+
"loss": 0.3717,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7669983416252073,
|
| 1308 |
+
"grad_norm": 1.997191295027296,
|
| 1309 |
+
"learning_rate": 1.5656729349189742e-06,
|
| 1310 |
+
"loss": 0.3702,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7711442786069652,
|
| 1315 |
+
"grad_norm": 1.9480055411783619,
|
| 1316 |
+
"learning_rate": 1.5134249681391416e-06,
|
| 1317 |
+
"loss": 0.3662,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.775290215588723,
|
| 1322 |
+
"grad_norm": 2.1202087648855708,
|
| 1323 |
+
"learning_rate": 1.4619077562118477e-06,
|
| 1324 |
+
"loss": 0.3624,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7794361525704809,
|
| 1329 |
+
"grad_norm": 2.0287381557784436,
|
| 1330 |
+
"learning_rate": 1.411132096682606e-06,
|
| 1331 |
+
"loss": 0.3647,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7835820895522388,
|
| 1336 |
+
"grad_norm": 2.1095441746045074,
|
| 1337 |
+
"learning_rate": 1.3611086316742057e-06,
|
| 1338 |
+
"loss": 0.3666,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7877280265339967,
|
| 1343 |
+
"grad_norm": 2.0271810521548077,
|
| 1344 |
+
"learning_rate": 1.3118478456562073e-06,
|
| 1345 |
+
"loss": 0.3644,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7918739635157546,
|
| 1350 |
+
"grad_norm": 2.017879350372662,
|
| 1351 |
+
"learning_rate": 1.2633600632474962e-06,
|
| 1352 |
+
"loss": 0.3555,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7960199004975125,
|
| 1357 |
+
"grad_norm": 2.0158894267700416,
|
| 1358 |
+
"learning_rate": 1.2156554470523364e-06,
|
| 1359 |
+
"loss": 0.3579,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.8001658374792703,
|
| 1364 |
+
"grad_norm": 2.135171063836108,
|
| 1365 |
+
"learning_rate": 1.1687439955303764e-06,
|
| 1366 |
+
"loss": 0.3549,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.8043117744610282,
|
| 1371 |
+
"grad_norm": 2.051748632880903,
|
| 1372 |
+
"learning_rate": 1.1226355409010686e-06,
|
| 1373 |
+
"loss": 0.3567,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8084577114427861,
|
| 1378 |
+
"grad_norm": 1.9895182337910944,
|
| 1379 |
+
"learning_rate": 1.0773397470829145e-06,
|
| 1380 |
+
"loss": 0.3506,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.8126036484245439,
|
| 1385 |
+
"grad_norm": 2.0734006194975025,
|
| 1386 |
+
"learning_rate": 1.032866107667999e-06,
|
| 1387 |
+
"loss": 0.3527,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8167495854063018,
|
| 1392 |
+
"grad_norm": 1.9987662217078168,
|
| 1393 |
+
"learning_rate": 9.892239439322243e-07,
|
| 1394 |
+
"loss": 0.3581,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8208955223880597,
|
| 1399 |
+
"grad_norm": 2.0823516989471083,
|
| 1400 |
+
"learning_rate": 9.464224028816427e-07,
|
| 1401 |
+
"loss": 0.3665,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.8250414593698175,
|
| 1406 |
+
"grad_norm": 2.051484969572584,
|
| 1407 |
+
"learning_rate": 9.044704553353323e-07,
|
| 1408 |
+
"loss": 0.3537,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8291873963515755,
|
| 1413 |
+
"grad_norm": 2.019033384913334,
|
| 1414 |
+
"learning_rate": 8.633768940451981e-07,
|
| 1415 |
+
"loss": 0.3484,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8333333333333334,
|
| 1420 |
+
"grad_norm": 2.071090455759551,
|
| 1421 |
+
"learning_rate": 8.231503318530814e-07,
|
| 1422 |
+
"loss": 0.3515,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8374792703150912,
|
| 1427 |
+
"grad_norm": 1.987573001153748,
|
| 1428 |
+
"learning_rate": 7.837991998855899e-07,
|
| 1429 |
+
"loss": 0.3458,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8416252072968491,
|
| 1434 |
+
"grad_norm": 2.1342753070628544,
|
| 1435 |
+
"learning_rate": 7.453317457870096e-07,
|
| 1436 |
+
"loss": 0.3481,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.845771144278607,
|
| 1441 |
+
"grad_norm": 1.9443796196582224,
|
| 1442 |
+
"learning_rate": 7.077560319906696e-07,
|
| 1443 |
+
"loss": 0.3468,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8499170812603648,
|
| 1448 |
+
"grad_norm": 2.1533371687865737,
|
| 1449 |
+
"learning_rate": 6.710799340291341e-07,
|
| 1450 |
+
"loss": 0.3516,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8540630182421227,
|
| 1455 |
+
"grad_norm": 2.046122945823328,
|
| 1456 |
+
"learning_rate": 6.353111388835564e-07,
|
| 1457 |
+
"loss": 0.348,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8582089552238806,
|
| 1462 |
+
"grad_norm": 2.0087071630631192,
|
| 1463 |
+
"learning_rate": 6.00457143372557e-07,
|
| 1464 |
+
"loss": 0.3464,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8623548922056384,
|
| 1469 |
+
"grad_norm": 2.031472375135797,
|
| 1470 |
+
"learning_rate": 5.665252525809583e-07,
|
| 1471 |
+
"loss": 0.351,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.8665008291873963,
|
| 1476 |
+
"grad_norm": 2.085600740397554,
|
| 1477 |
+
"learning_rate": 5.335225783287051e-07,
|
| 1478 |
+
"loss": 0.3522,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8706467661691543,
|
| 1483 |
+
"grad_norm": 2.0629885384115543,
|
| 1484 |
+
"learning_rate": 5.014560376802913e-07,
|
| 1485 |
+
"loss": 0.3507,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.8747927031509121,
|
| 1490 |
+
"grad_norm": 2.0514710921617265,
|
| 1491 |
+
"learning_rate": 4.703323514950042e-07,
|
| 1492 |
+
"loss": 0.3414,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.87893864013267,
|
| 1497 |
+
"grad_norm": 2.0910097285997415,
|
| 1498 |
+
"learning_rate": 4.401580430182928e-07,
|
| 1499 |
+
"loss": 0.3432,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8830845771144279,
|
| 1504 |
+
"grad_norm": 2.007785736762621,
|
| 1505 |
+
"learning_rate": 4.1093943651455305e-07,
|
| 1506 |
+
"loss": 0.3479,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8872305140961857,
|
| 1511 |
+
"grad_norm": 2.0849923696855326,
|
| 1512 |
+
"learning_rate": 3.826826559416219e-07,
|
| 1513 |
+
"loss": 0.3472,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8913764510779436,
|
| 1518 |
+
"grad_norm": 2.0132819974862843,
|
| 1519 |
+
"learning_rate": 3.5539362366724784e-07,
|
| 1520 |
+
"loss": 0.3399,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8955223880597015,
|
| 1525 |
+
"grad_norm": 2.074275526756306,
|
| 1526 |
+
"learning_rate": 3.290780592278148e-07,
|
| 1527 |
+
"loss": 0.3416,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8996683250414593,
|
| 1532 |
+
"grad_norm": 2.05724879908193,
|
| 1533 |
+
"learning_rate": 3.0374147812958387e-07,
|
| 1534 |
+
"loss": 0.3505,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.9038142620232172,
|
| 1539 |
+
"grad_norm": 2.157618991705193,
|
| 1540 |
+
"learning_rate": 2.7938919069268654e-07,
|
| 1541 |
+
"loss": 0.3505,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.9079601990049752,
|
| 1546 |
+
"grad_norm": 2.1499572962789775,
|
| 1547 |
+
"learning_rate": 2.5602630093813253e-07,
|
| 1548 |
+
"loss": 0.3546,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.912106135986733,
|
| 1553 |
+
"grad_norm": 2.0332963522319485,
|
| 1554 |
+
"learning_rate": 2.3365770551805223e-07,
|
| 1555 |
+
"loss": 0.3411,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.9162520729684909,
|
| 1560 |
+
"grad_norm": 1.9439734508841187,
|
| 1561 |
+
"learning_rate": 2.1228809268940164e-07,
|
| 1562 |
+
"loss": 0.3467,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9203980099502488,
|
| 1567 |
+
"grad_norm": 2.01922575745273,
|
| 1568 |
+
"learning_rate": 1.919219413313478e-07,
|
| 1569 |
+
"loss": 0.3456,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.9245439469320066,
|
| 1574 |
+
"grad_norm": 2.0723878543911143,
|
| 1575 |
+
"learning_rate": 1.725635200065323e-07,
|
| 1576 |
+
"loss": 0.3351,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.9286898839137645,
|
| 1581 |
+
"grad_norm": 2.02921530849707,
|
| 1582 |
+
"learning_rate": 1.5421688606642392e-07,
|
| 1583 |
+
"loss": 0.3399,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9328358208955224,
|
| 1588 |
+
"grad_norm": 2.0191708205978904,
|
| 1589 |
+
"learning_rate": 1.3688588480092913e-07,
|
| 1590 |
+
"loss": 0.3401,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.9369817578772802,
|
| 1595 |
+
"grad_norm": 2.1261706152763393,
|
| 1596 |
+
"learning_rate": 1.205741486324552e-07,
|
| 1597 |
+
"loss": 0.3417,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9411276948590381,
|
| 1602 |
+
"grad_norm": 2.1058790836653287,
|
| 1603 |
+
"learning_rate": 1.0528509635458873e-07,
|
| 1604 |
+
"loss": 0.3404,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.945273631840796,
|
| 1609 |
+
"grad_norm": 2.074684841336411,
|
| 1610 |
+
"learning_rate": 9.102193241554757e-08,
|
| 1611 |
+
"loss": 0.3375,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9494195688225538,
|
| 1616 |
+
"grad_norm": 1.991625852605994,
|
| 1617 |
+
"learning_rate": 7.778764624655433e-08,
|
| 1618 |
+
"loss": 0.3322,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9535655058043118,
|
| 1623 |
+
"grad_norm": 2.104924653809554,
|
| 1624 |
+
"learning_rate": 6.558501163527964e-08,
|
| 1625 |
+
"loss": 0.336,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9577114427860697,
|
| 1630 |
+
"grad_norm": 2.0384736104779226,
|
| 1631 |
+
"learning_rate": 5.44165861444812e-08,
|
| 1632 |
+
"loss": 0.3448,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9618573797678275,
|
| 1637 |
+
"grad_norm": 2.0936118845033405,
|
| 1638 |
+
"learning_rate": 4.428471057596362e-08,
|
| 1639 |
+
"loss": 0.3381,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.9660033167495854,
|
| 1644 |
+
"grad_norm": 2.0037182601593466,
|
| 1645 |
+
"learning_rate": 3.519150847996422e-08,
|
| 1646 |
+
"loss": 0.3457,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.9701492537313433,
|
| 1651 |
+
"grad_norm": 2.09482647445627,
|
| 1652 |
+
"learning_rate": 2.713888571007739e-08,
|
| 1653 |
+
"loss": 0.3398,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.9742951907131011,
|
| 1658 |
+
"grad_norm": 2.085192366585127,
|
| 1659 |
+
"learning_rate": 2.012853002380466e-08,
|
| 1660 |
+
"loss": 0.3442,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.978441127694859,
|
| 1665 |
+
"grad_norm": 1.9611348789077625,
|
| 1666 |
+
"learning_rate": 1.4161910728816009e-08,
|
| 1667 |
+
"loss": 0.3373,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9825870646766169,
|
| 1672 |
+
"grad_norm": 2.109979592147509,
|
| 1673 |
+
"learning_rate": 9.240278374995637e-09,
|
| 1674 |
+
"loss": 0.3365,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9867330016583747,
|
| 1679 |
+
"grad_norm": 2.112199186520632,
|
| 1680 |
+
"learning_rate": 5.364664492337746e-09,
|
| 1681 |
+
"loss": 0.3407,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9908789386401327,
|
| 1686 |
+
"grad_norm": 2.1961533969052236,
|
| 1687 |
+
"learning_rate": 2.5358813747500266e-09,
|
| 1688 |
+
"loss": 0.3407,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.9950248756218906,
|
| 1693 |
+
"grad_norm": 2.050107312724138,
|
| 1694 |
+
"learning_rate": 7.545219097987444e-10,
|
| 1695 |
+
"loss": 0.3484,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9991708126036484,
|
| 1700 |
+
"grad_norm": 1.9981992682594107,
|
| 1701 |
+
"learning_rate": 2.0959454449243076e-11,
|
| 1702 |
+
"loss": 0.345,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 1.0,
|
| 1707 |
+
"eval_loss": 0.3594575822353363,
|
| 1708 |
+
"eval_runtime": 0.9533,
|
| 1709 |
+
"eval_samples_per_second": 3.147,
|
| 1710 |
+
"eval_steps_per_second": 1.049,
|
| 1711 |
+
"step": 1206
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.0,
|
| 1715 |
+
"step": 1206,
|
| 1716 |
+
"total_flos": 252459519836160.0,
|
| 1717 |
+
"train_loss": 0.5262548677858033,
|
| 1718 |
+
"train_runtime": 29272.1392,
|
| 1719 |
+
"train_samples_per_second": 1.318,
|
| 1720 |
+
"train_steps_per_second": 0.041
|
| 1721 |
+
}
|
| 1722 |
+
],
|
| 1723 |
+
"logging_steps": 5,
|
| 1724 |
+
"max_steps": 1206,
|
| 1725 |
+
"num_input_tokens_seen": 0,
|
| 1726 |
+
"num_train_epochs": 1,
|
| 1727 |
+
"save_steps": 100,
|
| 1728 |
+
"stateful_callbacks": {
|
| 1729 |
+
"TrainerControl": {
|
| 1730 |
+
"args": {
|
| 1731 |
+
"should_epoch_stop": false,
|
| 1732 |
+
"should_evaluate": false,
|
| 1733 |
+
"should_log": false,
|
| 1734 |
+
"should_save": true,
|
| 1735 |
+
"should_training_stop": true
|
| 1736 |
+
},
|
| 1737 |
+
"attributes": {}
|
| 1738 |
+
}
|
| 1739 |
+
},
|
| 1740 |
+
"total_flos": 252459519836160.0,
|
| 1741 |
+
"train_batch_size": 2,
|
| 1742 |
+
"trial_name": null,
|
| 1743 |
+
"trial_params": null
|
| 1744 |
+
}
|