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 +1751 -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_stylemix
|
| 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_stylemix
|
| 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: 0.3757
|
| 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 |
+
| 0.3657 | 0.9996 | 1211 | 0.3757 |
|
| 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": 0.9995872884853487,
|
| 3 |
+
"total_flos": 253506418114560.0,
|
| 4 |
+
"train_loss": 0.5435279840826496,
|
| 5 |
+
"train_runtime": 29545.3245,
|
| 6 |
+
"train_samples": 98881,
|
| 7 |
+
"train_samples_per_second": 1.312,
|
| 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": 0.9995872884853487,
|
| 3 |
+
"total_flos": 253506418114560.0,
|
| 4 |
+
"train_loss": 0.5435279840826496,
|
| 5 |
+
"train_runtime": 29545.3245,
|
| 6 |
+
"train_samples": 98881,
|
| 7 |
+
"train_samples_per_second": 1.312,
|
| 8 |
+
"train_steps_per_second": 0.041
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1751 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9995872884853487,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1211,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0008254230293025176,
|
| 13 |
+
"grad_norm": 24.685088561067097,
|
| 14 |
+
"learning_rate": 8.19672131147541e-08,
|
| 15 |
+
"loss": 1.4397,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004127115146512587,
|
| 20 |
+
"grad_norm": 23.945153463040807,
|
| 21 |
+
"learning_rate": 4.0983606557377047e-07,
|
| 22 |
+
"loss": 1.4292,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008254230293025175,
|
| 27 |
+
"grad_norm": 8.201914286275175,
|
| 28 |
+
"learning_rate": 8.196721311475409e-07,
|
| 29 |
+
"loss": 1.3321,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.012381345439537762,
|
| 34 |
+
"grad_norm": 11.180188226746075,
|
| 35 |
+
"learning_rate": 1.2295081967213116e-06,
|
| 36 |
+
"loss": 1.1764,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01650846058605035,
|
| 41 |
+
"grad_norm": 3.0006985745643777,
|
| 42 |
+
"learning_rate": 1.6393442622950819e-06,
|
| 43 |
+
"loss": 1.0311,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.020635575732562937,
|
| 48 |
+
"grad_norm": 3.0231996454961116,
|
| 49 |
+
"learning_rate": 2.0491803278688526e-06,
|
| 50 |
+
"loss": 0.9824,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.024762690879075525,
|
| 55 |
+
"grad_norm": 2.394417208637789,
|
| 56 |
+
"learning_rate": 2.459016393442623e-06,
|
| 57 |
+
"loss": 0.9428,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.028889806025588115,
|
| 62 |
+
"grad_norm": 2.243762354492006,
|
| 63 |
+
"learning_rate": 2.8688524590163937e-06,
|
| 64 |
+
"loss": 0.929,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0330169211721007,
|
| 69 |
+
"grad_norm": 2.31711751154106,
|
| 70 |
+
"learning_rate": 3.2786885245901638e-06,
|
| 71 |
+
"loss": 0.9036,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03714403631861329,
|
| 76 |
+
"grad_norm": 2.717403904295213,
|
| 77 |
+
"learning_rate": 3.6885245901639347e-06,
|
| 78 |
+
"loss": 0.8884,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.041271151465125874,
|
| 83 |
+
"grad_norm": 2.274524441450386,
|
| 84 |
+
"learning_rate": 4.098360655737705e-06,
|
| 85 |
+
"loss": 0.885,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04539826661163846,
|
| 90 |
+
"grad_norm": 2.457481290190043,
|
| 91 |
+
"learning_rate": 4.508196721311476e-06,
|
| 92 |
+
"loss": 0.8721,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04952538175815105,
|
| 97 |
+
"grad_norm": 2.3198474831023175,
|
| 98 |
+
"learning_rate": 4.918032786885246e-06,
|
| 99 |
+
"loss": 0.8735,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.053652496904663644,
|
| 104 |
+
"grad_norm": 2.2677876278066753,
|
| 105 |
+
"learning_rate": 5.327868852459017e-06,
|
| 106 |
+
"loss": 0.8592,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05777961205117623,
|
| 111 |
+
"grad_norm": 2.334132338411567,
|
| 112 |
+
"learning_rate": 5.737704918032787e-06,
|
| 113 |
+
"loss": 0.8573,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06190672719768882,
|
| 118 |
+
"grad_norm": 2.3114481733901213,
|
| 119 |
+
"learning_rate": 6.147540983606558e-06,
|
| 120 |
+
"loss": 0.8511,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0660338423442014,
|
| 125 |
+
"grad_norm": 2.445435314197215,
|
| 126 |
+
"learning_rate": 6.5573770491803276e-06,
|
| 127 |
+
"loss": 0.8427,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.070160957490714,
|
| 132 |
+
"grad_norm": 2.3320755043408483,
|
| 133 |
+
"learning_rate": 6.967213114754099e-06,
|
| 134 |
+
"loss": 0.8494,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07428807263722657,
|
| 139 |
+
"grad_norm": 2.4947952053454916,
|
| 140 |
+
"learning_rate": 7.3770491803278695e-06,
|
| 141 |
+
"loss": 0.8334,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07841518778373917,
|
| 146 |
+
"grad_norm": 2.4255315812672666,
|
| 147 |
+
"learning_rate": 7.786885245901639e-06,
|
| 148 |
+
"loss": 0.8292,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08254230293025175,
|
| 153 |
+
"grad_norm": 2.5980903528360586,
|
| 154 |
+
"learning_rate": 8.19672131147541e-06,
|
| 155 |
+
"loss": 0.8182,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08666941807676434,
|
| 160 |
+
"grad_norm": 2.557969631480717,
|
| 161 |
+
"learning_rate": 8.60655737704918e-06,
|
| 162 |
+
"loss": 0.83,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.09079653322327692,
|
| 167 |
+
"grad_norm": 2.3958016117230363,
|
| 168 |
+
"learning_rate": 9.016393442622952e-06,
|
| 169 |
+
"loss": 0.8132,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09492364836978952,
|
| 174 |
+
"grad_norm": 2.47094805804508,
|
| 175 |
+
"learning_rate": 9.426229508196723e-06,
|
| 176 |
+
"loss": 0.8058,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0990507635163021,
|
| 181 |
+
"grad_norm": 2.2173686833192234,
|
| 182 |
+
"learning_rate": 9.836065573770493e-06,
|
| 183 |
+
"loss": 0.8129,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.10317787866281469,
|
| 188 |
+
"grad_norm": 2.442299703694065,
|
| 189 |
+
"learning_rate": 9.999812749151968e-06,
|
| 190 |
+
"loss": 0.8043,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10730499380932729,
|
| 195 |
+
"grad_norm": 2.70020809980298,
|
| 196 |
+
"learning_rate": 9.998668489204265e-06,
|
| 197 |
+
"loss": 0.8005,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.11143210895583987,
|
| 202 |
+
"grad_norm": 2.4881429411766014,
|
| 203 |
+
"learning_rate": 9.99648423533726e-06,
|
| 204 |
+
"loss": 0.7993,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11555922410235246,
|
| 209 |
+
"grad_norm": 2.3569754221609953,
|
| 210 |
+
"learning_rate": 9.993260441994116e-06,
|
| 211 |
+
"loss": 0.7966,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11968633924886504,
|
| 216 |
+
"grad_norm": 2.455448628360433,
|
| 217 |
+
"learning_rate": 9.988997779898544e-06,
|
| 218 |
+
"loss": 0.8019,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12381345439537764,
|
| 223 |
+
"grad_norm": 2.5006306104937623,
|
| 224 |
+
"learning_rate": 9.983697135915253e-06,
|
| 225 |
+
"loss": 0.7921,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12794056954189023,
|
| 230 |
+
"grad_norm": 2.5688606957093594,
|
| 231 |
+
"learning_rate": 9.977359612865424e-06,
|
| 232 |
+
"loss": 0.7825,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.1320676846884028,
|
| 237 |
+
"grad_norm": 2.416695522058028,
|
| 238 |
+
"learning_rate": 9.96998652929728e-06,
|
| 239 |
+
"loss": 0.7809,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1361947998349154,
|
| 244 |
+
"grad_norm": 2.2779935869735732,
|
| 245 |
+
"learning_rate": 9.961579419211741e-06,
|
| 246 |
+
"loss": 0.7767,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.140321914981428,
|
| 251 |
+
"grad_norm": 2.405259985341037,
|
| 252 |
+
"learning_rate": 9.952140031743282e-06,
|
| 253 |
+
"loss": 0.7696,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.14444903012794058,
|
| 258 |
+
"grad_norm": 2.2359965979949186,
|
| 259 |
+
"learning_rate": 9.941670330796007e-06,
|
| 260 |
+
"loss": 0.7571,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14857614527445315,
|
| 265 |
+
"grad_norm": 2.123914969069781,
|
| 266 |
+
"learning_rate": 9.930172494635058e-06,
|
| 267 |
+
"loss": 0.7545,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.15270326042096574,
|
| 272 |
+
"grad_norm": 2.3685141262823057,
|
| 273 |
+
"learning_rate": 9.917648915433413e-06,
|
| 274 |
+
"loss": 0.7577,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.15683037556747834,
|
| 279 |
+
"grad_norm": 2.288582295953061,
|
| 280 |
+
"learning_rate": 9.904102198774188e-06,
|
| 281 |
+
"loss": 0.752,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.16095749071399093,
|
| 286 |
+
"grad_norm": 2.237056392660144,
|
| 287 |
+
"learning_rate": 9.889535163108537e-06,
|
| 288 |
+
"loss": 0.7439,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1650846058605035,
|
| 293 |
+
"grad_norm": 2.4463435064141312,
|
| 294 |
+
"learning_rate": 9.873950839169248e-06,
|
| 295 |
+
"loss": 0.7302,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1692117210070161,
|
| 300 |
+
"grad_norm": 2.255727307108107,
|
| 301 |
+
"learning_rate": 9.857352469340204e-06,
|
| 302 |
+
"loss": 0.7332,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.17333883615352869,
|
| 307 |
+
"grad_norm": 2.315889036709674,
|
| 308 |
+
"learning_rate": 9.839743506981783e-06,
|
| 309 |
+
"loss": 0.7334,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.17746595130004128,
|
| 314 |
+
"grad_norm": 2.4721326878211363,
|
| 315 |
+
"learning_rate": 9.821127615712365e-06,
|
| 316 |
+
"loss": 0.7184,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.18159306644655385,
|
| 321 |
+
"grad_norm": 2.4415976276779277,
|
| 322 |
+
"learning_rate": 9.801508668646117e-06,
|
| 323 |
+
"loss": 0.7149,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.18572018159306644,
|
| 328 |
+
"grad_norm": 2.412349551631901,
|
| 329 |
+
"learning_rate": 9.780890747587163e-06,
|
| 330 |
+
"loss": 0.713,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.18984729673957904,
|
| 335 |
+
"grad_norm": 2.515848530060018,
|
| 336 |
+
"learning_rate": 9.759278142180348e-06,
|
| 337 |
+
"loss": 0.7172,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19397441188609163,
|
| 342 |
+
"grad_norm": 2.55352651790107,
|
| 343 |
+
"learning_rate": 9.736675349018768e-06,
|
| 344 |
+
"loss": 0.7042,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1981015270326042,
|
| 349 |
+
"grad_norm": 2.2895880245761515,
|
| 350 |
+
"learning_rate": 9.713087070708225e-06,
|
| 351 |
+
"loss": 0.7109,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2022286421791168,
|
| 356 |
+
"grad_norm": 2.3671724385797566,
|
| 357 |
+
"learning_rate": 9.688518214888836e-06,
|
| 358 |
+
"loss": 0.7001,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.20635575732562939,
|
| 363 |
+
"grad_norm": 2.3617892749786074,
|
| 364 |
+
"learning_rate": 9.662973893213976e-06,
|
| 365 |
+
"loss": 0.6979,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.21048287247214198,
|
| 370 |
+
"grad_norm": 2.254544557833255,
|
| 371 |
+
"learning_rate": 9.636459420286779e-06,
|
| 372 |
+
"loss": 0.6914,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.21460998761865457,
|
| 377 |
+
"grad_norm": 2.219547321736457,
|
| 378 |
+
"learning_rate": 9.60898031255441e-06,
|
| 379 |
+
"loss": 0.6964,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.21873710276516714,
|
| 384 |
+
"grad_norm": 2.211511469572542,
|
| 385 |
+
"learning_rate": 9.580542287160348e-06,
|
| 386 |
+
"loss": 0.6695,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.22286421791167974,
|
| 391 |
+
"grad_norm": 2.5041061805226787,
|
| 392 |
+
"learning_rate": 9.551151260754907e-06,
|
| 393 |
+
"loss": 0.6864,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.22699133305819233,
|
| 398 |
+
"grad_norm": 2.3511386929231097,
|
| 399 |
+
"learning_rate": 9.520813348264252e-06,
|
| 400 |
+
"loss": 0.6809,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.23111844820470492,
|
| 405 |
+
"grad_norm": 2.960097513048209,
|
| 406 |
+
"learning_rate": 9.489534861618168e-06,
|
| 407 |
+
"loss": 0.6555,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2352455633512175,
|
| 412 |
+
"grad_norm": 2.2082387429502712,
|
| 413 |
+
"learning_rate": 9.45732230843683e-06,
|
| 414 |
+
"loss": 0.6568,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.23937267849773008,
|
| 419 |
+
"grad_norm": 2.151410227639591,
|
| 420 |
+
"learning_rate": 9.424182390676872e-06,
|
| 421 |
+
"loss": 0.6645,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.24349979364424268,
|
| 426 |
+
"grad_norm": 2.2209736960884334,
|
| 427 |
+
"learning_rate": 9.39012200323701e-06,
|
| 428 |
+
"loss": 0.6705,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.24762690879075527,
|
| 433 |
+
"grad_norm": 2.414842272352699,
|
| 434 |
+
"learning_rate": 9.355148232523538e-06,
|
| 435 |
+
"loss": 0.6496,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.25175402393726787,
|
| 440 |
+
"grad_norm": 2.5739956654811236,
|
| 441 |
+
"learning_rate": 9.319268354975958e-06,
|
| 442 |
+
"loss": 0.657,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.25588113908378046,
|
| 447 |
+
"grad_norm": 2.618163978298313,
|
| 448 |
+
"learning_rate": 9.282489835553106e-06,
|
| 449 |
+
"loss": 0.6496,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.260008254230293,
|
| 454 |
+
"grad_norm": 2.281248176869453,
|
| 455 |
+
"learning_rate": 9.244820326180013e-06,
|
| 456 |
+
"loss": 0.6563,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2641353693768056,
|
| 461 |
+
"grad_norm": 2.297254363064407,
|
| 462 |
+
"learning_rate": 9.206267664155906e-06,
|
| 463 |
+
"loss": 0.6432,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2682624845233182,
|
| 468 |
+
"grad_norm": 2.3343424077594306,
|
| 469 |
+
"learning_rate": 9.166839870523627e-06,
|
| 470 |
+
"loss": 0.6352,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2723895996698308,
|
| 475 |
+
"grad_norm": 2.3001895352129904,
|
| 476 |
+
"learning_rate": 9.126545148400807e-06,
|
| 477 |
+
"loss": 0.6346,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2765167148163434,
|
| 482 |
+
"grad_norm": 2.335491244315102,
|
| 483 |
+
"learning_rate": 9.085391881273182e-06,
|
| 484 |
+
"loss": 0.6417,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.280643829962856,
|
| 489 |
+
"grad_norm": 2.8711945935898986,
|
| 490 |
+
"learning_rate": 9.04338863125038e-06,
|
| 491 |
+
"loss": 0.6203,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.28477094510936857,
|
| 496 |
+
"grad_norm": 2.4481124095885964,
|
| 497 |
+
"learning_rate": 9.00054413728452e-06,
|
| 498 |
+
"loss": 0.6253,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.28889806025588116,
|
| 503 |
+
"grad_norm": 2.208108640856046,
|
| 504 |
+
"learning_rate": 8.956867313352055e-06,
|
| 505 |
+
"loss": 0.6246,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2930251754023937,
|
| 510 |
+
"grad_norm": 2.2861750757235995,
|
| 511 |
+
"learning_rate": 8.912367246599175e-06,
|
| 512 |
+
"loss": 0.6152,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2971522905489063,
|
| 517 |
+
"grad_norm": 2.548396902849064,
|
| 518 |
+
"learning_rate": 8.867053195451182e-06,
|
| 519 |
+
"loss": 0.6262,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3012794056954189,
|
| 524 |
+
"grad_norm": 2.305963742549307,
|
| 525 |
+
"learning_rate": 8.820934587686247e-06,
|
| 526 |
+
"loss": 0.6151,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3054065208419315,
|
| 531 |
+
"grad_norm": 2.4126232699277157,
|
| 532 |
+
"learning_rate": 8.77402101847391e-06,
|
| 533 |
+
"loss": 0.6071,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3095336359884441,
|
| 538 |
+
"grad_norm": 2.2092911857967867,
|
| 539 |
+
"learning_rate": 8.726322248378775e-06,
|
| 540 |
+
"loss": 0.6144,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3136607511349567,
|
| 545 |
+
"grad_norm": 2.2414757536483125,
|
| 546 |
+
"learning_rate": 8.677848201329775e-06,
|
| 547 |
+
"loss": 0.5971,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.31778786628146927,
|
| 552 |
+
"grad_norm": 2.1301302453717423,
|
| 553 |
+
"learning_rate": 8.628608962555467e-06,
|
| 554 |
+
"loss": 0.6171,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.32191498142798186,
|
| 559 |
+
"grad_norm": 2.1138762246320373,
|
| 560 |
+
"learning_rate": 8.578614776485744e-06,
|
| 561 |
+
"loss": 0.5913,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.32604209657449446,
|
| 566 |
+
"grad_norm": 2.2822356894428277,
|
| 567 |
+
"learning_rate": 8.527876044620453e-06,
|
| 568 |
+
"loss": 0.5967,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.330169211721007,
|
| 573 |
+
"grad_norm": 2.241849162079401,
|
| 574 |
+
"learning_rate": 8.47640332336531e-06,
|
| 575 |
+
"loss": 0.6042,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3342963268675196,
|
| 580 |
+
"grad_norm": 2.2294939629300097,
|
| 581 |
+
"learning_rate": 8.424207321835598e-06,
|
| 582 |
+
"loss": 0.5824,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3384234420140322,
|
| 587 |
+
"grad_norm": 2.2031216423678543,
|
| 588 |
+
"learning_rate": 8.371298899628091e-06,
|
| 589 |
+
"loss": 0.5778,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3425505571605448,
|
| 594 |
+
"grad_norm": 2.33426131831025,
|
| 595 |
+
"learning_rate": 8.317689064561671e-06,
|
| 596 |
+
"loss": 0.5846,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.34667767230705737,
|
| 601 |
+
"grad_norm": 2.3437558474733895,
|
| 602 |
+
"learning_rate": 8.263388970387102e-06,
|
| 603 |
+
"loss": 0.5752,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.35080478745356997,
|
| 608 |
+
"grad_norm": 2.194788845362847,
|
| 609 |
+
"learning_rate": 8.20840991446645e-06,
|
| 610 |
+
"loss": 0.5796,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.35493190260008256,
|
| 615 |
+
"grad_norm": 2.3039527642284776,
|
| 616 |
+
"learning_rate": 8.152763335422612e-06,
|
| 617 |
+
"loss": 0.5768,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.35905901774659515,
|
| 622 |
+
"grad_norm": 2.271949627752587,
|
| 623 |
+
"learning_rate": 8.096460810759473e-06,
|
| 624 |
+
"loss": 0.5745,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3631861328931077,
|
| 629 |
+
"grad_norm": 2.299369126451729,
|
| 630 |
+
"learning_rate": 8.03951405445314e-06,
|
| 631 |
+
"loss": 0.5629,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3673132480396203,
|
| 636 |
+
"grad_norm": 2.701487019844598,
|
| 637 |
+
"learning_rate": 7.98193491451483e-06,
|
| 638 |
+
"loss": 0.5682,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3714403631861329,
|
| 643 |
+
"grad_norm": 2.329511707999701,
|
| 644 |
+
"learning_rate": 7.923735370525809e-06,
|
| 645 |
+
"loss": 0.5741,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3755674783326455,
|
| 650 |
+
"grad_norm": 2.2666378605955155,
|
| 651 |
+
"learning_rate": 7.864927531145012e-06,
|
| 652 |
+
"loss": 0.5613,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.37969459347915807,
|
| 657 |
+
"grad_norm": 2.44812863797748,
|
| 658 |
+
"learning_rate": 7.805523631589775e-06,
|
| 659 |
+
"loss": 0.5654,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.38382170862567067,
|
| 664 |
+
"grad_norm": 2.346335768826544,
|
| 665 |
+
"learning_rate": 7.745536031090252e-06,
|
| 666 |
+
"loss": 0.5577,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.38794882377218326,
|
| 671 |
+
"grad_norm": 2.340018514786721,
|
| 672 |
+
"learning_rate": 7.684977210318024e-06,
|
| 673 |
+
"loss": 0.5575,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.39207593891869585,
|
| 678 |
+
"grad_norm": 2.250185737483932,
|
| 679 |
+
"learning_rate": 7.623859768789441e-06,
|
| 680 |
+
"loss": 0.5456,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3962030540652084,
|
| 685 |
+
"grad_norm": 2.1535843417391938,
|
| 686 |
+
"learning_rate": 7.5621964222442455e-06,
|
| 687 |
+
"loss": 0.5504,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.400330169211721,
|
| 692 |
+
"grad_norm": 2.279014812989976,
|
| 693 |
+
"learning_rate": 7.500000000000001e-06,
|
| 694 |
+
"loss": 0.5418,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4044572843582336,
|
| 699 |
+
"grad_norm": 2.248120170776638,
|
| 700 |
+
"learning_rate": 7.437283442282903e-06,
|
| 701 |
+
"loss": 0.5299,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4085843995047462,
|
| 706 |
+
"grad_norm": 2.1290912998771443,
|
| 707 |
+
"learning_rate": 7.374059797535516e-06,
|
| 708 |
+
"loss": 0.5382,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.41271151465125877,
|
| 713 |
+
"grad_norm": 2.2318811155096325,
|
| 714 |
+
"learning_rate": 7.310342219701981e-06,
|
| 715 |
+
"loss": 0.5386,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.41683862979777137,
|
| 720 |
+
"grad_norm": 2.0172840744038445,
|
| 721 |
+
"learning_rate": 7.246143965491288e-06,
|
| 722 |
+
"loss": 0.532,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.42096574494428396,
|
| 727 |
+
"grad_norm": 2.0888969433680953,
|
| 728 |
+
"learning_rate": 7.181478391619162e-06,
|
| 729 |
+
"loss": 0.5205,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.42509286009079655,
|
| 734 |
+
"grad_norm": 2.15061141359648,
|
| 735 |
+
"learning_rate": 7.11635895202914e-06,
|
| 736 |
+
"loss": 0.5311,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.42921997523730915,
|
| 741 |
+
"grad_norm": 2.2686115013098704,
|
| 742 |
+
"learning_rate": 7.050799195093421e-06,
|
| 743 |
+
"loss": 0.5301,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4333470903838217,
|
| 748 |
+
"grad_norm": 2.341272332235732,
|
| 749 |
+
"learning_rate": 6.984812760794078e-06,
|
| 750 |
+
"loss": 0.5362,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4374742055303343,
|
| 755 |
+
"grad_norm": 2.166891534024358,
|
| 756 |
+
"learning_rate": 6.918413377885193e-06,
|
| 757 |
+
"loss": 0.5293,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4416013206768469,
|
| 762 |
+
"grad_norm": 2.2412320223529134,
|
| 763 |
+
"learning_rate": 6.851614861036533e-06,
|
| 764 |
+
"loss": 0.5283,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.44572843582335947,
|
| 769 |
+
"grad_norm": 2.2162169091736166,
|
| 770 |
+
"learning_rate": 6.78443110795936e-06,
|
| 771 |
+
"loss": 0.5187,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.44985555096987206,
|
| 776 |
+
"grad_norm": 2.1687285736037447,
|
| 777 |
+
"learning_rate": 6.716876096514944e-06,
|
| 778 |
+
"loss": 0.5208,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.45398266611638466,
|
| 783 |
+
"grad_norm": 2.1587915614775297,
|
| 784 |
+
"learning_rate": 6.648963881806411e-06,
|
| 785 |
+
"loss": 0.5058,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.45810978126289725,
|
| 790 |
+
"grad_norm": 2.05653346463343,
|
| 791 |
+
"learning_rate": 6.580708593254526e-06,
|
| 792 |
+
"loss": 0.5103,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.46223689640940985,
|
| 797 |
+
"grad_norm": 2.1531395501548642,
|
| 798 |
+
"learning_rate": 6.512124431658006e-06,
|
| 799 |
+
"loss": 0.5103,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4663640115559224,
|
| 804 |
+
"grad_norm": 2.176090356825283,
|
| 805 |
+
"learning_rate": 6.443225666238976e-06,
|
| 806 |
+
"loss": 0.5062,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.470491126702435,
|
| 811 |
+
"grad_norm": 2.209786359023677,
|
| 812 |
+
"learning_rate": 6.37402663167421e-06,
|
| 813 |
+
"loss": 0.5027,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4746182418489476,
|
| 818 |
+
"grad_norm": 2.056272842336906,
|
| 819 |
+
"learning_rate": 6.304541725112734e-06,
|
| 820 |
+
"loss": 0.5056,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.47874535699546017,
|
| 825 |
+
"grad_norm": 2.162812572951533,
|
| 826 |
+
"learning_rate": 6.234785403180438e-06,
|
| 827 |
+
"loss": 0.4928,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.48287247214197276,
|
| 832 |
+
"grad_norm": 2.211681052073838,
|
| 833 |
+
"learning_rate": 6.164772178972321e-06,
|
| 834 |
+
"loss": 0.4938,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.48699958728848536,
|
| 839 |
+
"grad_norm": 2.0594553618227613,
|
| 840 |
+
"learning_rate": 6.094516619032975e-06,
|
| 841 |
+
"loss": 0.5024,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.49112670243499795,
|
| 846 |
+
"grad_norm": 2.1196964005402803,
|
| 847 |
+
"learning_rate": 6.024033340325954e-06,
|
| 848 |
+
"loss": 0.4877,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.49525381758151055,
|
| 853 |
+
"grad_norm": 2.0731282875409787,
|
| 854 |
+
"learning_rate": 5.953337007192659e-06,
|
| 855 |
+
"loss": 0.4903,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4993809327280231,
|
| 860 |
+
"grad_norm": 2.072327686896433,
|
| 861 |
+
"learning_rate": 5.882442328301356e-06,
|
| 862 |
+
"loss": 0.4814,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5035080478745357,
|
| 867 |
+
"grad_norm": 2.118695720658721,
|
| 868 |
+
"learning_rate": 5.811364053586973e-06,
|
| 869 |
+
"loss": 0.4766,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5076351630210483,
|
| 874 |
+
"grad_norm": 2.119885860879392,
|
| 875 |
+
"learning_rate": 5.740116971182322e-06,
|
| 876 |
+
"loss": 0.4777,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5117622781675609,
|
| 881 |
+
"grad_norm": 2.073375813482556,
|
| 882 |
+
"learning_rate": 5.668715904341365e-06,
|
| 883 |
+
"loss": 0.4734,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5158893933140735,
|
| 888 |
+
"grad_norm": 2.0709791028178572,
|
| 889 |
+
"learning_rate": 5.5971757083551625e-06,
|
| 890 |
+
"loss": 0.4707,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.520016508460586,
|
| 895 |
+
"grad_norm": 2.0492205452436396,
|
| 896 |
+
"learning_rate": 5.525511267461186e-06,
|
| 897 |
+
"loss": 0.4736,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5241436236070987,
|
| 902 |
+
"grad_norm": 2.0319046135786953,
|
| 903 |
+
"learning_rate": 5.453737491746572e-06,
|
| 904 |
+
"loss": 0.4713,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5282707387536112,
|
| 909 |
+
"grad_norm": 2.0892666680641327,
|
| 910 |
+
"learning_rate": 5.381869314046031e-06,
|
| 911 |
+
"loss": 0.4764,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5323978539001238,
|
| 916 |
+
"grad_norm": 2.078568518806719,
|
| 917 |
+
"learning_rate": 5.3099216868349965e-06,
|
| 918 |
+
"loss": 0.4671,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5365249690466364,
|
| 923 |
+
"grad_norm": 2.0410694490990724,
|
| 924 |
+
"learning_rate": 5.237909579118713e-06,
|
| 925 |
+
"loss": 0.4778,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.540652084193149,
|
| 930 |
+
"grad_norm": 2.0761769744117444,
|
| 931 |
+
"learning_rate": 5.165847973317854e-06,
|
| 932 |
+
"loss": 0.4678,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5447791993396616,
|
| 937 |
+
"grad_norm": 2.130099579144156,
|
| 938 |
+
"learning_rate": 5.093751862151388e-06,
|
| 939 |
+
"loss": 0.4649,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5489063144861742,
|
| 944 |
+
"grad_norm": 2.078379103493414,
|
| 945 |
+
"learning_rate": 5.021636245517261e-06,
|
| 946 |
+
"loss": 0.4732,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5530334296326868,
|
| 951 |
+
"grad_norm": 2.0058829746015507,
|
| 952 |
+
"learning_rate": 4.9495161273716174e-06,
|
| 953 |
+
"loss": 0.4566,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5571605447791993,
|
| 958 |
+
"grad_norm": 2.22400636300283,
|
| 959 |
+
"learning_rate": 4.877406512607159e-06,
|
| 960 |
+
"loss": 0.4572,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.561287659925712,
|
| 965 |
+
"grad_norm": 2.115628462513928,
|
| 966 |
+
"learning_rate": 4.805322403931312e-06,
|
| 967 |
+
"loss": 0.4599,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5654147750722245,
|
| 972 |
+
"grad_norm": 2.0907681450444353,
|
| 973 |
+
"learning_rate": 4.73327879874486e-06,
|
| 974 |
+
"loss": 0.4647,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5695418902187371,
|
| 979 |
+
"grad_norm": 2.203495989612946,
|
| 980 |
+
"learning_rate": 4.661290686021661e-06,
|
| 981 |
+
"loss": 0.4631,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5736690053652497,
|
| 986 |
+
"grad_norm": 2.0602084077093235,
|
| 987 |
+
"learning_rate": 4.589373043190137e-06,
|
| 988 |
+
"loss": 0.4596,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5777961205117623,
|
| 993 |
+
"grad_norm": 2.08958806455361,
|
| 994 |
+
"learning_rate": 4.517540833017152e-06,
|
| 995 |
+
"loss": 0.4481,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5819232356582749,
|
| 1000 |
+
"grad_norm": 2.0835614126353112,
|
| 1001 |
+
"learning_rate": 4.445809000494945e-06,
|
| 1002 |
+
"loss": 0.4457,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5860503508047874,
|
| 1007 |
+
"grad_norm": 2.1121573908487057,
|
| 1008 |
+
"learning_rate": 4.374192469731771e-06,
|
| 1009 |
+
"loss": 0.4485,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5901774659513,
|
| 1014 |
+
"grad_norm": 2.0348765238388307,
|
| 1015 |
+
"learning_rate": 4.302706140846864e-06,
|
| 1016 |
+
"loss": 0.4531,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5943045810978126,
|
| 1021 |
+
"grad_norm": 2.097167725695452,
|
| 1022 |
+
"learning_rate": 4.231364886870417e-06,
|
| 1023 |
+
"loss": 0.432,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5984316962443252,
|
| 1028 |
+
"grad_norm": 2.035634145828901,
|
| 1029 |
+
"learning_rate": 4.160183550649176e-06,
|
| 1030 |
+
"loss": 0.4542,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6025588113908378,
|
| 1035 |
+
"grad_norm": 2.083465204457464,
|
| 1036 |
+
"learning_rate": 4.089176941758332e-06,
|
| 1037 |
+
"loss": 0.4373,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6066859265373504,
|
| 1042 |
+
"grad_norm": 2.302682902657171,
|
| 1043 |
+
"learning_rate": 4.018359833420323e-06,
|
| 1044 |
+
"loss": 0.4355,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.610813041683863,
|
| 1049 |
+
"grad_norm": 2.1209322204861167,
|
| 1050 |
+
"learning_rate": 3.9477469594311975e-06,
|
| 1051 |
+
"loss": 0.4382,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6149401568303756,
|
| 1056 |
+
"grad_norm": 2.1130799454748694,
|
| 1057 |
+
"learning_rate": 3.8773530110952e-06,
|
| 1058 |
+
"loss": 0.4303,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.6190672719768882,
|
| 1063 |
+
"grad_norm": 2.067330224712638,
|
| 1064 |
+
"learning_rate": 3.807192634168168e-06,
|
| 1065 |
+
"loss": 0.4282,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6231943871234007,
|
| 1070 |
+
"grad_norm": 2.121507686505271,
|
| 1071 |
+
"learning_rate": 3.7372804258104367e-06,
|
| 1072 |
+
"loss": 0.4398,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6273215022699133,
|
| 1077 |
+
"grad_norm": 2.081955763632376,
|
| 1078 |
+
"learning_rate": 3.667630931549826e-06,
|
| 1079 |
+
"loss": 0.4321,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6314486174164259,
|
| 1084 |
+
"grad_norm": 2.0671194356029936,
|
| 1085 |
+
"learning_rate": 3.598258642255387e-06,
|
| 1086 |
+
"loss": 0.4297,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6355757325629385,
|
| 1091 |
+
"grad_norm": 2.0745096928290274,
|
| 1092 |
+
"learning_rate": 3.529177991122519e-06,
|
| 1093 |
+
"loss": 0.4291,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6397028477094511,
|
| 1098 |
+
"grad_norm": 2.04047679569905,
|
| 1099 |
+
"learning_rate": 3.460403350670077e-06,
|
| 1100 |
+
"loss": 0.4234,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6438299628559637,
|
| 1105 |
+
"grad_norm": 2.0184632129343254,
|
| 1106 |
+
"learning_rate": 3.3919490297501167e-06,
|
| 1107 |
+
"loss": 0.4322,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6479570780024763,
|
| 1112 |
+
"grad_norm": 2.0467555064446983,
|
| 1113 |
+
"learning_rate": 3.3238292705708675e-06,
|
| 1114 |
+
"loss": 0.423,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6520841931489889,
|
| 1119 |
+
"grad_norm": 2.15202224763677,
|
| 1120 |
+
"learning_rate": 3.256058245733592e-06,
|
| 1121 |
+
"loss": 0.4189,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6562113082955014,
|
| 1126 |
+
"grad_norm": 1.9966574895688705,
|
| 1127 |
+
"learning_rate": 3.1886500552839105e-06,
|
| 1128 |
+
"loss": 0.4199,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.660338423442014,
|
| 1133 |
+
"grad_norm": 2.0485705146360753,
|
| 1134 |
+
"learning_rate": 3.121618723778225e-06,
|
| 1135 |
+
"loss": 0.4106,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6644655385885266,
|
| 1140 |
+
"grad_norm": 1.939157035039348,
|
| 1141 |
+
"learning_rate": 3.054978197365861e-06,
|
| 1142 |
+
"loss": 0.4164,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6685926537350392,
|
| 1147 |
+
"grad_norm": 1.9248864355584618,
|
| 1148 |
+
"learning_rate": 2.9887423408875056e-06,
|
| 1149 |
+
"loss": 0.4082,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6727197688815518,
|
| 1154 |
+
"grad_norm": 2.0698662946290423,
|
| 1155 |
+
"learning_rate": 2.9229249349905686e-06,
|
| 1156 |
+
"loss": 0.415,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6768468840280644,
|
| 1161 |
+
"grad_norm": 2.1520358829070094,
|
| 1162 |
+
"learning_rate": 2.8575396732620673e-06,
|
| 1163 |
+
"loss": 0.4003,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.680973999174577,
|
| 1168 |
+
"grad_norm": 1.9676814885778169,
|
| 1169 |
+
"learning_rate": 2.7926001593796114e-06,
|
| 1170 |
+
"loss": 0.4173,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6851011143210896,
|
| 1175 |
+
"grad_norm": 2.0886545446660363,
|
| 1176 |
+
"learning_rate": 2.728119904281105e-06,
|
| 1177 |
+
"loss": 0.4174,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6892282294676021,
|
| 1182 |
+
"grad_norm": 2.1021768729202637,
|
| 1183 |
+
"learning_rate": 2.6641123233537395e-06,
|
| 1184 |
+
"loss": 0.4009,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6933553446141147,
|
| 1189 |
+
"grad_norm": 2.164281085927058,
|
| 1190 |
+
"learning_rate": 2.6005907336428615e-06,
|
| 1191 |
+
"loss": 0.4017,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6974824597606273,
|
| 1196 |
+
"grad_norm": 2.092014051056356,
|
| 1197 |
+
"learning_rate": 2.537568351081311e-06,
|
| 1198 |
+
"loss": 0.4046,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7016095749071399,
|
| 1203 |
+
"grad_norm": 1.9368454491782257,
|
| 1204 |
+
"learning_rate": 2.4750582877397933e-06,
|
| 1205 |
+
"loss": 0.4073,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7057366900536525,
|
| 1210 |
+
"grad_norm": 2.019204414322489,
|
| 1211 |
+
"learning_rate": 2.4130735490988457e-06,
|
| 1212 |
+
"loss": 0.4019,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7098638052001651,
|
| 1217 |
+
"grad_norm": 1.9404031254190708,
|
| 1218 |
+
"learning_rate": 2.3516270313430085e-06,
|
| 1219 |
+
"loss": 0.3965,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7139909203466777,
|
| 1224 |
+
"grad_norm": 1.9803275733019916,
|
| 1225 |
+
"learning_rate": 2.2907315186777057e-06,
|
| 1226 |
+
"loss": 0.4009,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7181180354931903,
|
| 1231 |
+
"grad_norm": 1.9105419735279907,
|
| 1232 |
+
"learning_rate": 2.230399680669449e-06,
|
| 1233 |
+
"loss": 0.3925,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7222451506397028,
|
| 1238 |
+
"grad_norm": 2.11701521254791,
|
| 1239 |
+
"learning_rate": 2.170644069609876e-06,
|
| 1240 |
+
"loss": 0.3878,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7263722657862154,
|
| 1245 |
+
"grad_norm": 2.009828108641354,
|
| 1246 |
+
"learning_rate": 2.1114771179041894e-06,
|
| 1247 |
+
"loss": 0.3939,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.730499380932728,
|
| 1252 |
+
"grad_norm": 2.0525311494906417,
|
| 1253 |
+
"learning_rate": 2.052911135484551e-06,
|
| 1254 |
+
"loss": 0.3983,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7346264960792406,
|
| 1259 |
+
"grad_norm": 2.0024645682646214,
|
| 1260 |
+
"learning_rate": 1.9949583072489455e-06,
|
| 1261 |
+
"loss": 0.3991,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7387536112257532,
|
| 1266 |
+
"grad_norm": 1.983708310096234,
|
| 1267 |
+
"learning_rate": 1.9376306905260604e-06,
|
| 1268 |
+
"loss": 0.3908,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.7428807263722658,
|
| 1273 |
+
"grad_norm": 1.9201326193885593,
|
| 1274 |
+
"learning_rate": 1.8809402125667064e-06,
|
| 1275 |
+
"loss": 0.3894,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7470078415187784,
|
| 1280 |
+
"grad_norm": 2.0461993333856068,
|
| 1281 |
+
"learning_rate": 1.8248986680623077e-06,
|
| 1282 |
+
"loss": 0.3845,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.751134956665291,
|
| 1287 |
+
"grad_norm": 1.9895671478806258,
|
| 1288 |
+
"learning_rate": 1.7695177166909672e-06,
|
| 1289 |
+
"loss": 0.3924,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7552620718118036,
|
| 1294 |
+
"grad_norm": 2.057619939386755,
|
| 1295 |
+
"learning_rate": 1.7148088806916114e-06,
|
| 1296 |
+
"loss": 0.3895,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7593891869583161,
|
| 1301 |
+
"grad_norm": 2.0151906155411354,
|
| 1302 |
+
"learning_rate": 1.6607835424667578e-06,
|
| 1303 |
+
"loss": 0.3865,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7635163021048287,
|
| 1308 |
+
"grad_norm": 1.925068927133916,
|
| 1309 |
+
"learning_rate": 1.6074529422143398e-06,
|
| 1310 |
+
"loss": 0.3852,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7676434172513413,
|
| 1315 |
+
"grad_norm": 2.0136796449917256,
|
| 1316 |
+
"learning_rate": 1.554828175589151e-06,
|
| 1317 |
+
"loss": 0.3787,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7717705323978539,
|
| 1322 |
+
"grad_norm": 1.9559990017194175,
|
| 1323 |
+
"learning_rate": 1.5029201913943425e-06,
|
| 1324 |
+
"loss": 0.3796,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7758976475443665,
|
| 1329 |
+
"grad_norm": 2.1521890680893585,
|
| 1330 |
+
"learning_rate": 1.4517397893034718e-06,
|
| 1331 |
+
"loss": 0.3842,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7800247626908791,
|
| 1336 |
+
"grad_norm": 2.0029729095922213,
|
| 1337 |
+
"learning_rate": 1.4012976176135978e-06,
|
| 1338 |
+
"loss": 0.393,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7841518778373917,
|
| 1343 |
+
"grad_norm": 2.021018024866405,
|
| 1344 |
+
"learning_rate": 1.35160417102985e-06,
|
| 1345 |
+
"loss": 0.3873,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7882789929839042,
|
| 1350 |
+
"grad_norm": 1.9835697101743557,
|
| 1351 |
+
"learning_rate": 1.3026697884819644e-06,
|
| 1352 |
+
"loss": 0.3861,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7924061081304168,
|
| 1357 |
+
"grad_norm": 1.9340723815291019,
|
| 1358 |
+
"learning_rate": 1.25450465097322e-06,
|
| 1359 |
+
"loss": 0.3903,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7965332232769294,
|
| 1364 |
+
"grad_norm": 2.1191758392362035,
|
| 1365 |
+
"learning_rate": 1.207118779462248e-06,
|
| 1366 |
+
"loss": 0.3749,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.800660338423442,
|
| 1371 |
+
"grad_norm": 2.132380601853603,
|
| 1372 |
+
"learning_rate": 1.160522032778123e-06,
|
| 1373 |
+
"loss": 0.3782,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8047874535699546,
|
| 1378 |
+
"grad_norm": 1.9708201131489755,
|
| 1379 |
+
"learning_rate": 1.1147241055691909e-06,
|
| 1380 |
+
"loss": 0.3705,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.8089145687164672,
|
| 1385 |
+
"grad_norm": 2.01028167948197,
|
| 1386 |
+
"learning_rate": 1.0697345262860638e-06,
|
| 1387 |
+
"loss": 0.3742,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8130416838629798,
|
| 1392 |
+
"grad_norm": 1.9991004397644925,
|
| 1393 |
+
"learning_rate": 1.0255626551991765e-06,
|
| 1394 |
+
"loss": 0.3806,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8171687990094924,
|
| 1399 |
+
"grad_norm": 1.9398586585837028,
|
| 1400 |
+
"learning_rate": 9.822176824513524e-07,
|
| 1401 |
+
"loss": 0.3731,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.821295914156005,
|
| 1406 |
+
"grad_norm": 1.96994225142757,
|
| 1407 |
+
"learning_rate": 9.397086261457511e-07,
|
| 1408 |
+
"loss": 0.3776,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8254230293025175,
|
| 1413 |
+
"grad_norm": 2.003600819188701,
|
| 1414 |
+
"learning_rate": 8.980443304696107e-07,
|
| 1415 |
+
"loss": 0.3819,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8295501444490301,
|
| 1420 |
+
"grad_norm": 2.0351798859032484,
|
| 1421 |
+
"learning_rate": 8.572334638541857e-07,
|
| 1422 |
+
"loss": 0.3774,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8336772595955427,
|
| 1427 |
+
"grad_norm": 2.1430516379993403,
|
| 1428 |
+
"learning_rate": 8.172845171712379e-07,
|
| 1429 |
+
"loss": 0.3737,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8378043747420553,
|
| 1434 |
+
"grad_norm": 2.03671544389429,
|
| 1435 |
+
"learning_rate": 7.782058019664773e-07,
|
| 1436 |
+
"loss": 0.3709,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8419314898885679,
|
| 1441 |
+
"grad_norm": 2.014237343029339,
|
| 1442 |
+
"learning_rate": 7.40005448730306e-07,
|
| 1443 |
+
"loss": 0.3717,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8460586050350805,
|
| 1448 |
+
"grad_norm": 2.001733960692795,
|
| 1449 |
+
"learning_rate": 7.026914052062433e-07,
|
| 1450 |
+
"loss": 0.378,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8501857201815931,
|
| 1455 |
+
"grad_norm": 2.0314797478672864,
|
| 1456 |
+
"learning_rate": 6.662714347373589e-07,
|
| 1457 |
+
"loss": 0.3718,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8543128353281056,
|
| 1462 |
+
"grad_norm": 2.0398788512991985,
|
| 1463 |
+
"learning_rate": 6.307531146510754e-07,
|
| 1464 |
+
"loss": 0.3802,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8584399504746183,
|
| 1469 |
+
"grad_norm": 2.0401593697279616,
|
| 1470 |
+
"learning_rate": 5.961438346826792e-07,
|
| 1471 |
+
"loss": 0.3723,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.8625670656211308,
|
| 1476 |
+
"grad_norm": 2.105641702439893,
|
| 1477 |
+
"learning_rate": 5.6245079543785e-07,
|
| 1478 |
+
"loss": 0.3674,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8666941807676434,
|
| 1483 |
+
"grad_norm": 2.0475120583137962,
|
| 1484 |
+
"learning_rate": 5.296810068945474e-07,
|
| 1485 |
+
"loss": 0.3767,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.870821295914156,
|
| 1490 |
+
"grad_norm": 2.037325684321563,
|
| 1491 |
+
"learning_rate": 4.97841286944557e-07,
|
| 1492 |
+
"loss": 0.3717,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8749484110606686,
|
| 1497 |
+
"grad_norm": 1.8826839605258934,
|
| 1498 |
+
"learning_rate": 4.6693825997499274e-07,
|
| 1499 |
+
"loss": 0.366,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8790755262071812,
|
| 1504 |
+
"grad_norm": 2.1270152173670294,
|
| 1505 |
+
"learning_rate": 4.3697835549007474e-07,
|
| 1506 |
+
"loss": 0.3657,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8832026413536938,
|
| 1511 |
+
"grad_norm": 1.986884523560406,
|
| 1512 |
+
"learning_rate": 4.0796780677343606e-07,
|
| 1513 |
+
"loss": 0.3687,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8873297565002064,
|
| 1518 |
+
"grad_norm": 2.0992498730608453,
|
| 1519 |
+
"learning_rate": 3.799126495912686e-07,
|
| 1520 |
+
"loss": 0.3638,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8914568716467189,
|
| 1525 |
+
"grad_norm": 2.028752109036286,
|
| 1526 |
+
"learning_rate": 3.528187209365486e-07,
|
| 1527 |
+
"loss": 0.3699,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8955839867932315,
|
| 1532 |
+
"grad_norm": 2.0346601882201707,
|
| 1533 |
+
"learning_rate": 3.26691657814634e-07,
|
| 1534 |
+
"loss": 0.3703,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8997111019397441,
|
| 1539 |
+
"grad_norm": 2.0217798718878717,
|
| 1540 |
+
"learning_rate": 3.015368960704584e-07,
|
| 1541 |
+
"loss": 0.3604,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.9038382170862567,
|
| 1546 |
+
"grad_norm": 1.9484068285089462,
|
| 1547 |
+
"learning_rate": 2.7735966925757807e-07,
|
| 1548 |
+
"loss": 0.3582,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.9079653322327693,
|
| 1553 |
+
"grad_norm": 1.9903170219755175,
|
| 1554 |
+
"learning_rate": 2.5416500754931294e-07,
|
| 1555 |
+
"loss": 0.3641,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.9120924473792819,
|
| 1560 |
+
"grad_norm": 2.025285970039861,
|
| 1561 |
+
"learning_rate": 2.3195773669219767e-07,
|
| 1562 |
+
"loss": 0.3651,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9162195625257945,
|
| 1567 |
+
"grad_norm": 1.9954116267972222,
|
| 1568 |
+
"learning_rate": 2.1074247700196337e-07,
|
| 1569 |
+
"loss": 0.3665,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.920346677672307,
|
| 1574 |
+
"grad_norm": 2.0132907410321588,
|
| 1575 |
+
"learning_rate": 1.905236424022633e-07,
|
| 1576 |
+
"loss": 0.3586,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.9244737928188197,
|
| 1581 |
+
"grad_norm": 1.9966146901368165,
|
| 1582 |
+
"learning_rate": 1.7130543950633405e-07,
|
| 1583 |
+
"loss": 0.3588,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9286009079653322,
|
| 1588 |
+
"grad_norm": 2.0300570451418314,
|
| 1589 |
+
"learning_rate": 1.5309186674179633e-07,
|
| 1590 |
+
"loss": 0.3634,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.9327280231118448,
|
| 1595 |
+
"grad_norm": 2.1175207366778146,
|
| 1596 |
+
"learning_rate": 1.358867135187636e-07,
|
| 1597 |
+
"loss": 0.3734,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9368551382583574,
|
| 1602 |
+
"grad_norm": 2.0296632257291956,
|
| 1603 |
+
"learning_rate": 1.1969355944143834e-07,
|
| 1604 |
+
"loss": 0.3632,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.94098225340487,
|
| 1609 |
+
"grad_norm": 1.9716530471763118,
|
| 1610 |
+
"learning_rate": 1.0451577356336118e-07,
|
| 1611 |
+
"loss": 0.3556,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9451093685513826,
|
| 1616 |
+
"grad_norm": 2.079524156836045,
|
| 1617 |
+
"learning_rate": 9.035651368646647e-08,
|
| 1618 |
+
"loss": 0.3627,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9492364836978952,
|
| 1623 |
+
"grad_norm": 2.063707259064606,
|
| 1624 |
+
"learning_rate": 7.721872570408684e-08,
|
| 1625 |
+
"loss": 0.3633,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9533635988444078,
|
| 1630 |
+
"grad_norm": 2.007674392778877,
|
| 1631 |
+
"learning_rate": 6.510514298804838e-08,
|
| 1632 |
+
"loss": 0.3622,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9574907139909203,
|
| 1637 |
+
"grad_norm": 2.0337463475612534,
|
| 1638 |
+
"learning_rate": 5.401828581997948e-08,
|
| 1639 |
+
"loss": 0.3576,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.961617829137433,
|
| 1644 |
+
"grad_norm": 1.994551313475225,
|
| 1645 |
+
"learning_rate": 4.396046086695915e-08,
|
| 1646 |
+
"loss": 0.3569,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.9657449442839455,
|
| 1651 |
+
"grad_norm": 2.016111248082697,
|
| 1652 |
+
"learning_rate": 3.4933760701602595e-08,
|
| 1653 |
+
"loss": 0.3578,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.9698720594304581,
|
| 1658 |
+
"grad_norm": 2.061506062561967,
|
| 1659 |
+
"learning_rate": 2.6940063366693303e-08,
|
| 1660 |
+
"loss": 0.3642,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.9739991745769707,
|
| 1665 |
+
"grad_norm": 1.9380243614271475,
|
| 1666 |
+
"learning_rate": 1.998103198444845e-08,
|
| 1667 |
+
"loss": 0.3682,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9781262897234833,
|
| 1672 |
+
"grad_norm": 2.0523022237333506,
|
| 1673 |
+
"learning_rate": 1.405811441049898e-08,
|
| 1674 |
+
"loss": 0.3558,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9822534048699959,
|
| 1679 |
+
"grad_norm": 1.9847092252520235,
|
| 1680 |
+
"learning_rate": 9.1725429326589e-09,
|
| 1681 |
+
"loss": 0.3513,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9863805200165084,
|
| 1686 |
+
"grad_norm": 1.9944372302390396,
|
| 1687 |
+
"learning_rate": 5.325334014539829e-09,
|
| 1688 |
+
"loss": 0.3682,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.9905076351630211,
|
| 1693 |
+
"grad_norm": 2.0557893820556847,
|
| 1694 |
+
"learning_rate": 2.5172880840745873e-09,
|
| 1695 |
+
"loss": 0.3642,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9946347503095336,
|
| 1700 |
+
"grad_norm": 1.9769612777056,
|
| 1701 |
+
"learning_rate": 7.48989366980979e-10,
|
| 1702 |
+
"loss": 0.368,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.9987618654560462,
|
| 1707 |
+
"grad_norm": 2.1452580734926343,
|
| 1708 |
+
"learning_rate": 2.080576521568123e-11,
|
| 1709 |
+
"loss": 0.3657,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.9995872884853487,
|
| 1714 |
+
"eval_loss": 0.3757357895374298,
|
| 1715 |
+
"eval_runtime": 0.963,
|
| 1716 |
+
"eval_samples_per_second": 3.115,
|
| 1717 |
+
"eval_steps_per_second": 1.038,
|
| 1718 |
+
"step": 1211
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9995872884853487,
|
| 1722 |
+
"step": 1211,
|
| 1723 |
+
"total_flos": 253506418114560.0,
|
| 1724 |
+
"train_loss": 0.5435279840826496,
|
| 1725 |
+
"train_runtime": 29545.3245,
|
| 1726 |
+
"train_samples_per_second": 1.312,
|
| 1727 |
+
"train_steps_per_second": 0.041
|
| 1728 |
+
}
|
| 1729 |
+
],
|
| 1730 |
+
"logging_steps": 5,
|
| 1731 |
+
"max_steps": 1211,
|
| 1732 |
+
"num_input_tokens_seen": 0,
|
| 1733 |
+
"num_train_epochs": 1,
|
| 1734 |
+
"save_steps": 100,
|
| 1735 |
+
"stateful_callbacks": {
|
| 1736 |
+
"TrainerControl": {
|
| 1737 |
+
"args": {
|
| 1738 |
+
"should_epoch_stop": false,
|
| 1739 |
+
"should_evaluate": false,
|
| 1740 |
+
"should_log": false,
|
| 1741 |
+
"should_save": true,
|
| 1742 |
+
"should_training_stop": true
|
| 1743 |
+
},
|
| 1744 |
+
"attributes": {}
|
| 1745 |
+
}
|
| 1746 |
+
},
|
| 1747 |
+
"total_flos": 253506418114560.0,
|
| 1748 |
+
"train_batch_size": 2,
|
| 1749 |
+
"trial_name": null,
|
| 1750 |
+
"trial_params": null
|
| 1751 |
+
}
|