Model save
Browse files- README.md +66 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +2858 -0
README.md
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: barc0/Llama-3.1-ARC-Potpourri-Transduction-8B
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: problem184_model
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# problem184_model
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [barc0/Llama-3.1-ARC-Potpourri-Transduction-8B](https://huggingface.co/barc0/Llama-3.1-ARC-Potpourri-Transduction-8B) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0047
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0001
|
| 41 |
+
- train_batch_size: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 2
|
| 47 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 2
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 0.0009 | 1.0 | 200 | 0.0760 |
|
| 57 |
+
| 0.0004 | 2.0 | 400 | 0.0047 |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- PEFT 0.13.2
|
| 63 |
+
- Transformers 4.46.2
|
| 64 |
+
- Pytorch 2.4.0+cu121
|
| 65 |
+
- Datasets 3.1.0
|
| 66 |
+
- Tokenizers 0.20.3
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 6612300644352.0,
|
| 4 |
+
"train_loss": 0.033545353031595365,
|
| 5 |
+
"train_runtime": 998.7883,
|
| 6 |
+
"train_samples": 400,
|
| 7 |
+
"train_samples_per_second": 0.801,
|
| 8 |
+
"train_steps_per_second": 0.4
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 6612300644352.0,
|
| 4 |
+
"train_loss": 0.033545353031595365,
|
| 5 |
+
"train_runtime": 998.7883,
|
| 6 |
+
"train_samples": 400,
|
| 7 |
+
"train_samples_per_second": 0.801,
|
| 8 |
+
"train_steps_per_second": 0.4
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2858 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005,
|
| 13 |
+
"grad_norm": 0.07553869201667132,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 0.277,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01,
|
| 20 |
+
"grad_norm": 0.07142321749040205,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.1731,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.015,
|
| 27 |
+
"grad_norm": 0.04004541380672789,
|
| 28 |
+
"learning_rate": 7.5e-06,
|
| 29 |
+
"loss": 0.2066,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02,
|
| 34 |
+
"grad_norm": 0.03135466258081705,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 0.1541,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.025,
|
| 41 |
+
"grad_norm": 0.08107204814222205,
|
| 42 |
+
"learning_rate": 1.25e-05,
|
| 43 |
+
"loss": 0.3396,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03,
|
| 48 |
+
"grad_norm": 0.02826228780285631,
|
| 49 |
+
"learning_rate": 1.5e-05,
|
| 50 |
+
"loss": 0.0857,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.035,
|
| 55 |
+
"grad_norm": 0.04534872682356579,
|
| 56 |
+
"learning_rate": 1.75e-05,
|
| 57 |
+
"loss": 0.1668,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.04,
|
| 62 |
+
"grad_norm": 0.08062512421235096,
|
| 63 |
+
"learning_rate": 2e-05,
|
| 64 |
+
"loss": 0.307,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.045,
|
| 69 |
+
"grad_norm": 0.09855642598201231,
|
| 70 |
+
"learning_rate": 2.25e-05,
|
| 71 |
+
"loss": 0.3094,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.05,
|
| 76 |
+
"grad_norm": 0.08223399648677109,
|
| 77 |
+
"learning_rate": 2.5e-05,
|
| 78 |
+
"loss": 0.3169,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.055,
|
| 83 |
+
"grad_norm": 0.036306345795043035,
|
| 84 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 85 |
+
"loss": 0.156,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06,
|
| 90 |
+
"grad_norm": 0.0446200343132829,
|
| 91 |
+
"learning_rate": 3e-05,
|
| 92 |
+
"loss": 0.1902,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.065,
|
| 97 |
+
"grad_norm": 0.0394950521156506,
|
| 98 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 99 |
+
"loss": 0.1665,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.07,
|
| 104 |
+
"grad_norm": 0.0779690335299316,
|
| 105 |
+
"learning_rate": 3.5e-05,
|
| 106 |
+
"loss": 0.3144,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.075,
|
| 111 |
+
"grad_norm": 0.0646426759636167,
|
| 112 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 113 |
+
"loss": 0.2018,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08,
|
| 118 |
+
"grad_norm": 0.08984226365886303,
|
| 119 |
+
"learning_rate": 4e-05,
|
| 120 |
+
"loss": 0.2255,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.085,
|
| 125 |
+
"grad_norm": 0.036674267927634084,
|
| 126 |
+
"learning_rate": 4.25e-05,
|
| 127 |
+
"loss": 0.1264,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.09,
|
| 132 |
+
"grad_norm": 0.09371135142216022,
|
| 133 |
+
"learning_rate": 4.5e-05,
|
| 134 |
+
"loss": 0.2776,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.095,
|
| 139 |
+
"grad_norm": 0.04744368538635616,
|
| 140 |
+
"learning_rate": 4.75e-05,
|
| 141 |
+
"loss": 0.149,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1,
|
| 146 |
+
"grad_norm": 0.11052474076313488,
|
| 147 |
+
"learning_rate": 5e-05,
|
| 148 |
+
"loss": 0.3257,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.105,
|
| 153 |
+
"grad_norm": 0.12235330629107824,
|
| 154 |
+
"learning_rate": 5.25e-05,
|
| 155 |
+
"loss": 0.2811,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.11,
|
| 160 |
+
"grad_norm": 0.09475772803286764,
|
| 161 |
+
"learning_rate": 5.500000000000001e-05,
|
| 162 |
+
"loss": 0.2566,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.115,
|
| 167 |
+
"grad_norm": 0.07682060292069236,
|
| 168 |
+
"learning_rate": 5.7499999999999995e-05,
|
| 169 |
+
"loss": 0.194,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12,
|
| 174 |
+
"grad_norm": 0.10253318753424252,
|
| 175 |
+
"learning_rate": 6e-05,
|
| 176 |
+
"loss": 0.1742,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.125,
|
| 181 |
+
"grad_norm": 0.15452516312260753,
|
| 182 |
+
"learning_rate": 6.25e-05,
|
| 183 |
+
"loss": 0.2426,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13,
|
| 188 |
+
"grad_norm": 0.06980713124278756,
|
| 189 |
+
"learning_rate": 6.500000000000001e-05,
|
| 190 |
+
"loss": 0.1524,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.135,
|
| 195 |
+
"grad_norm": 0.17169291075340687,
|
| 196 |
+
"learning_rate": 6.750000000000001e-05,
|
| 197 |
+
"loss": 0.2302,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14,
|
| 202 |
+
"grad_norm": 0.12097272092886306,
|
| 203 |
+
"learning_rate": 7e-05,
|
| 204 |
+
"loss": 0.1751,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.145,
|
| 209 |
+
"grad_norm": 0.04894250463264245,
|
| 210 |
+
"learning_rate": 7.25e-05,
|
| 211 |
+
"loss": 0.0896,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15,
|
| 216 |
+
"grad_norm": 0.16842709573150588,
|
| 217 |
+
"learning_rate": 7.500000000000001e-05,
|
| 218 |
+
"loss": 0.3866,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.155,
|
| 223 |
+
"grad_norm": 0.10293501286832617,
|
| 224 |
+
"learning_rate": 7.75e-05,
|
| 225 |
+
"loss": 0.166,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16,
|
| 230 |
+
"grad_norm": 0.06255095679441007,
|
| 231 |
+
"learning_rate": 8e-05,
|
| 232 |
+
"loss": 0.1388,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.165,
|
| 237 |
+
"grad_norm": 0.09759267630202532,
|
| 238 |
+
"learning_rate": 8.25e-05,
|
| 239 |
+
"loss": 0.1998,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17,
|
| 244 |
+
"grad_norm": 0.03725031482525615,
|
| 245 |
+
"learning_rate": 8.5e-05,
|
| 246 |
+
"loss": 0.0808,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.175,
|
| 251 |
+
"grad_norm": 0.06548370439282793,
|
| 252 |
+
"learning_rate": 8.75e-05,
|
| 253 |
+
"loss": 0.1329,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.18,
|
| 258 |
+
"grad_norm": 0.06363300533898003,
|
| 259 |
+
"learning_rate": 9e-05,
|
| 260 |
+
"loss": 0.0765,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.185,
|
| 265 |
+
"grad_norm": 0.1319969217444984,
|
| 266 |
+
"learning_rate": 9.250000000000001e-05,
|
| 267 |
+
"loss": 0.1583,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19,
|
| 272 |
+
"grad_norm": 0.1570940458256374,
|
| 273 |
+
"learning_rate": 9.5e-05,
|
| 274 |
+
"loss": 0.1703,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.195,
|
| 279 |
+
"grad_norm": 0.17816543762213244,
|
| 280 |
+
"learning_rate": 9.75e-05,
|
| 281 |
+
"loss": 0.1677,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2,
|
| 286 |
+
"grad_norm": 0.06814489472510708,
|
| 287 |
+
"learning_rate": 0.0001,
|
| 288 |
+
"loss": 0.0813,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.205,
|
| 293 |
+
"grad_norm": 0.12234791580019863,
|
| 294 |
+
"learning_rate": 9.999809615320856e-05,
|
| 295 |
+
"loss": 0.067,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.21,
|
| 300 |
+
"grad_norm": 0.11898491410474155,
|
| 301 |
+
"learning_rate": 9.999238475781957e-05,
|
| 302 |
+
"loss": 0.0832,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.215,
|
| 307 |
+
"grad_norm": 0.11599178809474275,
|
| 308 |
+
"learning_rate": 9.998286624877786e-05,
|
| 309 |
+
"loss": 0.1261,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.22,
|
| 314 |
+
"grad_norm": 0.18501882395233082,
|
| 315 |
+
"learning_rate": 9.99695413509548e-05,
|
| 316 |
+
"loss": 0.1908,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.225,
|
| 321 |
+
"grad_norm": 0.06856454706988713,
|
| 322 |
+
"learning_rate": 9.99524110790929e-05,
|
| 323 |
+
"loss": 0.0672,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.23,
|
| 328 |
+
"grad_norm": 0.05947222817167692,
|
| 329 |
+
"learning_rate": 9.99314767377287e-05,
|
| 330 |
+
"loss": 0.0745,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.235,
|
| 335 |
+
"grad_norm": 0.07765190202488637,
|
| 336 |
+
"learning_rate": 9.990673992109335e-05,
|
| 337 |
+
"loss": 0.0756,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.24,
|
| 342 |
+
"grad_norm": 0.1359994526177061,
|
| 343 |
+
"learning_rate": 9.987820251299122e-05,
|
| 344 |
+
"loss": 0.0331,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.245,
|
| 349 |
+
"grad_norm": 0.1373406192095304,
|
| 350 |
+
"learning_rate": 9.98458666866564e-05,
|
| 351 |
+
"loss": 0.0806,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.25,
|
| 356 |
+
"grad_norm": 0.05473571476326683,
|
| 357 |
+
"learning_rate": 9.980973490458728e-05,
|
| 358 |
+
"loss": 0.0522,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.255,
|
| 363 |
+
"grad_norm": 0.1169729050564611,
|
| 364 |
+
"learning_rate": 9.976980991835894e-05,
|
| 365 |
+
"loss": 0.0762,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.26,
|
| 370 |
+
"grad_norm": 0.05579810258223367,
|
| 371 |
+
"learning_rate": 9.972609476841367e-05,
|
| 372 |
+
"loss": 0.0345,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.265,
|
| 377 |
+
"grad_norm": 0.09381823707624487,
|
| 378 |
+
"learning_rate": 9.967859278382938e-05,
|
| 379 |
+
"loss": 0.0897,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.27,
|
| 384 |
+
"grad_norm": 0.07938166641979207,
|
| 385 |
+
"learning_rate": 9.962730758206611e-05,
|
| 386 |
+
"loss": 0.0654,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.275,
|
| 391 |
+
"grad_norm": 0.13087865330833595,
|
| 392 |
+
"learning_rate": 9.957224306869053e-05,
|
| 393 |
+
"loss": 0.1015,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.28,
|
| 398 |
+
"grad_norm": 0.263321695571175,
|
| 399 |
+
"learning_rate": 9.951340343707852e-05,
|
| 400 |
+
"loss": 0.0519,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.285,
|
| 405 |
+
"grad_norm": 0.14202447905316087,
|
| 406 |
+
"learning_rate": 9.945079316809585e-05,
|
| 407 |
+
"loss": 0.0368,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.29,
|
| 412 |
+
"grad_norm": 0.21296861819635188,
|
| 413 |
+
"learning_rate": 9.938441702975689e-05,
|
| 414 |
+
"loss": 0.0699,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.295,
|
| 419 |
+
"grad_norm": 0.17479932116014785,
|
| 420 |
+
"learning_rate": 9.931428007686158e-05,
|
| 421 |
+
"loss": 0.0383,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3,
|
| 426 |
+
"grad_norm": 0.19384996245201125,
|
| 427 |
+
"learning_rate": 9.924038765061042e-05,
|
| 428 |
+
"loss": 0.0977,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.305,
|
| 433 |
+
"grad_norm": 0.3132458011578011,
|
| 434 |
+
"learning_rate": 9.916274537819775e-05,
|
| 435 |
+
"loss": 0.0783,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.31,
|
| 440 |
+
"grad_norm": 0.24343342969724435,
|
| 441 |
+
"learning_rate": 9.908135917238321e-05,
|
| 442 |
+
"loss": 0.1309,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.315,
|
| 447 |
+
"grad_norm": 0.06358039936658727,
|
| 448 |
+
"learning_rate": 9.899623523104149e-05,
|
| 449 |
+
"loss": 0.0341,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.32,
|
| 454 |
+
"grad_norm": 0.2220768799042642,
|
| 455 |
+
"learning_rate": 9.890738003669029e-05,
|
| 456 |
+
"loss": 0.1244,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.325,
|
| 461 |
+
"grad_norm": 0.09877556684729816,
|
| 462 |
+
"learning_rate": 9.881480035599667e-05,
|
| 463 |
+
"loss": 0.0584,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.33,
|
| 468 |
+
"grad_norm": 0.07125822760195885,
|
| 469 |
+
"learning_rate": 9.871850323926177e-05,
|
| 470 |
+
"loss": 0.0099,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.335,
|
| 475 |
+
"grad_norm": 0.19991568339646582,
|
| 476 |
+
"learning_rate": 9.861849601988383e-05,
|
| 477 |
+
"loss": 0.0532,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.34,
|
| 482 |
+
"grad_norm": 0.18115871524580987,
|
| 483 |
+
"learning_rate": 9.851478631379982e-05,
|
| 484 |
+
"loss": 0.1011,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.345,
|
| 489 |
+
"grad_norm": 0.07567960238386034,
|
| 490 |
+
"learning_rate": 9.84073820189054e-05,
|
| 491 |
+
"loss": 0.034,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.35,
|
| 496 |
+
"grad_norm": 0.15451087209839287,
|
| 497 |
+
"learning_rate": 9.829629131445342e-05,
|
| 498 |
+
"loss": 0.0348,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.355,
|
| 503 |
+
"grad_norm": 0.08313091042199042,
|
| 504 |
+
"learning_rate": 9.818152266043114e-05,
|
| 505 |
+
"loss": 0.0422,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.36,
|
| 510 |
+
"grad_norm": 0.07302944049536657,
|
| 511 |
+
"learning_rate": 9.806308479691595e-05,
|
| 512 |
+
"loss": 0.0417,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.365,
|
| 517 |
+
"grad_norm": 0.041309166149277605,
|
| 518 |
+
"learning_rate": 9.794098674340965e-05,
|
| 519 |
+
"loss": 0.0231,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.37,
|
| 524 |
+
"grad_norm": 0.05571619251756244,
|
| 525 |
+
"learning_rate": 9.781523779815179e-05,
|
| 526 |
+
"loss": 0.0517,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.375,
|
| 531 |
+
"grad_norm": 0.07780175573372976,
|
| 532 |
+
"learning_rate": 9.768584753741134e-05,
|
| 533 |
+
"loss": 0.0312,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.38,
|
| 538 |
+
"grad_norm": 0.046226872712960713,
|
| 539 |
+
"learning_rate": 9.755282581475769e-05,
|
| 540 |
+
"loss": 0.0418,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.385,
|
| 545 |
+
"grad_norm": 0.08057703631440487,
|
| 546 |
+
"learning_rate": 9.741618276030997e-05,
|
| 547 |
+
"loss": 0.0419,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.39,
|
| 552 |
+
"grad_norm": 0.05050489987091237,
|
| 553 |
+
"learning_rate": 9.727592877996585e-05,
|
| 554 |
+
"loss": 0.0163,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.395,
|
| 559 |
+
"grad_norm": 0.0959178937528156,
|
| 560 |
+
"learning_rate": 9.713207455460894e-05,
|
| 561 |
+
"loss": 0.0428,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4,
|
| 566 |
+
"grad_norm": 0.08603826631317996,
|
| 567 |
+
"learning_rate": 9.698463103929542e-05,
|
| 568 |
+
"loss": 0.0674,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.405,
|
| 573 |
+
"grad_norm": 0.044173037654917646,
|
| 574 |
+
"learning_rate": 9.683360946241989e-05,
|
| 575 |
+
"loss": 0.0425,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.41,
|
| 580 |
+
"grad_norm": 0.026290493271921827,
|
| 581 |
+
"learning_rate": 9.667902132486009e-05,
|
| 582 |
+
"loss": 0.0053,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.415,
|
| 587 |
+
"grad_norm": 0.09325040701230163,
|
| 588 |
+
"learning_rate": 9.652087839910124e-05,
|
| 589 |
+
"loss": 0.0505,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.42,
|
| 594 |
+
"grad_norm": 0.11764293119485195,
|
| 595 |
+
"learning_rate": 9.635919272833938e-05,
|
| 596 |
+
"loss": 0.0305,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.425,
|
| 601 |
+
"grad_norm": 0.18654991334295415,
|
| 602 |
+
"learning_rate": 9.619397662556435e-05,
|
| 603 |
+
"loss": 0.011,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.43,
|
| 608 |
+
"grad_norm": 0.08098006638815154,
|
| 609 |
+
"learning_rate": 9.602524267262203e-05,
|
| 610 |
+
"loss": 0.0183,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.435,
|
| 615 |
+
"grad_norm": 0.07659797438155175,
|
| 616 |
+
"learning_rate": 9.58530037192562e-05,
|
| 617 |
+
"loss": 0.0274,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.44,
|
| 622 |
+
"grad_norm": 0.16347900939631616,
|
| 623 |
+
"learning_rate": 9.567727288213005e-05,
|
| 624 |
+
"loss": 0.0859,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.445,
|
| 629 |
+
"grad_norm": 0.15150817750296133,
|
| 630 |
+
"learning_rate": 9.549806354382717e-05,
|
| 631 |
+
"loss": 0.0333,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.45,
|
| 636 |
+
"grad_norm": 0.07123424851384792,
|
| 637 |
+
"learning_rate": 9.53153893518325e-05,
|
| 638 |
+
"loss": 0.0405,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.455,
|
| 643 |
+
"grad_norm": 0.11026163678942051,
|
| 644 |
+
"learning_rate": 9.512926421749304e-05,
|
| 645 |
+
"loss": 0.0462,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.46,
|
| 650 |
+
"grad_norm": 0.07331190265762698,
|
| 651 |
+
"learning_rate": 9.493970231495835e-05,
|
| 652 |
+
"loss": 0.0069,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.465,
|
| 657 |
+
"grad_norm": 0.06418964403024571,
|
| 658 |
+
"learning_rate": 9.474671808010126e-05,
|
| 659 |
+
"loss": 0.0441,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.47,
|
| 664 |
+
"grad_norm": 0.04366811050044381,
|
| 665 |
+
"learning_rate": 9.45503262094184e-05,
|
| 666 |
+
"loss": 0.0321,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.475,
|
| 671 |
+
"grad_norm": 0.06891688314378318,
|
| 672 |
+
"learning_rate": 9.435054165891109e-05,
|
| 673 |
+
"loss": 0.0197,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.48,
|
| 678 |
+
"grad_norm": 0.0791719137172165,
|
| 679 |
+
"learning_rate": 9.414737964294636e-05,
|
| 680 |
+
"loss": 0.0211,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.485,
|
| 685 |
+
"grad_norm": 0.09055479977019616,
|
| 686 |
+
"learning_rate": 9.394085563309827e-05,
|
| 687 |
+
"loss": 0.0368,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.49,
|
| 692 |
+
"grad_norm": 0.1078342902713965,
|
| 693 |
+
"learning_rate": 9.373098535696979e-05,
|
| 694 |
+
"loss": 0.0402,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.495,
|
| 699 |
+
"grad_norm": 0.04234438785284557,
|
| 700 |
+
"learning_rate": 9.351778479699499e-05,
|
| 701 |
+
"loss": 0.0155,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5,
|
| 706 |
+
"grad_norm": 0.32259634488597166,
|
| 707 |
+
"learning_rate": 9.330127018922194e-05,
|
| 708 |
+
"loss": 0.0234,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.505,
|
| 713 |
+
"grad_norm": 0.057346774722805624,
|
| 714 |
+
"learning_rate": 9.308145802207629e-05,
|
| 715 |
+
"loss": 0.0057,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.51,
|
| 720 |
+
"grad_norm": 0.06845584491597821,
|
| 721 |
+
"learning_rate": 9.285836503510562e-05,
|
| 722 |
+
"loss": 0.0239,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.515,
|
| 727 |
+
"grad_norm": 0.027458411814212238,
|
| 728 |
+
"learning_rate": 9.263200821770461e-05,
|
| 729 |
+
"loss": 0.0046,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.52,
|
| 734 |
+
"grad_norm": 0.04119439564225103,
|
| 735 |
+
"learning_rate": 9.24024048078213e-05,
|
| 736 |
+
"loss": 0.0067,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.525,
|
| 741 |
+
"grad_norm": 0.25272039680541203,
|
| 742 |
+
"learning_rate": 9.21695722906443e-05,
|
| 743 |
+
"loss": 0.1068,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.53,
|
| 748 |
+
"grad_norm": 0.15642301404795694,
|
| 749 |
+
"learning_rate": 9.193352839727121e-05,
|
| 750 |
+
"loss": 0.0487,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.535,
|
| 755 |
+
"grad_norm": 0.020234092439841273,
|
| 756 |
+
"learning_rate": 9.169429110335841e-05,
|
| 757 |
+
"loss": 0.0044,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.54,
|
| 762 |
+
"grad_norm": 0.021594134995858247,
|
| 763 |
+
"learning_rate": 9.145187862775209e-05,
|
| 764 |
+
"loss": 0.0023,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.545,
|
| 769 |
+
"grad_norm": 0.11956523872488599,
|
| 770 |
+
"learning_rate": 9.120630943110077e-05,
|
| 771 |
+
"loss": 0.0647,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.55,
|
| 776 |
+
"grad_norm": 0.06453797114816638,
|
| 777 |
+
"learning_rate": 9.09576022144496e-05,
|
| 778 |
+
"loss": 0.0378,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.555,
|
| 783 |
+
"grad_norm": 0.040619091199357006,
|
| 784 |
+
"learning_rate": 9.070577591781597e-05,
|
| 785 |
+
"loss": 0.0101,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.56,
|
| 790 |
+
"grad_norm": 0.012631286216702108,
|
| 791 |
+
"learning_rate": 9.045084971874738e-05,
|
| 792 |
+
"loss": 0.0025,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.565,
|
| 797 |
+
"grad_norm": 0.11250625624010586,
|
| 798 |
+
"learning_rate": 9.019284303086087e-05,
|
| 799 |
+
"loss": 0.0326,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.57,
|
| 804 |
+
"grad_norm": 0.008498513085188694,
|
| 805 |
+
"learning_rate": 8.993177550236464e-05,
|
| 806 |
+
"loss": 0.0019,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.575,
|
| 811 |
+
"grad_norm": 0.06549524971538588,
|
| 812 |
+
"learning_rate": 8.966766701456177e-05,
|
| 813 |
+
"loss": 0.0144,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.58,
|
| 818 |
+
"grad_norm": 0.1243637807598006,
|
| 819 |
+
"learning_rate": 8.940053768033609e-05,
|
| 820 |
+
"loss": 0.0427,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.585,
|
| 825 |
+
"grad_norm": 0.4551297436540559,
|
| 826 |
+
"learning_rate": 8.91304078426207e-05,
|
| 827 |
+
"loss": 0.1364,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.59,
|
| 832 |
+
"grad_norm": 0.14895611964997935,
|
| 833 |
+
"learning_rate": 8.885729807284856e-05,
|
| 834 |
+
"loss": 0.0516,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.595,
|
| 839 |
+
"grad_norm": 0.03168114639692374,
|
| 840 |
+
"learning_rate": 8.858122916938601e-05,
|
| 841 |
+
"loss": 0.0038,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6,
|
| 846 |
+
"grad_norm": 0.08370376943417981,
|
| 847 |
+
"learning_rate": 8.83022221559489e-05,
|
| 848 |
+
"loss": 0.0068,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.605,
|
| 853 |
+
"grad_norm": 0.07476474940113496,
|
| 854 |
+
"learning_rate": 8.802029828000156e-05,
|
| 855 |
+
"loss": 0.0182,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.61,
|
| 860 |
+
"grad_norm": 0.13159976173852048,
|
| 861 |
+
"learning_rate": 8.773547901113862e-05,
|
| 862 |
+
"loss": 0.0352,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.615,
|
| 867 |
+
"grad_norm": 0.022707973634618142,
|
| 868 |
+
"learning_rate": 8.744778603945011e-05,
|
| 869 |
+
"loss": 0.0033,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.62,
|
| 874 |
+
"grad_norm": 0.12482944752372985,
|
| 875 |
+
"learning_rate": 8.715724127386972e-05,
|
| 876 |
+
"loss": 0.0202,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.625,
|
| 881 |
+
"grad_norm": 0.39302174930812633,
|
| 882 |
+
"learning_rate": 8.68638668405062e-05,
|
| 883 |
+
"loss": 0.0758,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.63,
|
| 888 |
+
"grad_norm": 0.19383130479510832,
|
| 889 |
+
"learning_rate": 8.656768508095853e-05,
|
| 890 |
+
"loss": 0.0325,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.635,
|
| 895 |
+
"grad_norm": 0.03039287191356843,
|
| 896 |
+
"learning_rate": 8.626871855061438e-05,
|
| 897 |
+
"loss": 0.0041,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.64,
|
| 902 |
+
"grad_norm": 0.048215778020076296,
|
| 903 |
+
"learning_rate": 8.596699001693255e-05,
|
| 904 |
+
"loss": 0.0273,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.645,
|
| 909 |
+
"grad_norm": 0.1392522953938403,
|
| 910 |
+
"learning_rate": 8.566252245770909e-05,
|
| 911 |
+
"loss": 0.0622,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.65,
|
| 916 |
+
"grad_norm": 0.07411597583466574,
|
| 917 |
+
"learning_rate": 8.535533905932738e-05,
|
| 918 |
+
"loss": 0.0116,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.655,
|
| 923 |
+
"grad_norm": 0.08394509913329051,
|
| 924 |
+
"learning_rate": 8.504546321499255e-05,
|
| 925 |
+
"loss": 0.0454,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.66,
|
| 930 |
+
"grad_norm": 0.04530702990326196,
|
| 931 |
+
"learning_rate": 8.473291852294987e-05,
|
| 932 |
+
"loss": 0.0185,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.665,
|
| 937 |
+
"grad_norm": 0.03765639754984064,
|
| 938 |
+
"learning_rate": 8.44177287846877e-05,
|
| 939 |
+
"loss": 0.0065,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.67,
|
| 944 |
+
"grad_norm": 0.031586807052073985,
|
| 945 |
+
"learning_rate": 8.409991800312493e-05,
|
| 946 |
+
"loss": 0.0064,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.675,
|
| 951 |
+
"grad_norm": 0.08297489591002068,
|
| 952 |
+
"learning_rate": 8.377951038078302e-05,
|
| 953 |
+
"loss": 0.0086,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.68,
|
| 958 |
+
"grad_norm": 0.03562288518144821,
|
| 959 |
+
"learning_rate": 8.345653031794292e-05,
|
| 960 |
+
"loss": 0.0117,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.685,
|
| 965 |
+
"grad_norm": 0.031042156897553914,
|
| 966 |
+
"learning_rate": 8.313100241078689e-05,
|
| 967 |
+
"loss": 0.0032,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.69,
|
| 972 |
+
"grad_norm": 0.02483745350649097,
|
| 973 |
+
"learning_rate": 8.280295144952536e-05,
|
| 974 |
+
"loss": 0.0116,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.695,
|
| 979 |
+
"grad_norm": 0.009338647544721107,
|
| 980 |
+
"learning_rate": 8.247240241650918e-05,
|
| 981 |
+
"loss": 0.0015,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7,
|
| 986 |
+
"grad_norm": 0.34270075427008806,
|
| 987 |
+
"learning_rate": 8.213938048432697e-05,
|
| 988 |
+
"loss": 0.0789,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.705,
|
| 993 |
+
"grad_norm": 0.024842574848788327,
|
| 994 |
+
"learning_rate": 8.18039110138882e-05,
|
| 995 |
+
"loss": 0.0039,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.71,
|
| 1000 |
+
"grad_norm": 0.12421960639233866,
|
| 1001 |
+
"learning_rate": 8.146601955249188e-05,
|
| 1002 |
+
"loss": 0.0398,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.715,
|
| 1007 |
+
"grad_norm": 0.056187242189785436,
|
| 1008 |
+
"learning_rate": 8.112573183188099e-05,
|
| 1009 |
+
"loss": 0.0103,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.72,
|
| 1014 |
+
"grad_norm": 0.023766584635126216,
|
| 1015 |
+
"learning_rate": 8.07830737662829e-05,
|
| 1016 |
+
"loss": 0.0049,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.725,
|
| 1021 |
+
"grad_norm": 0.04512771394673685,
|
| 1022 |
+
"learning_rate": 8.043807145043604e-05,
|
| 1023 |
+
"loss": 0.0219,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.73,
|
| 1028 |
+
"grad_norm": 0.0274526387971708,
|
| 1029 |
+
"learning_rate": 8.009075115760243e-05,
|
| 1030 |
+
"loss": 0.0033,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.735,
|
| 1035 |
+
"grad_norm": 0.19532934746359332,
|
| 1036 |
+
"learning_rate": 7.974113933756707e-05,
|
| 1037 |
+
"loss": 0.0734,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.74,
|
| 1042 |
+
"grad_norm": 0.032852874131568864,
|
| 1043 |
+
"learning_rate": 7.938926261462366e-05,
|
| 1044 |
+
"loss": 0.0042,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.745,
|
| 1049 |
+
"grad_norm": 0.0640117280012966,
|
| 1050 |
+
"learning_rate": 7.903514778554699e-05,
|
| 1051 |
+
"loss": 0.0156,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.75,
|
| 1056 |
+
"grad_norm": 0.047410567888246616,
|
| 1057 |
+
"learning_rate": 7.86788218175523e-05,
|
| 1058 |
+
"loss": 0.0074,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.755,
|
| 1063 |
+
"grad_norm": 0.16842524790743002,
|
| 1064 |
+
"learning_rate": 7.832031184624164e-05,
|
| 1065 |
+
"loss": 0.0398,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.76,
|
| 1070 |
+
"grad_norm": 0.006147996063681932,
|
| 1071 |
+
"learning_rate": 7.795964517353735e-05,
|
| 1072 |
+
"loss": 0.001,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.765,
|
| 1077 |
+
"grad_norm": 0.04008165019001057,
|
| 1078 |
+
"learning_rate": 7.75968492656029e-05,
|
| 1079 |
+
"loss": 0.0059,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.77,
|
| 1084 |
+
"grad_norm": 0.007292282430214955,
|
| 1085 |
+
"learning_rate": 7.723195175075136e-05,
|
| 1086 |
+
"loss": 0.0012,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.775,
|
| 1091 |
+
"grad_norm": 0.14138265996109384,
|
| 1092 |
+
"learning_rate": 7.68649804173412e-05,
|
| 1093 |
+
"loss": 0.0259,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.78,
|
| 1098 |
+
"grad_norm": 0.07468592471462034,
|
| 1099 |
+
"learning_rate": 7.649596321166024e-05,
|
| 1100 |
+
"loss": 0.0192,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.785,
|
| 1105 |
+
"grad_norm": 0.014395933526058188,
|
| 1106 |
+
"learning_rate": 7.612492823579745e-05,
|
| 1107 |
+
"loss": 0.0025,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.79,
|
| 1112 |
+
"grad_norm": 0.0038502942774920196,
|
| 1113 |
+
"learning_rate": 7.575190374550272e-05,
|
| 1114 |
+
"loss": 0.0005,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.795,
|
| 1119 |
+
"grad_norm": 0.10226426219442211,
|
| 1120 |
+
"learning_rate": 7.537691814803521e-05,
|
| 1121 |
+
"loss": 0.0123,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8,
|
| 1126 |
+
"grad_norm": 0.012762312166514576,
|
| 1127 |
+
"learning_rate": 7.500000000000001e-05,
|
| 1128 |
+
"loss": 0.0019,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.805,
|
| 1133 |
+
"grad_norm": 0.01630848631501936,
|
| 1134 |
+
"learning_rate": 7.462117800517336e-05,
|
| 1135 |
+
"loss": 0.0019,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.81,
|
| 1140 |
+
"grad_norm": 0.09121920855984286,
|
| 1141 |
+
"learning_rate": 7.424048101231686e-05,
|
| 1142 |
+
"loss": 0.0252,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.815,
|
| 1147 |
+
"grad_norm": 0.044779381951263734,
|
| 1148 |
+
"learning_rate": 7.385793801298042e-05,
|
| 1149 |
+
"loss": 0.0049,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.82,
|
| 1154 |
+
"grad_norm": 0.0069701432925383874,
|
| 1155 |
+
"learning_rate": 7.347357813929454e-05,
|
| 1156 |
+
"loss": 0.0005,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.825,
|
| 1161 |
+
"grad_norm": 0.04833507572943274,
|
| 1162 |
+
"learning_rate": 7.308743066175172e-05,
|
| 1163 |
+
"loss": 0.0127,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.83,
|
| 1168 |
+
"grad_norm": 0.001782479532615867,
|
| 1169 |
+
"learning_rate": 7.269952498697734e-05,
|
| 1170 |
+
"loss": 0.0003,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.835,
|
| 1175 |
+
"grad_norm": 0.1033970098216112,
|
| 1176 |
+
"learning_rate": 7.230989065549044e-05,
|
| 1177 |
+
"loss": 0.0331,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.84,
|
| 1182 |
+
"grad_norm": 0.050963861639285846,
|
| 1183 |
+
"learning_rate": 7.191855733945387e-05,
|
| 1184 |
+
"loss": 0.0078,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.845,
|
| 1189 |
+
"grad_norm": 0.025336676884289144,
|
| 1190 |
+
"learning_rate": 7.152555484041476e-05,
|
| 1191 |
+
"loss": 0.0037,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.85,
|
| 1196 |
+
"grad_norm": 0.04631516809520592,
|
| 1197 |
+
"learning_rate": 7.113091308703498e-05,
|
| 1198 |
+
"loss": 0.0026,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.855,
|
| 1203 |
+
"grad_norm": 0.024526402505777593,
|
| 1204 |
+
"learning_rate": 7.073466213281196e-05,
|
| 1205 |
+
"loss": 0.0032,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.86,
|
| 1210 |
+
"grad_norm": 0.12794480290400578,
|
| 1211 |
+
"learning_rate": 7.033683215379002e-05,
|
| 1212 |
+
"loss": 0.0286,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.865,
|
| 1217 |
+
"grad_norm": 0.05388064509201144,
|
| 1218 |
+
"learning_rate": 6.993745344626231e-05,
|
| 1219 |
+
"loss": 0.019,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.87,
|
| 1224 |
+
"grad_norm": 0.029754195193650093,
|
| 1225 |
+
"learning_rate": 6.953655642446368e-05,
|
| 1226 |
+
"loss": 0.0127,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.875,
|
| 1231 |
+
"grad_norm": 0.11351797783250586,
|
| 1232 |
+
"learning_rate": 6.91341716182545e-05,
|
| 1233 |
+
"loss": 0.0276,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.88,
|
| 1238 |
+
"grad_norm": 0.04238335598804278,
|
| 1239 |
+
"learning_rate": 6.873032967079561e-05,
|
| 1240 |
+
"loss": 0.0044,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.885,
|
| 1245 |
+
"grad_norm": 0.07192092013374167,
|
| 1246 |
+
"learning_rate": 6.832506133621487e-05,
|
| 1247 |
+
"loss": 0.0033,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.89,
|
| 1252 |
+
"grad_norm": 0.086298683194853,
|
| 1253 |
+
"learning_rate": 6.7918397477265e-05,
|
| 1254 |
+
"loss": 0.0255,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.895,
|
| 1259 |
+
"grad_norm": 0.010115510174389889,
|
| 1260 |
+
"learning_rate": 6.751036906297337e-05,
|
| 1261 |
+
"loss": 0.0005,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9,
|
| 1266 |
+
"grad_norm": 0.11331127094042127,
|
| 1267 |
+
"learning_rate": 6.710100716628344e-05,
|
| 1268 |
+
"loss": 0.0334,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.905,
|
| 1273 |
+
"grad_norm": 0.00359905539491798,
|
| 1274 |
+
"learning_rate": 6.669034296168855e-05,
|
| 1275 |
+
"loss": 0.0005,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.91,
|
| 1280 |
+
"grad_norm": 0.003786666725911578,
|
| 1281 |
+
"learning_rate": 6.627840772285784e-05,
|
| 1282 |
+
"loss": 0.0004,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.915,
|
| 1287 |
+
"grad_norm": 0.1178858284146152,
|
| 1288 |
+
"learning_rate": 6.586523282025462e-05,
|
| 1289 |
+
"loss": 0.0059,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.92,
|
| 1294 |
+
"grad_norm": 0.0028456285153728687,
|
| 1295 |
+
"learning_rate": 6.545084971874738e-05,
|
| 1296 |
+
"loss": 0.0004,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.925,
|
| 1301 |
+
"grad_norm": 0.005713089625735049,
|
| 1302 |
+
"learning_rate": 6.503528997521366e-05,
|
| 1303 |
+
"loss": 0.0012,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.93,
|
| 1308 |
+
"grad_norm": 0.0030406845588623957,
|
| 1309 |
+
"learning_rate": 6.461858523613684e-05,
|
| 1310 |
+
"loss": 0.0003,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.935,
|
| 1315 |
+
"grad_norm": 0.0028405909976545836,
|
| 1316 |
+
"learning_rate": 6.420076723519614e-05,
|
| 1317 |
+
"loss": 0.0003,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.94,
|
| 1322 |
+
"grad_norm": 0.2531768340842614,
|
| 1323 |
+
"learning_rate": 6.378186779084995e-05,
|
| 1324 |
+
"loss": 0.0043,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.945,
|
| 1329 |
+
"grad_norm": 0.05452635559394549,
|
| 1330 |
+
"learning_rate": 6.336191880391284e-05,
|
| 1331 |
+
"loss": 0.0288,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.95,
|
| 1336 |
+
"grad_norm": 0.009714923881054332,
|
| 1337 |
+
"learning_rate": 6.294095225512603e-05,
|
| 1338 |
+
"loss": 0.0012,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.955,
|
| 1343 |
+
"grad_norm": 0.004170292898958103,
|
| 1344 |
+
"learning_rate": 6.251900020272208e-05,
|
| 1345 |
+
"loss": 0.0004,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.96,
|
| 1350 |
+
"grad_norm": 0.06545271449024238,
|
| 1351 |
+
"learning_rate": 6.209609477998338e-05,
|
| 1352 |
+
"loss": 0.0316,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.965,
|
| 1357 |
+
"grad_norm": 0.002583766103443765,
|
| 1358 |
+
"learning_rate": 6.167226819279528e-05,
|
| 1359 |
+
"loss": 0.0005,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.97,
|
| 1364 |
+
"grad_norm": 0.05335515958281829,
|
| 1365 |
+
"learning_rate": 6.124755271719325e-05,
|
| 1366 |
+
"loss": 0.0248,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.975,
|
| 1371 |
+
"grad_norm": 0.11111023851134917,
|
| 1372 |
+
"learning_rate": 6.0821980696905146e-05,
|
| 1373 |
+
"loss": 0.015,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.98,
|
| 1378 |
+
"grad_norm": 0.0237783109632992,
|
| 1379 |
+
"learning_rate": 6.0395584540887963e-05,
|
| 1380 |
+
"loss": 0.0053,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.985,
|
| 1385 |
+
"grad_norm": 0.04143947097844851,
|
| 1386 |
+
"learning_rate": 5.9968396720859864e-05,
|
| 1387 |
+
"loss": 0.0051,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.99,
|
| 1392 |
+
"grad_norm": 0.02380124179400986,
|
| 1393 |
+
"learning_rate": 5.9540449768827246e-05,
|
| 1394 |
+
"loss": 0.0035,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.995,
|
| 1399 |
+
"grad_norm": 0.01783821356262124,
|
| 1400 |
+
"learning_rate": 5.911177627460739e-05,
|
| 1401 |
+
"loss": 0.0019,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.0,
|
| 1406 |
+
"grad_norm": 0.012578119490010059,
|
| 1407 |
+
"learning_rate": 5.868240888334653e-05,
|
| 1408 |
+
"loss": 0.0009,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.0,
|
| 1413 |
+
"eval_loss": 0.07603452354669571,
|
| 1414 |
+
"eval_runtime": 0.4047,
|
| 1415 |
+
"eval_samples_per_second": 2.471,
|
| 1416 |
+
"eval_steps_per_second": 2.471,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.005,
|
| 1421 |
+
"grad_norm": 0.014839351675487417,
|
| 1422 |
+
"learning_rate": 5.8252380293033884e-05,
|
| 1423 |
+
"loss": 0.0006,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.01,
|
| 1428 |
+
"grad_norm": 0.012314267589470526,
|
| 1429 |
+
"learning_rate": 5.782172325201155e-05,
|
| 1430 |
+
"loss": 0.0029,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.015,
|
| 1435 |
+
"grad_norm": 0.003125090751647675,
|
| 1436 |
+
"learning_rate": 5.7390470556480545e-05,
|
| 1437 |
+
"loss": 0.0004,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.02,
|
| 1442 |
+
"grad_norm": 0.007073162659384819,
|
| 1443 |
+
"learning_rate": 5.695865504800327e-05,
|
| 1444 |
+
"loss": 0.001,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.025,
|
| 1449 |
+
"grad_norm": 0.31379793587721094,
|
| 1450 |
+
"learning_rate": 5.6526309611002594e-05,
|
| 1451 |
+
"loss": 0.0057,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.03,
|
| 1456 |
+
"grad_norm": 0.0011295952381376085,
|
| 1457 |
+
"learning_rate": 5.6093467170257374e-05,
|
| 1458 |
+
"loss": 0.0001,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.035,
|
| 1463 |
+
"grad_norm": 0.0046728430974052725,
|
| 1464 |
+
"learning_rate": 5.566016068839535e-05,
|
| 1465 |
+
"loss": 0.0003,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.04,
|
| 1470 |
+
"grad_norm": 0.0055708348232121965,
|
| 1471 |
+
"learning_rate": 5.522642316338268e-05,
|
| 1472 |
+
"loss": 0.0006,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.045,
|
| 1477 |
+
"grad_norm": 0.0465350248080129,
|
| 1478 |
+
"learning_rate": 5.4792287626011204e-05,
|
| 1479 |
+
"loss": 0.0012,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.05,
|
| 1484 |
+
"grad_norm": 0.0008151862390823683,
|
| 1485 |
+
"learning_rate": 5.435778713738292e-05,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.055,
|
| 1491 |
+
"grad_norm": 0.0017764023093753461,
|
| 1492 |
+
"learning_rate": 5.392295478639225e-05,
|
| 1493 |
+
"loss": 0.0003,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.06,
|
| 1498 |
+
"grad_norm": 0.028033413085950722,
|
| 1499 |
+
"learning_rate": 5.348782368720626e-05,
|
| 1500 |
+
"loss": 0.0065,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.065,
|
| 1505 |
+
"grad_norm": 0.00419815372280561,
|
| 1506 |
+
"learning_rate": 5.3052426976742855e-05,
|
| 1507 |
+
"loss": 0.0004,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.07,
|
| 1512 |
+
"grad_norm": 0.0102127421097092,
|
| 1513 |
+
"learning_rate": 5.26167978121472e-05,
|
| 1514 |
+
"loss": 0.0021,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.075,
|
| 1519 |
+
"grad_norm": 0.0015729817142162348,
|
| 1520 |
+
"learning_rate": 5.218096936826681e-05,
|
| 1521 |
+
"loss": 0.0003,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.08,
|
| 1526 |
+
"grad_norm": 0.004004133543726314,
|
| 1527 |
+
"learning_rate": 5.174497483512506e-05,
|
| 1528 |
+
"loss": 0.0002,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.085,
|
| 1533 |
+
"grad_norm": 0.0012454184210680926,
|
| 1534 |
+
"learning_rate": 5.1308847415393666e-05,
|
| 1535 |
+
"loss": 0.0002,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.09,
|
| 1540 |
+
"grad_norm": 0.0005409115041498422,
|
| 1541 |
+
"learning_rate": 5.0872620321864185e-05,
|
| 1542 |
+
"loss": 0.0001,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.095,
|
| 1547 |
+
"grad_norm": 0.001726174471896587,
|
| 1548 |
+
"learning_rate": 5.04363267749187e-05,
|
| 1549 |
+
"loss": 0.0002,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.1,
|
| 1554 |
+
"grad_norm": 0.0010814524798174311,
|
| 1555 |
+
"learning_rate": 5e-05,
|
| 1556 |
+
"loss": 0.0001,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.105,
|
| 1561 |
+
"grad_norm": 0.009560946952238247,
|
| 1562 |
+
"learning_rate": 4.9563673225081314e-05,
|
| 1563 |
+
"loss": 0.0008,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.11,
|
| 1568 |
+
"grad_norm": 0.01441628431952276,
|
| 1569 |
+
"learning_rate": 4.912737967813583e-05,
|
| 1570 |
+
"loss": 0.0009,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.115,
|
| 1575 |
+
"grad_norm": 0.0023532660086251268,
|
| 1576 |
+
"learning_rate": 4.869115258460635e-05,
|
| 1577 |
+
"loss": 0.0003,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.12,
|
| 1582 |
+
"grad_norm": 0.057060873899210225,
|
| 1583 |
+
"learning_rate": 4.825502516487497e-05,
|
| 1584 |
+
"loss": 0.0041,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.125,
|
| 1589 |
+
"grad_norm": 0.020282977300679454,
|
| 1590 |
+
"learning_rate": 4.781903063173321e-05,
|
| 1591 |
+
"loss": 0.002,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.13,
|
| 1596 |
+
"grad_norm": 0.02046603304331513,
|
| 1597 |
+
"learning_rate": 4.738320218785281e-05,
|
| 1598 |
+
"loss": 0.0075,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.135,
|
| 1603 |
+
"grad_norm": 0.013990089439389753,
|
| 1604 |
+
"learning_rate": 4.694757302325715e-05,
|
| 1605 |
+
"loss": 0.0014,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.1400000000000001,
|
| 1610 |
+
"grad_norm": 0.211826851392036,
|
| 1611 |
+
"learning_rate": 4.6512176312793736e-05,
|
| 1612 |
+
"loss": 0.0054,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.145,
|
| 1617 |
+
"grad_norm": 0.0064942039937938645,
|
| 1618 |
+
"learning_rate": 4.607704521360776e-05,
|
| 1619 |
+
"loss": 0.0011,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.15,
|
| 1624 |
+
"grad_norm": 0.028422826475287776,
|
| 1625 |
+
"learning_rate": 4.564221286261709e-05,
|
| 1626 |
+
"loss": 0.0109,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.155,
|
| 1631 |
+
"grad_norm": 0.0002586519968099824,
|
| 1632 |
+
"learning_rate": 4.52077123739888e-05,
|
| 1633 |
+
"loss": 0.0,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.16,
|
| 1638 |
+
"grad_norm": 0.13429703504492657,
|
| 1639 |
+
"learning_rate": 4.477357683661734e-05,
|
| 1640 |
+
"loss": 0.0067,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.165,
|
| 1645 |
+
"grad_norm": 0.0009260106161090144,
|
| 1646 |
+
"learning_rate": 4.433983931160467e-05,
|
| 1647 |
+
"loss": 0.0001,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.17,
|
| 1652 |
+
"grad_norm": 0.2565611405752989,
|
| 1653 |
+
"learning_rate": 4.390653282974264e-05,
|
| 1654 |
+
"loss": 0.0069,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.175,
|
| 1659 |
+
"grad_norm": 0.026210448606146238,
|
| 1660 |
+
"learning_rate": 4.347369038899744e-05,
|
| 1661 |
+
"loss": 0.002,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.18,
|
| 1666 |
+
"grad_norm": 0.0012302259036583447,
|
| 1667 |
+
"learning_rate": 4.3041344951996746e-05,
|
| 1668 |
+
"loss": 0.0002,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.185,
|
| 1673 |
+
"grad_norm": 0.02625494458907875,
|
| 1674 |
+
"learning_rate": 4.260952944351947e-05,
|
| 1675 |
+
"loss": 0.0011,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.19,
|
| 1680 |
+
"grad_norm": 0.000887755762843119,
|
| 1681 |
+
"learning_rate": 4.2178276747988446e-05,
|
| 1682 |
+
"loss": 0.0001,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.195,
|
| 1687 |
+
"grad_norm": 0.003911862624404058,
|
| 1688 |
+
"learning_rate": 4.174761970696612e-05,
|
| 1689 |
+
"loss": 0.0002,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.2,
|
| 1694 |
+
"grad_norm": 0.03517802115885231,
|
| 1695 |
+
"learning_rate": 4.131759111665349e-05,
|
| 1696 |
+
"loss": 0.0019,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.205,
|
| 1701 |
+
"grad_norm": 0.020725215351612047,
|
| 1702 |
+
"learning_rate": 4.088822372539263e-05,
|
| 1703 |
+
"loss": 0.001,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.21,
|
| 1708 |
+
"grad_norm": 0.0002202527290049613,
|
| 1709 |
+
"learning_rate": 4.045955023117276e-05,
|
| 1710 |
+
"loss": 0.0,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.215,
|
| 1715 |
+
"grad_norm": 0.004526177203391082,
|
| 1716 |
+
"learning_rate": 4.003160327914015e-05,
|
| 1717 |
+
"loss": 0.0005,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.22,
|
| 1722 |
+
"grad_norm": 0.018305572530131792,
|
| 1723 |
+
"learning_rate": 3.960441545911204e-05,
|
| 1724 |
+
"loss": 0.0005,
|
| 1725 |
+
"step": 244
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.225,
|
| 1729 |
+
"grad_norm": 0.0007986471410436935,
|
| 1730 |
+
"learning_rate": 3.917801930309486e-05,
|
| 1731 |
+
"loss": 0.0001,
|
| 1732 |
+
"step": 245
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.23,
|
| 1736 |
+
"grad_norm": 0.0259210834529961,
|
| 1737 |
+
"learning_rate": 3.875244728280676e-05,
|
| 1738 |
+
"loss": 0.0015,
|
| 1739 |
+
"step": 246
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.2349999999999999,
|
| 1743 |
+
"grad_norm": 0.007431515309959642,
|
| 1744 |
+
"learning_rate": 3.832773180720475e-05,
|
| 1745 |
+
"loss": 0.0008,
|
| 1746 |
+
"step": 247
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.24,
|
| 1750 |
+
"grad_norm": 0.21370639527155416,
|
| 1751 |
+
"learning_rate": 3.790390522001662e-05,
|
| 1752 |
+
"loss": 0.0296,
|
| 1753 |
+
"step": 248
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.245,
|
| 1757 |
+
"grad_norm": 0.0017749674756192052,
|
| 1758 |
+
"learning_rate": 3.748099979727792e-05,
|
| 1759 |
+
"loss": 0.0002,
|
| 1760 |
+
"step": 249
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.25,
|
| 1764 |
+
"grad_norm": 0.06429161818678288,
|
| 1765 |
+
"learning_rate": 3.705904774487396e-05,
|
| 1766 |
+
"loss": 0.004,
|
| 1767 |
+
"step": 250
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.255,
|
| 1771 |
+
"grad_norm": 0.4999355458993204,
|
| 1772 |
+
"learning_rate": 3.663808119608716e-05,
|
| 1773 |
+
"loss": 0.0118,
|
| 1774 |
+
"step": 251
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.26,
|
| 1778 |
+
"grad_norm": 0.004634517265489021,
|
| 1779 |
+
"learning_rate": 3.6218132209150045e-05,
|
| 1780 |
+
"loss": 0.0002,
|
| 1781 |
+
"step": 252
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.2650000000000001,
|
| 1785 |
+
"grad_norm": 0.10035773018671189,
|
| 1786 |
+
"learning_rate": 3.579923276480387e-05,
|
| 1787 |
+
"loss": 0.0132,
|
| 1788 |
+
"step": 253
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.27,
|
| 1792 |
+
"grad_norm": 0.014262309364614909,
|
| 1793 |
+
"learning_rate": 3.5381414763863166e-05,
|
| 1794 |
+
"loss": 0.0006,
|
| 1795 |
+
"step": 254
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.275,
|
| 1799 |
+
"grad_norm": 0.0006263407491042736,
|
| 1800 |
+
"learning_rate": 3.4964710024786354e-05,
|
| 1801 |
+
"loss": 0.0001,
|
| 1802 |
+
"step": 255
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.28,
|
| 1806 |
+
"grad_norm": 0.0018123347337400675,
|
| 1807 |
+
"learning_rate": 3.4549150281252636e-05,
|
| 1808 |
+
"loss": 0.0003,
|
| 1809 |
+
"step": 256
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.285,
|
| 1813 |
+
"grad_norm": 0.0180314931604898,
|
| 1814 |
+
"learning_rate": 3.4134767179745406e-05,
|
| 1815 |
+
"loss": 0.0056,
|
| 1816 |
+
"step": 257
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.29,
|
| 1820 |
+
"grad_norm": 0.005834199925826024,
|
| 1821 |
+
"learning_rate": 3.372159227714218e-05,
|
| 1822 |
+
"loss": 0.0007,
|
| 1823 |
+
"step": 258
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.295,
|
| 1827 |
+
"grad_norm": 0.008235805677850457,
|
| 1828 |
+
"learning_rate": 3.330965703831146e-05,
|
| 1829 |
+
"loss": 0.0014,
|
| 1830 |
+
"step": 259
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.3,
|
| 1834 |
+
"grad_norm": 0.01238354074430894,
|
| 1835 |
+
"learning_rate": 3.289899283371657e-05,
|
| 1836 |
+
"loss": 0.0027,
|
| 1837 |
+
"step": 260
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.305,
|
| 1841 |
+
"grad_norm": 0.01737694828749958,
|
| 1842 |
+
"learning_rate": 3.248963093702663e-05,
|
| 1843 |
+
"loss": 0.0022,
|
| 1844 |
+
"step": 261
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.31,
|
| 1848 |
+
"grad_norm": 0.0019556398997595315,
|
| 1849 |
+
"learning_rate": 3.2081602522734986e-05,
|
| 1850 |
+
"loss": 0.0002,
|
| 1851 |
+
"step": 262
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.315,
|
| 1855 |
+
"grad_norm": 0.011691972118475544,
|
| 1856 |
+
"learning_rate": 3.167493866378514e-05,
|
| 1857 |
+
"loss": 0.0009,
|
| 1858 |
+
"step": 263
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.32,
|
| 1862 |
+
"grad_norm": 0.004834944193413171,
|
| 1863 |
+
"learning_rate": 3.12696703292044e-05,
|
| 1864 |
+
"loss": 0.0004,
|
| 1865 |
+
"step": 264
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.325,
|
| 1869 |
+
"grad_norm": 0.00037022329873905106,
|
| 1870 |
+
"learning_rate": 3.086582838174551e-05,
|
| 1871 |
+
"loss": 0.0,
|
| 1872 |
+
"step": 265
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.33,
|
| 1876 |
+
"grad_norm": 0.017585669474956853,
|
| 1877 |
+
"learning_rate": 3.046344357553632e-05,
|
| 1878 |
+
"loss": 0.0034,
|
| 1879 |
+
"step": 266
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.335,
|
| 1883 |
+
"grad_norm": 0.0015713497036280754,
|
| 1884 |
+
"learning_rate": 3.006254655373769e-05,
|
| 1885 |
+
"loss": 0.0002,
|
| 1886 |
+
"step": 267
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.34,
|
| 1890 |
+
"grad_norm": 0.006894231638829978,
|
| 1891 |
+
"learning_rate": 2.9663167846209998e-05,
|
| 1892 |
+
"loss": 0.0012,
|
| 1893 |
+
"step": 268
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.345,
|
| 1897 |
+
"grad_norm": 0.0024889803245449374,
|
| 1898 |
+
"learning_rate": 2.926533786718806e-05,
|
| 1899 |
+
"loss": 0.0003,
|
| 1900 |
+
"step": 269
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.35,
|
| 1904 |
+
"grad_norm": 0.196326209710552,
|
| 1905 |
+
"learning_rate": 2.886908691296504e-05,
|
| 1906 |
+
"loss": 0.0108,
|
| 1907 |
+
"step": 270
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.355,
|
| 1911 |
+
"grad_norm": 0.0006519053745859627,
|
| 1912 |
+
"learning_rate": 2.8474445159585235e-05,
|
| 1913 |
+
"loss": 0.0001,
|
| 1914 |
+
"step": 271
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.3599999999999999,
|
| 1918 |
+
"grad_norm": 0.017091849830244706,
|
| 1919 |
+
"learning_rate": 2.8081442660546125e-05,
|
| 1920 |
+
"loss": 0.0017,
|
| 1921 |
+
"step": 272
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.365,
|
| 1925 |
+
"grad_norm": 0.000303651475823324,
|
| 1926 |
+
"learning_rate": 2.7690109344509563e-05,
|
| 1927 |
+
"loss": 0.0001,
|
| 1928 |
+
"step": 273
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.37,
|
| 1932 |
+
"grad_norm": 0.026097473151970325,
|
| 1933 |
+
"learning_rate": 2.7300475013022663e-05,
|
| 1934 |
+
"loss": 0.0078,
|
| 1935 |
+
"step": 274
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.375,
|
| 1939 |
+
"grad_norm": 0.00044755689357070186,
|
| 1940 |
+
"learning_rate": 2.6912569338248315e-05,
|
| 1941 |
+
"loss": 0.0001,
|
| 1942 |
+
"step": 275
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.38,
|
| 1946 |
+
"grad_norm": 0.0005042582378594239,
|
| 1947 |
+
"learning_rate": 2.6526421860705473e-05,
|
| 1948 |
+
"loss": 0.0,
|
| 1949 |
+
"step": 276
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.385,
|
| 1953 |
+
"grad_norm": 0.003238731213585818,
|
| 1954 |
+
"learning_rate": 2.6142061987019577e-05,
|
| 1955 |
+
"loss": 0.0001,
|
| 1956 |
+
"step": 277
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.3900000000000001,
|
| 1960 |
+
"grad_norm": 0.014755686528500189,
|
| 1961 |
+
"learning_rate": 2.575951898768315e-05,
|
| 1962 |
+
"loss": 0.0013,
|
| 1963 |
+
"step": 278
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.395,
|
| 1967 |
+
"grad_norm": 0.00553162806351944,
|
| 1968 |
+
"learning_rate": 2.537882199482665e-05,
|
| 1969 |
+
"loss": 0.0005,
|
| 1970 |
+
"step": 279
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.4,
|
| 1974 |
+
"grad_norm": 0.0005184577628493705,
|
| 1975 |
+
"learning_rate": 2.500000000000001e-05,
|
| 1976 |
+
"loss": 0.0001,
|
| 1977 |
+
"step": 280
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.405,
|
| 1981 |
+
"grad_norm": 0.0037574278569336966,
|
| 1982 |
+
"learning_rate": 2.4623081851964806e-05,
|
| 1983 |
+
"loss": 0.0002,
|
| 1984 |
+
"step": 281
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.41,
|
| 1988 |
+
"grad_norm": 0.0013208423085230652,
|
| 1989 |
+
"learning_rate": 2.4248096254497288e-05,
|
| 1990 |
+
"loss": 0.0001,
|
| 1991 |
+
"step": 282
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.415,
|
| 1995 |
+
"grad_norm": 0.008749035967488314,
|
| 1996 |
+
"learning_rate": 2.3875071764202563e-05,
|
| 1997 |
+
"loss": 0.0006,
|
| 1998 |
+
"step": 283
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.42,
|
| 2002 |
+
"grad_norm": 0.1377433041169159,
|
| 2003 |
+
"learning_rate": 2.350403678833976e-05,
|
| 2004 |
+
"loss": 0.0129,
|
| 2005 |
+
"step": 284
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.425,
|
| 2009 |
+
"grad_norm": 0.0010673749455521134,
|
| 2010 |
+
"learning_rate": 2.3135019582658802e-05,
|
| 2011 |
+
"loss": 0.0001,
|
| 2012 |
+
"step": 285
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.43,
|
| 2016 |
+
"grad_norm": 0.0014952971304103484,
|
| 2017 |
+
"learning_rate": 2.2768048249248648e-05,
|
| 2018 |
+
"loss": 0.0001,
|
| 2019 |
+
"step": 286
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.435,
|
| 2023 |
+
"grad_norm": 0.0037047451799289005,
|
| 2024 |
+
"learning_rate": 2.2403150734397094e-05,
|
| 2025 |
+
"loss": 0.0003,
|
| 2026 |
+
"step": 287
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.44,
|
| 2030 |
+
"grad_norm": 0.0016251406693602976,
|
| 2031 |
+
"learning_rate": 2.2040354826462668e-05,
|
| 2032 |
+
"loss": 0.0002,
|
| 2033 |
+
"step": 288
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.445,
|
| 2037 |
+
"grad_norm": 0.00028939999776533094,
|
| 2038 |
+
"learning_rate": 2.167968815375837e-05,
|
| 2039 |
+
"loss": 0.0,
|
| 2040 |
+
"step": 289
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.45,
|
| 2044 |
+
"grad_norm": 0.001349118937608808,
|
| 2045 |
+
"learning_rate": 2.132117818244771e-05,
|
| 2046 |
+
"loss": 0.0001,
|
| 2047 |
+
"step": 290
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.455,
|
| 2051 |
+
"grad_norm": 0.0067421886978353855,
|
| 2052 |
+
"learning_rate": 2.0964852214453013e-05,
|
| 2053 |
+
"loss": 0.0008,
|
| 2054 |
+
"step": 291
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.46,
|
| 2058 |
+
"grad_norm": 0.0011089997483127705,
|
| 2059 |
+
"learning_rate": 2.061073738537635e-05,
|
| 2060 |
+
"loss": 0.0002,
|
| 2061 |
+
"step": 292
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.465,
|
| 2065 |
+
"grad_norm": 0.0028423391897432755,
|
| 2066 |
+
"learning_rate": 2.0258860662432942e-05,
|
| 2067 |
+
"loss": 0.0002,
|
| 2068 |
+
"step": 293
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.47,
|
| 2072 |
+
"grad_norm": 0.01152426090888302,
|
| 2073 |
+
"learning_rate": 1.9909248842397584e-05,
|
| 2074 |
+
"loss": 0.0005,
|
| 2075 |
+
"step": 294
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.475,
|
| 2079 |
+
"grad_norm": 0.0674872940289676,
|
| 2080 |
+
"learning_rate": 1.9561928549563968e-05,
|
| 2081 |
+
"loss": 0.0089,
|
| 2082 |
+
"step": 295
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.48,
|
| 2086 |
+
"grad_norm": 0.0014300801418590321,
|
| 2087 |
+
"learning_rate": 1.9216926233717085e-05,
|
| 2088 |
+
"loss": 0.0001,
|
| 2089 |
+
"step": 296
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.4849999999999999,
|
| 2093 |
+
"grad_norm": 0.0009000382689770731,
|
| 2094 |
+
"learning_rate": 1.887426816811903e-05,
|
| 2095 |
+
"loss": 0.0001,
|
| 2096 |
+
"step": 297
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.49,
|
| 2100 |
+
"grad_norm": 0.0037992575880660263,
|
| 2101 |
+
"learning_rate": 1.8533980447508137e-05,
|
| 2102 |
+
"loss": 0.0005,
|
| 2103 |
+
"step": 298
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.495,
|
| 2107 |
+
"grad_norm": 0.22933461676514044,
|
| 2108 |
+
"learning_rate": 1.8196088986111797e-05,
|
| 2109 |
+
"loss": 0.0094,
|
| 2110 |
+
"step": 299
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.5,
|
| 2114 |
+
"grad_norm": 0.042524052580419906,
|
| 2115 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 2116 |
+
"loss": 0.0092,
|
| 2117 |
+
"step": 300
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.505,
|
| 2121 |
+
"grad_norm": 0.0023691615908781444,
|
| 2122 |
+
"learning_rate": 1.7527597583490822e-05,
|
| 2123 |
+
"loss": 0.0001,
|
| 2124 |
+
"step": 301
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.51,
|
| 2128 |
+
"grad_norm": 0.019487602890200652,
|
| 2129 |
+
"learning_rate": 1.7197048550474643e-05,
|
| 2130 |
+
"loss": 0.0033,
|
| 2131 |
+
"step": 302
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.5150000000000001,
|
| 2135 |
+
"grad_norm": 0.0016513291656156698,
|
| 2136 |
+
"learning_rate": 1.6868997589213136e-05,
|
| 2137 |
+
"loss": 0.0003,
|
| 2138 |
+
"step": 303
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.52,
|
| 2142 |
+
"grad_norm": 0.0014481350536131741,
|
| 2143 |
+
"learning_rate": 1.6543469682057106e-05,
|
| 2144 |
+
"loss": 0.0002,
|
| 2145 |
+
"step": 304
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.525,
|
| 2149 |
+
"grad_norm": 0.030859003830790453,
|
| 2150 |
+
"learning_rate": 1.622048961921699e-05,
|
| 2151 |
+
"loss": 0.0047,
|
| 2152 |
+
"step": 305
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.53,
|
| 2156 |
+
"grad_norm": 0.0013227578422165872,
|
| 2157 |
+
"learning_rate": 1.5900081996875083e-05,
|
| 2158 |
+
"loss": 0.0002,
|
| 2159 |
+
"step": 306
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.5350000000000001,
|
| 2163 |
+
"grad_norm": 0.040650034143242364,
|
| 2164 |
+
"learning_rate": 1.5582271215312294e-05,
|
| 2165 |
+
"loss": 0.0058,
|
| 2166 |
+
"step": 307
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.54,
|
| 2170 |
+
"grad_norm": 0.010752068089793936,
|
| 2171 |
+
"learning_rate": 1.526708147705013e-05,
|
| 2172 |
+
"loss": 0.0004,
|
| 2173 |
+
"step": 308
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.545,
|
| 2177 |
+
"grad_norm": 0.0007336372616353323,
|
| 2178 |
+
"learning_rate": 1.4954536785007456e-05,
|
| 2179 |
+
"loss": 0.0001,
|
| 2180 |
+
"step": 309
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.55,
|
| 2184 |
+
"grad_norm": 0.0006185076286575476,
|
| 2185 |
+
"learning_rate": 1.4644660940672627e-05,
|
| 2186 |
+
"loss": 0.0001,
|
| 2187 |
+
"step": 310
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.5550000000000002,
|
| 2191 |
+
"grad_norm": 0.0012108807503260438,
|
| 2192 |
+
"learning_rate": 1.4337477542290928e-05,
|
| 2193 |
+
"loss": 0.0001,
|
| 2194 |
+
"step": 311
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.56,
|
| 2198 |
+
"grad_norm": 0.014653087670058953,
|
| 2199 |
+
"learning_rate": 1.4033009983067452e-05,
|
| 2200 |
+
"loss": 0.0038,
|
| 2201 |
+
"step": 312
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.565,
|
| 2205 |
+
"grad_norm": 0.00070106922262732,
|
| 2206 |
+
"learning_rate": 1.373128144938563e-05,
|
| 2207 |
+
"loss": 0.0001,
|
| 2208 |
+
"step": 313
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.5699999999999998,
|
| 2212 |
+
"grad_norm": 0.022894525200356754,
|
| 2213 |
+
"learning_rate": 1.3432314919041478e-05,
|
| 2214 |
+
"loss": 0.0034,
|
| 2215 |
+
"step": 314
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.575,
|
| 2219 |
+
"grad_norm": 0.15319546037835274,
|
| 2220 |
+
"learning_rate": 1.3136133159493802e-05,
|
| 2221 |
+
"loss": 0.005,
|
| 2222 |
+
"step": 315
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.58,
|
| 2226 |
+
"grad_norm": 0.02298142323498265,
|
| 2227 |
+
"learning_rate": 1.2842758726130283e-05,
|
| 2228 |
+
"loss": 0.0036,
|
| 2229 |
+
"step": 316
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.585,
|
| 2233 |
+
"grad_norm": 0.000645097633470432,
|
| 2234 |
+
"learning_rate": 1.2552213960549891e-05,
|
| 2235 |
+
"loss": 0.0001,
|
| 2236 |
+
"step": 317
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.5899999999999999,
|
| 2240 |
+
"grad_norm": 0.08053780392470371,
|
| 2241 |
+
"learning_rate": 1.22645209888614e-05,
|
| 2242 |
+
"loss": 0.0249,
|
| 2243 |
+
"step": 318
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.595,
|
| 2247 |
+
"grad_norm": 0.00023214176871761666,
|
| 2248 |
+
"learning_rate": 1.1979701719998453e-05,
|
| 2249 |
+
"loss": 0.0,
|
| 2250 |
+
"step": 319
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.6,
|
| 2254 |
+
"grad_norm": 0.0008201475435205818,
|
| 2255 |
+
"learning_rate": 1.1697777844051105e-05,
|
| 2256 |
+
"loss": 0.0001,
|
| 2257 |
+
"step": 320
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.605,
|
| 2261 |
+
"grad_norm": 0.00476872158921415,
|
| 2262 |
+
"learning_rate": 1.1418770830614013e-05,
|
| 2263 |
+
"loss": 0.0002,
|
| 2264 |
+
"step": 321
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.6099999999999999,
|
| 2268 |
+
"grad_norm": 0.001393468029081844,
|
| 2269 |
+
"learning_rate": 1.1142701927151456e-05,
|
| 2270 |
+
"loss": 0.0001,
|
| 2271 |
+
"step": 322
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.615,
|
| 2275 |
+
"grad_norm": 0.003859464277860662,
|
| 2276 |
+
"learning_rate": 1.0869592157379304e-05,
|
| 2277 |
+
"loss": 0.0001,
|
| 2278 |
+
"step": 323
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.62,
|
| 2282 |
+
"grad_norm": 0.004353663579813155,
|
| 2283 |
+
"learning_rate": 1.0599462319663905e-05,
|
| 2284 |
+
"loss": 0.0003,
|
| 2285 |
+
"step": 324
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.625,
|
| 2289 |
+
"grad_norm": 0.013200524828428624,
|
| 2290 |
+
"learning_rate": 1.0332332985438248e-05,
|
| 2291 |
+
"loss": 0.0018,
|
| 2292 |
+
"step": 325
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.63,
|
| 2296 |
+
"grad_norm": 0.0024888356543417843,
|
| 2297 |
+
"learning_rate": 1.006822449763537e-05,
|
| 2298 |
+
"loss": 0.0001,
|
| 2299 |
+
"step": 326
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.635,
|
| 2303 |
+
"grad_norm": 0.02372982952309258,
|
| 2304 |
+
"learning_rate": 9.807156969139136e-06,
|
| 2305 |
+
"loss": 0.0031,
|
| 2306 |
+
"step": 327
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.6400000000000001,
|
| 2310 |
+
"grad_norm": 0.0227337977382863,
|
| 2311 |
+
"learning_rate": 9.549150281252633e-06,
|
| 2312 |
+
"loss": 0.0032,
|
| 2313 |
+
"step": 328
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.645,
|
| 2317 |
+
"grad_norm": 0.0006468911312178352,
|
| 2318 |
+
"learning_rate": 9.294224082184045e-06,
|
| 2319 |
+
"loss": 0.0001,
|
| 2320 |
+
"step": 329
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.65,
|
| 2324 |
+
"grad_norm": 0.00031790071448860156,
|
| 2325 |
+
"learning_rate": 9.042397785550405e-06,
|
| 2326 |
+
"loss": 0.0,
|
| 2327 |
+
"step": 330
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.655,
|
| 2331 |
+
"grad_norm": 0.014069307621453687,
|
| 2332 |
+
"learning_rate": 8.793690568899216e-06,
|
| 2333 |
+
"loss": 0.001,
|
| 2334 |
+
"step": 331
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.6600000000000001,
|
| 2338 |
+
"grad_norm": 0.002913520292376533,
|
| 2339 |
+
"learning_rate": 8.548121372247918e-06,
|
| 2340 |
+
"loss": 0.0001,
|
| 2341 |
+
"step": 332
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.665,
|
| 2345 |
+
"grad_norm": 0.010269178065459858,
|
| 2346 |
+
"learning_rate": 8.305708896641594e-06,
|
| 2347 |
+
"loss": 0.0008,
|
| 2348 |
+
"step": 333
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.67,
|
| 2352 |
+
"grad_norm": 0.004202310016346467,
|
| 2353 |
+
"learning_rate": 8.066471602728803e-06,
|
| 2354 |
+
"loss": 0.0002,
|
| 2355 |
+
"step": 334
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.675,
|
| 2359 |
+
"grad_norm": 0.06791650915734657,
|
| 2360 |
+
"learning_rate": 7.830427709355725e-06,
|
| 2361 |
+
"loss": 0.0089,
|
| 2362 |
+
"step": 335
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.6800000000000002,
|
| 2366 |
+
"grad_norm": 0.00023193543187296342,
|
| 2367 |
+
"learning_rate": 7.597595192178702e-06,
|
| 2368 |
+
"loss": 0.0,
|
| 2369 |
+
"step": 336
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.685,
|
| 2373 |
+
"grad_norm": 0.009522222092425871,
|
| 2374 |
+
"learning_rate": 7.367991782295391e-06,
|
| 2375 |
+
"loss": 0.0005,
|
| 2376 |
+
"step": 337
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.69,
|
| 2380 |
+
"grad_norm": 0.0445881491075994,
|
| 2381 |
+
"learning_rate": 7.1416349648943894e-06,
|
| 2382 |
+
"loss": 0.0016,
|
| 2383 |
+
"step": 338
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.6949999999999998,
|
| 2387 |
+
"grad_norm": 0.000599146744339273,
|
| 2388 |
+
"learning_rate": 6.918541977923709e-06,
|
| 2389 |
+
"loss": 0.0001,
|
| 2390 |
+
"step": 339
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.7,
|
| 2394 |
+
"grad_norm": 0.027249503153038576,
|
| 2395 |
+
"learning_rate": 6.698729810778065e-06,
|
| 2396 |
+
"loss": 0.0021,
|
| 2397 |
+
"step": 340
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.705,
|
| 2401 |
+
"grad_norm": 0.002016792070660535,
|
| 2402 |
+
"learning_rate": 6.482215203005015e-06,
|
| 2403 |
+
"loss": 0.0002,
|
| 2404 |
+
"step": 341
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.71,
|
| 2408 |
+
"grad_norm": 0.007484120928358062,
|
| 2409 |
+
"learning_rate": 6.269014643030213e-06,
|
| 2410 |
+
"loss": 0.0003,
|
| 2411 |
+
"step": 342
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.7149999999999999,
|
| 2415 |
+
"grad_norm": 0.0005063050902788136,
|
| 2416 |
+
"learning_rate": 6.059144366901736e-06,
|
| 2417 |
+
"loss": 0.0,
|
| 2418 |
+
"step": 343
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.72,
|
| 2422 |
+
"grad_norm": 0.0009197549253660609,
|
| 2423 |
+
"learning_rate": 5.852620357053651e-06,
|
| 2424 |
+
"loss": 0.0001,
|
| 2425 |
+
"step": 344
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.725,
|
| 2429 |
+
"grad_norm": 0.003191114724532278,
|
| 2430 |
+
"learning_rate": 5.649458341088915e-06,
|
| 2431 |
+
"loss": 0.0005,
|
| 2432 |
+
"step": 345
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.73,
|
| 2436 |
+
"grad_norm": 0.0014447492382083482,
|
| 2437 |
+
"learning_rate": 5.449673790581611e-06,
|
| 2438 |
+
"loss": 0.0002,
|
| 2439 |
+
"step": 346
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.7349999999999999,
|
| 2443 |
+
"grad_norm": 0.004059252437141361,
|
| 2444 |
+
"learning_rate": 5.2532819198987506e-06,
|
| 2445 |
+
"loss": 0.0006,
|
| 2446 |
+
"step": 347
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.74,
|
| 2450 |
+
"grad_norm": 0.02120955388832761,
|
| 2451 |
+
"learning_rate": 5.060297685041659e-06,
|
| 2452 |
+
"loss": 0.0007,
|
| 2453 |
+
"step": 348
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.745,
|
| 2457 |
+
"grad_norm": 0.0011651360195809543,
|
| 2458 |
+
"learning_rate": 4.87073578250698e-06,
|
| 2459 |
+
"loss": 0.0001,
|
| 2460 |
+
"step": 349
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.75,
|
| 2464 |
+
"grad_norm": 0.0010159633731216518,
|
| 2465 |
+
"learning_rate": 4.684610648167503e-06,
|
| 2466 |
+
"loss": 0.0001,
|
| 2467 |
+
"step": 350
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.755,
|
| 2471 |
+
"grad_norm": 0.04624949210450388,
|
| 2472 |
+
"learning_rate": 4.501936456172845e-06,
|
| 2473 |
+
"loss": 0.0023,
|
| 2474 |
+
"step": 351
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.76,
|
| 2478 |
+
"grad_norm": 0.013865761867621462,
|
| 2479 |
+
"learning_rate": 4.322727117869951e-06,
|
| 2480 |
+
"loss": 0.0008,
|
| 2481 |
+
"step": 352
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.7650000000000001,
|
| 2485 |
+
"grad_norm": 0.0006593659250274612,
|
| 2486 |
+
"learning_rate": 4.146996280743798e-06,
|
| 2487 |
+
"loss": 0.0001,
|
| 2488 |
+
"step": 353
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.77,
|
| 2492 |
+
"grad_norm": 0.0006699609878223177,
|
| 2493 |
+
"learning_rate": 3.974757327377981e-06,
|
| 2494 |
+
"loss": 0.0001,
|
| 2495 |
+
"step": 354
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.775,
|
| 2499 |
+
"grad_norm": 0.010073962298842973,
|
| 2500 |
+
"learning_rate": 3.8060233744356633e-06,
|
| 2501 |
+
"loss": 0.0014,
|
| 2502 |
+
"step": 355
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.78,
|
| 2506 |
+
"grad_norm": 0.00046017499160588973,
|
| 2507 |
+
"learning_rate": 3.6408072716606346e-06,
|
| 2508 |
+
"loss": 0.0001,
|
| 2509 |
+
"step": 356
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.7850000000000001,
|
| 2513 |
+
"grad_norm": 0.001412185181637751,
|
| 2514 |
+
"learning_rate": 3.479121600898777e-06,
|
| 2515 |
+
"loss": 0.0001,
|
| 2516 |
+
"step": 357
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.79,
|
| 2520 |
+
"grad_norm": 0.0010948577791496526,
|
| 2521 |
+
"learning_rate": 3.3209786751399187e-06,
|
| 2522 |
+
"loss": 0.0001,
|
| 2523 |
+
"step": 358
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.795,
|
| 2527 |
+
"grad_norm": 0.0027495650503305998,
|
| 2528 |
+
"learning_rate": 3.1663905375801216e-06,
|
| 2529 |
+
"loss": 0.0003,
|
| 2530 |
+
"step": 359
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 1.8,
|
| 2534 |
+
"grad_norm": 0.00497037863878421,
|
| 2535 |
+
"learning_rate": 3.0153689607045845e-06,
|
| 2536 |
+
"loss": 0.0004,
|
| 2537 |
+
"step": 360
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 1.8050000000000002,
|
| 2541 |
+
"grad_norm": 0.04826775060358748,
|
| 2542 |
+
"learning_rate": 2.8679254453910785e-06,
|
| 2543 |
+
"loss": 0.0052,
|
| 2544 |
+
"step": 361
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 1.81,
|
| 2548 |
+
"grad_norm": 0.0002190733399899643,
|
| 2549 |
+
"learning_rate": 2.724071220034158e-06,
|
| 2550 |
+
"loss": 0.0,
|
| 2551 |
+
"step": 362
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 1.815,
|
| 2555 |
+
"grad_norm": 0.03983428885724064,
|
| 2556 |
+
"learning_rate": 2.583817239690034e-06,
|
| 2557 |
+
"loss": 0.0108,
|
| 2558 |
+
"step": 363
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 1.8199999999999998,
|
| 2562 |
+
"grad_norm": 0.005353636861069711,
|
| 2563 |
+
"learning_rate": 2.4471741852423237e-06,
|
| 2564 |
+
"loss": 0.0004,
|
| 2565 |
+
"step": 364
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 1.825,
|
| 2569 |
+
"grad_norm": 0.007365720628728885,
|
| 2570 |
+
"learning_rate": 2.314152462588659e-06,
|
| 2571 |
+
"loss": 0.0009,
|
| 2572 |
+
"step": 365
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 1.83,
|
| 2576 |
+
"grad_norm": 0.0020943736140563596,
|
| 2577 |
+
"learning_rate": 2.1847622018482283e-06,
|
| 2578 |
+
"loss": 0.0002,
|
| 2579 |
+
"step": 366
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 1.835,
|
| 2583 |
+
"grad_norm": 0.049181169958885725,
|
| 2584 |
+
"learning_rate": 2.0590132565903476e-06,
|
| 2585 |
+
"loss": 0.0066,
|
| 2586 |
+
"step": 367
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 1.8399999999999999,
|
| 2590 |
+
"grad_norm": 0.008505272014854959,
|
| 2591 |
+
"learning_rate": 1.9369152030840556e-06,
|
| 2592 |
+
"loss": 0.001,
|
| 2593 |
+
"step": 368
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 1.845,
|
| 2597 |
+
"grad_norm": 0.015026960520249318,
|
| 2598 |
+
"learning_rate": 1.8184773395688526e-06,
|
| 2599 |
+
"loss": 0.0027,
|
| 2600 |
+
"step": 369
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 1.85,
|
| 2604 |
+
"grad_norm": 0.0012138520031968582,
|
| 2605 |
+
"learning_rate": 1.70370868554659e-06,
|
| 2606 |
+
"loss": 0.0002,
|
| 2607 |
+
"step": 370
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 1.855,
|
| 2611 |
+
"grad_norm": 0.00024125978466275796,
|
| 2612 |
+
"learning_rate": 1.5926179810946184e-06,
|
| 2613 |
+
"loss": 0.0001,
|
| 2614 |
+
"step": 371
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 1.8599999999999999,
|
| 2618 |
+
"grad_norm": 0.07024226815561596,
|
| 2619 |
+
"learning_rate": 1.4852136862001764e-06,
|
| 2620 |
+
"loss": 0.004,
|
| 2621 |
+
"step": 372
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 1.865,
|
| 2625 |
+
"grad_norm": 0.0021247803242464055,
|
| 2626 |
+
"learning_rate": 1.3815039801161721e-06,
|
| 2627 |
+
"loss": 0.0002,
|
| 2628 |
+
"step": 373
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.87,
|
| 2632 |
+
"grad_norm": 0.04502769124604199,
|
| 2633 |
+
"learning_rate": 1.2814967607382432e-06,
|
| 2634 |
+
"loss": 0.0122,
|
| 2635 |
+
"step": 374
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 1.875,
|
| 2639 |
+
"grad_norm": 0.007452172380573964,
|
| 2640 |
+
"learning_rate": 1.1851996440033319e-06,
|
| 2641 |
+
"loss": 0.001,
|
| 2642 |
+
"step": 375
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 1.88,
|
| 2646 |
+
"grad_norm": 0.0015481238834281481,
|
| 2647 |
+
"learning_rate": 1.0926199633097157e-06,
|
| 2648 |
+
"loss": 0.0002,
|
| 2649 |
+
"step": 376
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 1.885,
|
| 2653 |
+
"grad_norm": 0.005530884948723974,
|
| 2654 |
+
"learning_rate": 1.0037647689585206e-06,
|
| 2655 |
+
"loss": 0.0009,
|
| 2656 |
+
"step": 377
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 1.8900000000000001,
|
| 2660 |
+
"grad_norm": 0.0067523371656823305,
|
| 2661 |
+
"learning_rate": 9.186408276168013e-07,
|
| 2662 |
+
"loss": 0.0005,
|
| 2663 |
+
"step": 378
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 1.895,
|
| 2667 |
+
"grad_norm": 0.003506175781494403,
|
| 2668 |
+
"learning_rate": 8.372546218022747e-07,
|
| 2669 |
+
"loss": 0.0002,
|
| 2670 |
+
"step": 379
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 1.9,
|
| 2674 |
+
"grad_norm": 0.0019742849650606926,
|
| 2675 |
+
"learning_rate": 7.596123493895991e-07,
|
| 2676 |
+
"loss": 0.0003,
|
| 2677 |
+
"step": 380
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 1.905,
|
| 2681 |
+
"grad_norm": 0.0009738360342464184,
|
| 2682 |
+
"learning_rate": 6.857199231384282e-07,
|
| 2683 |
+
"loss": 0.0001,
|
| 2684 |
+
"step": 381
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.9100000000000001,
|
| 2688 |
+
"grad_norm": 0.0004408482841272174,
|
| 2689 |
+
"learning_rate": 6.15582970243117e-07,
|
| 2690 |
+
"loss": 0.0,
|
| 2691 |
+
"step": 382
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 1.915,
|
| 2695 |
+
"grad_norm": 0.010244955433424012,
|
| 2696 |
+
"learning_rate": 5.492068319041588e-07,
|
| 2697 |
+
"loss": 0.0013,
|
| 2698 |
+
"step": 383
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 1.92,
|
| 2702 |
+
"grad_norm": 0.0521719111078745,
|
| 2703 |
+
"learning_rate": 4.865965629214819e-07,
|
| 2704 |
+
"loss": 0.0071,
|
| 2705 |
+
"step": 384
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 1.925,
|
| 2709 |
+
"grad_norm": 0.0012864130857160657,
|
| 2710 |
+
"learning_rate": 4.277569313094809e-07,
|
| 2711 |
+
"loss": 0.0002,
|
| 2712 |
+
"step": 385
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 1.9300000000000002,
|
| 2716 |
+
"grad_norm": 0.2423177287343396,
|
| 2717 |
+
"learning_rate": 3.7269241793390085e-07,
|
| 2718 |
+
"loss": 0.028,
|
| 2719 |
+
"step": 386
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 1.935,
|
| 2723 |
+
"grad_norm": 0.010456316918486782,
|
| 2724 |
+
"learning_rate": 3.214072161706272e-07,
|
| 2725 |
+
"loss": 0.0007,
|
| 2726 |
+
"step": 387
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 1.94,
|
| 2730 |
+
"grad_norm": 0.0005706094269043214,
|
| 2731 |
+
"learning_rate": 2.7390523158633554e-07,
|
| 2732 |
+
"loss": 0.0001,
|
| 2733 |
+
"step": 388
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 1.9449999999999998,
|
| 2737 |
+
"grad_norm": 0.013608987991299646,
|
| 2738 |
+
"learning_rate": 2.3019008164105738e-07,
|
| 2739 |
+
"loss": 0.0026,
|
| 2740 |
+
"step": 389
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 1.95,
|
| 2744 |
+
"grad_norm": 0.01259450581057638,
|
| 2745 |
+
"learning_rate": 1.9026509541272275e-07,
|
| 2746 |
+
"loss": 0.0006,
|
| 2747 |
+
"step": 390
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 1.955,
|
| 2751 |
+
"grad_norm": 0.006309175090935663,
|
| 2752 |
+
"learning_rate": 1.5413331334360182e-07,
|
| 2753 |
+
"loss": 0.0006,
|
| 2754 |
+
"step": 391
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 1.96,
|
| 2758 |
+
"grad_norm": 0.00027718336636984805,
|
| 2759 |
+
"learning_rate": 1.2179748700879012e-07,
|
| 2760 |
+
"loss": 0.0001,
|
| 2761 |
+
"step": 392
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 1.9649999999999999,
|
| 2765 |
+
"grad_norm": 0.01973445145502157,
|
| 2766 |
+
"learning_rate": 9.3260078906654e-08,
|
| 2767 |
+
"loss": 0.0058,
|
| 2768 |
+
"step": 393
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 1.97,
|
| 2772 |
+
"grad_norm": 0.0050223114988680664,
|
| 2773 |
+
"learning_rate": 6.852326227130834e-08,
|
| 2774 |
+
"loss": 0.0007,
|
| 2775 |
+
"step": 394
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 1.975,
|
| 2779 |
+
"grad_norm": 0.041307491240042314,
|
| 2780 |
+
"learning_rate": 4.7588920907110094e-08,
|
| 2781 |
+
"loss": 0.004,
|
| 2782 |
+
"step": 395
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 1.98,
|
| 2786 |
+
"grad_norm": 0.0019487309746442478,
|
| 2787 |
+
"learning_rate": 3.04586490452119e-08,
|
| 2788 |
+
"loss": 0.0001,
|
| 2789 |
+
"step": 396
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 1.9849999999999999,
|
| 2793 |
+
"grad_norm": 0.00051503727945184,
|
| 2794 |
+
"learning_rate": 1.7133751222137007e-08,
|
| 2795 |
+
"loss": 0.0001,
|
| 2796 |
+
"step": 397
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 1.99,
|
| 2800 |
+
"grad_norm": 0.02902500982586163,
|
| 2801 |
+
"learning_rate": 7.615242180436522e-09,
|
| 2802 |
+
"loss": 0.003,
|
| 2803 |
+
"step": 398
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 1.995,
|
| 2807 |
+
"grad_norm": 0.02095204026685112,
|
| 2808 |
+
"learning_rate": 1.903846791434516e-09,
|
| 2809 |
+
"loss": 0.0024,
|
| 2810 |
+
"step": 399
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 2.0,
|
| 2814 |
+
"grad_norm": 0.006401152408197018,
|
| 2815 |
+
"learning_rate": 0.0,
|
| 2816 |
+
"loss": 0.0004,
|
| 2817 |
+
"step": 400
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 2.0,
|
| 2821 |
+
"eval_loss": 0.0046582599170506,
|
| 2822 |
+
"eval_runtime": 1.5245,
|
| 2823 |
+
"eval_samples_per_second": 0.656,
|
| 2824 |
+
"eval_steps_per_second": 0.656,
|
| 2825 |
+
"step": 400
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 2.0,
|
| 2829 |
+
"step": 400,
|
| 2830 |
+
"total_flos": 6612300644352.0,
|
| 2831 |
+
"train_loss": 0.033545353031595365,
|
| 2832 |
+
"train_runtime": 998.7883,
|
| 2833 |
+
"train_samples_per_second": 0.801,
|
| 2834 |
+
"train_steps_per_second": 0.4
|
| 2835 |
+
}
|
| 2836 |
+
],
|
| 2837 |
+
"logging_steps": 1,
|
| 2838 |
+
"max_steps": 400,
|
| 2839 |
+
"num_input_tokens_seen": 0,
|
| 2840 |
+
"num_train_epochs": 2,
|
| 2841 |
+
"save_steps": 500,
|
| 2842 |
+
"stateful_callbacks": {
|
| 2843 |
+
"TrainerControl": {
|
| 2844 |
+
"args": {
|
| 2845 |
+
"should_epoch_stop": false,
|
| 2846 |
+
"should_evaluate": false,
|
| 2847 |
+
"should_log": false,
|
| 2848 |
+
"should_save": true,
|
| 2849 |
+
"should_training_stop": true
|
| 2850 |
+
},
|
| 2851 |
+
"attributes": {}
|
| 2852 |
+
}
|
| 2853 |
+
},
|
| 2854 |
+
"total_flos": 6612300644352.0,
|
| 2855 |
+
"train_batch_size": 1,
|
| 2856 |
+
"trial_name": null,
|
| 2857 |
+
"trial_params": null
|
| 2858 |
+
}
|