Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +8 -0
- train_results.json +9 -0
- trainer_state.json +2759 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-2-9b
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Gemma-2-9B-sft-ultrachat
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Gemma-2-9B-sft-ultrachat
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b).
|
| 15 |
+
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 16 |
+
|
| 17 |
+
## Quick start
|
| 18 |
+
|
| 19 |
+
```python
|
| 20 |
+
from transformers import pipeline
|
| 21 |
+
|
| 22 |
+
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
| 23 |
+
generator = pipeline("text-generation", model="AmberYifan/Gemma-2-9B-sft-ultrachat", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.2
|
| 37 |
+
- Transformers: 4.46.3
|
| 38 |
+
- Pytorch: 2.5.1+cu118
|
| 39 |
+
- Datasets: 3.2.0
|
| 40 |
+
- Tokenizers: 0.20.3
|
| 41 |
+
|
| 42 |
+
## Citations
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
Cite TRL as:
|
| 47 |
+
|
| 48 |
+
```bibtex
|
| 49 |
+
@misc{vonwerra2022trl,
|
| 50 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 51 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 52 |
+
year = 2020,
|
| 53 |
+
journal = {GitHub repository},
|
| 54 |
+
publisher = {GitHub},
|
| 55 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 56 |
+
}
|
| 57 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 230071600152576.0,
|
| 4 |
+
"train_loss": 1.2287210629593512,
|
| 5 |
+
"train_runtime": 18714.7709,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 3.303,
|
| 8 |
+
"train_steps_per_second": 0.103
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": 1,
|
| 6 |
+
"pad_token_id": 0,
|
| 7 |
+
"transformers_version": "4.46.3"
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 230071600152576.0,
|
| 4 |
+
"train_loss": 1.2287210629593512,
|
| 5 |
+
"train_runtime": 18714.7709,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 3.303,
|
| 8 |
+
"train_steps_per_second": 0.103
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2759 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1932,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005175983436853002,
|
| 13 |
+
"grad_norm": 118.52372570307482,
|
| 14 |
+
"learning_rate": 1.0309278350515465e-07,
|
| 15 |
+
"loss": 3.0995,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002587991718426501,
|
| 20 |
+
"grad_norm": 106.06460811089278,
|
| 21 |
+
"learning_rate": 5.154639175257732e-07,
|
| 22 |
+
"loss": 3.2515,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005175983436853002,
|
| 27 |
+
"grad_norm": 45.432580654463486,
|
| 28 |
+
"learning_rate": 1.0309278350515464e-06,
|
| 29 |
+
"loss": 2.5474,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007763975155279503,
|
| 34 |
+
"grad_norm": 5.707645113932802,
|
| 35 |
+
"learning_rate": 1.5463917525773197e-06,
|
| 36 |
+
"loss": 1.6318,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.010351966873706004,
|
| 41 |
+
"grad_norm": 4.4628083580213325,
|
| 42 |
+
"learning_rate": 2.061855670103093e-06,
|
| 43 |
+
"loss": 1.3823,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.012939958592132506,
|
| 48 |
+
"grad_norm": 3.761000828527356,
|
| 49 |
+
"learning_rate": 2.577319587628866e-06,
|
| 50 |
+
"loss": 1.3176,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.015527950310559006,
|
| 55 |
+
"grad_norm": 3.0497152788250235,
|
| 56 |
+
"learning_rate": 3.0927835051546395e-06,
|
| 57 |
+
"loss": 1.1812,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018115942028985508,
|
| 62 |
+
"grad_norm": 2.696900373411112,
|
| 63 |
+
"learning_rate": 3.6082474226804126e-06,
|
| 64 |
+
"loss": 1.274,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.020703933747412008,
|
| 69 |
+
"grad_norm": 2.363119260182366,
|
| 70 |
+
"learning_rate": 4.123711340206186e-06,
|
| 71 |
+
"loss": 1.2032,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023291925465838508,
|
| 76 |
+
"grad_norm": 2.549100089774392,
|
| 77 |
+
"learning_rate": 4.639175257731959e-06,
|
| 78 |
+
"loss": 1.2152,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.025879917184265012,
|
| 83 |
+
"grad_norm": 2.8330026439235567,
|
| 84 |
+
"learning_rate": 5.154639175257732e-06,
|
| 85 |
+
"loss": 1.1962,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.028467908902691512,
|
| 90 |
+
"grad_norm": 2.4165498298939156,
|
| 91 |
+
"learning_rate": 5.670103092783505e-06,
|
| 92 |
+
"loss": 1.211,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.031055900621118012,
|
| 97 |
+
"grad_norm": 2.409145711425535,
|
| 98 |
+
"learning_rate": 6.185567010309279e-06,
|
| 99 |
+
"loss": 1.2127,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03364389233954451,
|
| 104 |
+
"grad_norm": 2.5761976363169863,
|
| 105 |
+
"learning_rate": 6.701030927835052e-06,
|
| 106 |
+
"loss": 1.2057,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.036231884057971016,
|
| 111 |
+
"grad_norm": 2.476050413527567,
|
| 112 |
+
"learning_rate": 7.216494845360825e-06,
|
| 113 |
+
"loss": 1.2286,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03881987577639751,
|
| 118 |
+
"grad_norm": 2.467699606641844,
|
| 119 |
+
"learning_rate": 7.731958762886599e-06,
|
| 120 |
+
"loss": 1.2134,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.041407867494824016,
|
| 125 |
+
"grad_norm": 2.2939966741651387,
|
| 126 |
+
"learning_rate": 8.247422680412371e-06,
|
| 127 |
+
"loss": 1.1893,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04399585921325052,
|
| 132 |
+
"grad_norm": 2.762524555569291,
|
| 133 |
+
"learning_rate": 8.762886597938146e-06,
|
| 134 |
+
"loss": 1.2244,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.046583850931677016,
|
| 139 |
+
"grad_norm": 2.1801186709719245,
|
| 140 |
+
"learning_rate": 9.278350515463918e-06,
|
| 141 |
+
"loss": 1.1464,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.04917184265010352,
|
| 146 |
+
"grad_norm": 2.1622793216147542,
|
| 147 |
+
"learning_rate": 9.793814432989691e-06,
|
| 148 |
+
"loss": 1.1941,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.051759834368530024,
|
| 153 |
+
"grad_norm": 2.327680697741413,
|
| 154 |
+
"learning_rate": 1.0309278350515464e-05,
|
| 155 |
+
"loss": 1.2271,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05434782608695652,
|
| 160 |
+
"grad_norm": 2.417174235586904,
|
| 161 |
+
"learning_rate": 1.0824742268041238e-05,
|
| 162 |
+
"loss": 1.2522,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.056935817805383024,
|
| 167 |
+
"grad_norm": 2.1527719549162843,
|
| 168 |
+
"learning_rate": 1.134020618556701e-05,
|
| 169 |
+
"loss": 1.2158,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05952380952380952,
|
| 174 |
+
"grad_norm": 2.089884090927776,
|
| 175 |
+
"learning_rate": 1.1855670103092785e-05,
|
| 176 |
+
"loss": 1.1686,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.062111801242236024,
|
| 181 |
+
"grad_norm": 2.087619755097997,
|
| 182 |
+
"learning_rate": 1.2371134020618558e-05,
|
| 183 |
+
"loss": 1.1924,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06469979296066253,
|
| 188 |
+
"grad_norm": 2.159539140677949,
|
| 189 |
+
"learning_rate": 1.2886597938144332e-05,
|
| 190 |
+
"loss": 1.2169,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06728778467908902,
|
| 195 |
+
"grad_norm": 2.7014157342913996,
|
| 196 |
+
"learning_rate": 1.3402061855670103e-05,
|
| 197 |
+
"loss": 1.2805,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06987577639751552,
|
| 202 |
+
"grad_norm": 2.0065957039857434,
|
| 203 |
+
"learning_rate": 1.3917525773195878e-05,
|
| 204 |
+
"loss": 1.1845,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07246376811594203,
|
| 209 |
+
"grad_norm": 2.6446222341165266,
|
| 210 |
+
"learning_rate": 1.443298969072165e-05,
|
| 211 |
+
"loss": 1.2289,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07505175983436853,
|
| 216 |
+
"grad_norm": 2.115637544048349,
|
| 217 |
+
"learning_rate": 1.4948453608247425e-05,
|
| 218 |
+
"loss": 1.2695,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07763975155279502,
|
| 223 |
+
"grad_norm": 2.0387064738115095,
|
| 224 |
+
"learning_rate": 1.5463917525773197e-05,
|
| 225 |
+
"loss": 1.2664,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08022774327122154,
|
| 230 |
+
"grad_norm": 2.074102510035905,
|
| 231 |
+
"learning_rate": 1.597938144329897e-05,
|
| 232 |
+
"loss": 1.1976,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08281573498964803,
|
| 237 |
+
"grad_norm": 2.871459263076559,
|
| 238 |
+
"learning_rate": 1.6494845360824743e-05,
|
| 239 |
+
"loss": 1.2579,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08540372670807453,
|
| 244 |
+
"grad_norm": 1.9920241256865008,
|
| 245 |
+
"learning_rate": 1.7010309278350517e-05,
|
| 246 |
+
"loss": 1.3046,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08799171842650104,
|
| 251 |
+
"grad_norm": 2.0573548308344707,
|
| 252 |
+
"learning_rate": 1.752577319587629e-05,
|
| 253 |
+
"loss": 1.2717,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09057971014492754,
|
| 258 |
+
"grad_norm": 1.8934736779126353,
|
| 259 |
+
"learning_rate": 1.8041237113402062e-05,
|
| 260 |
+
"loss": 1.2516,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09316770186335403,
|
| 265 |
+
"grad_norm": 1.9513462230819247,
|
| 266 |
+
"learning_rate": 1.8556701030927837e-05,
|
| 267 |
+
"loss": 1.275,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09575569358178054,
|
| 272 |
+
"grad_norm": 1.9681874455253996,
|
| 273 |
+
"learning_rate": 1.907216494845361e-05,
|
| 274 |
+
"loss": 1.2622,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.09834368530020704,
|
| 279 |
+
"grad_norm": 2.168981469623358,
|
| 280 |
+
"learning_rate": 1.9587628865979382e-05,
|
| 281 |
+
"loss": 1.2687,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10093167701863354,
|
| 286 |
+
"grad_norm": 1.9357981604266234,
|
| 287 |
+
"learning_rate": 1.999998366308358e-05,
|
| 288 |
+
"loss": 1.2902,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.10351966873706005,
|
| 293 |
+
"grad_norm": 1.9716824691797907,
|
| 294 |
+
"learning_rate": 1.9999411876613578e-05,
|
| 295 |
+
"loss": 1.2445,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10610766045548654,
|
| 300 |
+
"grad_norm": 1.9437244964639826,
|
| 301 |
+
"learning_rate": 1.9998023297700656e-05,
|
| 302 |
+
"loss": 1.2473,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.10869565217391304,
|
| 307 |
+
"grad_norm": 2.0447501542042135,
|
| 308 |
+
"learning_rate": 1.9995818039769564e-05,
|
| 309 |
+
"loss": 1.3562,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11128364389233955,
|
| 314 |
+
"grad_norm": 2.136938650583081,
|
| 315 |
+
"learning_rate": 1.999279628295469e-05,
|
| 316 |
+
"loss": 1.2552,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11387163561076605,
|
| 321 |
+
"grad_norm": 1.9517996473885892,
|
| 322 |
+
"learning_rate": 1.9988958274085367e-05,
|
| 323 |
+
"loss": 1.2911,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11645962732919254,
|
| 328 |
+
"grad_norm": 1.7952101956790416,
|
| 329 |
+
"learning_rate": 1.99843043266657e-05,
|
| 330 |
+
"loss": 1.2789,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.11904761904761904,
|
| 335 |
+
"grad_norm": 2.0601437201646275,
|
| 336 |
+
"learning_rate": 1.9978834820848952e-05,
|
| 337 |
+
"loss": 1.3249,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12163561076604555,
|
| 342 |
+
"grad_norm": 1.9179791275324676,
|
| 343 |
+
"learning_rate": 1.99725502034065e-05,
|
| 344 |
+
"loss": 1.2859,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12422360248447205,
|
| 349 |
+
"grad_norm": 2.146343845103461,
|
| 350 |
+
"learning_rate": 1.996545098769134e-05,
|
| 351 |
+
"loss": 1.259,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.12681159420289856,
|
| 356 |
+
"grad_norm": 1.9311811617661367,
|
| 357 |
+
"learning_rate": 1.995753775359615e-05,
|
| 358 |
+
"loss": 1.2773,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.12939958592132506,
|
| 363 |
+
"grad_norm": 1.8858776139378066,
|
| 364 |
+
"learning_rate": 1.994881114750593e-05,
|
| 365 |
+
"loss": 1.2574,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13198757763975155,
|
| 370 |
+
"grad_norm": 1.8059315541403178,
|
| 371 |
+
"learning_rate": 1.993927188224519e-05,
|
| 372 |
+
"loss": 1.2613,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13457556935817805,
|
| 377 |
+
"grad_norm": 1.7923594093741397,
|
| 378 |
+
"learning_rate": 1.9928920737019735e-05,
|
| 379 |
+
"loss": 1.2602,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.13716356107660455,
|
| 384 |
+
"grad_norm": 1.774108800205075,
|
| 385 |
+
"learning_rate": 1.991775855735301e-05,
|
| 386 |
+
"loss": 1.3387,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13975155279503104,
|
| 391 |
+
"grad_norm": 1.913850446875917,
|
| 392 |
+
"learning_rate": 1.9905786255017046e-05,
|
| 393 |
+
"loss": 1.3221,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14233954451345757,
|
| 398 |
+
"grad_norm": 1.9693465712493503,
|
| 399 |
+
"learning_rate": 1.9893004807957957e-05,
|
| 400 |
+
"loss": 1.2983,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14492753623188406,
|
| 405 |
+
"grad_norm": 1.790233528375858,
|
| 406 |
+
"learning_rate": 1.987941526021609e-05,
|
| 407 |
+
"loss": 1.2954,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.14751552795031056,
|
| 412 |
+
"grad_norm": 2.8906112799819574,
|
| 413 |
+
"learning_rate": 1.9865018721840708e-05,
|
| 414 |
+
"loss": 1.2964,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.15010351966873706,
|
| 419 |
+
"grad_norm": 1.9226491916325386,
|
| 420 |
+
"learning_rate": 1.9849816368799356e-05,
|
| 421 |
+
"loss": 1.299,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15269151138716355,
|
| 426 |
+
"grad_norm": 1.9512005440074536,
|
| 427 |
+
"learning_rate": 1.9833809442881776e-05,
|
| 428 |
+
"loss": 1.323,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15527950310559005,
|
| 433 |
+
"grad_norm": 1.7853859665144927,
|
| 434 |
+
"learning_rate": 1.981699925159847e-05,
|
| 435 |
+
"loss": 1.3069,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.15786749482401657,
|
| 440 |
+
"grad_norm": 1.7547957709103652,
|
| 441 |
+
"learning_rate": 1.979938716807392e-05,
|
| 442 |
+
"loss": 1.3264,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16045548654244307,
|
| 447 |
+
"grad_norm": 1.852378431981816,
|
| 448 |
+
"learning_rate": 1.978097463093441e-05,
|
| 449 |
+
"loss": 1.2946,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16304347826086957,
|
| 454 |
+
"grad_norm": 1.8256299629827946,
|
| 455 |
+
"learning_rate": 1.976176314419051e-05,
|
| 456 |
+
"loss": 1.2683,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16563146997929606,
|
| 461 |
+
"grad_norm": 1.8017450986442927,
|
| 462 |
+
"learning_rate": 1.974175427711423e-05,
|
| 463 |
+
"loss": 1.3127,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.16821946169772256,
|
| 468 |
+
"grad_norm": 1.8676977703064013,
|
| 469 |
+
"learning_rate": 1.9720949664110843e-05,
|
| 470 |
+
"loss": 1.3071,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17080745341614906,
|
| 475 |
+
"grad_norm": 1.806177669035387,
|
| 476 |
+
"learning_rate": 1.9699351004585354e-05,
|
| 477 |
+
"loss": 1.2485,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17339544513457558,
|
| 482 |
+
"grad_norm": 1.7075013810392663,
|
| 483 |
+
"learning_rate": 1.9676960062803714e-05,
|
| 484 |
+
"loss": 1.2904,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17598343685300208,
|
| 489 |
+
"grad_norm": 1.7968766744551163,
|
| 490 |
+
"learning_rate": 1.9653778667748695e-05,
|
| 491 |
+
"loss": 1.3346,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.17857142857142858,
|
| 496 |
+
"grad_norm": 2.2546697403830587,
|
| 497 |
+
"learning_rate": 1.962980871297049e-05,
|
| 498 |
+
"loss": 1.2952,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18115942028985507,
|
| 503 |
+
"grad_norm": 1.855772798318876,
|
| 504 |
+
"learning_rate": 1.9605052156432042e-05,
|
| 505 |
+
"loss": 1.243,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18374741200828157,
|
| 510 |
+
"grad_norm": 1.6468966576864474,
|
| 511 |
+
"learning_rate": 1.9579511020349117e-05,
|
| 512 |
+
"loss": 1.2897,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18633540372670807,
|
| 517 |
+
"grad_norm": 1.7882764084793803,
|
| 518 |
+
"learning_rate": 1.9553187391025102e-05,
|
| 519 |
+
"loss": 1.2678,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.18892339544513456,
|
| 524 |
+
"grad_norm": 1.8772543986524286,
|
| 525 |
+
"learning_rate": 1.952608341868061e-05,
|
| 526 |
+
"loss": 1.3172,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.1915113871635611,
|
| 531 |
+
"grad_norm": 1.648896837108699,
|
| 532 |
+
"learning_rate": 1.949820131727783e-05,
|
| 533 |
+
"loss": 1.2923,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19409937888198758,
|
| 538 |
+
"grad_norm": 1.7068857061711549,
|
| 539 |
+
"learning_rate": 1.9469543364339674e-05,
|
| 540 |
+
"loss": 1.3209,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.19668737060041408,
|
| 545 |
+
"grad_norm": 1.6145254855564042,
|
| 546 |
+
"learning_rate": 1.9440111900763758e-05,
|
| 547 |
+
"loss": 1.273,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.19927536231884058,
|
| 552 |
+
"grad_norm": 1.7464698166979602,
|
| 553 |
+
"learning_rate": 1.940990933063118e-05,
|
| 554 |
+
"loss": 1.3115,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20186335403726707,
|
| 559 |
+
"grad_norm": 2.237780229221515,
|
| 560 |
+
"learning_rate": 1.9378938121010128e-05,
|
| 561 |
+
"loss": 1.3093,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20445134575569357,
|
| 566 |
+
"grad_norm": 1.8816106738657112,
|
| 567 |
+
"learning_rate": 1.9347200801754395e-05,
|
| 568 |
+
"loss": 1.2999,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2070393374741201,
|
| 573 |
+
"grad_norm": 1.7444438072126305,
|
| 574 |
+
"learning_rate": 1.9314699965296694e-05,
|
| 575 |
+
"loss": 1.2824,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2096273291925466,
|
| 580 |
+
"grad_norm": 2.048386739159661,
|
| 581 |
+
"learning_rate": 1.9281438266436923e-05,
|
| 582 |
+
"loss": 1.2933,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2122153209109731,
|
| 587 |
+
"grad_norm": 3.306515280854177,
|
| 588 |
+
"learning_rate": 1.92474184221253e-05,
|
| 589 |
+
"loss": 1.295,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21480331262939958,
|
| 594 |
+
"grad_norm": 1.5615788627848504,
|
| 595 |
+
"learning_rate": 1.9212643211240433e-05,
|
| 596 |
+
"loss": 1.2794,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.21739130434782608,
|
| 601 |
+
"grad_norm": 1.6191612683725647,
|
| 602 |
+
"learning_rate": 1.917711547436232e-05,
|
| 603 |
+
"loss": 1.2846,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.21997929606625258,
|
| 608 |
+
"grad_norm": 1.7354290362475042,
|
| 609 |
+
"learning_rate": 1.9140838113540347e-05,
|
| 610 |
+
"loss": 1.2621,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2225672877846791,
|
| 615 |
+
"grad_norm": 1.779070935639009,
|
| 616 |
+
"learning_rate": 1.9103814092056205e-05,
|
| 617 |
+
"loss": 1.3176,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2251552795031056,
|
| 622 |
+
"grad_norm": 1.9188634938137341,
|
| 623 |
+
"learning_rate": 1.9066046434181854e-05,
|
| 624 |
+
"loss": 1.3437,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2277432712215321,
|
| 629 |
+
"grad_norm": 1.6905846039991461,
|
| 630 |
+
"learning_rate": 1.902753822493248e-05,
|
| 631 |
+
"loss": 1.2722,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2303312629399586,
|
| 636 |
+
"grad_norm": 1.7568380824359298,
|
| 637 |
+
"learning_rate": 1.8988292609814513e-05,
|
| 638 |
+
"loss": 1.2728,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2329192546583851,
|
| 643 |
+
"grad_norm": 1.664244033715747,
|
| 644 |
+
"learning_rate": 1.8948312794568674e-05,
|
| 645 |
+
"loss": 1.2558,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.23550724637681159,
|
| 650 |
+
"grad_norm": 1.6016572217912088,
|
| 651 |
+
"learning_rate": 1.8907602044908125e-05,
|
| 652 |
+
"loss": 1.2908,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.23809523809523808,
|
| 657 |
+
"grad_norm": 1.7484345463876294,
|
| 658 |
+
"learning_rate": 1.886616368625171e-05,
|
| 659 |
+
"loss": 1.3261,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2406832298136646,
|
| 664 |
+
"grad_norm": 1.7212899167763012,
|
| 665 |
+
"learning_rate": 1.8824001103452316e-05,
|
| 666 |
+
"loss": 1.2625,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2432712215320911,
|
| 671 |
+
"grad_norm": 1.6482436683937527,
|
| 672 |
+
"learning_rate": 1.8781117740520386e-05,
|
| 673 |
+
"loss": 1.2785,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2458592132505176,
|
| 678 |
+
"grad_norm": 1.5757960105455753,
|
| 679 |
+
"learning_rate": 1.8737517100342624e-05,
|
| 680 |
+
"loss": 1.2733,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2484472049689441,
|
| 685 |
+
"grad_norm": 1.64845152675654,
|
| 686 |
+
"learning_rate": 1.869320274439583e-05,
|
| 687 |
+
"loss": 1.3033,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2510351966873706,
|
| 692 |
+
"grad_norm": 1.8462235169663426,
|
| 693 |
+
"learning_rate": 1.864817829245598e-05,
|
| 694 |
+
"loss": 1.3084,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2536231884057971,
|
| 699 |
+
"grad_norm": 1.7739351798022618,
|
| 700 |
+
"learning_rate": 1.86024474223026e-05,
|
| 701 |
+
"loss": 1.2938,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2562111801242236,
|
| 706 |
+
"grad_norm": 1.628868267103692,
|
| 707 |
+
"learning_rate": 1.8556013869418282e-05,
|
| 708 |
+
"loss": 1.2852,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2587991718426501,
|
| 713 |
+
"grad_norm": 1.6457493283632671,
|
| 714 |
+
"learning_rate": 1.8508881426683618e-05,
|
| 715 |
+
"loss": 1.2341,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2613871635610766,
|
| 720 |
+
"grad_norm": 1.5598114432949137,
|
| 721 |
+
"learning_rate": 1.8461053944067324e-05,
|
| 722 |
+
"loss": 1.3068,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2639751552795031,
|
| 727 |
+
"grad_norm": 1.7427944903352592,
|
| 728 |
+
"learning_rate": 1.8412535328311813e-05,
|
| 729 |
+
"loss": 1.2866,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2665631469979296,
|
| 734 |
+
"grad_norm": 1.7159655860891432,
|
| 735 |
+
"learning_rate": 1.8363329542614033e-05,
|
| 736 |
+
"loss": 1.2987,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.2691511387163561,
|
| 741 |
+
"grad_norm": 1.6937281668589332,
|
| 742 |
+
"learning_rate": 1.831344060630176e-05,
|
| 743 |
+
"loss": 1.2861,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.2717391304347826,
|
| 748 |
+
"grad_norm": 1.7317450117613962,
|
| 749 |
+
"learning_rate": 1.826287259450528e-05,
|
| 750 |
+
"loss": 1.3135,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2743271221532091,
|
| 755 |
+
"grad_norm": 1.736815145364657,
|
| 756 |
+
"learning_rate": 1.8211629637824516e-05,
|
| 757 |
+
"loss": 1.3229,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2769151138716356,
|
| 762 |
+
"grad_norm": 1.717368110143028,
|
| 763 |
+
"learning_rate": 1.8159715921991612e-05,
|
| 764 |
+
"loss": 1.3,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2795031055900621,
|
| 769 |
+
"grad_norm": 1.7858279813792615,
|
| 770 |
+
"learning_rate": 1.8107135687529044e-05,
|
| 771 |
+
"loss": 1.304,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.28209109730848864,
|
| 776 |
+
"grad_norm": 1.6420365658498397,
|
| 777 |
+
"learning_rate": 1.8053893229403218e-05,
|
| 778 |
+
"loss": 1.2746,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.28467908902691513,
|
| 783 |
+
"grad_norm": 1.781865206777271,
|
| 784 |
+
"learning_rate": 1.7999992896673667e-05,
|
| 785 |
+
"loss": 1.2941,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.28726708074534163,
|
| 790 |
+
"grad_norm": 1.717576281479697,
|
| 791 |
+
"learning_rate": 1.7945439092137763e-05,
|
| 792 |
+
"loss": 1.2934,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.2898550724637681,
|
| 797 |
+
"grad_norm": 1.8605583484947545,
|
| 798 |
+
"learning_rate": 1.789023627197112e-05,
|
| 799 |
+
"loss": 1.2609,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2924430641821946,
|
| 804 |
+
"grad_norm": 1.7448230138089384,
|
| 805 |
+
"learning_rate": 1.783438894536357e-05,
|
| 806 |
+
"loss": 1.297,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2950310559006211,
|
| 811 |
+
"grad_norm": 1.7954544554359166,
|
| 812 |
+
"learning_rate": 1.7777901674150827e-05,
|
| 813 |
+
"loss": 1.2839,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2976190476190476,
|
| 818 |
+
"grad_norm": 1.82466567221307,
|
| 819 |
+
"learning_rate": 1.7720779072441897e-05,
|
| 820 |
+
"loss": 1.2602,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3002070393374741,
|
| 825 |
+
"grad_norm": 1.6162229238306363,
|
| 826 |
+
"learning_rate": 1.7663025806242126e-05,
|
| 827 |
+
"loss": 1.2819,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3027950310559006,
|
| 832 |
+
"grad_norm": 1.7224353038885882,
|
| 833 |
+
"learning_rate": 1.7604646593072107e-05,
|
| 834 |
+
"loss": 1.2789,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3053830227743271,
|
| 839 |
+
"grad_norm": 1.5302928069449526,
|
| 840 |
+
"learning_rate": 1.7545646201582304e-05,
|
| 841 |
+
"loss": 1.2741,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3079710144927536,
|
| 846 |
+
"grad_norm": 1.7278169638412462,
|
| 847 |
+
"learning_rate": 1.7486029451163552e-05,
|
| 848 |
+
"loss": 1.2761,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3105590062111801,
|
| 853 |
+
"grad_norm": 7.66316697414024,
|
| 854 |
+
"learning_rate": 1.742580121155337e-05,
|
| 855 |
+
"loss": 1.2848,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.31314699792960665,
|
| 860 |
+
"grad_norm": 1.7170749939899528,
|
| 861 |
+
"learning_rate": 1.73649664024382e-05,
|
| 862 |
+
"loss": 1.3145,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.31573498964803315,
|
| 867 |
+
"grad_norm": 1.5882256935177697,
|
| 868 |
+
"learning_rate": 1.730352999305152e-05,
|
| 869 |
+
"loss": 1.2437,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.31832298136645965,
|
| 874 |
+
"grad_norm": 1.6451039487590284,
|
| 875 |
+
"learning_rate": 1.7241497001767967e-05,
|
| 876 |
+
"loss": 1.3017,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32091097308488614,
|
| 881 |
+
"grad_norm": 1.6109005034073363,
|
| 882 |
+
"learning_rate": 1.7178872495693397e-05,
|
| 883 |
+
"loss": 1.264,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32349896480331264,
|
| 888 |
+
"grad_norm": 1.641706658754867,
|
| 889 |
+
"learning_rate": 1.7115661590250992e-05,
|
| 890 |
+
"loss": 1.2764,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.32608695652173914,
|
| 895 |
+
"grad_norm": 1.8516711248065172,
|
| 896 |
+
"learning_rate": 1.7051869448763406e-05,
|
| 897 |
+
"loss": 1.276,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.32867494824016563,
|
| 902 |
+
"grad_norm": 1.5631168769858184,
|
| 903 |
+
"learning_rate": 1.6987501282031013e-05,
|
| 904 |
+
"loss": 1.2148,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.33126293995859213,
|
| 909 |
+
"grad_norm": 1.6459960001672511,
|
| 910 |
+
"learning_rate": 1.692256234790624e-05,
|
| 911 |
+
"loss": 1.2473,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3338509316770186,
|
| 916 |
+
"grad_norm": 2.9555007934323703,
|
| 917 |
+
"learning_rate": 1.6857057950864134e-05,
|
| 918 |
+
"loss": 1.2874,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3364389233954451,
|
| 923 |
+
"grad_norm": 1.612379368244149,
|
| 924 |
+
"learning_rate": 1.6790993441569022e-05,
|
| 925 |
+
"loss": 1.2373,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3390269151138716,
|
| 930 |
+
"grad_norm": 1.6757461678281282,
|
| 931 |
+
"learning_rate": 1.6724374216437475e-05,
|
| 932 |
+
"loss": 1.2801,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3416149068322981,
|
| 937 |
+
"grad_norm": 1.5726408014891122,
|
| 938 |
+
"learning_rate": 1.6657205717197495e-05,
|
| 939 |
+
"loss": 1.2065,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3442028985507246,
|
| 944 |
+
"grad_norm": 1.6029270709660246,
|
| 945 |
+
"learning_rate": 1.6589493430444026e-05,
|
| 946 |
+
"loss": 1.2928,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.34679089026915116,
|
| 951 |
+
"grad_norm": 1.7396806928714428,
|
| 952 |
+
"learning_rate": 1.6521242887190764e-05,
|
| 953 |
+
"loss": 1.3018,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.34937888198757766,
|
| 958 |
+
"grad_norm": 1.683802020776293,
|
| 959 |
+
"learning_rate": 1.6452459662418386e-05,
|
| 960 |
+
"loss": 1.2703,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.35196687370600416,
|
| 965 |
+
"grad_norm": 1.6299055475198938,
|
| 966 |
+
"learning_rate": 1.638314937461915e-05,
|
| 967 |
+
"loss": 1.2465,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.35455486542443065,
|
| 972 |
+
"grad_norm": 1.5793607569305301,
|
| 973 |
+
"learning_rate": 1.6313317685337947e-05,
|
| 974 |
+
"loss": 1.2774,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.35714285714285715,
|
| 979 |
+
"grad_norm": 1.6509230878640204,
|
| 980 |
+
"learning_rate": 1.6242970298709867e-05,
|
| 981 |
+
"loss": 1.2394,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.35973084886128365,
|
| 986 |
+
"grad_norm": 1.6564302207780428,
|
| 987 |
+
"learning_rate": 1.6172112960994234e-05,
|
| 988 |
+
"loss": 1.3113,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.36231884057971014,
|
| 993 |
+
"grad_norm": 1.720021595292221,
|
| 994 |
+
"learning_rate": 1.6100751460105244e-05,
|
| 995 |
+
"loss": 1.2588,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.36490683229813664,
|
| 1000 |
+
"grad_norm": 1.5844840946364063,
|
| 1001 |
+
"learning_rate": 1.602889162513919e-05,
|
| 1002 |
+
"loss": 1.2807,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.36749482401656314,
|
| 1007 |
+
"grad_norm": 1.6839767345265815,
|
| 1008 |
+
"learning_rate": 1.59565393258983e-05,
|
| 1009 |
+
"loss": 1.2607,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.37008281573498963,
|
| 1014 |
+
"grad_norm": 1.7860415375986487,
|
| 1015 |
+
"learning_rate": 1.588370047241128e-05,
|
| 1016 |
+
"loss": 1.256,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.37267080745341613,
|
| 1021 |
+
"grad_norm": 1.7117135290100387,
|
| 1022 |
+
"learning_rate": 1.5810381014450557e-05,
|
| 1023 |
+
"loss": 1.2263,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3752587991718426,
|
| 1028 |
+
"grad_norm": 1.5845047105100838,
|
| 1029 |
+
"learning_rate": 1.5736586941046285e-05,
|
| 1030 |
+
"loss": 1.242,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3778467908902691,
|
| 1035 |
+
"grad_norm": 1.80500551409116,
|
| 1036 |
+
"learning_rate": 1.5662324279997115e-05,
|
| 1037 |
+
"loss": 1.2394,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.3804347826086957,
|
| 1042 |
+
"grad_norm": 1.5963745820078468,
|
| 1043 |
+
"learning_rate": 1.5587599097377845e-05,
|
| 1044 |
+
"loss": 1.2124,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3830227743271222,
|
| 1049 |
+
"grad_norm": 1.7129591053368485,
|
| 1050 |
+
"learning_rate": 1.5512417497043898e-05,
|
| 1051 |
+
"loss": 1.2285,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.38561076604554867,
|
| 1056 |
+
"grad_norm": 1.947935106696502,
|
| 1057 |
+
"learning_rate": 1.5436785620132756e-05,
|
| 1058 |
+
"loss": 1.2297,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.38819875776397517,
|
| 1063 |
+
"grad_norm": 1.5854599615971694,
|
| 1064 |
+
"learning_rate": 1.5360709644562296e-05,
|
| 1065 |
+
"loss": 1.2551,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.39078674948240166,
|
| 1070 |
+
"grad_norm": 1.6101311120937332,
|
| 1071 |
+
"learning_rate": 1.5284195784526196e-05,
|
| 1072 |
+
"loss": 1.2708,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.39337474120082816,
|
| 1077 |
+
"grad_norm": 1.647707331905149,
|
| 1078 |
+
"learning_rate": 1.520725028998629e-05,
|
| 1079 |
+
"loss": 1.266,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.39596273291925466,
|
| 1084 |
+
"grad_norm": 1.9215587673185384,
|
| 1085 |
+
"learning_rate": 1.512987944616207e-05,
|
| 1086 |
+
"loss": 1.2563,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.39855072463768115,
|
| 1091 |
+
"grad_norm": 1.5923791612452982,
|
| 1092 |
+
"learning_rate": 1.5052089573017283e-05,
|
| 1093 |
+
"loss": 1.2951,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.40113871635610765,
|
| 1098 |
+
"grad_norm": 1.533522718790619,
|
| 1099 |
+
"learning_rate": 1.4973887024743686e-05,
|
| 1100 |
+
"loss": 1.2629,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.40372670807453415,
|
| 1105 |
+
"grad_norm": 1.5392604333005226,
|
| 1106 |
+
"learning_rate": 1.4895278189242017e-05,
|
| 1107 |
+
"loss": 1.1853,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.40631469979296064,
|
| 1112 |
+
"grad_norm": 1.5911615984318879,
|
| 1113 |
+
"learning_rate": 1.4816269487600184e-05,
|
| 1114 |
+
"loss": 1.2374,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.40890269151138714,
|
| 1119 |
+
"grad_norm": 1.540407306505209,
|
| 1120 |
+
"learning_rate": 1.4736867373568795e-05,
|
| 1121 |
+
"loss": 1.2208,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4114906832298137,
|
| 1126 |
+
"grad_norm": 1.5985263367678237,
|
| 1127 |
+
"learning_rate": 1.4657078333033967e-05,
|
| 1128 |
+
"loss": 1.2569,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4140786749482402,
|
| 1133 |
+
"grad_norm": 1.5598791416023927,
|
| 1134 |
+
"learning_rate": 1.4576908883487549e-05,
|
| 1135 |
+
"loss": 1.184,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4166666666666667,
|
| 1140 |
+
"grad_norm": 1.564166816397901,
|
| 1141 |
+
"learning_rate": 1.4496365573494724e-05,
|
| 1142 |
+
"loss": 1.278,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4192546583850932,
|
| 1147 |
+
"grad_norm": 1.5214555732261263,
|
| 1148 |
+
"learning_rate": 1.4415454982159121e-05,
|
| 1149 |
+
"loss": 1.2756,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4218426501035197,
|
| 1154 |
+
"grad_norm": 1.566924379219736,
|
| 1155 |
+
"learning_rate": 1.433418371858539e-05,
|
| 1156 |
+
"loss": 1.2208,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4244306418219462,
|
| 1161 |
+
"grad_norm": 1.693241805167907,
|
| 1162 |
+
"learning_rate": 1.4252558421339355e-05,
|
| 1163 |
+
"loss": 1.2649,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.42701863354037267,
|
| 1168 |
+
"grad_norm": 1.845359674560561,
|
| 1169 |
+
"learning_rate": 1.4170585757905742e-05,
|
| 1170 |
+
"loss": 1.2263,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.42960662525879917,
|
| 1175 |
+
"grad_norm": 1.5444000254202592,
|
| 1176 |
+
"learning_rate": 1.4088272424143546e-05,
|
| 1177 |
+
"loss": 1.2764,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.43219461697722567,
|
| 1182 |
+
"grad_norm": 1.5904821648663396,
|
| 1183 |
+
"learning_rate": 1.4005625143739103e-05,
|
| 1184 |
+
"loss": 1.2467,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.43478260869565216,
|
| 1189 |
+
"grad_norm": 1.6406582849973572,
|
| 1190 |
+
"learning_rate": 1.3922650667656861e-05,
|
| 1191 |
+
"loss": 1.2466,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.43737060041407866,
|
| 1196 |
+
"grad_norm": 1.5874801423879288,
|
| 1197 |
+
"learning_rate": 1.3839355773587932e-05,
|
| 1198 |
+
"loss": 1.2818,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.43995859213250516,
|
| 1203 |
+
"grad_norm": 1.5664894322663914,
|
| 1204 |
+
"learning_rate": 1.3755747265396466e-05,
|
| 1205 |
+
"loss": 1.2569,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.44254658385093165,
|
| 1210 |
+
"grad_norm": 1.639337291197075,
|
| 1211 |
+
"learning_rate": 1.3671831972563889e-05,
|
| 1212 |
+
"loss": 1.2466,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4451345755693582,
|
| 1217 |
+
"grad_norm": 1.4882976719575092,
|
| 1218 |
+
"learning_rate": 1.3587616749631037e-05,
|
| 1219 |
+
"loss": 1.2262,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4477225672877847,
|
| 1224 |
+
"grad_norm": 1.5575004320774404,
|
| 1225 |
+
"learning_rate": 1.3503108475638244e-05,
|
| 1226 |
+
"loss": 1.2685,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4503105590062112,
|
| 1231 |
+
"grad_norm": 1.5995590657708154,
|
| 1232 |
+
"learning_rate": 1.3418314053563455e-05,
|
| 1233 |
+
"loss": 1.251,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.4528985507246377,
|
| 1238 |
+
"grad_norm": 1.560631857872418,
|
| 1239 |
+
"learning_rate": 1.3333240409758328e-05,
|
| 1240 |
+
"loss": 1.2201,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4554865424430642,
|
| 1245 |
+
"grad_norm": 1.5801386912905093,
|
| 1246 |
+
"learning_rate": 1.3247894493382507e-05,
|
| 1247 |
+
"loss": 1.2186,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4580745341614907,
|
| 1252 |
+
"grad_norm": 1.749634243011622,
|
| 1253 |
+
"learning_rate": 1.3162283275835938e-05,
|
| 1254 |
+
"loss": 1.1815,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4606625258799172,
|
| 1259 |
+
"grad_norm": 1.6614862019498566,
|
| 1260 |
+
"learning_rate": 1.3076413750189468e-05,
|
| 1261 |
+
"loss": 1.2601,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4632505175983437,
|
| 1266 |
+
"grad_norm": 1.6371821901498413,
|
| 1267 |
+
"learning_rate": 1.2990292930613559e-05,
|
| 1268 |
+
"loss": 1.2368,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4658385093167702,
|
| 1273 |
+
"grad_norm": 1.56278259362623,
|
| 1274 |
+
"learning_rate": 1.2903927851805416e-05,
|
| 1275 |
+
"loss": 1.1672,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4684265010351967,
|
| 1280 |
+
"grad_norm": 1.4842721179109506,
|
| 1281 |
+
"learning_rate": 1.2817325568414299e-05,
|
| 1282 |
+
"loss": 1.2425,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.47101449275362317,
|
| 1287 |
+
"grad_norm": 1.5573631764732299,
|
| 1288 |
+
"learning_rate": 1.2730493154465311e-05,
|
| 1289 |
+
"loss": 1.2584,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.47360248447204967,
|
| 1294 |
+
"grad_norm": 1.5747650735756835,
|
| 1295 |
+
"learning_rate": 1.2643437702781559e-05,
|
| 1296 |
+
"loss": 1.2027,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.47619047619047616,
|
| 1301 |
+
"grad_norm": 1.7147334990616894,
|
| 1302 |
+
"learning_rate": 1.2556166324404747e-05,
|
| 1303 |
+
"loss": 1.2369,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.4787784679089027,
|
| 1308 |
+
"grad_norm": 1.6173781131907976,
|
| 1309 |
+
"learning_rate": 1.2468686148014379e-05,
|
| 1310 |
+
"loss": 1.2348,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.4813664596273292,
|
| 1315 |
+
"grad_norm": 1.6483571836594604,
|
| 1316 |
+
"learning_rate": 1.2381004319345402e-05,
|
| 1317 |
+
"loss": 1.3268,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4839544513457557,
|
| 1322 |
+
"grad_norm": 1.5421617996607333,
|
| 1323 |
+
"learning_rate": 1.2293128000604552e-05,
|
| 1324 |
+
"loss": 1.2666,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4865424430641822,
|
| 1329 |
+
"grad_norm": 1.618282632952526,
|
| 1330 |
+
"learning_rate": 1.2205064369885291e-05,
|
| 1331 |
+
"loss": 1.2037,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.4891304347826087,
|
| 1336 |
+
"grad_norm": 1.6786332269059123,
|
| 1337 |
+
"learning_rate": 1.2116820620581486e-05,
|
| 1338 |
+
"loss": 1.2555,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.4917184265010352,
|
| 1343 |
+
"grad_norm": 1.6174195289732147,
|
| 1344 |
+
"learning_rate": 1.2028403960799821e-05,
|
| 1345 |
+
"loss": 1.2518,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.4943064182194617,
|
| 1350 |
+
"grad_norm": 1.49372149520001,
|
| 1351 |
+
"learning_rate": 1.1939821612771008e-05,
|
| 1352 |
+
"loss": 1.2294,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.4968944099378882,
|
| 1357 |
+
"grad_norm": 1.5736827426176978,
|
| 1358 |
+
"learning_rate": 1.1851080812259839e-05,
|
| 1359 |
+
"loss": 1.216,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.4994824016563147,
|
| 1364 |
+
"grad_norm": 1.6096915420259137,
|
| 1365 |
+
"learning_rate": 1.1762188807974153e-05,
|
| 1366 |
+
"loss": 1.2033,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5020703933747412,
|
| 1371 |
+
"grad_norm": 1.6642406229664035,
|
| 1372 |
+
"learning_rate": 1.1673152860972725e-05,
|
| 1373 |
+
"loss": 1.2276,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5046583850931677,
|
| 1378 |
+
"grad_norm": 1.5472858925733155,
|
| 1379 |
+
"learning_rate": 1.158398024407215e-05,
|
| 1380 |
+
"loss": 1.2135,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5072463768115942,
|
| 1385 |
+
"grad_norm": 1.6378928317247718,
|
| 1386 |
+
"learning_rate": 1.1494678241252781e-05,
|
| 1387 |
+
"loss": 1.1857,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5098343685300207,
|
| 1392 |
+
"grad_norm": 1.669332160090618,
|
| 1393 |
+
"learning_rate": 1.1405254147063722e-05,
|
| 1394 |
+
"loss": 1.2294,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5124223602484472,
|
| 1399 |
+
"grad_norm": 1.5329461024041375,
|
| 1400 |
+
"learning_rate": 1.1315715266027014e-05,
|
| 1401 |
+
"loss": 1.2223,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5150103519668737,
|
| 1406 |
+
"grad_norm": 1.643890352883545,
|
| 1407 |
+
"learning_rate": 1.1226068912040935e-05,
|
| 1408 |
+
"loss": 1.223,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5175983436853002,
|
| 1413 |
+
"grad_norm": 1.7480566033309979,
|
| 1414 |
+
"learning_rate": 1.1136322407782603e-05,
|
| 1415 |
+
"loss": 1.2288,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5201863354037267,
|
| 1420 |
+
"grad_norm": 1.5968227515636302,
|
| 1421 |
+
"learning_rate": 1.1046483084109801e-05,
|
| 1422 |
+
"loss": 1.2106,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5227743271221532,
|
| 1427 |
+
"grad_norm": 1.6044766188232418,
|
| 1428 |
+
"learning_rate": 1.0956558279462188e-05,
|
| 1429 |
+
"loss": 1.2208,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5253623188405797,
|
| 1434 |
+
"grad_norm": 1.6067811085133883,
|
| 1435 |
+
"learning_rate": 1.0866555339261854e-05,
|
| 1436 |
+
"loss": 1.2516,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5279503105590062,
|
| 1441 |
+
"grad_norm": 1.7393965202791053,
|
| 1442 |
+
"learning_rate": 1.0776481615313314e-05,
|
| 1443 |
+
"loss": 1.2248,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5305383022774327,
|
| 1448 |
+
"grad_norm": 1.5858453669945456,
|
| 1449 |
+
"learning_rate": 1.0686344465202988e-05,
|
| 1450 |
+
"loss": 1.2176,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5331262939958592,
|
| 1455 |
+
"grad_norm": 1.5018326905549035,
|
| 1456 |
+
"learning_rate": 1.05961512516982e-05,
|
| 1457 |
+
"loss": 1.1795,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5357142857142857,
|
| 1462 |
+
"grad_norm": 1.7095471413310692,
|
| 1463 |
+
"learning_rate": 1.050590934214576e-05,
|
| 1464 |
+
"loss": 1.2241,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5383022774327122,
|
| 1469 |
+
"grad_norm": 1.566999738860566,
|
| 1470 |
+
"learning_rate": 1.0415626107870171e-05,
|
| 1471 |
+
"loss": 1.1825,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5408902691511387,
|
| 1476 |
+
"grad_norm": 1.6457686223760508,
|
| 1477 |
+
"learning_rate": 1.0325308923571506e-05,
|
| 1478 |
+
"loss": 1.284,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5434782608695652,
|
| 1483 |
+
"grad_norm": 1.6348224943577552,
|
| 1484 |
+
"learning_rate": 1.0234965166723003e-05,
|
| 1485 |
+
"loss": 1.2698,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5460662525879917,
|
| 1490 |
+
"grad_norm": 1.5465862470654987,
|
| 1491 |
+
"learning_rate": 1.014460221696847e-05,
|
| 1492 |
+
"loss": 1.1963,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5486542443064182,
|
| 1497 |
+
"grad_norm": 1.767248102032812,
|
| 1498 |
+
"learning_rate": 1.0054227455519459e-05,
|
| 1499 |
+
"loss": 1.1638,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5512422360248447,
|
| 1504 |
+
"grad_norm": 2.0136126055316383,
|
| 1505 |
+
"learning_rate": 9.963848264552351e-06,
|
| 1506 |
+
"loss": 1.2307,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5538302277432712,
|
| 1511 |
+
"grad_norm": 1.5145725671730654,
|
| 1512 |
+
"learning_rate": 9.873472026605347e-06,
|
| 1513 |
+
"loss": 1.2352,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5564182194616977,
|
| 1518 |
+
"grad_norm": 1.5882467088516106,
|
| 1519 |
+
"learning_rate": 9.783106123975436e-06,
|
| 1520 |
+
"loss": 1.2108,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5590062111801242,
|
| 1525 |
+
"grad_norm": 1.5181691898121104,
|
| 1526 |
+
"learning_rate": 9.692757938115376e-06,
|
| 1527 |
+
"loss": 1.1861,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5615942028985508,
|
| 1532 |
+
"grad_norm": 1.6410339653531256,
|
| 1533 |
+
"learning_rate": 9.602434849030747e-06,
|
| 1534 |
+
"loss": 1.2465,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5641821946169773,
|
| 1539 |
+
"grad_norm": 1.5661030436893173,
|
| 1540 |
+
"learning_rate": 9.512144234677116e-06,
|
| 1541 |
+
"loss": 1.243,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5667701863354038,
|
| 1546 |
+
"grad_norm": 1.8049727192206169,
|
| 1547 |
+
"learning_rate": 9.421893470357397e-06,
|
| 1548 |
+
"loss": 1.2283,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5693581780538303,
|
| 1553 |
+
"grad_norm": 1.8382407169522683,
|
| 1554 |
+
"learning_rate": 9.331689928119382e-06,
|
| 1555 |
+
"loss": 1.2307,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5719461697722568,
|
| 1560 |
+
"grad_norm": 1.541503454678279,
|
| 1561 |
+
"learning_rate": 9.24154097615357e-06,
|
| 1562 |
+
"loss": 1.157,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5745341614906833,
|
| 1567 |
+
"grad_norm": 1.5831300806041508,
|
| 1568 |
+
"learning_rate": 9.151453978191312e-06,
|
| 1569 |
+
"loss": 1.2318,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5771221532091098,
|
| 1574 |
+
"grad_norm": 1.5551918280787969,
|
| 1575 |
+
"learning_rate": 9.061436292903303e-06,
|
| 1576 |
+
"loss": 1.1745,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5797101449275363,
|
| 1581 |
+
"grad_norm": 1.4974161909157238,
|
| 1582 |
+
"learning_rate": 8.9714952732985e-06,
|
| 1583 |
+
"loss": 1.1957,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5822981366459627,
|
| 1588 |
+
"grad_norm": 1.6338432021840628,
|
| 1589 |
+
"learning_rate": 8.88163826612349e-06,
|
| 1590 |
+
"loss": 1.2651,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5848861283643892,
|
| 1595 |
+
"grad_norm": 1.605496912027716,
|
| 1596 |
+
"learning_rate": 8.791872611262393e-06,
|
| 1597 |
+
"loss": 1.183,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5874741200828157,
|
| 1602 |
+
"grad_norm": 6.292512705322076,
|
| 1603 |
+
"learning_rate": 8.702205641137293e-06,
|
| 1604 |
+
"loss": 1.214,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5900621118012422,
|
| 1609 |
+
"grad_norm": 1.5795886266757169,
|
| 1610 |
+
"learning_rate": 8.61264468010932e-06,
|
| 1611 |
+
"loss": 1.1703,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5926501035196687,
|
| 1616 |
+
"grad_norm": 1.6798626881987566,
|
| 1617 |
+
"learning_rate": 8.523197043880334e-06,
|
| 1618 |
+
"loss": 1.1722,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5952380952380952,
|
| 1623 |
+
"grad_norm": 1.4739551385813945,
|
| 1624 |
+
"learning_rate": 8.433870038895377e-06,
|
| 1625 |
+
"loss": 1.2068,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5978260869565217,
|
| 1630 |
+
"grad_norm": 1.539016959394501,
|
| 1631 |
+
"learning_rate": 8.34467096174584e-06,
|
| 1632 |
+
"loss": 1.1899,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6004140786749482,
|
| 1637 |
+
"grad_norm": 1.7032826938631798,
|
| 1638 |
+
"learning_rate": 8.255607098573454e-06,
|
| 1639 |
+
"loss": 1.2071,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6030020703933747,
|
| 1644 |
+
"grad_norm": 1.6061844765987872,
|
| 1645 |
+
"learning_rate": 8.166685724475123e-06,
|
| 1646 |
+
"loss": 1.2302,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6055900621118012,
|
| 1651 |
+
"grad_norm": 1.5665329666940486,
|
| 1652 |
+
"learning_rate": 8.077914102908661e-06,
|
| 1653 |
+
"loss": 1.2037,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6081780538302277,
|
| 1658 |
+
"grad_norm": 1.6218099795936876,
|
| 1659 |
+
"learning_rate": 7.989299485099498e-06,
|
| 1660 |
+
"loss": 1.2411,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6107660455486542,
|
| 1665 |
+
"grad_norm": 1.544888953496484,
|
| 1666 |
+
"learning_rate": 7.90084910944835e-06,
|
| 1667 |
+
"loss": 1.163,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6133540372670807,
|
| 1672 |
+
"grad_norm": 1.5468464434079643,
|
| 1673 |
+
"learning_rate": 7.812570200939985e-06,
|
| 1674 |
+
"loss": 1.2069,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6159420289855072,
|
| 1679 |
+
"grad_norm": 1.6191217485063485,
|
| 1680 |
+
"learning_rate": 7.724469970553028e-06,
|
| 1681 |
+
"loss": 1.212,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6185300207039337,
|
| 1686 |
+
"grad_norm": 1.6584669157155563,
|
| 1687 |
+
"learning_rate": 7.636555614670953e-06,
|
| 1688 |
+
"loss": 1.2359,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6211180124223602,
|
| 1693 |
+
"grad_norm": 1.6065749708841677,
|
| 1694 |
+
"learning_rate": 7.548834314494254e-06,
|
| 1695 |
+
"loss": 1.1731,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6237060041407867,
|
| 1700 |
+
"grad_norm": 1.580536951203462,
|
| 1701 |
+
"learning_rate": 7.461313235453858e-06,
|
| 1702 |
+
"loss": 1.2095,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6262939958592133,
|
| 1707 |
+
"grad_norm": 1.5890916003841715,
|
| 1708 |
+
"learning_rate": 7.37399952662581e-06,
|
| 1709 |
+
"loss": 1.1903,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6288819875776398,
|
| 1714 |
+
"grad_norm": 1.6317367363623994,
|
| 1715 |
+
"learning_rate": 7.2869003201473235e-06,
|
| 1716 |
+
"loss": 1.1615,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6314699792960663,
|
| 1721 |
+
"grad_norm": 1.5877174955747102,
|
| 1722 |
+
"learning_rate": 7.200022730634188e-06,
|
| 1723 |
+
"loss": 1.2261,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6340579710144928,
|
| 1728 |
+
"grad_norm": 1.6863972908974845,
|
| 1729 |
+
"learning_rate": 7.113373854599621e-06,
|
| 1730 |
+
"loss": 1.1994,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6366459627329193,
|
| 1735 |
+
"grad_norm": 1.6649535109927227,
|
| 1736 |
+
"learning_rate": 7.026960769874605e-06,
|
| 1737 |
+
"loss": 1.1562,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6392339544513458,
|
| 1742 |
+
"grad_norm": 1.6139061664993306,
|
| 1743 |
+
"learning_rate": 6.940790535029726e-06,
|
| 1744 |
+
"loss": 1.2049,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6418219461697723,
|
| 1749 |
+
"grad_norm": 2.0575985271226043,
|
| 1750 |
+
"learning_rate": 6.85487018879861e-06,
|
| 1751 |
+
"loss": 1.1997,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6444099378881988,
|
| 1756 |
+
"grad_norm": 1.5693541403629991,
|
| 1757 |
+
"learning_rate": 6.76920674950297e-06,
|
| 1758 |
+
"loss": 1.1713,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6469979296066253,
|
| 1763 |
+
"grad_norm": 1.4730612685033373,
|
| 1764 |
+
"learning_rate": 6.683807214479323e-06,
|
| 1765 |
+
"loss": 1.2526,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6495859213250518,
|
| 1770 |
+
"grad_norm": 1.6427505615358542,
|
| 1771 |
+
"learning_rate": 6.598678559507414e-06,
|
| 1772 |
+
"loss": 1.2033,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6521739130434783,
|
| 1777 |
+
"grad_norm": 1.4335417526682601,
|
| 1778 |
+
"learning_rate": 6.513827738240408e-06,
|
| 1779 |
+
"loss": 1.1668,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6547619047619048,
|
| 1784 |
+
"grad_norm": 1.5130023195633595,
|
| 1785 |
+
"learning_rate": 6.429261681636889e-06,
|
| 1786 |
+
"loss": 1.2091,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6573498964803313,
|
| 1791 |
+
"grad_norm": 1.5309037687636573,
|
| 1792 |
+
"learning_rate": 6.344987297394713e-06,
|
| 1793 |
+
"loss": 1.2039,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6599378881987578,
|
| 1798 |
+
"grad_norm": 1.5308412500551256,
|
| 1799 |
+
"learning_rate": 6.261011469386755e-06,
|
| 1800 |
+
"loss": 1.1725,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6625258799171843,
|
| 1805 |
+
"grad_norm": 4.125033452457996,
|
| 1806 |
+
"learning_rate": 6.177341057098607e-06,
|
| 1807 |
+
"loss": 1.2281,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6651138716356108,
|
| 1812 |
+
"grad_norm": 1.6368311674369278,
|
| 1813 |
+
"learning_rate": 6.09398289506827e-06,
|
| 1814 |
+
"loss": 1.198,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6677018633540373,
|
| 1819 |
+
"grad_norm": 1.543486534141238,
|
| 1820 |
+
"learning_rate": 6.010943792327875e-06,
|
| 1821 |
+
"loss": 1.1872,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6702898550724637,
|
| 1826 |
+
"grad_norm": 1.6262983681397423,
|
| 1827 |
+
"learning_rate": 5.92823053184751e-06,
|
| 1828 |
+
"loss": 1.2178,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6728778467908902,
|
| 1833 |
+
"grad_norm": 1.4789073947225242,
|
| 1834 |
+
"learning_rate": 5.845849869981137e-06,
|
| 1835 |
+
"loss": 1.1271,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6754658385093167,
|
| 1840 |
+
"grad_norm": 1.7175017242675854,
|
| 1841 |
+
"learning_rate": 5.7638085359147235e-06,
|
| 1842 |
+
"loss": 1.1758,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6780538302277432,
|
| 1847 |
+
"grad_norm": 1.621323772156704,
|
| 1848 |
+
"learning_rate": 5.682113231116557e-06,
|
| 1849 |
+
"loss": 1.1944,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6806418219461697,
|
| 1854 |
+
"grad_norm": 1.4735479464826196,
|
| 1855 |
+
"learning_rate": 5.600770628789872e-06,
|
| 1856 |
+
"loss": 1.1943,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6832298136645962,
|
| 1861 |
+
"grad_norm": 1.539494955374121,
|
| 1862 |
+
"learning_rate": 5.519787373327725e-06,
|
| 1863 |
+
"loss": 1.1556,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6858178053830227,
|
| 1868 |
+
"grad_norm": 1.6129814926752055,
|
| 1869 |
+
"learning_rate": 5.439170079770269e-06,
|
| 1870 |
+
"loss": 1.2101,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6884057971014492,
|
| 1875 |
+
"grad_norm": 1.6647287957418007,
|
| 1876 |
+
"learning_rate": 5.358925333264403e-06,
|
| 1877 |
+
"loss": 1.1943,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6909937888198758,
|
| 1882 |
+
"grad_norm": 1.5220188441031306,
|
| 1883 |
+
"learning_rate": 5.279059688525878e-06,
|
| 1884 |
+
"loss": 1.1722,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6935817805383023,
|
| 1889 |
+
"grad_norm": 1.6621504122467943,
|
| 1890 |
+
"learning_rate": 5.199579669303885e-06,
|
| 1891 |
+
"loss": 1.1982,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6961697722567288,
|
| 1896 |
+
"grad_norm": 1.5286471811019702,
|
| 1897 |
+
"learning_rate": 5.1204917678481525e-06,
|
| 1898 |
+
"loss": 1.1371,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6987577639751553,
|
| 1903 |
+
"grad_norm": 1.5940066135025674,
|
| 1904 |
+
"learning_rate": 5.0418024443786395e-06,
|
| 1905 |
+
"loss": 1.1547,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7013457556935818,
|
| 1910 |
+
"grad_norm": 1.8246740368118712,
|
| 1911 |
+
"learning_rate": 4.963518126557846e-06,
|
| 1912 |
+
"loss": 1.1669,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7039337474120083,
|
| 1917 |
+
"grad_norm": 1.583850157022352,
|
| 1918 |
+
"learning_rate": 4.885645208965779e-06,
|
| 1919 |
+
"loss": 1.1952,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7065217391304348,
|
| 1924 |
+
"grad_norm": 1.826552327257342,
|
| 1925 |
+
"learning_rate": 4.8081900525775984e-06,
|
| 1926 |
+
"loss": 1.1775,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7091097308488613,
|
| 1931 |
+
"grad_norm": 1.6908047939700899,
|
| 1932 |
+
"learning_rate": 4.731158984244042e-06,
|
| 1933 |
+
"loss": 1.2053,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7116977225672878,
|
| 1938 |
+
"grad_norm": 1.577115477870209,
|
| 1939 |
+
"learning_rate": 4.654558296174617e-06,
|
| 1940 |
+
"loss": 1.1756,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.7142857142857143,
|
| 1945 |
+
"grad_norm": 1.583278280234861,
|
| 1946 |
+
"learning_rate": 4.578394245423626e-06,
|
| 1947 |
+
"loss": 1.1486,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7168737060041408,
|
| 1952 |
+
"grad_norm": 1.5474720201156653,
|
| 1953 |
+
"learning_rate": 4.502673053379077e-06,
|
| 1954 |
+
"loss": 1.2029,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7194616977225673,
|
| 1959 |
+
"grad_norm": 1.9218329295357086,
|
| 1960 |
+
"learning_rate": 4.427400905254483e-06,
|
| 1961 |
+
"loss": 1.1592,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7220496894409938,
|
| 1966 |
+
"grad_norm": 1.5807402259350405,
|
| 1967 |
+
"learning_rate": 4.352583949583619e-06,
|
| 1968 |
+
"loss": 1.1443,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7246376811594203,
|
| 1973 |
+
"grad_norm": 1.5984439783246036,
|
| 1974 |
+
"learning_rate": 4.278228297718307e-06,
|
| 1975 |
+
"loss": 1.1852,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7272256728778468,
|
| 1980 |
+
"grad_norm": 1.562314289203796,
|
| 1981 |
+
"learning_rate": 4.204340023329215e-06,
|
| 1982 |
+
"loss": 1.1463,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7298136645962733,
|
| 1987 |
+
"grad_norm": 1.6156395775351333,
|
| 1988 |
+
"learning_rate": 4.130925161909716e-06,
|
| 1989 |
+
"loss": 1.1512,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7324016563146998,
|
| 1994 |
+
"grad_norm": 1.5829034816425451,
|
| 1995 |
+
"learning_rate": 4.057989710282897e-06,
|
| 1996 |
+
"loss": 1.1246,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7349896480331263,
|
| 2001 |
+
"grad_norm": 1.523735802005734,
|
| 2002 |
+
"learning_rate": 3.985539626111708e-06,
|
| 2003 |
+
"loss": 1.1668,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7375776397515528,
|
| 2008 |
+
"grad_norm": 1.505474487450996,
|
| 2009 |
+
"learning_rate": 3.913580827412334e-06,
|
| 2010 |
+
"loss": 1.1391,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7401656314699793,
|
| 2015 |
+
"grad_norm": 1.6890981988605818,
|
| 2016 |
+
"learning_rate": 3.842119192070762e-06,
|
| 2017 |
+
"loss": 1.151,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7427536231884058,
|
| 2022 |
+
"grad_norm": 1.6496446204834274,
|
| 2023 |
+
"learning_rate": 3.77116055736267e-06,
|
| 2024 |
+
"loss": 1.1906,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7453416149068323,
|
| 2029 |
+
"grad_norm": 1.62048985165388,
|
| 2030 |
+
"learning_rate": 3.7007107194765945e-06,
|
| 2031 |
+
"loss": 1.1676,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7479296066252588,
|
| 2036 |
+
"grad_norm": 1.6419977108085304,
|
| 2037 |
+
"learning_rate": 3.630775433040502e-06,
|
| 2038 |
+
"loss": 1.1397,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7505175983436853,
|
| 2043 |
+
"grad_norm": 1.5143054947159678,
|
| 2044 |
+
"learning_rate": 3.561360410651713e-06,
|
| 2045 |
+
"loss": 1.1569,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7531055900621118,
|
| 2050 |
+
"grad_norm": 1.5184069642552582,
|
| 2051 |
+
"learning_rate": 3.492471322410268e-06,
|
| 2052 |
+
"loss": 1.1661,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7556935817805382,
|
| 2057 |
+
"grad_norm": 1.740302385789126,
|
| 2058 |
+
"learning_rate": 3.4241137954557792e-06,
|
| 2059 |
+
"loss": 1.2001,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7582815734989649,
|
| 2064 |
+
"grad_norm": 1.735136651036564,
|
| 2065 |
+
"learning_rate": 3.3562934135077794e-06,
|
| 2066 |
+
"loss": 1.2071,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7608695652173914,
|
| 2071 |
+
"grad_norm": 1.512017941498727,
|
| 2072 |
+
"learning_rate": 3.2890157164096315e-06,
|
| 2073 |
+
"loss": 1.1562,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7634575569358178,
|
| 2078 |
+
"grad_norm": 1.56398858453209,
|
| 2079 |
+
"learning_rate": 3.2222861996759912e-06,
|
| 2080 |
+
"loss": 1.1755,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7660455486542443,
|
| 2085 |
+
"grad_norm": 1.8429752138603448,
|
| 2086 |
+
"learning_rate": 3.156110314043933e-06,
|
| 2087 |
+
"loss": 1.1382,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7686335403726708,
|
| 2092 |
+
"grad_norm": 1.531763749709719,
|
| 2093 |
+
"learning_rate": 3.0904934650276897e-06,
|
| 2094 |
+
"loss": 1.1716,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7712215320910973,
|
| 2099 |
+
"grad_norm": 1.570674065613626,
|
| 2100 |
+
"learning_rate": 3.025441012477126e-06,
|
| 2101 |
+
"loss": 1.1911,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7738095238095238,
|
| 2106 |
+
"grad_norm": 1.508664008364339,
|
| 2107 |
+
"learning_rate": 2.9609582701399266e-06,
|
| 2108 |
+
"loss": 1.1716,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7763975155279503,
|
| 2113 |
+
"grad_norm": 1.4777031849165594,
|
| 2114 |
+
"learning_rate": 2.89705050522753e-06,
|
| 2115 |
+
"loss": 1.1456,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7789855072463768,
|
| 2120 |
+
"grad_norm": 1.5683218260891874,
|
| 2121 |
+
"learning_rate": 2.833722937984893e-06,
|
| 2122 |
+
"loss": 1.1936,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7815734989648033,
|
| 2127 |
+
"grad_norm": 1.558998170252945,
|
| 2128 |
+
"learning_rate": 2.7709807412640745e-06,
|
| 2129 |
+
"loss": 1.1532,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7841614906832298,
|
| 2134 |
+
"grad_norm": 1.5325074641404255,
|
| 2135 |
+
"learning_rate": 2.70882904010171e-06,
|
| 2136 |
+
"loss": 1.1544,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7867494824016563,
|
| 2141 |
+
"grad_norm": 1.587667874017625,
|
| 2142 |
+
"learning_rate": 2.6472729113003614e-06,
|
| 2143 |
+
"loss": 1.1363,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7893374741200828,
|
| 2148 |
+
"grad_norm": 1.487145300264583,
|
| 2149 |
+
"learning_rate": 2.5863173830138212e-06,
|
| 2150 |
+
"loss": 1.1609,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7919254658385093,
|
| 2155 |
+
"grad_norm": 1.5296852825320244,
|
| 2156 |
+
"learning_rate": 2.5259674343364104e-06,
|
| 2157 |
+
"loss": 1.1166,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7945134575569358,
|
| 2162 |
+
"grad_norm": 1.523019173708786,
|
| 2163 |
+
"learning_rate": 2.4662279948962497e-06,
|
| 2164 |
+
"loss": 1.1443,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7971014492753623,
|
| 2169 |
+
"grad_norm": 1.6276678280152375,
|
| 2170 |
+
"learning_rate": 2.4071039444526046e-06,
|
| 2171 |
+
"loss": 1.2047,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7996894409937888,
|
| 2176 |
+
"grad_norm": 1.5518526603934675,
|
| 2177 |
+
"learning_rate": 2.3486001124972636e-06,
|
| 2178 |
+
"loss": 1.1357,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8022774327122153,
|
| 2183 |
+
"grad_norm": 1.5395591809545814,
|
| 2184 |
+
"learning_rate": 2.290721277860064e-06,
|
| 2185 |
+
"loss": 1.1562,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8048654244306418,
|
| 2190 |
+
"grad_norm": 1.5155042472429905,
|
| 2191 |
+
"learning_rate": 2.233472168318529e-06,
|
| 2192 |
+
"loss": 1.1257,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8074534161490683,
|
| 2197 |
+
"grad_norm": 1.6546895809171993,
|
| 2198 |
+
"learning_rate": 2.176857460211693e-06,
|
| 2199 |
+
"loss": 1.1813,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8100414078674948,
|
| 2204 |
+
"grad_norm": 1.8271870019910332,
|
| 2205 |
+
"learning_rate": 2.120881778058109e-06,
|
| 2206 |
+
"loss": 1.1749,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.8126293995859213,
|
| 2211 |
+
"grad_norm": 1.5721705885784398,
|
| 2212 |
+
"learning_rate": 2.0655496941780993e-06,
|
| 2213 |
+
"loss": 1.1332,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8152173913043478,
|
| 2218 |
+
"grad_norm": 1.5777293602086202,
|
| 2219 |
+
"learning_rate": 2.0108657283202783e-06,
|
| 2220 |
+
"loss": 1.1468,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8178053830227743,
|
| 2225 |
+
"grad_norm": 1.5754598191510423,
|
| 2226 |
+
"learning_rate": 1.9568343472923524e-06,
|
| 2227 |
+
"loss": 1.1793,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8203933747412008,
|
| 2232 |
+
"grad_norm": 1.6259405743934179,
|
| 2233 |
+
"learning_rate": 1.903459964596267e-06,
|
| 2234 |
+
"loss": 1.1499,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8229813664596274,
|
| 2239 |
+
"grad_norm": 1.6594307558296557,
|
| 2240 |
+
"learning_rate": 1.8507469400676736e-06,
|
| 2241 |
+
"loss": 1.2668,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8255693581780539,
|
| 2246 |
+
"grad_norm": 1.5346716651192356,
|
| 2247 |
+
"learning_rate": 1.798699579519817e-06,
|
| 2248 |
+
"loss": 1.1057,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8281573498964804,
|
| 2253 |
+
"grad_norm": 1.5898990173488856,
|
| 2254 |
+
"learning_rate": 1.7473221343918057e-06,
|
| 2255 |
+
"loss": 1.154,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8307453416149069,
|
| 2260 |
+
"grad_norm": 1.6218963246045284,
|
| 2261 |
+
"learning_rate": 1.6966188014013574e-06,
|
| 2262 |
+
"loss": 1.203,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8333333333333334,
|
| 2267 |
+
"grad_norm": 1.5327256183704974,
|
| 2268 |
+
"learning_rate": 1.6465937222019745e-06,
|
| 2269 |
+
"loss": 1.0827,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8359213250517599,
|
| 2274 |
+
"grad_norm": 1.5484614153825313,
|
| 2275 |
+
"learning_rate": 1.597250983044637e-06,
|
| 2276 |
+
"loss": 1.1512,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8385093167701864,
|
| 2281 |
+
"grad_norm": 1.547251548669474,
|
| 2282 |
+
"learning_rate": 1.5485946144440357e-06,
|
| 2283 |
+
"loss": 1.1621,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8410973084886129,
|
| 2288 |
+
"grad_norm": 1.5301077366867935,
|
| 2289 |
+
"learning_rate": 1.500628590849339e-06,
|
| 2290 |
+
"loss": 1.1,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8436853002070394,
|
| 2295 |
+
"grad_norm": 1.5775881041342432,
|
| 2296 |
+
"learning_rate": 1.4533568303195333e-06,
|
| 2297 |
+
"loss": 1.1629,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8462732919254659,
|
| 2302 |
+
"grad_norm": 1.6513355081710792,
|
| 2303 |
+
"learning_rate": 1.4067831942033904e-06,
|
| 2304 |
+
"loss": 1.1563,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8488612836438924,
|
| 2309 |
+
"grad_norm": 1.6098056906806337,
|
| 2310 |
+
"learning_rate": 1.3609114868240537e-06,
|
| 2311 |
+
"loss": 1.1659,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8514492753623188,
|
| 2316 |
+
"grad_norm": 1.5232813594236112,
|
| 2317 |
+
"learning_rate": 1.3157454551682824e-06,
|
| 2318 |
+
"loss": 1.1068,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8540372670807453,
|
| 2323 |
+
"grad_norm": 1.596056108948684,
|
| 2324 |
+
"learning_rate": 1.2712887885803937e-06,
|
| 2325 |
+
"loss": 1.1769,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8566252587991718,
|
| 2330 |
+
"grad_norm": 1.5685789393387255,
|
| 2331 |
+
"learning_rate": 1.227545118460889e-06,
|
| 2332 |
+
"loss": 1.1644,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8592132505175983,
|
| 2337 |
+
"grad_norm": 1.6464311208376787,
|
| 2338 |
+
"learning_rate": 1.184518017969829e-06,
|
| 2339 |
+
"loss": 1.1533,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8618012422360248,
|
| 2344 |
+
"grad_norm": 1.5274475944853032,
|
| 2345 |
+
"learning_rate": 1.1422110017349676e-06,
|
| 2346 |
+
"loss": 1.1709,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8643892339544513,
|
| 2351 |
+
"grad_norm": 1.5509872198392205,
|
| 2352 |
+
"learning_rate": 1.100627525564668e-06,
|
| 2353 |
+
"loss": 1.1344,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8669772256728778,
|
| 2358 |
+
"grad_norm": 1.504837308033305,
|
| 2359 |
+
"learning_rate": 1.0597709861656057e-06,
|
| 2360 |
+
"loss": 1.1358,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8695652173913043,
|
| 2365 |
+
"grad_norm": 1.6991797160404174,
|
| 2366 |
+
"learning_rate": 1.0196447208653193e-06,
|
| 2367 |
+
"loss": 1.1857,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8721532091097308,
|
| 2372 |
+
"grad_norm": 1.5275470265589042,
|
| 2373 |
+
"learning_rate": 9.802520073396016e-07,
|
| 2374 |
+
"loss": 1.1314,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8747412008281573,
|
| 2379 |
+
"grad_norm": 1.6006710258152612,
|
| 2380 |
+
"learning_rate": 9.415960633447674e-07,
|
| 2381 |
+
"loss": 1.1662,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8773291925465838,
|
| 2386 |
+
"grad_norm": 1.716257348826808,
|
| 2387 |
+
"learning_rate": 9.036800464548157e-07,
|
| 2388 |
+
"loss": 1.1509,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8799171842650103,
|
| 2393 |
+
"grad_norm": 1.5117187651990462,
|
| 2394 |
+
"learning_rate": 8.665070538035037e-07,
|
| 2395 |
+
"loss": 1.1591,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8825051759834368,
|
| 2400 |
+
"grad_norm": 1.5458076964833938,
|
| 2401 |
+
"learning_rate": 8.300801218313548e-07,
|
| 2402 |
+
"loss": 1.1849,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.8850931677018633,
|
| 2407 |
+
"grad_norm": 1.6346788638439136,
|
| 2408 |
+
"learning_rate": 7.944022260376416e-07,
|
| 2409 |
+
"loss": 1.1733,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8876811594202898,
|
| 2414 |
+
"grad_norm": 1.4857294789989146,
|
| 2415 |
+
"learning_rate": 7.594762807373313e-07,
|
| 2416 |
+
"loss": 1.1091,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8902691511387164,
|
| 2421 |
+
"grad_norm": 1.544124199759589,
|
| 2422 |
+
"learning_rate": 7.253051388230248e-07,
|
| 2423 |
+
"loss": 1.1374,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8928571428571429,
|
| 2428 |
+
"grad_norm": 1.8377202213987396,
|
| 2429 |
+
"learning_rate": 6.918915915319302e-07,
|
| 2430 |
+
"loss": 1.1643,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8954451345755694,
|
| 2435 |
+
"grad_norm": 1.5475759630185042,
|
| 2436 |
+
"learning_rate": 6.592383682178593e-07,
|
| 2437 |
+
"loss": 1.154,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8980331262939959,
|
| 2442 |
+
"grad_norm": 1.521052252376937,
|
| 2443 |
+
"learning_rate": 6.273481361282807e-07,
|
| 2444 |
+
"loss": 1.1293,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9006211180124224,
|
| 2449 |
+
"grad_norm": 1.5185508439469686,
|
| 2450 |
+
"learning_rate": 5.96223500186458e-07,
|
| 2451 |
+
"loss": 1.1744,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9032091097308489,
|
| 2456 |
+
"grad_norm": 1.7082962342593018,
|
| 2457 |
+
"learning_rate": 5.658670027786561e-07,
|
| 2458 |
+
"loss": 1.1756,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9057971014492754,
|
| 2463 |
+
"grad_norm": 1.5757735686564247,
|
| 2464 |
+
"learning_rate": 5.362811235464727e-07,
|
| 2465 |
+
"loss": 1.1849,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9083850931677019,
|
| 2470 |
+
"grad_norm": 1.62779706084884,
|
| 2471 |
+
"learning_rate": 5.074682791842988e-07,
|
| 2472 |
+
"loss": 1.156,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9109730848861284,
|
| 2477 |
+
"grad_norm": 1.5312273750679573,
|
| 2478 |
+
"learning_rate": 4.794308232419065e-07,
|
| 2479 |
+
"loss": 1.1497,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9135610766045549,
|
| 2484 |
+
"grad_norm": 1.5488311929264333,
|
| 2485 |
+
"learning_rate": 4.521710459322015e-07,
|
| 2486 |
+
"loss": 1.1561,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9161490683229814,
|
| 2491 |
+
"grad_norm": 1.6641638382041457,
|
| 2492 |
+
"learning_rate": 4.256911739441505e-07,
|
| 2493 |
+
"loss": 1.1556,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9187370600414079,
|
| 2498 |
+
"grad_norm": 1.6032525408960623,
|
| 2499 |
+
"learning_rate": 3.999933702608949e-07,
|
| 2500 |
+
"loss": 1.1168,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9213250517598344,
|
| 2505 |
+
"grad_norm": 1.5699960465562155,
|
| 2506 |
+
"learning_rate": 3.7507973398307584e-07,
|
| 2507 |
+
"loss": 1.1501,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9239130434782609,
|
| 2512 |
+
"grad_norm": 1.5994478108257122,
|
| 2513 |
+
"learning_rate": 3.5095230015736113e-07,
|
| 2514 |
+
"loss": 1.1159,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9265010351966874,
|
| 2519 |
+
"grad_norm": 1.5126252156346516,
|
| 2520 |
+
"learning_rate": 3.276130396102217e-07,
|
| 2521 |
+
"loss": 1.094,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9290890269151139,
|
| 2526 |
+
"grad_norm": 1.5178216629033467,
|
| 2527 |
+
"learning_rate": 3.0506385878694255e-07,
|
| 2528 |
+
"loss": 1.1273,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9316770186335404,
|
| 2533 |
+
"grad_norm": 1.5185613841579717,
|
| 2534 |
+
"learning_rate": 2.8330659959589944e-07,
|
| 2535 |
+
"loss": 1.1456,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9342650103519669,
|
| 2540 |
+
"grad_norm": 1.6347573627683816,
|
| 2541 |
+
"learning_rate": 2.623430392581061e-07,
|
| 2542 |
+
"loss": 1.1372,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9368530020703933,
|
| 2547 |
+
"grad_norm": 1.5838574431621013,
|
| 2548 |
+
"learning_rate": 2.421748901620369e-07,
|
| 2549 |
+
"loss": 1.1273,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9394409937888198,
|
| 2554 |
+
"grad_norm": 1.6461360935258826,
|
| 2555 |
+
"learning_rate": 2.2280379972375643e-07,
|
| 2556 |
+
"loss": 1.1199,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9420289855072463,
|
| 2561 |
+
"grad_norm": 1.5251560302510245,
|
| 2562 |
+
"learning_rate": 2.0423135025235298e-07,
|
| 2563 |
+
"loss": 1.1869,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9446169772256728,
|
| 2568 |
+
"grad_norm": 1.74800052983838,
|
| 2569 |
+
"learning_rate": 1.8645905882068605e-07,
|
| 2570 |
+
"loss": 1.1277,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9472049689440993,
|
| 2575 |
+
"grad_norm": 1.6647572293390658,
|
| 2576 |
+
"learning_rate": 1.6948837714146793e-07,
|
| 2577 |
+
"loss": 1.1026,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9497929606625258,
|
| 2582 |
+
"grad_norm": 1.6498964813278796,
|
| 2583 |
+
"learning_rate": 1.533206914486818e-07,
|
| 2584 |
+
"loss": 1.1413,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9523809523809523,
|
| 2589 |
+
"grad_norm": 1.6314577909166261,
|
| 2590 |
+
"learning_rate": 1.3795732238434557e-07,
|
| 2591 |
+
"loss": 1.1441,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9549689440993789,
|
| 2596 |
+
"grad_norm": 1.5601815262952223,
|
| 2597 |
+
"learning_rate": 1.2339952489064056e-07,
|
| 2598 |
+
"loss": 1.2041,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9575569358178054,
|
| 2603 |
+
"grad_norm": 1.5625160012256416,
|
| 2604 |
+
"learning_rate": 1.0964848810740114e-07,
|
| 2605 |
+
"loss": 1.1631,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9601449275362319,
|
| 2610 |
+
"grad_norm": 1.7039383904038539,
|
| 2611 |
+
"learning_rate": 9.670533527498139e-08,
|
| 2612 |
+
"loss": 1.1729,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9627329192546584,
|
| 2617 |
+
"grad_norm": 1.59975036577853,
|
| 2618 |
+
"learning_rate": 8.457112364250064e-08,
|
| 2619 |
+
"loss": 1.166,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9653209109730849,
|
| 2624 |
+
"grad_norm": 1.601576491104471,
|
| 2625 |
+
"learning_rate": 7.324684438148933e-08,
|
| 2626 |
+
"loss": 1.152,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9679089026915114,
|
| 2631 |
+
"grad_norm": 1.5522923849319805,
|
| 2632 |
+
"learning_rate": 6.273342250492254e-08,
|
| 2633 |
+
"loss": 1.1144,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9704968944099379,
|
| 2638 |
+
"grad_norm": 1.5598616648811972,
|
| 2639 |
+
"learning_rate": 5.303171679166053e-08,
|
| 2640 |
+
"loss": 1.2196,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9730848861283644,
|
| 2645 |
+
"grad_norm": 1.6532449106020801,
|
| 2646 |
+
"learning_rate": 4.414251971630035e-08,
|
| 2647 |
+
"loss": 1.1882,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9756728778467909,
|
| 2652 |
+
"grad_norm": 1.5427460516528153,
|
| 2653 |
+
"learning_rate": 3.606655738444653e-08,
|
| 2654 |
+
"loss": 1.2124,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9782608695652174,
|
| 2659 |
+
"grad_norm": 1.5671272147865478,
|
| 2660 |
+
"learning_rate": 2.8804489473395205e-08,
|
| 2661 |
+
"loss": 1.1492,
|
| 2662 |
+
"step": 1890
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9808488612836439,
|
| 2666 |
+
"grad_norm": 1.5788966872678778,
|
| 2667 |
+
"learning_rate": 2.235690917825384e-08,
|
| 2668 |
+
"loss": 1.1441,
|
| 2669 |
+
"step": 1895
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.9834368530020704,
|
| 2673 |
+
"grad_norm": 1.6331717365418186,
|
| 2674 |
+
"learning_rate": 1.672434316348337e-08,
|
| 2675 |
+
"loss": 1.1179,
|
| 2676 |
+
"step": 1900
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9860248447204969,
|
| 2680 |
+
"grad_norm": 1.4925310986828344,
|
| 2681 |
+
"learning_rate": 1.1907251519877039e-08,
|
| 2682 |
+
"loss": 1.1612,
|
| 2683 |
+
"step": 1905
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9886128364389234,
|
| 2687 |
+
"grad_norm": 1.7342337869712523,
|
| 2688 |
+
"learning_rate": 7.906027726981568e-09,
|
| 2689 |
+
"loss": 1.1252,
|
| 2690 |
+
"step": 1910
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.9912008281573499,
|
| 2694 |
+
"grad_norm": 1.6269325139131876,
|
| 2695 |
+
"learning_rate": 4.720998620955097e-09,
|
| 2696 |
+
"loss": 1.1918,
|
| 2697 |
+
"step": 1915
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9937888198757764,
|
| 2701 |
+
"grad_norm": 1.4814594112612938,
|
| 2702 |
+
"learning_rate": 2.3524243678685333e-09,
|
| 2703 |
+
"loss": 1.1315,
|
| 2704 |
+
"step": 1920
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.9963768115942029,
|
| 2708 |
+
"grad_norm": 1.5876922772379496,
|
| 2709 |
+
"learning_rate": 8.004984424558881e-10,
|
| 2710 |
+
"loss": 1.1227,
|
| 2711 |
+
"step": 1925
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.9989648033126294,
|
| 2715 |
+
"grad_norm": 1.694840692501587,
|
| 2716 |
+
"learning_rate": 6.534761230914122e-11,
|
| 2717 |
+
"loss": 1.1366,
|
| 2718 |
+
"step": 1930
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 1.0,
|
| 2722 |
+
"eval_loss": 1.1816973686218262,
|
| 2723 |
+
"eval_runtime": 1548.7687,
|
| 2724 |
+
"eval_samples_per_second": 17.673,
|
| 2725 |
+
"eval_steps_per_second": 0.553,
|
| 2726 |
+
"step": 1932
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 1.0,
|
| 2730 |
+
"step": 1932,
|
| 2731 |
+
"total_flos": 230071600152576.0,
|
| 2732 |
+
"train_loss": 1.2287210629593512,
|
| 2733 |
+
"train_runtime": 18714.7709,
|
| 2734 |
+
"train_samples_per_second": 3.303,
|
| 2735 |
+
"train_steps_per_second": 0.103
|
| 2736 |
+
}
|
| 2737 |
+
],
|
| 2738 |
+
"logging_steps": 5,
|
| 2739 |
+
"max_steps": 1932,
|
| 2740 |
+
"num_input_tokens_seen": 0,
|
| 2741 |
+
"num_train_epochs": 1,
|
| 2742 |
+
"save_steps": 500,
|
| 2743 |
+
"stateful_callbacks": {
|
| 2744 |
+
"TrainerControl": {
|
| 2745 |
+
"args": {
|
| 2746 |
+
"should_epoch_stop": false,
|
| 2747 |
+
"should_evaluate": false,
|
| 2748 |
+
"should_log": false,
|
| 2749 |
+
"should_save": true,
|
| 2750 |
+
"should_training_stop": true
|
| 2751 |
+
},
|
| 2752 |
+
"attributes": {}
|
| 2753 |
+
}
|
| 2754 |
+
},
|
| 2755 |
+
"total_flos": 230071600152576.0,
|
| 2756 |
+
"train_batch_size": 8,
|
| 2757 |
+
"trial_name": null,
|
| 2758 |
+
"trial_params": null
|
| 2759 |
+
}
|