Commit
•
84ca869
1
Parent(s):
3378c66
Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +10 -0
- train_results.json +9 -0
- trainer_state.json +2325 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-2-7b-hf
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: llama2-7b-sft-ultrachat
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for llama2-7b-sft-ultrachat
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf).
|
| 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/llama2-7b-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 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yifanwang/huggingface/runs/hevvxhft)
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.0
|
| 37 |
+
- Transformers: 4.46.3
|
| 38 |
+
- Pytorch: 2.1.2+cu121
|
| 39 |
+
- Datasets: 3.1.0
|
| 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": 165422873640960.0,
|
| 4 |
+
"train_loss": 0.9319284432040059,
|
| 5 |
+
"train_runtime": 12380.9765,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 4.197,
|
| 8 |
+
"train_steps_per_second": 0.131
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"max_length": 4096,
|
| 6 |
+
"pad_token_id": 0,
|
| 7 |
+
"temperature": 0.6,
|
| 8 |
+
"top_p": 0.9,
|
| 9 |
+
"transformers_version": "4.46.3"
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 165422873640960.0,
|
| 4 |
+
"train_loss": 0.9319284432040059,
|
| 5 |
+
"train_runtime": 12380.9765,
|
| 6 |
+
"train_samples": 51966,
|
| 7 |
+
"train_samples_per_second": 4.197,
|
| 8 |
+
"train_steps_per_second": 0.131
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2325 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1624,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006157635467980296,
|
| 13 |
+
"grad_norm": 7.310926153142698,
|
| 14 |
+
"learning_rate": 1.226993865030675e-07,
|
| 15 |
+
"loss": 1.5491,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003078817733990148,
|
| 20 |
+
"grad_norm": 7.35403837138417,
|
| 21 |
+
"learning_rate": 6.134969325153375e-07,
|
| 22 |
+
"loss": 1.5843,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006157635467980296,
|
| 27 |
+
"grad_norm": 6.803167318550182,
|
| 28 |
+
"learning_rate": 1.226993865030675e-06,
|
| 29 |
+
"loss": 1.5309,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009236453201970444,
|
| 34 |
+
"grad_norm": 5.835727118488878,
|
| 35 |
+
"learning_rate": 1.8404907975460124e-06,
|
| 36 |
+
"loss": 1.3594,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012315270935960592,
|
| 41 |
+
"grad_norm": 3.698538754344085,
|
| 42 |
+
"learning_rate": 2.45398773006135e-06,
|
| 43 |
+
"loss": 1.2405,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01539408866995074,
|
| 48 |
+
"grad_norm": 2.041108035069763,
|
| 49 |
+
"learning_rate": 3.0674846625766875e-06,
|
| 50 |
+
"loss": 1.1387,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01847290640394089,
|
| 55 |
+
"grad_norm": 1.0814828345339402,
|
| 56 |
+
"learning_rate": 3.680981595092025e-06,
|
| 57 |
+
"loss": 1.0108,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.021551724137931036,
|
| 62 |
+
"grad_norm": 0.8687795931835679,
|
| 63 |
+
"learning_rate": 4.294478527607362e-06,
|
| 64 |
+
"loss": 0.9621,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.024630541871921183,
|
| 69 |
+
"grad_norm": 0.8664847350475609,
|
| 70 |
+
"learning_rate": 4.9079754601227e-06,
|
| 71 |
+
"loss": 1.0134,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02770935960591133,
|
| 76 |
+
"grad_norm": 0.8801651222007403,
|
| 77 |
+
"learning_rate": 5.521472392638038e-06,
|
| 78 |
+
"loss": 0.9283,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03078817733990148,
|
| 83 |
+
"grad_norm": 0.8636788681790382,
|
| 84 |
+
"learning_rate": 6.134969325153375e-06,
|
| 85 |
+
"loss": 0.9328,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.033866995073891626,
|
| 90 |
+
"grad_norm": 0.8253490914438462,
|
| 91 |
+
"learning_rate": 6.748466257668712e-06,
|
| 92 |
+
"loss": 0.9051,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03694581280788178,
|
| 97 |
+
"grad_norm": 0.8184367283533114,
|
| 98 |
+
"learning_rate": 7.36196319018405e-06,
|
| 99 |
+
"loss": 1.0101,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04002463054187192,
|
| 104 |
+
"grad_norm": 0.8465714119368565,
|
| 105 |
+
"learning_rate": 7.975460122699386e-06,
|
| 106 |
+
"loss": 0.9229,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04310344827586207,
|
| 111 |
+
"grad_norm": 0.8285519257452998,
|
| 112 |
+
"learning_rate": 8.588957055214725e-06,
|
| 113 |
+
"loss": 0.9491,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.046182266009852216,
|
| 118 |
+
"grad_norm": 0.769823593105267,
|
| 119 |
+
"learning_rate": 9.202453987730062e-06,
|
| 120 |
+
"loss": 0.9726,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04926108374384237,
|
| 125 |
+
"grad_norm": 0.8064693479221889,
|
| 126 |
+
"learning_rate": 9.8159509202454e-06,
|
| 127 |
+
"loss": 0.9667,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05233990147783251,
|
| 132 |
+
"grad_norm": 1.0363791467221333,
|
| 133 |
+
"learning_rate": 1.0429447852760737e-05,
|
| 134 |
+
"loss": 0.9209,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05541871921182266,
|
| 139 |
+
"grad_norm": 0.7842863481909221,
|
| 140 |
+
"learning_rate": 1.1042944785276076e-05,
|
| 141 |
+
"loss": 0.9398,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.058497536945812806,
|
| 146 |
+
"grad_norm": 0.7367025450271165,
|
| 147 |
+
"learning_rate": 1.1656441717791411e-05,
|
| 148 |
+
"loss": 0.9461,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06157635467980296,
|
| 153 |
+
"grad_norm": 0.7689269543001077,
|
| 154 |
+
"learning_rate": 1.226993865030675e-05,
|
| 155 |
+
"loss": 0.9177,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06465517241379311,
|
| 160 |
+
"grad_norm": 0.8274857270444617,
|
| 161 |
+
"learning_rate": 1.2883435582822085e-05,
|
| 162 |
+
"loss": 0.9525,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06773399014778325,
|
| 167 |
+
"grad_norm": 0.7618255279710626,
|
| 168 |
+
"learning_rate": 1.3496932515337424e-05,
|
| 169 |
+
"loss": 0.9264,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0708128078817734,
|
| 174 |
+
"grad_norm": 0.7980402273299848,
|
| 175 |
+
"learning_rate": 1.4110429447852763e-05,
|
| 176 |
+
"loss": 0.9339,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07389162561576355,
|
| 181 |
+
"grad_norm": 0.8376644789397331,
|
| 182 |
+
"learning_rate": 1.47239263803681e-05,
|
| 183 |
+
"loss": 0.8859,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0769704433497537,
|
| 188 |
+
"grad_norm": 0.8130407079299852,
|
| 189 |
+
"learning_rate": 1.5337423312883436e-05,
|
| 190 |
+
"loss": 0.9146,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08004926108374384,
|
| 195 |
+
"grad_norm": 0.7956012283141837,
|
| 196 |
+
"learning_rate": 1.5950920245398772e-05,
|
| 197 |
+
"loss": 0.9154,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08312807881773399,
|
| 202 |
+
"grad_norm": 0.7948481430201038,
|
| 203 |
+
"learning_rate": 1.656441717791411e-05,
|
| 204 |
+
"loss": 0.9576,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08620689655172414,
|
| 209 |
+
"grad_norm": 0.8257741416380704,
|
| 210 |
+
"learning_rate": 1.717791411042945e-05,
|
| 211 |
+
"loss": 0.9515,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08928571428571429,
|
| 216 |
+
"grad_norm": 0.8290236573203579,
|
| 217 |
+
"learning_rate": 1.7791411042944788e-05,
|
| 218 |
+
"loss": 0.9678,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09236453201970443,
|
| 223 |
+
"grad_norm": 0.9224233612252317,
|
| 224 |
+
"learning_rate": 1.8404907975460123e-05,
|
| 225 |
+
"loss": 0.9644,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09544334975369458,
|
| 230 |
+
"grad_norm": 0.8138389103383028,
|
| 231 |
+
"learning_rate": 1.9018404907975462e-05,
|
| 232 |
+
"loss": 0.9028,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09852216748768473,
|
| 237 |
+
"grad_norm": 0.8149251514397249,
|
| 238 |
+
"learning_rate": 1.96319018404908e-05,
|
| 239 |
+
"loss": 0.9381,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10160098522167488,
|
| 244 |
+
"grad_norm": 0.857572912046182,
|
| 245 |
+
"learning_rate": 1.999990752408443e-05,
|
| 246 |
+
"loss": 0.9526,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10467980295566502,
|
| 251 |
+
"grad_norm": 0.8424271480084681,
|
| 252 |
+
"learning_rate": 1.9998867189676517e-05,
|
| 253 |
+
"loss": 0.9598,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10775862068965517,
|
| 258 |
+
"grad_norm": 0.7367129010600535,
|
| 259 |
+
"learning_rate": 1.999667104662347e-05,
|
| 260 |
+
"loss": 0.9543,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11083743842364532,
|
| 265 |
+
"grad_norm": 0.7601180219708648,
|
| 266 |
+
"learning_rate": 1.9993319348786157e-05,
|
| 267 |
+
"loss": 0.9672,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11391625615763547,
|
| 272 |
+
"grad_norm": 0.796186238392561,
|
| 273 |
+
"learning_rate": 1.9988812483600597e-05,
|
| 274 |
+
"loss": 0.8957,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11699507389162561,
|
| 279 |
+
"grad_norm": 0.7747918983305362,
|
| 280 |
+
"learning_rate": 1.9983150972033186e-05,
|
| 281 |
+
"loss": 0.963,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12007389162561577,
|
| 286 |
+
"grad_norm": 0.8300480182123051,
|
| 287 |
+
"learning_rate": 1.9976335468520452e-05,
|
| 288 |
+
"loss": 0.9824,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12315270935960591,
|
| 293 |
+
"grad_norm": 0.8310354348747966,
|
| 294 |
+
"learning_rate": 1.9968366760893437e-05,
|
| 295 |
+
"loss": 0.9692,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12623152709359606,
|
| 300 |
+
"grad_norm": 0.8445303140068814,
|
| 301 |
+
"learning_rate": 1.9959245770286602e-05,
|
| 302 |
+
"loss": 0.9546,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12931034482758622,
|
| 307 |
+
"grad_norm": 0.8979903129953134,
|
| 308 |
+
"learning_rate": 1.994897355103136e-05,
|
| 309 |
+
"loss": 0.9375,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.13238916256157635,
|
| 314 |
+
"grad_norm": 0.7465524065375957,
|
| 315 |
+
"learning_rate": 1.9937551290534208e-05,
|
| 316 |
+
"loss": 0.9057,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1354679802955665,
|
| 321 |
+
"grad_norm": 0.7497704866674313,
|
| 322 |
+
"learning_rate": 1.9924980309139455e-05,
|
| 323 |
+
"loss": 0.9434,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.13854679802955666,
|
| 328 |
+
"grad_norm": 0.7076359322846222,
|
| 329 |
+
"learning_rate": 1.9911262059976614e-05,
|
| 330 |
+
"loss": 0.9462,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1416256157635468,
|
| 335 |
+
"grad_norm": 0.7848088777070403,
|
| 336 |
+
"learning_rate": 1.9896398128792413e-05,
|
| 337 |
+
"loss": 0.9476,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14470443349753695,
|
| 342 |
+
"grad_norm": 0.8165026350741112,
|
| 343 |
+
"learning_rate": 1.988039023376751e-05,
|
| 344 |
+
"loss": 0.9527,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1477832512315271,
|
| 349 |
+
"grad_norm": 0.8498025037832465,
|
| 350 |
+
"learning_rate": 1.9863240225317868e-05,
|
| 351 |
+
"loss": 1.0053,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15086206896551724,
|
| 356 |
+
"grad_norm": 0.7353554935083384,
|
| 357 |
+
"learning_rate": 1.984495008588086e-05,
|
| 358 |
+
"loss": 0.9222,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1539408866995074,
|
| 363 |
+
"grad_norm": 0.7553279060128398,
|
| 364 |
+
"learning_rate": 1.982552192968612e-05,
|
| 365 |
+
"loss": 0.9732,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15701970443349753,
|
| 370 |
+
"grad_norm": 0.7548683527271752,
|
| 371 |
+
"learning_rate": 1.9804958002511137e-05,
|
| 372 |
+
"loss": 0.9459,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.16009852216748768,
|
| 377 |
+
"grad_norm": 0.7794156820504012,
|
| 378 |
+
"learning_rate": 1.9783260681421667e-05,
|
| 379 |
+
"loss": 0.9573,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16317733990147784,
|
| 384 |
+
"grad_norm": 0.7422263502441188,
|
| 385 |
+
"learning_rate": 1.9760432474496963e-05,
|
| 386 |
+
"loss": 0.9375,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16625615763546797,
|
| 391 |
+
"grad_norm": 0.7259884159865896,
|
| 392 |
+
"learning_rate": 1.973647602053984e-05,
|
| 393 |
+
"loss": 0.9483,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16933497536945813,
|
| 398 |
+
"grad_norm": 0.7801247475320573,
|
| 399 |
+
"learning_rate": 1.9711394088771658e-05,
|
| 400 |
+
"loss": 0.9663,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.1724137931034483,
|
| 405 |
+
"grad_norm": 0.8468086332749512,
|
| 406 |
+
"learning_rate": 1.9685189578512206e-05,
|
| 407 |
+
"loss": 0.9581,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.17549261083743842,
|
| 412 |
+
"grad_norm": 0.760222004142156,
|
| 413 |
+
"learning_rate": 1.9657865518844578e-05,
|
| 414 |
+
"loss": 0.9405,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17857142857142858,
|
| 419 |
+
"grad_norm": 0.818552353689237,
|
| 420 |
+
"learning_rate": 1.962942506826501e-05,
|
| 421 |
+
"loss": 0.9436,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1816502463054187,
|
| 426 |
+
"grad_norm": 0.7270444014459365,
|
| 427 |
+
"learning_rate": 1.9599871514317785e-05,
|
| 428 |
+
"loss": 0.9587,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.18472906403940886,
|
| 433 |
+
"grad_norm": 0.7569082732767817,
|
| 434 |
+
"learning_rate": 1.9569208273215204e-05,
|
| 435 |
+
"loss": 0.9092,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18780788177339902,
|
| 440 |
+
"grad_norm": 0.8035781870618336,
|
| 441 |
+
"learning_rate": 1.953743888944271e-05,
|
| 442 |
+
"loss": 0.9207,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.19088669950738915,
|
| 447 |
+
"grad_norm": 0.798197452115362,
|
| 448 |
+
"learning_rate": 1.950456703534915e-05,
|
| 449 |
+
"loss": 0.9722,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1939655172413793,
|
| 454 |
+
"grad_norm": 0.7911708031305178,
|
| 455 |
+
"learning_rate": 1.9470596510722285e-05,
|
| 456 |
+
"loss": 0.9088,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.19704433497536947,
|
| 461 |
+
"grad_norm": 0.7671171753975288,
|
| 462 |
+
"learning_rate": 1.9435531242349545e-05,
|
| 463 |
+
"loss": 1.0116,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2001231527093596,
|
| 468 |
+
"grad_norm": 0.7471976542017915,
|
| 469 |
+
"learning_rate": 1.9399375283564134e-05,
|
| 470 |
+
"loss": 0.9443,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.20320197044334976,
|
| 475 |
+
"grad_norm": 0.7932187536453967,
|
| 476 |
+
"learning_rate": 1.9362132813776472e-05,
|
| 477 |
+
"loss": 0.937,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2062807881773399,
|
| 482 |
+
"grad_norm": 0.7683161798892411,
|
| 483 |
+
"learning_rate": 1.9323808137991084e-05,
|
| 484 |
+
"loss": 0.9485,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.20935960591133004,
|
| 489 |
+
"grad_norm": 0.7661725970320424,
|
| 490 |
+
"learning_rate": 1.9284405686308982e-05,
|
| 491 |
+
"loss": 0.9133,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2124384236453202,
|
| 496 |
+
"grad_norm": 0.7873862980673396,
|
| 497 |
+
"learning_rate": 1.924393001341555e-05,
|
| 498 |
+
"loss": 0.9207,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.21551724137931033,
|
| 503 |
+
"grad_norm": 0.7452478426256695,
|
| 504 |
+
"learning_rate": 1.9202385798054073e-05,
|
| 505 |
+
"loss": 0.9087,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2185960591133005,
|
| 510 |
+
"grad_norm": 0.7793073146063183,
|
| 511 |
+
"learning_rate": 1.9159777842484878e-05,
|
| 512 |
+
"loss": 0.9345,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.22167487684729065,
|
| 517 |
+
"grad_norm": 0.8169813616813546,
|
| 518 |
+
"learning_rate": 1.911611107193024e-05,
|
| 519 |
+
"loss": 0.9382,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.22475369458128078,
|
| 524 |
+
"grad_norm": 0.7266251337678646,
|
| 525 |
+
"learning_rate": 1.9071390534005045e-05,
|
| 526 |
+
"loss": 0.9732,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.22783251231527094,
|
| 531 |
+
"grad_norm": 0.7477667607003808,
|
| 532 |
+
"learning_rate": 1.9025621398133333e-05,
|
| 533 |
+
"loss": 0.9518,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2309113300492611,
|
| 538 |
+
"grad_norm": 0.9165568422039077,
|
| 539 |
+
"learning_rate": 1.8978808954950722e-05,
|
| 540 |
+
"loss": 0.9503,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.23399014778325122,
|
| 545 |
+
"grad_norm": 0.7529486101747109,
|
| 546 |
+
"learning_rate": 1.8930958615692854e-05,
|
| 547 |
+
"loss": 0.9591,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23706896551724138,
|
| 552 |
+
"grad_norm": 0.6888920479042022,
|
| 553 |
+
"learning_rate": 1.8882075911569887e-05,
|
| 554 |
+
"loss": 0.88,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.24014778325123154,
|
| 559 |
+
"grad_norm": 0.7255523898786278,
|
| 560 |
+
"learning_rate": 1.8832166493127128e-05,
|
| 561 |
+
"loss": 0.933,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.24322660098522167,
|
| 566 |
+
"grad_norm": 0.7128259336980561,
|
| 567 |
+
"learning_rate": 1.8781236129591847e-05,
|
| 568 |
+
"loss": 0.9078,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.24630541871921183,
|
| 573 |
+
"grad_norm": 0.7070843217854903,
|
| 574 |
+
"learning_rate": 1.8729290708206412e-05,
|
| 575 |
+
"loss": 0.9636,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24938423645320196,
|
| 580 |
+
"grad_norm": 0.718867215532029,
|
| 581 |
+
"learning_rate": 1.867633623354773e-05,
|
| 582 |
+
"loss": 0.9556,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2524630541871921,
|
| 587 |
+
"grad_norm": 0.7805465337550137,
|
| 588 |
+
"learning_rate": 1.8622378826833186e-05,
|
| 589 |
+
"loss": 0.9685,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2555418719211823,
|
| 594 |
+
"grad_norm": 0.7719203711056516,
|
| 595 |
+
"learning_rate": 1.856742472521304e-05,
|
| 596 |
+
"loss": 0.9524,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.25862068965517243,
|
| 601 |
+
"grad_norm": 0.7192834205111203,
|
| 602 |
+
"learning_rate": 1.8511480281049475e-05,
|
| 603 |
+
"loss": 0.9246,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2616995073891626,
|
| 608 |
+
"grad_norm": 0.6723358874994304,
|
| 609 |
+
"learning_rate": 1.8454551961182276e-05,
|
| 610 |
+
"loss": 0.9173,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2647783251231527,
|
| 615 |
+
"grad_norm": 0.6817001490385899,
|
| 616 |
+
"learning_rate": 1.8396646346181327e-05,
|
| 617 |
+
"loss": 0.9389,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.26785714285714285,
|
| 622 |
+
"grad_norm": 0.7207425588978127,
|
| 623 |
+
"learning_rate": 1.8337770129585918e-05,
|
| 624 |
+
"loss": 0.9731,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.270935960591133,
|
| 629 |
+
"grad_norm": 0.6937128223592749,
|
| 630 |
+
"learning_rate": 1.8277930117131025e-05,
|
| 631 |
+
"loss": 0.9666,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.27401477832512317,
|
| 636 |
+
"grad_norm": 0.7206019687649675,
|
| 637 |
+
"learning_rate": 1.8217133225960597e-05,
|
| 638 |
+
"loss": 0.9047,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2770935960591133,
|
| 643 |
+
"grad_norm": 0.6589164154450292,
|
| 644 |
+
"learning_rate": 1.8155386483827995e-05,
|
| 645 |
+
"loss": 0.9323,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2801724137931034,
|
| 650 |
+
"grad_norm": 0.7477503173033859,
|
| 651 |
+
"learning_rate": 1.8092697028283598e-05,
|
| 652 |
+
"loss": 0.9434,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2832512315270936,
|
| 657 |
+
"grad_norm": 0.6946681549404662,
|
| 658 |
+
"learning_rate": 1.8029072105849767e-05,
|
| 659 |
+
"loss": 0.9097,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.28633004926108374,
|
| 664 |
+
"grad_norm": 0.7993317655881385,
|
| 665 |
+
"learning_rate": 1.7964519071183188e-05,
|
| 666 |
+
"loss": 0.9484,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2894088669950739,
|
| 671 |
+
"grad_norm": 0.7061937279236051,
|
| 672 |
+
"learning_rate": 1.789904538622471e-05,
|
| 673 |
+
"loss": 0.9456,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.29248768472906406,
|
| 678 |
+
"grad_norm": 0.7072130152863852,
|
| 679 |
+
"learning_rate": 1.7832658619336794e-05,
|
| 680 |
+
"loss": 0.9524,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2955665024630542,
|
| 685 |
+
"grad_norm": 0.714294689243862,
|
| 686 |
+
"learning_rate": 1.7765366444428655e-05,
|
| 687 |
+
"loss": 0.9547,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2986453201970443,
|
| 692 |
+
"grad_norm": 0.7609301009818032,
|
| 693 |
+
"learning_rate": 1.7697176640069217e-05,
|
| 694 |
+
"loss": 0.9146,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3017241379310345,
|
| 699 |
+
"grad_norm": 0.7537368897585695,
|
| 700 |
+
"learning_rate": 1.762809708858793e-05,
|
| 701 |
+
"loss": 0.9451,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.30480295566502463,
|
| 706 |
+
"grad_norm": 0.7477622773826048,
|
| 707 |
+
"learning_rate": 1.7558135775163645e-05,
|
| 708 |
+
"loss": 0.9489,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3078817733990148,
|
| 713 |
+
"grad_norm": 0.8035379494578148,
|
| 714 |
+
"learning_rate": 1.7487300786901568e-05,
|
| 715 |
+
"loss": 0.887,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.31096059113300495,
|
| 720 |
+
"grad_norm": 0.8085018346753247,
|
| 721 |
+
"learning_rate": 1.7415600311898436e-05,
|
| 722 |
+
"loss": 0.9788,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.31403940886699505,
|
| 727 |
+
"grad_norm": 0.7199766001172444,
|
| 728 |
+
"learning_rate": 1.734304263829602e-05,
|
| 729 |
+
"loss": 0.9482,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3171182266009852,
|
| 734 |
+
"grad_norm": 0.7155949547205896,
|
| 735 |
+
"learning_rate": 1.726963615332308e-05,
|
| 736 |
+
"loss": 0.9638,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.32019704433497537,
|
| 741 |
+
"grad_norm": 0.8023351717912234,
|
| 742 |
+
"learning_rate": 1.7195389342325843e-05,
|
| 743 |
+
"loss": 0.9418,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3232758620689655,
|
| 748 |
+
"grad_norm": 0.6707946282641698,
|
| 749 |
+
"learning_rate": 1.7120310787787136e-05,
|
| 750 |
+
"loss": 0.9051,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3263546798029557,
|
| 755 |
+
"grad_norm": 0.7628307879605529,
|
| 756 |
+
"learning_rate": 1.7044409168334327e-05,
|
| 757 |
+
"loss": 0.9301,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3294334975369458,
|
| 762 |
+
"grad_norm": 0.7018750333060788,
|
| 763 |
+
"learning_rate": 1.696769325773611e-05,
|
| 764 |
+
"loss": 0.9203,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.33251231527093594,
|
| 769 |
+
"grad_norm": 0.8985483854038159,
|
| 770 |
+
"learning_rate": 1.6890171923888323e-05,
|
| 771 |
+
"loss": 0.942,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3355911330049261,
|
| 776 |
+
"grad_norm": 0.7095099515570173,
|
| 777 |
+
"learning_rate": 1.6811854127788857e-05,
|
| 778 |
+
"loss": 0.9883,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.33866995073891626,
|
| 783 |
+
"grad_norm": 0.663301100449732,
|
| 784 |
+
"learning_rate": 1.6732748922501832e-05,
|
| 785 |
+
"loss": 0.9958,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3417487684729064,
|
| 790 |
+
"grad_norm": 0.7007829684906022,
|
| 791 |
+
"learning_rate": 1.6652865452111115e-05,
|
| 792 |
+
"loss": 0.8977,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3448275862068966,
|
| 797 |
+
"grad_norm": 0.7794993793101852,
|
| 798 |
+
"learning_rate": 1.657221295066332e-05,
|
| 799 |
+
"loss": 0.963,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3479064039408867,
|
| 804 |
+
"grad_norm": 0.6817973499177562,
|
| 805 |
+
"learning_rate": 1.6490800741100396e-05,
|
| 806 |
+
"loss": 0.9448,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.35098522167487683,
|
| 811 |
+
"grad_norm": 0.7464691369559269,
|
| 812 |
+
"learning_rate": 1.6408638234181975e-05,
|
| 813 |
+
"loss": 0.9295,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.354064039408867,
|
| 818 |
+
"grad_norm": 0.6661854545324757,
|
| 819 |
+
"learning_rate": 1.6325734927397514e-05,
|
| 820 |
+
"loss": 0.943,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.35714285714285715,
|
| 825 |
+
"grad_norm": 0.6704430428924031,
|
| 826 |
+
"learning_rate": 1.6242100403868472e-05,
|
| 827 |
+
"loss": 0.9223,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3602216748768473,
|
| 832 |
+
"grad_norm": 0.7450322676169298,
|
| 833 |
+
"learning_rate": 1.615774433124054e-05,
|
| 834 |
+
"loss": 0.9296,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3633004926108374,
|
| 839 |
+
"grad_norm": 0.785984380526607,
|
| 840 |
+
"learning_rate": 1.6072676460566136e-05,
|
| 841 |
+
"loss": 0.952,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.36637931034482757,
|
| 846 |
+
"grad_norm": 0.7156872420189895,
|
| 847 |
+
"learning_rate": 1.5986906625177215e-05,
|
| 848 |
+
"loss": 0.9382,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3694581280788177,
|
| 853 |
+
"grad_norm": 0.7208351590256996,
|
| 854 |
+
"learning_rate": 1.590044473954863e-05,
|
| 855 |
+
"loss": 0.9951,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3725369458128079,
|
| 860 |
+
"grad_norm": 0.6980642220237895,
|
| 861 |
+
"learning_rate": 1.5813300798152048e-05,
|
| 862 |
+
"loss": 0.9297,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.37561576354679804,
|
| 867 |
+
"grad_norm": 0.6658042870926646,
|
| 868 |
+
"learning_rate": 1.5725484874300673e-05,
|
| 869 |
+
"loss": 0.9697,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3786945812807882,
|
| 874 |
+
"grad_norm": 0.6989438524636783,
|
| 875 |
+
"learning_rate": 1.5637007118984814e-05,
|
| 876 |
+
"loss": 0.9356,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3817733990147783,
|
| 881 |
+
"grad_norm": 0.6554764404296011,
|
| 882 |
+
"learning_rate": 1.5547877759698498e-05,
|
| 883 |
+
"loss": 0.9367,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.38485221674876846,
|
| 888 |
+
"grad_norm": 0.7212087304435114,
|
| 889 |
+
"learning_rate": 1.5458107099257245e-05,
|
| 890 |
+
"loss": 0.9277,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3879310344827586,
|
| 895 |
+
"grad_norm": 0.7411621278427354,
|
| 896 |
+
"learning_rate": 1.5367705514607107e-05,
|
| 897 |
+
"loss": 0.9152,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3910098522167488,
|
| 902 |
+
"grad_norm": 0.7201507693917704,
|
| 903 |
+
"learning_rate": 1.527668345562516e-05,
|
| 904 |
+
"loss": 0.9169,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.39408866995073893,
|
| 909 |
+
"grad_norm": 0.7894484158745393,
|
| 910 |
+
"learning_rate": 1.518505144391157e-05,
|
| 911 |
+
"loss": 0.946,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.39716748768472904,
|
| 916 |
+
"grad_norm": 0.7132866192747803,
|
| 917 |
+
"learning_rate": 1.5092820071573358e-05,
|
| 918 |
+
"loss": 0.9612,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4002463054187192,
|
| 923 |
+
"grad_norm": 0.8036672482141785,
|
| 924 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 925 |
+
"loss": 0.936,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.40332512315270935,
|
| 930 |
+
"grad_norm": 0.730609032181912,
|
| 931 |
+
"learning_rate": 1.490660195863106e-05,
|
| 932 |
+
"loss": 0.9155,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4064039408866995,
|
| 937 |
+
"grad_norm": 0.7875912602755151,
|
| 938 |
+
"learning_rate": 1.4812636743715912e-05,
|
| 939 |
+
"loss": 0.9289,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.40948275862068967,
|
| 944 |
+
"grad_norm": 0.7504791630648989,
|
| 945 |
+
"learning_rate": 1.4718115217065766e-05,
|
| 946 |
+
"loss": 0.9457,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4125615763546798,
|
| 951 |
+
"grad_norm": 0.7348305998616946,
|
| 952 |
+
"learning_rate": 1.462304830479811e-05,
|
| 953 |
+
"loss": 0.9072,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.41564039408866993,
|
| 958 |
+
"grad_norm": 0.758969021564288,
|
| 959 |
+
"learning_rate": 1.4527446996073714e-05,
|
| 960 |
+
"loss": 0.9177,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4187192118226601,
|
| 965 |
+
"grad_norm": 0.7012954160981419,
|
| 966 |
+
"learning_rate": 1.4431322341826348e-05,
|
| 967 |
+
"loss": 0.9025,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.42179802955665024,
|
| 972 |
+
"grad_norm": 0.8675884737141334,
|
| 973 |
+
"learning_rate": 1.433468545348537e-05,
|
| 974 |
+
"loss": 0.9515,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4248768472906404,
|
| 979 |
+
"grad_norm": 0.6907521835103683,
|
| 980 |
+
"learning_rate": 1.4237547501691298e-05,
|
| 981 |
+
"loss": 0.9369,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.42795566502463056,
|
| 986 |
+
"grad_norm": 0.8760691875396653,
|
| 987 |
+
"learning_rate": 1.4139919715004558e-05,
|
| 988 |
+
"loss": 0.9336,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.43103448275862066,
|
| 993 |
+
"grad_norm": 0.7605520563419292,
|
| 994 |
+
"learning_rate": 1.4041813378607534e-05,
|
| 995 |
+
"loss": 0.9459,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4341133004926108,
|
| 1000 |
+
"grad_norm": 0.7482466819256248,
|
| 1001 |
+
"learning_rate": 1.3943239833000068e-05,
|
| 1002 |
+
"loss": 0.9281,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.437192118226601,
|
| 1007 |
+
"grad_norm": 0.6853008905243693,
|
| 1008 |
+
"learning_rate": 1.3844210472688557e-05,
|
| 1009 |
+
"loss": 0.9115,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.44027093596059114,
|
| 1014 |
+
"grad_norm": 0.6785408874646308,
|
| 1015 |
+
"learning_rate": 1.3744736744868832e-05,
|
| 1016 |
+
"loss": 0.9147,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4433497536945813,
|
| 1021 |
+
"grad_norm": 0.7667132134232257,
|
| 1022 |
+
"learning_rate": 1.3644830148102915e-05,
|
| 1023 |
+
"loss": 0.9454,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.44642857142857145,
|
| 1028 |
+
"grad_norm": 0.7343585908583702,
|
| 1029 |
+
"learning_rate": 1.3544502230989868e-05,
|
| 1030 |
+
"loss": 0.927,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.44950738916256155,
|
| 1035 |
+
"grad_norm": 0.7135467850082023,
|
| 1036 |
+
"learning_rate": 1.344376459083085e-05,
|
| 1037 |
+
"loss": 0.9248,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4525862068965517,
|
| 1042 |
+
"grad_norm": 0.7291722090004472,
|
| 1043 |
+
"learning_rate": 1.3342628872288505e-05,
|
| 1044 |
+
"loss": 0.9401,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.45566502463054187,
|
| 1049 |
+
"grad_norm": 0.7200530793723301,
|
| 1050 |
+
"learning_rate": 1.3241106766040957e-05,
|
| 1051 |
+
"loss": 0.9097,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.45874384236453203,
|
| 1056 |
+
"grad_norm": 0.9856265254605108,
|
| 1057 |
+
"learning_rate": 1.3139210007430404e-05,
|
| 1058 |
+
"loss": 0.9283,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4618226600985222,
|
| 1063 |
+
"grad_norm": 0.7341322171237868,
|
| 1064 |
+
"learning_rate": 1.3036950375106588e-05,
|
| 1065 |
+
"loss": 0.9333,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4649014778325123,
|
| 1070 |
+
"grad_norm": 0.7546656555280624,
|
| 1071 |
+
"learning_rate": 1.2934339689665274e-05,
|
| 1072 |
+
"loss": 0.9042,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.46798029556650245,
|
| 1077 |
+
"grad_norm": 0.7226811365920998,
|
| 1078 |
+
"learning_rate": 1.283138981228184e-05,
|
| 1079 |
+
"loss": 0.9139,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4710591133004926,
|
| 1084 |
+
"grad_norm": 0.7181811040603533,
|
| 1085 |
+
"learning_rate": 1.2728112643340201e-05,
|
| 1086 |
+
"loss": 0.9264,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.47413793103448276,
|
| 1091 |
+
"grad_norm": 0.8849529024867829,
|
| 1092 |
+
"learning_rate": 1.26245201210572e-05,
|
| 1093 |
+
"loss": 0.9177,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4772167487684729,
|
| 1098 |
+
"grad_norm": 0.7087551590617035,
|
| 1099 |
+
"learning_rate": 1.2520624220102623e-05,
|
| 1100 |
+
"loss": 0.921,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4802955665024631,
|
| 1105 |
+
"grad_norm": 0.6935685016158297,
|
| 1106 |
+
"learning_rate": 1.2416436950215001e-05,
|
| 1107 |
+
"loss": 0.9406,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4833743842364532,
|
| 1112 |
+
"grad_norm": 0.7473725653174663,
|
| 1113 |
+
"learning_rate": 1.2311970354813345e-05,
|
| 1114 |
+
"loss": 0.9873,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.48645320197044334,
|
| 1119 |
+
"grad_norm": 0.657674984254088,
|
| 1120 |
+
"learning_rate": 1.220723650960502e-05,
|
| 1121 |
+
"loss": 0.927,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4895320197044335,
|
| 1126 |
+
"grad_norm": 0.7186327207648633,
|
| 1127 |
+
"learning_rate": 1.2102247521189838e-05,
|
| 1128 |
+
"loss": 0.9004,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.49261083743842365,
|
| 1133 |
+
"grad_norm": 0.6823249047599709,
|
| 1134 |
+
"learning_rate": 1.199701552566064e-05,
|
| 1135 |
+
"loss": 0.8879,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4956896551724138,
|
| 1140 |
+
"grad_norm": 0.705573871553639,
|
| 1141 |
+
"learning_rate": 1.1891552687200414e-05,
|
| 1142 |
+
"loss": 0.9649,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4987684729064039,
|
| 1147 |
+
"grad_norm": 0.7071856320020758,
|
| 1148 |
+
"learning_rate": 1.1785871196676196e-05,
|
| 1149 |
+
"loss": 0.9731,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5018472906403941,
|
| 1154 |
+
"grad_norm": 0.6605795316984785,
|
| 1155 |
+
"learning_rate": 1.167998327022988e-05,
|
| 1156 |
+
"loss": 0.9767,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5049261083743842,
|
| 1161 |
+
"grad_norm": 0.689384383606719,
|
| 1162 |
+
"learning_rate": 1.1573901147866108e-05,
|
| 1163 |
+
"loss": 0.929,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5080049261083743,
|
| 1168 |
+
"grad_norm": 0.7016418038725151,
|
| 1169 |
+
"learning_rate": 1.1467637092037399e-05,
|
| 1170 |
+
"loss": 0.9208,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5110837438423645,
|
| 1175 |
+
"grad_norm": 0.6435950510096468,
|
| 1176 |
+
"learning_rate": 1.1361203386226672e-05,
|
| 1177 |
+
"loss": 0.933,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5141625615763546,
|
| 1182 |
+
"grad_norm": 0.6915860359426802,
|
| 1183 |
+
"learning_rate": 1.1254612333527368e-05,
|
| 1184 |
+
"loss": 0.904,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5172413793103449,
|
| 1189 |
+
"grad_norm": 0.7278487014487579,
|
| 1190 |
+
"learning_rate": 1.1147876255221274e-05,
|
| 1191 |
+
"loss": 0.926,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.520320197044335,
|
| 1196 |
+
"grad_norm": 0.8402556179449276,
|
| 1197 |
+
"learning_rate": 1.1041007489354263e-05,
|
| 1198 |
+
"loss": 0.91,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5233990147783252,
|
| 1203 |
+
"grad_norm": 0.6913958208278679,
|
| 1204 |
+
"learning_rate": 1.093401838931009e-05,
|
| 1205 |
+
"loss": 0.9298,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5264778325123153,
|
| 1210 |
+
"grad_norm": 0.6942000092673813,
|
| 1211 |
+
"learning_rate": 1.0826921322382407e-05,
|
| 1212 |
+
"loss": 0.9254,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5295566502463054,
|
| 1217 |
+
"grad_norm": 0.6863019990229392,
|
| 1218 |
+
"learning_rate": 1.071972866834519e-05,
|
| 1219 |
+
"loss": 0.8875,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5326354679802956,
|
| 1224 |
+
"grad_norm": 0.7444864272324827,
|
| 1225 |
+
"learning_rate": 1.061245281802171e-05,
|
| 1226 |
+
"loss": 0.9348,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5357142857142857,
|
| 1231 |
+
"grad_norm": 0.7472300249765285,
|
| 1232 |
+
"learning_rate": 1.0505106171852226e-05,
|
| 1233 |
+
"loss": 0.8728,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5387931034482759,
|
| 1238 |
+
"grad_norm": 0.7421508787127711,
|
| 1239 |
+
"learning_rate": 1.039770113846056e-05,
|
| 1240 |
+
"loss": 0.897,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.541871921182266,
|
| 1245 |
+
"grad_norm": 0.6931950229252386,
|
| 1246 |
+
"learning_rate": 1.0290250133219755e-05,
|
| 1247 |
+
"loss": 0.9215,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5449507389162561,
|
| 1252 |
+
"grad_norm": 0.7648333450579011,
|
| 1253 |
+
"learning_rate": 1.0182765576816916e-05,
|
| 1254 |
+
"loss": 0.9577,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5480295566502463,
|
| 1259 |
+
"grad_norm": 0.7006731215648692,
|
| 1260 |
+
"learning_rate": 1.0075259893817465e-05,
|
| 1261 |
+
"loss": 0.9245,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5511083743842364,
|
| 1266 |
+
"grad_norm": 0.6687572466831397,
|
| 1267 |
+
"learning_rate": 9.967745511228922e-06,
|
| 1268 |
+
"loss": 0.8813,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5541871921182266,
|
| 1273 |
+
"grad_norm": 0.7175380682948211,
|
| 1274 |
+
"learning_rate": 9.860234857064439e-06,
|
| 1275 |
+
"loss": 0.9319,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5572660098522167,
|
| 1280 |
+
"grad_norm": 0.6948450711755244,
|
| 1281 |
+
"learning_rate": 9.752740358906168e-06,
|
| 1282 |
+
"loss": 0.9341,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5603448275862069,
|
| 1287 |
+
"grad_norm": 0.6892353302488635,
|
| 1288 |
+
"learning_rate": 9.645274442468746e-06,
|
| 1289 |
+
"loss": 0.9269,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5634236453201971,
|
| 1294 |
+
"grad_norm": 0.7325769139031748,
|
| 1295 |
+
"learning_rate": 9.537849530162915e-06,
|
| 1296 |
+
"loss": 0.9228,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5665024630541872,
|
| 1301 |
+
"grad_norm": 0.7282730155939394,
|
| 1302 |
+
"learning_rate": 9.430478039659604e-06,
|
| 1303 |
+
"loss": 0.9138,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5695812807881774,
|
| 1308 |
+
"grad_norm": 0.6428816898450572,
|
| 1309 |
+
"learning_rate": 9.323172382454494e-06,
|
| 1310 |
+
"loss": 0.9274,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5726600985221675,
|
| 1315 |
+
"grad_norm": 0.6703704767112079,
|
| 1316 |
+
"learning_rate": 9.215944962433329e-06,
|
| 1317 |
+
"loss": 0.855,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5757389162561576,
|
| 1322 |
+
"grad_norm": 0.7944858406137029,
|
| 1323 |
+
"learning_rate": 9.108808174438128e-06,
|
| 1324 |
+
"loss": 0.9416,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5788177339901478,
|
| 1329 |
+
"grad_norm": 0.721760753318748,
|
| 1330 |
+
"learning_rate": 9.001774402834375e-06,
|
| 1331 |
+
"loss": 0.8895,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5818965517241379,
|
| 1336 |
+
"grad_norm": 0.7210922094960251,
|
| 1337 |
+
"learning_rate": 8.89485602007949e-06,
|
| 1338 |
+
"loss": 0.952,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5849753694581281,
|
| 1343 |
+
"grad_norm": 0.7216379603390005,
|
| 1344 |
+
"learning_rate": 8.788065385292637e-06,
|
| 1345 |
+
"loss": 0.9067,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5880541871921182,
|
| 1350 |
+
"grad_norm": 0.6880918231702351,
|
| 1351 |
+
"learning_rate": 8.68141484282609e-06,
|
| 1352 |
+
"loss": 0.9327,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5911330049261084,
|
| 1357 |
+
"grad_norm": 0.7777971176157483,
|
| 1358 |
+
"learning_rate": 8.574916720838293e-06,
|
| 1359 |
+
"loss": 0.9514,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5942118226600985,
|
| 1364 |
+
"grad_norm": 0.7227495926203028,
|
| 1365 |
+
"learning_rate": 8.468583329868805e-06,
|
| 1366 |
+
"loss": 0.9463,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5972906403940886,
|
| 1371 |
+
"grad_norm": 0.7219855521048865,
|
| 1372 |
+
"learning_rate": 8.362426961415274e-06,
|
| 1373 |
+
"loss": 0.944,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6003694581280788,
|
| 1378 |
+
"grad_norm": 0.7155939312082987,
|
| 1379 |
+
"learning_rate": 8.256459886512618e-06,
|
| 1380 |
+
"loss": 0.9108,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.603448275862069,
|
| 1385 |
+
"grad_norm": 0.6993953380271827,
|
| 1386 |
+
"learning_rate": 8.150694354314556e-06,
|
| 1387 |
+
"loss": 0.9557,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6065270935960592,
|
| 1392 |
+
"grad_norm": 0.7638296699531455,
|
| 1393 |
+
"learning_rate": 8.0451425906777e-06,
|
| 1394 |
+
"loss": 0.9312,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6096059113300493,
|
| 1399 |
+
"grad_norm": 0.6831250724265701,
|
| 1400 |
+
"learning_rate": 7.939816796748296e-06,
|
| 1401 |
+
"loss": 0.8998,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6126847290640394,
|
| 1406 |
+
"grad_norm": 0.7313287861208433,
|
| 1407 |
+
"learning_rate": 7.834729147551858e-06,
|
| 1408 |
+
"loss": 0.9255,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6157635467980296,
|
| 1413 |
+
"grad_norm": 0.6590166847034968,
|
| 1414 |
+
"learning_rate": 7.729891790585817e-06,
|
| 1415 |
+
"loss": 0.8842,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6188423645320197,
|
| 1420 |
+
"grad_norm": 0.7323284885675908,
|
| 1421 |
+
"learning_rate": 7.625316844415327e-06,
|
| 1422 |
+
"loss": 0.8812,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6219211822660099,
|
| 1427 |
+
"grad_norm": 0.7215619643224764,
|
| 1428 |
+
"learning_rate": 7.521016397272436e-06,
|
| 1429 |
+
"loss": 0.9061,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.625,
|
| 1434 |
+
"grad_norm": 0.7213661635662646,
|
| 1435 |
+
"learning_rate": 7.417002505658773e-06,
|
| 1436 |
+
"loss": 0.9604,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6280788177339901,
|
| 1441 |
+
"grad_norm": 0.6664085186795183,
|
| 1442 |
+
"learning_rate": 7.313287192951866e-06,
|
| 1443 |
+
"loss": 0.9152,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6311576354679803,
|
| 1448 |
+
"grad_norm": 0.8899370388302164,
|
| 1449 |
+
"learning_rate": 7.209882448015338e-06,
|
| 1450 |
+
"loss": 0.8997,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6342364532019704,
|
| 1455 |
+
"grad_norm": 0.6925853341918634,
|
| 1456 |
+
"learning_rate": 7.1068002238130465e-06,
|
| 1457 |
+
"loss": 0.9186,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6373152709359606,
|
| 1462 |
+
"grad_norm": 0.7526266211171548,
|
| 1463 |
+
"learning_rate": 7.004052436027397e-06,
|
| 1464 |
+
"loss": 0.907,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6403940886699507,
|
| 1469 |
+
"grad_norm": 0.7116871357451918,
|
| 1470 |
+
"learning_rate": 6.901650961681976e-06,
|
| 1471 |
+
"loss": 0.9136,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6434729064039408,
|
| 1476 |
+
"grad_norm": 0.7181997436838037,
|
| 1477 |
+
"learning_rate": 6.799607637768621e-06,
|
| 1478 |
+
"loss": 0.9165,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.646551724137931,
|
| 1483 |
+
"grad_norm": 0.7424070514212068,
|
| 1484 |
+
"learning_rate": 6.6979342598791395e-06,
|
| 1485 |
+
"loss": 0.9401,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6496305418719212,
|
| 1490 |
+
"grad_norm": 0.7056624976371111,
|
| 1491 |
+
"learning_rate": 6.596642580841827e-06,
|
| 1492 |
+
"loss": 0.9151,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6527093596059114,
|
| 1497 |
+
"grad_norm": 0.7679067760249104,
|
| 1498 |
+
"learning_rate": 6.495744309362879e-06,
|
| 1499 |
+
"loss": 0.8805,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6557881773399015,
|
| 1504 |
+
"grad_norm": 0.7261332639619558,
|
| 1505 |
+
"learning_rate": 6.395251108672975e-06,
|
| 1506 |
+
"loss": 0.8987,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6588669950738916,
|
| 1511 |
+
"grad_norm": 0.6531278111009767,
|
| 1512 |
+
"learning_rate": 6.295174595179041e-06,
|
| 1513 |
+
"loss": 0.9147,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6619458128078818,
|
| 1518 |
+
"grad_norm": 0.74223655982831,
|
| 1519 |
+
"learning_rate": 6.195526337121483e-06,
|
| 1520 |
+
"loss": 0.9432,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6650246305418719,
|
| 1525 |
+
"grad_norm": 0.6775090824932495,
|
| 1526 |
+
"learning_rate": 6.096317853236975e-06,
|
| 1527 |
+
"loss": 0.8558,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6681034482758621,
|
| 1532 |
+
"grad_norm": 0.7451875244452356,
|
| 1533 |
+
"learning_rate": 5.997560611426947e-06,
|
| 1534 |
+
"loss": 0.9129,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6711822660098522,
|
| 1539 |
+
"grad_norm": 0.775303530121011,
|
| 1540 |
+
"learning_rate": 5.899266027431965e-06,
|
| 1541 |
+
"loss": 0.9302,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6742610837438424,
|
| 1546 |
+
"grad_norm": 0.6681173244510462,
|
| 1547 |
+
"learning_rate": 5.80144546351216e-06,
|
| 1548 |
+
"loss": 0.8755,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6773399014778325,
|
| 1553 |
+
"grad_norm": 0.6460985572337176,
|
| 1554 |
+
"learning_rate": 5.704110227133792e-06,
|
| 1555 |
+
"loss": 0.916,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6804187192118226,
|
| 1560 |
+
"grad_norm": 0.6903928180028339,
|
| 1561 |
+
"learning_rate": 5.607271569662203e-06,
|
| 1562 |
+
"loss": 0.8914,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6834975369458128,
|
| 1567 |
+
"grad_norm": 0.7795810817387983,
|
| 1568 |
+
"learning_rate": 5.510940685061202e-06,
|
| 1569 |
+
"loss": 0.8998,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6865763546798029,
|
| 1574 |
+
"grad_norm": 0.7672315116978631,
|
| 1575 |
+
"learning_rate": 5.41512870859912e-06,
|
| 1576 |
+
"loss": 0.9303,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6896551724137931,
|
| 1581 |
+
"grad_norm": 0.7806157930588039,
|
| 1582 |
+
"learning_rate": 5.319846715561656e-06,
|
| 1583 |
+
"loss": 0.9332,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6927339901477833,
|
| 1588 |
+
"grad_norm": 0.7003891392810607,
|
| 1589 |
+
"learning_rate": 5.225105719971615e-06,
|
| 1590 |
+
"loss": 0.8993,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6958128078817734,
|
| 1595 |
+
"grad_norm": 0.6721487834927411,
|
| 1596 |
+
"learning_rate": 5.130916673315762e-06,
|
| 1597 |
+
"loss": 0.941,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6988916256157636,
|
| 1602 |
+
"grad_norm": 0.6757485350567788,
|
| 1603 |
+
"learning_rate": 5.037290463278914e-06,
|
| 1604 |
+
"loss": 0.9011,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7019704433497537,
|
| 1609 |
+
"grad_norm": 0.7048595319775736,
|
| 1610 |
+
"learning_rate": 4.94423791248536e-06,
|
| 1611 |
+
"loss": 0.8987,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7050492610837439,
|
| 1616 |
+
"grad_norm": 0.6775506981745996,
|
| 1617 |
+
"learning_rate": 4.851769777247857e-06,
|
| 1618 |
+
"loss": 0.9616,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.708128078817734,
|
| 1623 |
+
"grad_norm": 0.7082392404353789,
|
| 1624 |
+
"learning_rate": 4.759896746324247e-06,
|
| 1625 |
+
"loss": 0.9016,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7112068965517241,
|
| 1630 |
+
"grad_norm": 0.6819223359256046,
|
| 1631 |
+
"learning_rate": 4.668629439681907e-06,
|
| 1632 |
+
"loss": 0.9272,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7142857142857143,
|
| 1637 |
+
"grad_norm": 0.6845551815152396,
|
| 1638 |
+
"learning_rate": 4.577978407270156e-06,
|
| 1639 |
+
"loss": 0.933,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7173645320197044,
|
| 1644 |
+
"grad_norm": 0.6950058521766465,
|
| 1645 |
+
"learning_rate": 4.487954127800726e-06,
|
| 1646 |
+
"loss": 0.9041,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7204433497536946,
|
| 1651 |
+
"grad_norm": 0.6867684845183903,
|
| 1652 |
+
"learning_rate": 4.398567007536493e-06,
|
| 1653 |
+
"loss": 0.9612,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7235221674876847,
|
| 1658 |
+
"grad_norm": 0.6586394312886726,
|
| 1659 |
+
"learning_rate": 4.309827379088589e-06,
|
| 1660 |
+
"loss": 0.9089,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7266009852216748,
|
| 1665 |
+
"grad_norm": 0.7365265974356552,
|
| 1666 |
+
"learning_rate": 4.221745500221992e-06,
|
| 1667 |
+
"loss": 0.8724,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.729679802955665,
|
| 1672 |
+
"grad_norm": 0.7378780810314269,
|
| 1673 |
+
"learning_rate": 4.134331552669812e-06,
|
| 1674 |
+
"loss": 0.9275,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7327586206896551,
|
| 1679 |
+
"grad_norm": 0.6956027586420739,
|
| 1680 |
+
"learning_rate": 4.047595640956326e-06,
|
| 1681 |
+
"loss": 0.9094,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7358374384236454,
|
| 1686 |
+
"grad_norm": 0.6644083331902493,
|
| 1687 |
+
"learning_rate": 3.961547791228963e-06,
|
| 1688 |
+
"loss": 0.8809,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7389162561576355,
|
| 1693 |
+
"grad_norm": 0.6780105644457819,
|
| 1694 |
+
"learning_rate": 3.876197950099351e-06,
|
| 1695 |
+
"loss": 0.8894,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7419950738916257,
|
| 1700 |
+
"grad_norm": 0.7238626484216509,
|
| 1701 |
+
"learning_rate": 3.7915559834935355e-06,
|
| 1702 |
+
"loss": 0.9182,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7450738916256158,
|
| 1707 |
+
"grad_norm": 0.677354013936962,
|
| 1708 |
+
"learning_rate": 3.7076316755115407e-06,
|
| 1709 |
+
"loss": 0.9056,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7481527093596059,
|
| 1714 |
+
"grad_norm": 0.6887868120830557,
|
| 1715 |
+
"learning_rate": 3.6244347272963974e-06,
|
| 1716 |
+
"loss": 0.9033,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7512315270935961,
|
| 1721 |
+
"grad_norm": 0.7935141959330477,
|
| 1722 |
+
"learning_rate": 3.5419747559127294e-06,
|
| 1723 |
+
"loss": 0.9077,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7543103448275862,
|
| 1728 |
+
"grad_norm": 0.717043682095087,
|
| 1729 |
+
"learning_rate": 3.4602612932351065e-06,
|
| 1730 |
+
"loss": 0.9383,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7573891625615764,
|
| 1735 |
+
"grad_norm": 0.7125434144608334,
|
| 1736 |
+
"learning_rate": 3.3793037848461873e-06,
|
| 1737 |
+
"loss": 0.9025,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7604679802955665,
|
| 1742 |
+
"grad_norm": 0.6915559899322756,
|
| 1743 |
+
"learning_rate": 3.2991115889448877e-06,
|
| 1744 |
+
"loss": 0.9113,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7635467980295566,
|
| 1749 |
+
"grad_norm": 0.6806388512684862,
|
| 1750 |
+
"learning_rate": 3.2196939752646183e-06,
|
| 1751 |
+
"loss": 0.9104,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7666256157635468,
|
| 1756 |
+
"grad_norm": 0.7595959827137008,
|
| 1757 |
+
"learning_rate": 3.141060124001776e-06,
|
| 1758 |
+
"loss": 0.9138,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7697044334975369,
|
| 1763 |
+
"grad_norm": 0.7234170553960368,
|
| 1764 |
+
"learning_rate": 3.063219124754543e-06,
|
| 1765 |
+
"loss": 0.9132,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7727832512315271,
|
| 1770 |
+
"grad_norm": 0.7023061276615833,
|
| 1771 |
+
"learning_rate": 2.9861799754722033e-06,
|
| 1772 |
+
"loss": 0.892,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7758620689655172,
|
| 1777 |
+
"grad_norm": 0.6711943725081508,
|
| 1778 |
+
"learning_rate": 2.9099515814150335e-06,
|
| 1779 |
+
"loss": 0.887,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7789408866995073,
|
| 1784 |
+
"grad_norm": 0.725196707404757,
|
| 1785 |
+
"learning_rate": 2.8345427541248993e-06,
|
| 1786 |
+
"loss": 0.8776,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7820197044334976,
|
| 1791 |
+
"grad_norm": 0.6608109215361615,
|
| 1792 |
+
"learning_rate": 2.7599622104066937e-06,
|
| 1793 |
+
"loss": 0.9427,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7850985221674877,
|
| 1798 |
+
"grad_norm": 0.6996055449820648,
|
| 1799 |
+
"learning_rate": 2.6862185713207467e-06,
|
| 1800 |
+
"loss": 0.9142,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7881773399014779,
|
| 1805 |
+
"grad_norm": 0.7432961342047957,
|
| 1806 |
+
"learning_rate": 2.6133203611862554e-06,
|
| 1807 |
+
"loss": 0.9064,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.791256157635468,
|
| 1812 |
+
"grad_norm": 0.6236987444510391,
|
| 1813 |
+
"learning_rate": 2.5412760065959386e-06,
|
| 1814 |
+
"loss": 0.9048,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7943349753694581,
|
| 1819 |
+
"grad_norm": 0.6220890988258023,
|
| 1820 |
+
"learning_rate": 2.4700938354419823e-06,
|
| 1821 |
+
"loss": 0.8903,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7974137931034483,
|
| 1826 |
+
"grad_norm": 0.6892863270775752,
|
| 1827 |
+
"learning_rate": 2.3997820759533654e-06,
|
| 1828 |
+
"loss": 0.913,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8004926108374384,
|
| 1833 |
+
"grad_norm": 1.6773601433209218,
|
| 1834 |
+
"learning_rate": 2.3303488557447374e-06,
|
| 1835 |
+
"loss": 0.8753,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8035714285714286,
|
| 1840 |
+
"grad_norm": 0.722872865650384,
|
| 1841 |
+
"learning_rate": 2.26180220087692e-06,
|
| 1842 |
+
"loss": 0.8957,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8066502463054187,
|
| 1847 |
+
"grad_norm": 0.6613242490598888,
|
| 1848 |
+
"learning_rate": 2.194150034929133e-06,
|
| 1849 |
+
"loss": 0.9054,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8097290640394089,
|
| 1854 |
+
"grad_norm": 0.7330833022368549,
|
| 1855 |
+
"learning_rate": 2.1274001780830776e-06,
|
| 1856 |
+
"loss": 0.9283,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.812807881773399,
|
| 1861 |
+
"grad_norm": 0.6778392727939891,
|
| 1862 |
+
"learning_rate": 2.0615603462189824e-06,
|
| 1863 |
+
"loss": 0.9288,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8158866995073891,
|
| 1868 |
+
"grad_norm": 0.7707106273789677,
|
| 1869 |
+
"learning_rate": 1.9966381500236786e-06,
|
| 1870 |
+
"loss": 0.875,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8189655172413793,
|
| 1875 |
+
"grad_norm": 0.6432858185048476,
|
| 1876 |
+
"learning_rate": 1.932641094110855e-06,
|
| 1877 |
+
"loss": 0.8971,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8220443349753694,
|
| 1882 |
+
"grad_norm": 0.6914786964712628,
|
| 1883 |
+
"learning_rate": 1.869576576153581e-06,
|
| 1884 |
+
"loss": 0.8654,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8251231527093597,
|
| 1889 |
+
"grad_norm": 0.7215233738169529,
|
| 1890 |
+
"learning_rate": 1.8074518860291646e-06,
|
| 1891 |
+
"loss": 0.909,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8282019704433498,
|
| 1896 |
+
"grad_norm": 0.8465873318075495,
|
| 1897 |
+
"learning_rate": 1.746274204976498e-06,
|
| 1898 |
+
"loss": 0.9038,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8312807881773399,
|
| 1903 |
+
"grad_norm": 0.7216800410068818,
|
| 1904 |
+
"learning_rate": 1.6860506047659442e-06,
|
| 1905 |
+
"loss": 0.9222,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8343596059113301,
|
| 1910 |
+
"grad_norm": 0.7685628442751503,
|
| 1911 |
+
"learning_rate": 1.6267880468818787e-06,
|
| 1912 |
+
"loss": 0.8895,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8374384236453202,
|
| 1917 |
+
"grad_norm": 0.7314941509944337,
|
| 1918 |
+
"learning_rate": 1.5684933817180014e-06,
|
| 1919 |
+
"loss": 0.9106,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8405172413793104,
|
| 1924 |
+
"grad_norm": 0.7642178298055531,
|
| 1925 |
+
"learning_rate": 1.5111733477854507e-06,
|
| 1926 |
+
"loss": 0.9374,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8435960591133005,
|
| 1931 |
+
"grad_norm": 0.6424365034435878,
|
| 1932 |
+
"learning_rate": 1.454834570933884e-06,
|
| 1933 |
+
"loss": 0.899,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8466748768472906,
|
| 1938 |
+
"grad_norm": 0.7054107402438561,
|
| 1939 |
+
"learning_rate": 1.399483563585573e-06,
|
| 1940 |
+
"loss": 0.8976,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8497536945812808,
|
| 1945 |
+
"grad_norm": 0.7944345419145749,
|
| 1946 |
+
"learning_rate": 1.345126723982594e-06,
|
| 1947 |
+
"loss": 0.9645,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8528325123152709,
|
| 1952 |
+
"grad_norm": 0.7297116413505407,
|
| 1953 |
+
"learning_rate": 1.2917703354472467e-06,
|
| 1954 |
+
"loss": 0.9013,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8559113300492611,
|
| 1959 |
+
"grad_norm": 0.7016058795668882,
|
| 1960 |
+
"learning_rate": 1.2394205656557224e-06,
|
| 1961 |
+
"loss": 0.886,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8589901477832512,
|
| 1966 |
+
"grad_norm": 0.6817325792524672,
|
| 1967 |
+
"learning_rate": 1.1880834659251706e-06,
|
| 1968 |
+
"loss": 0.9106,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8620689655172413,
|
| 1973 |
+
"grad_norm": 0.7175638109399047,
|
| 1974 |
+
"learning_rate": 1.1377649705142012e-06,
|
| 1975 |
+
"loss": 0.9046,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8651477832512315,
|
| 1980 |
+
"grad_norm": 0.7325286340177533,
|
| 1981 |
+
"learning_rate": 1.0884708959369116e-06,
|
| 1982 |
+
"loss": 0.9344,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8682266009852216,
|
| 1987 |
+
"grad_norm": 0.6995057655558119,
|
| 1988 |
+
"learning_rate": 1.040206940290547e-06,
|
| 1989 |
+
"loss": 0.8886,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8713054187192119,
|
| 1994 |
+
"grad_norm": 0.6821188835840444,
|
| 1995 |
+
"learning_rate": 9.929786825968213e-07,
|
| 1996 |
+
"loss": 0.8976,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.874384236453202,
|
| 2001 |
+
"grad_norm": 0.7376721682017519,
|
| 2002 |
+
"learning_rate": 9.467915821570228e-07,
|
| 2003 |
+
"loss": 0.9492,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8774630541871922,
|
| 2008 |
+
"grad_norm": 0.9328210953314316,
|
| 2009 |
+
"learning_rate": 9.016509779209536e-07,
|
| 2010 |
+
"loss": 0.9074,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8805418719211823,
|
| 2015 |
+
"grad_norm": 0.7340819327835291,
|
| 2016 |
+
"learning_rate": 8.575620878697744e-07,
|
| 2017 |
+
"loss": 0.8837,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8836206896551724,
|
| 2022 |
+
"grad_norm": 0.6845434866282742,
|
| 2023 |
+
"learning_rate": 8.145300084128349e-07,
|
| 2024 |
+
"loss": 0.9185,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8866995073891626,
|
| 2029 |
+
"grad_norm": 0.6628364676048413,
|
| 2030 |
+
"learning_rate": 7.725597137985741e-07,
|
| 2031 |
+
"loss": 0.8584,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8897783251231527,
|
| 2036 |
+
"grad_norm": 0.7208226072354926,
|
| 2037 |
+
"learning_rate": 7.316560555395069e-07,
|
| 2038 |
+
"loss": 0.8826,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8928571428571429,
|
| 2043 |
+
"grad_norm": 0.6773256291080844,
|
| 2044 |
+
"learning_rate": 6.918237618514378e-07,
|
| 2045 |
+
"loss": 0.856,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.895935960591133,
|
| 2050 |
+
"grad_norm": 0.7034404298614597,
|
| 2051 |
+
"learning_rate": 6.530674371068946e-07,
|
| 2052 |
+
"loss": 0.9128,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8990147783251231,
|
| 2057 |
+
"grad_norm": 0.7210199225257817,
|
| 2058 |
+
"learning_rate": 6.153915613028915e-07,
|
| 2059 |
+
"loss": 0.9038,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9020935960591133,
|
| 2064 |
+
"grad_norm": 0.7406834860134297,
|
| 2065 |
+
"learning_rate": 5.788004895430799e-07,
|
| 2066 |
+
"loss": 0.8682,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9051724137931034,
|
| 2071 |
+
"grad_norm": 0.7117319738607917,
|
| 2072 |
+
"learning_rate": 5.43298451534312e-07,
|
| 2073 |
+
"loss": 0.9407,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9082512315270936,
|
| 2078 |
+
"grad_norm": 0.6909093195401991,
|
| 2079 |
+
"learning_rate": 5.088895510977154e-07,
|
| 2080 |
+
"loss": 0.8934,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9113300492610837,
|
| 2085 |
+
"grad_norm": 0.6395163285948796,
|
| 2086 |
+
"learning_rate": 4.755777656943239e-07,
|
| 2087 |
+
"loss": 0.9065,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9144088669950738,
|
| 2092 |
+
"grad_norm": 0.7727497202740083,
|
| 2093 |
+
"learning_rate": 4.433669459652945e-07,
|
| 2094 |
+
"loss": 0.8946,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9174876847290641,
|
| 2099 |
+
"grad_norm": 0.6727790753972084,
|
| 2100 |
+
"learning_rate": 4.1226081528680907e-07,
|
| 2101 |
+
"loss": 0.9206,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9205665024630542,
|
| 2106 |
+
"grad_norm": 0.6849296173953379,
|
| 2107 |
+
"learning_rate": 3.822629693396651e-07,
|
| 2108 |
+
"loss": 0.8824,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9236453201970444,
|
| 2113 |
+
"grad_norm": 0.7395506995675738,
|
| 2114 |
+
"learning_rate": 3.5337687569363734e-07,
|
| 2115 |
+
"loss": 0.9219,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9267241379310345,
|
| 2120 |
+
"grad_norm": 0.7042269789976827,
|
| 2121 |
+
"learning_rate": 3.2560587340665694e-07,
|
| 2122 |
+
"loss": 0.9093,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9298029556650246,
|
| 2127 |
+
"grad_norm": 0.7584121398933832,
|
| 2128 |
+
"learning_rate": 2.989531726388262e-07,
|
| 2129 |
+
"loss": 0.9112,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9328817733990148,
|
| 2134 |
+
"grad_norm": 0.7779893269073751,
|
| 2135 |
+
"learning_rate": 2.7342185428134604e-07,
|
| 2136 |
+
"loss": 0.9122,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9359605911330049,
|
| 2141 |
+
"grad_norm": 0.6361716458037133,
|
| 2142 |
+
"learning_rate": 2.4901486960039025e-07,
|
| 2143 |
+
"loss": 0.9178,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9390394088669951,
|
| 2148 |
+
"grad_norm": 0.7272001913384153,
|
| 2149 |
+
"learning_rate": 2.257350398959457e-07,
|
| 2150 |
+
"loss": 0.8953,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9421182266009852,
|
| 2155 |
+
"grad_norm": 0.6848577166162307,
|
| 2156 |
+
"learning_rate": 2.035850561756969e-07,
|
| 2157 |
+
"loss": 0.9155,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9451970443349754,
|
| 2162 |
+
"grad_norm": 0.7198842919309125,
|
| 2163 |
+
"learning_rate": 1.8256747884395577e-07,
|
| 2164 |
+
"loss": 0.9037,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9482758620689655,
|
| 2169 |
+
"grad_norm": 0.7115101379396237,
|
| 2170 |
+
"learning_rate": 1.6268473740569723e-07,
|
| 2171 |
+
"loss": 0.9647,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9513546798029556,
|
| 2176 |
+
"grad_norm": 0.7368827581632996,
|
| 2177 |
+
"learning_rate": 1.4393913018572182e-07,
|
| 2178 |
+
"loss": 0.8805,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9544334975369458,
|
| 2183 |
+
"grad_norm": 0.8079433748319568,
|
| 2184 |
+
"learning_rate": 1.2633282406298576e-07,
|
| 2185 |
+
"loss": 0.896,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9575123152709359,
|
| 2190 |
+
"grad_norm": 0.6587864044048146,
|
| 2191 |
+
"learning_rate": 1.0986785422011593e-07,
|
| 2192 |
+
"loss": 0.8944,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9605911330049262,
|
| 2197 |
+
"grad_norm": 0.6943529244689446,
|
| 2198 |
+
"learning_rate": 9.454612390816686e-08,
|
| 2199 |
+
"loss": 0.892,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9636699507389163,
|
| 2204 |
+
"grad_norm": 0.7475181568465236,
|
| 2205 |
+
"learning_rate": 8.036940422660345e-08,
|
| 2206 |
+
"loss": 0.9109,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9667487684729064,
|
| 2211 |
+
"grad_norm": 0.712566295810641,
|
| 2212 |
+
"learning_rate": 6.733933391858238e-08,
|
| 2213 |
+
"loss": 0.9387,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9698275862068966,
|
| 2218 |
+
"grad_norm": 0.6953912672416684,
|
| 2219 |
+
"learning_rate": 5.5457419181517145e-08,
|
| 2220 |
+
"loss": 0.9009,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9729064039408867,
|
| 2225 |
+
"grad_norm": 0.6683739761168683,
|
| 2226 |
+
"learning_rate": 4.472503349297497e-08,
|
| 2227 |
+
"loss": 0.8673,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9759852216748769,
|
| 2232 |
+
"grad_norm": 0.69079632147327,
|
| 2233 |
+
"learning_rate": 3.5143417451907195e-08,
|
| 2234 |
+
"loss": 0.8563,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.979064039408867,
|
| 2239 |
+
"grad_norm": 0.708420869441777,
|
| 2240 |
+
"learning_rate": 2.671367863524732e-08,
|
| 2241 |
+
"loss": 0.9274,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9821428571428571,
|
| 2246 |
+
"grad_norm": 0.8174030990687192,
|
| 2247 |
+
"learning_rate": 1.943679146988009e-08,
|
| 2248 |
+
"loss": 0.9223,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9852216748768473,
|
| 2253 |
+
"grad_norm": 0.742245024367394,
|
| 2254 |
+
"learning_rate": 1.3313597120002686e-08,
|
| 2255 |
+
"loss": 0.9255,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9883004926108374,
|
| 2260 |
+
"grad_norm": 0.6809579676933898,
|
| 2261 |
+
"learning_rate": 8.34480338989141e-09,
|
| 2262 |
+
"loss": 0.9343,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9913793103448276,
|
| 2267 |
+
"grad_norm": 0.6818330666493024,
|
| 2268 |
+
"learning_rate": 4.530984642087122e-09,
|
| 2269 |
+
"loss": 0.8425,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9944581280788177,
|
| 2274 |
+
"grad_norm": 0.6517282911169013,
|
| 2275 |
+
"learning_rate": 1.872581730997247e-09,
|
| 2276 |
+
"loss": 0.9258,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9975369458128078,
|
| 2281 |
+
"grad_norm": 0.7361227785975518,
|
| 2282 |
+
"learning_rate": 3.699019519387559e-10,
|
| 2283 |
+
"loss": 0.8655,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 1.0,
|
| 2288 |
+
"eval_loss": 0.9033653736114502,
|
| 2289 |
+
"eval_runtime": 1302.7533,
|
| 2290 |
+
"eval_samples_per_second": 17.739,
|
| 2291 |
+
"eval_steps_per_second": 0.555,
|
| 2292 |
+
"step": 1624
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.0,
|
| 2296 |
+
"step": 1624,
|
| 2297 |
+
"total_flos": 165422873640960.0,
|
| 2298 |
+
"train_loss": 0.9319284432040059,
|
| 2299 |
+
"train_runtime": 12380.9765,
|
| 2300 |
+
"train_samples_per_second": 4.197,
|
| 2301 |
+
"train_steps_per_second": 0.131
|
| 2302 |
+
}
|
| 2303 |
+
],
|
| 2304 |
+
"logging_steps": 5,
|
| 2305 |
+
"max_steps": 1624,
|
| 2306 |
+
"num_input_tokens_seen": 0,
|
| 2307 |
+
"num_train_epochs": 1,
|
| 2308 |
+
"save_steps": 500,
|
| 2309 |
+
"stateful_callbacks": {
|
| 2310 |
+
"TrainerControl": {
|
| 2311 |
+
"args": {
|
| 2312 |
+
"should_epoch_stop": false,
|
| 2313 |
+
"should_evaluate": false,
|
| 2314 |
+
"should_log": false,
|
| 2315 |
+
"should_save": true,
|
| 2316 |
+
"should_training_stop": true
|
| 2317 |
+
},
|
| 2318 |
+
"attributes": {}
|
| 2319 |
+
}
|
| 2320 |
+
},
|
| 2321 |
+
"total_flos": 165422873640960.0,
|
| 2322 |
+
"train_batch_size": 8,
|
| 2323 |
+
"trial_name": null,
|
| 2324 |
+
"trial_params": null
|
| 2325 |
+
}
|