Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +2220 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: AmberYifan/mistral-v0.1-7b-sft-ultrachat
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: mistral-v0.1-7b-sft-ultrachat-safeRLHF
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for mistral-v0.1-7b-sft-ultrachat-safeRLHF
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [AmberYifan/mistral-v0.1-7b-sft-ultrachat](https://huggingface.co/AmberYifan/mistral-v0.1-7b-sft-ultrachat).
|
| 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/mistral-v0.1-7b-sft-ultrachat-safeRLHF", 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/8am492yp)
|
| 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": 19248551362560.0,
|
| 4 |
+
"train_loss": 1.5783487337492912,
|
| 5 |
+
"train_runtime": 6620.0514,
|
| 6 |
+
"train_samples": 49550,
|
| 7 |
+
"train_samples_per_second": 7.485,
|
| 8 |
+
"train_steps_per_second": 0.234
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.46.3"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 19248551362560.0,
|
| 4 |
+
"train_loss": 1.5783487337492912,
|
| 5 |
+
"train_runtime": 6620.0514,
|
| 6 |
+
"train_samples": 49550,
|
| 7 |
+
"train_samples_per_second": 7.485,
|
| 8 |
+
"train_steps_per_second": 0.234
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1549,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006455777921239509,
|
| 13 |
+
"grad_norm": 16.205452234411737,
|
| 14 |
+
"learning_rate": 1.2903225806451614e-07,
|
| 15 |
+
"loss": 1.8573,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0032278889606197547,
|
| 20 |
+
"grad_norm": 14.995431561267777,
|
| 21 |
+
"learning_rate": 6.451612903225807e-07,
|
| 22 |
+
"loss": 1.8993,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006455777921239509,
|
| 27 |
+
"grad_norm": 10.81513181649353,
|
| 28 |
+
"learning_rate": 1.2903225806451614e-06,
|
| 29 |
+
"loss": 1.8257,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009683666881859263,
|
| 34 |
+
"grad_norm": 10.205458746998783,
|
| 35 |
+
"learning_rate": 1.935483870967742e-06,
|
| 36 |
+
"loss": 1.7098,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012911555842479019,
|
| 41 |
+
"grad_norm": 9.227538444236034,
|
| 42 |
+
"learning_rate": 2.580645161290323e-06,
|
| 43 |
+
"loss": 1.63,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.016139444803098774,
|
| 48 |
+
"grad_norm": 8.636180793854093,
|
| 49 |
+
"learning_rate": 3.225806451612903e-06,
|
| 50 |
+
"loss": 1.6054,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.019367333763718526,
|
| 55 |
+
"grad_norm": 7.836281445290913,
|
| 56 |
+
"learning_rate": 3.870967741935484e-06,
|
| 57 |
+
"loss": 1.554,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.022595222724338282,
|
| 62 |
+
"grad_norm": 7.718242026474227,
|
| 63 |
+
"learning_rate": 4.516129032258065e-06,
|
| 64 |
+
"loss": 1.5009,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.025823111684958037,
|
| 69 |
+
"grad_norm": 7.637693613009242,
|
| 70 |
+
"learning_rate": 5.161290322580646e-06,
|
| 71 |
+
"loss": 1.5568,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.029051000645577793,
|
| 76 |
+
"grad_norm": 8.037677797603621,
|
| 77 |
+
"learning_rate": 5.806451612903226e-06,
|
| 78 |
+
"loss": 1.509,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03227888960619755,
|
| 83 |
+
"grad_norm": 8.050943021309022,
|
| 84 |
+
"learning_rate": 6.451612903225806e-06,
|
| 85 |
+
"loss": 1.5483,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.035506778566817304,
|
| 90 |
+
"grad_norm": 7.8773672608999075,
|
| 91 |
+
"learning_rate": 7.096774193548388e-06,
|
| 92 |
+
"loss": 1.5449,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03873466752743705,
|
| 97 |
+
"grad_norm": 7.8185673315827104,
|
| 98 |
+
"learning_rate": 7.741935483870968e-06,
|
| 99 |
+
"loss": 1.5615,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04196255648805681,
|
| 104 |
+
"grad_norm": 7.288430315557615,
|
| 105 |
+
"learning_rate": 8.387096774193549e-06,
|
| 106 |
+
"loss": 1.5535,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.045190445448676564,
|
| 111 |
+
"grad_norm": 7.4125851014724224,
|
| 112 |
+
"learning_rate": 9.03225806451613e-06,
|
| 113 |
+
"loss": 1.56,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04841833440929632,
|
| 118 |
+
"grad_norm": 7.384090809487316,
|
| 119 |
+
"learning_rate": 9.67741935483871e-06,
|
| 120 |
+
"loss": 1.5751,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.051646223369916075,
|
| 125 |
+
"grad_norm": 7.186612923240028,
|
| 126 |
+
"learning_rate": 1.0322580645161291e-05,
|
| 127 |
+
"loss": 1.5353,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05487411233053583,
|
| 132 |
+
"grad_norm": 7.54719691241193,
|
| 133 |
+
"learning_rate": 1.096774193548387e-05,
|
| 134 |
+
"loss": 1.536,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.058102001291155586,
|
| 139 |
+
"grad_norm": 6.841634960153724,
|
| 140 |
+
"learning_rate": 1.1612903225806453e-05,
|
| 141 |
+
"loss": 1.5638,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06132989025177534,
|
| 146 |
+
"grad_norm": 6.884784528712181,
|
| 147 |
+
"learning_rate": 1.2258064516129034e-05,
|
| 148 |
+
"loss": 1.5449,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0645577792123951,
|
| 153 |
+
"grad_norm": 7.5067298597661525,
|
| 154 |
+
"learning_rate": 1.2903225806451613e-05,
|
| 155 |
+
"loss": 1.595,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06778566817301485,
|
| 160 |
+
"grad_norm": 7.371893992584374,
|
| 161 |
+
"learning_rate": 1.3548387096774194e-05,
|
| 162 |
+
"loss": 1.6045,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.07101355713363461,
|
| 167 |
+
"grad_norm": 7.220506075154664,
|
| 168 |
+
"learning_rate": 1.4193548387096776e-05,
|
| 169 |
+
"loss": 1.5886,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07424144609425436,
|
| 174 |
+
"grad_norm": 7.354995573259938,
|
| 175 |
+
"learning_rate": 1.4838709677419357e-05,
|
| 176 |
+
"loss": 1.6884,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0774693350548741,
|
| 181 |
+
"grad_norm": 7.431127341541042,
|
| 182 |
+
"learning_rate": 1.5483870967741936e-05,
|
| 183 |
+
"loss": 1.5783,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08069722401549387,
|
| 188 |
+
"grad_norm": 7.617961920752755,
|
| 189 |
+
"learning_rate": 1.6129032258064517e-05,
|
| 190 |
+
"loss": 1.5722,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08392511297611362,
|
| 195 |
+
"grad_norm": 7.2187575554924495,
|
| 196 |
+
"learning_rate": 1.6774193548387098e-05,
|
| 197 |
+
"loss": 1.6404,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08715300193673338,
|
| 202 |
+
"grad_norm": 8.229669273322479,
|
| 203 |
+
"learning_rate": 1.741935483870968e-05,
|
| 204 |
+
"loss": 1.6631,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.09038089089735313,
|
| 209 |
+
"grad_norm": 7.475319901543287,
|
| 210 |
+
"learning_rate": 1.806451612903226e-05,
|
| 211 |
+
"loss": 1.6931,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09360877985797289,
|
| 216 |
+
"grad_norm": 7.238180827010263,
|
| 217 |
+
"learning_rate": 1.870967741935484e-05,
|
| 218 |
+
"loss": 1.6595,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09683666881859264,
|
| 223 |
+
"grad_norm": 7.477385100319548,
|
| 224 |
+
"learning_rate": 1.935483870967742e-05,
|
| 225 |
+
"loss": 1.7362,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.1000645577792124,
|
| 230 |
+
"grad_norm": 6.8926903445205125,
|
| 231 |
+
"learning_rate": 2e-05,
|
| 232 |
+
"loss": 1.7156,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10329244673983215,
|
| 237 |
+
"grad_norm": 7.118422267427338,
|
| 238 |
+
"learning_rate": 1.9999365137586883e-05,
|
| 239 |
+
"loss": 1.7117,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1065203357004519,
|
| 244 |
+
"grad_norm": 7.359581007823435,
|
| 245 |
+
"learning_rate": 1.9997460630957587e-05,
|
| 246 |
+
"loss": 1.6915,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10974822466107166,
|
| 251 |
+
"grad_norm": 7.06877768743029,
|
| 252 |
+
"learning_rate": 1.9994286721932043e-05,
|
| 253 |
+
"loss": 1.7439,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11297611362169141,
|
| 258 |
+
"grad_norm": 7.218247396787645,
|
| 259 |
+
"learning_rate": 1.9989843813509366e-05,
|
| 260 |
+
"loss": 1.7124,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11620400258231117,
|
| 265 |
+
"grad_norm": 16.159481189289234,
|
| 266 |
+
"learning_rate": 1.998413246981666e-05,
|
| 267 |
+
"loss": 1.7058,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11943189154293092,
|
| 272 |
+
"grad_norm": 46.181125374261654,
|
| 273 |
+
"learning_rate": 1.9977153416037424e-05,
|
| 274 |
+
"loss": 1.7839,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12265978050355068,
|
| 279 |
+
"grad_norm": 41.37204715355036,
|
| 280 |
+
"learning_rate": 1.9968907538319433e-05,
|
| 281 |
+
"loss": 1.917,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12588766946417043,
|
| 286 |
+
"grad_norm": 44.987565321564276,
|
| 287 |
+
"learning_rate": 1.9959395883662257e-05,
|
| 288 |
+
"loss": 1.7653,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1291155584247902,
|
| 293 |
+
"grad_norm": 7.194696914719683,
|
| 294 |
+
"learning_rate": 1.99486196597843e-05,
|
| 295 |
+
"loss": 1.7191,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.13234344738540993,
|
| 300 |
+
"grad_norm": 7.430734307814263,
|
| 301 |
+
"learning_rate": 1.9936580234969464e-05,
|
| 302 |
+
"loss": 1.7557,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1355713363460297,
|
| 307 |
+
"grad_norm": 7.334778943984345,
|
| 308 |
+
"learning_rate": 1.9923279137893405e-05,
|
| 309 |
+
"loss": 1.7207,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.13879922530664945,
|
| 314 |
+
"grad_norm": 6.935862923382574,
|
| 315 |
+
"learning_rate": 1.990871805742944e-05,
|
| 316 |
+
"loss": 1.6912,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.14202711426726922,
|
| 321 |
+
"grad_norm": 6.43836756283986,
|
| 322 |
+
"learning_rate": 1.9892898842434106e-05,
|
| 323 |
+
"loss": 1.7097,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.14525500322788895,
|
| 328 |
+
"grad_norm": 6.790579150198959,
|
| 329 |
+
"learning_rate": 1.98758235015124e-05,
|
| 330 |
+
"loss": 1.6824,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1484828921885087,
|
| 335 |
+
"grad_norm": 7.563145193822757,
|
| 336 |
+
"learning_rate": 1.9857494202762758e-05,
|
| 337 |
+
"loss": 1.7204,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15171078114912848,
|
| 342 |
+
"grad_norm": 7.21569143328697,
|
| 343 |
+
"learning_rate": 1.983791327350174e-05,
|
| 344 |
+
"loss": 1.724,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1549386701097482,
|
| 349 |
+
"grad_norm": 7.396066704475681,
|
| 350 |
+
"learning_rate": 1.9817083199968552e-05,
|
| 351 |
+
"loss": 1.74,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15816655907036797,
|
| 356 |
+
"grad_norm": 6.409702062362999,
|
| 357 |
+
"learning_rate": 1.979500662700934e-05,
|
| 358 |
+
"loss": 1.733,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.16139444803098774,
|
| 363 |
+
"grad_norm": 7.33040263293043,
|
| 364 |
+
"learning_rate": 1.977168635774138e-05,
|
| 365 |
+
"loss": 1.7073,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1646223369916075,
|
| 370 |
+
"grad_norm": 6.627407014857698,
|
| 371 |
+
"learning_rate": 1.974712535319716e-05,
|
| 372 |
+
"loss": 1.7714,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.16785022595222723,
|
| 377 |
+
"grad_norm": 9.128235977291943,
|
| 378 |
+
"learning_rate": 1.97213267319484e-05,
|
| 379 |
+
"loss": 1.7289,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.171078114912847,
|
| 384 |
+
"grad_norm": 10.132426302191169,
|
| 385 |
+
"learning_rate": 1.969429376971009e-05,
|
| 386 |
+
"loss": 1.7112,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.17430600387346676,
|
| 391 |
+
"grad_norm": 7.187125458351554,
|
| 392 |
+
"learning_rate": 1.9666029898924558e-05,
|
| 393 |
+
"loss": 1.7521,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.17753389283408652,
|
| 398 |
+
"grad_norm": 6.559644342179301,
|
| 399 |
+
"learning_rate": 1.9636538708325642e-05,
|
| 400 |
+
"loss": 1.6834,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.18076178179470626,
|
| 405 |
+
"grad_norm": 6.619411966067338,
|
| 406 |
+
"learning_rate": 1.9605823942483034e-05,
|
| 407 |
+
"loss": 1.7152,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.18398967075532602,
|
| 412 |
+
"grad_norm": 7.081708830235818,
|
| 413 |
+
"learning_rate": 1.9573889501326803e-05,
|
| 414 |
+
"loss": 1.7742,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.18721755971594578,
|
| 419 |
+
"grad_norm": 6.868160684535405,
|
| 420 |
+
"learning_rate": 1.9540739439652222e-05,
|
| 421 |
+
"loss": 1.6998,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.19044544867656552,
|
| 426 |
+
"grad_norm": 6.1971536722827185,
|
| 427 |
+
"learning_rate": 1.9506377966604923e-05,
|
| 428 |
+
"loss": 1.7291,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.19367333763718528,
|
| 433 |
+
"grad_norm": 6.757970184040331,
|
| 434 |
+
"learning_rate": 1.9470809445146447e-05,
|
| 435 |
+
"loss": 1.7305,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.19690122659780504,
|
| 440 |
+
"grad_norm": 7.044564037259735,
|
| 441 |
+
"learning_rate": 1.9434038391500266e-05,
|
| 442 |
+
"loss": 1.7215,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2001291155584248,
|
| 447 |
+
"grad_norm": 7.414383701728508,
|
| 448 |
+
"learning_rate": 1.9396069474578348e-05,
|
| 449 |
+
"loss": 1.6677,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.20335700451904454,
|
| 454 |
+
"grad_norm": 6.557964172050999,
|
| 455 |
+
"learning_rate": 1.935690751538834e-05,
|
| 456 |
+
"loss": 1.7939,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2065848934796643,
|
| 461 |
+
"grad_norm": 6.33082377097435,
|
| 462 |
+
"learning_rate": 1.9316557486421423e-05,
|
| 463 |
+
"loss": 1.7408,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.20981278244028406,
|
| 468 |
+
"grad_norm": 6.73298193200644,
|
| 469 |
+
"learning_rate": 1.927502451102095e-05,
|
| 470 |
+
"loss": 1.6857,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2130406714009038,
|
| 475 |
+
"grad_norm": 7.006395729787068,
|
| 476 |
+
"learning_rate": 1.9232313862731915e-05,
|
| 477 |
+
"loss": 1.6956,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.21626856036152356,
|
| 482 |
+
"grad_norm": 6.4834753162658805,
|
| 483 |
+
"learning_rate": 1.918843096463137e-05,
|
| 484 |
+
"loss": 1.6794,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.21949644932214332,
|
| 489 |
+
"grad_norm": 6.987943551728841,
|
| 490 |
+
"learning_rate": 1.914338138863983e-05,
|
| 491 |
+
"loss": 1.6969,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.22272433828276308,
|
| 496 |
+
"grad_norm": 6.192316163330726,
|
| 497 |
+
"learning_rate": 1.9097170854813797e-05,
|
| 498 |
+
"loss": 1.7381,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.22595222724338282,
|
| 503 |
+
"grad_norm": 6.4383979807985465,
|
| 504 |
+
"learning_rate": 1.904980523061948e-05,
|
| 505 |
+
"loss": 1.7072,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.22918011620400258,
|
| 510 |
+
"grad_norm": 6.443512513966274,
|
| 511 |
+
"learning_rate": 1.9001290530187768e-05,
|
| 512 |
+
"loss": 1.7401,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.23240800516462234,
|
| 517 |
+
"grad_norm": 5.964542231058296,
|
| 518 |
+
"learning_rate": 1.8951632913550625e-05,
|
| 519 |
+
"loss": 1.7035,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.23563589412524208,
|
| 524 |
+
"grad_norm": 7.1697046567958935,
|
| 525 |
+
"learning_rate": 1.8900838685858913e-05,
|
| 526 |
+
"loss": 1.7233,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.23886378308586184,
|
| 531 |
+
"grad_norm": 6.169546810862376,
|
| 532 |
+
"learning_rate": 1.884891429658182e-05,
|
| 533 |
+
"loss": 1.7021,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2420916720464816,
|
| 538 |
+
"grad_norm": 5.973896098846117,
|
| 539 |
+
"learning_rate": 1.8795866338687968e-05,
|
| 540 |
+
"loss": 1.7463,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.24531956100710137,
|
| 545 |
+
"grad_norm": 6.896426991389762,
|
| 546 |
+
"learning_rate": 1.8741701547808272e-05,
|
| 547 |
+
"loss": 1.6499,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2485474499677211,
|
| 552 |
+
"grad_norm": 6.952076945810176,
|
| 553 |
+
"learning_rate": 1.868642680138069e-05,
|
| 554 |
+
"loss": 1.767,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.25177533892834086,
|
| 559 |
+
"grad_norm": 6.18518307255409,
|
| 560 |
+
"learning_rate": 1.863004911777701e-05,
|
| 561 |
+
"loss": 1.7078,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2550032278889606,
|
| 566 |
+
"grad_norm": 6.632738286453584,
|
| 567 |
+
"learning_rate": 1.8572575655411683e-05,
|
| 568 |
+
"loss": 1.6845,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2582311168495804,
|
| 573 |
+
"grad_norm": 5.978535399494513,
|
| 574 |
+
"learning_rate": 1.851401371183291e-05,
|
| 575 |
+
"loss": 1.707,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.26145900581020015,
|
| 580 |
+
"grad_norm": 6.13506677347622,
|
| 581 |
+
"learning_rate": 1.8454370722796052e-05,
|
| 582 |
+
"loss": 1.6807,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.26468689477081986,
|
| 587 |
+
"grad_norm": 6.556694455290259,
|
| 588 |
+
"learning_rate": 1.8393654261319504e-05,
|
| 589 |
+
"loss": 1.7744,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2679147837314396,
|
| 594 |
+
"grad_norm": 6.113175455632477,
|
| 595 |
+
"learning_rate": 1.8331872036723103e-05,
|
| 596 |
+
"loss": 1.7065,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2711426726920594,
|
| 601 |
+
"grad_norm": 6.482825433419425,
|
| 602 |
+
"learning_rate": 1.8269031893649306e-05,
|
| 603 |
+
"loss": 1.6784,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.27437056165267915,
|
| 608 |
+
"grad_norm": 6.329309102633813,
|
| 609 |
+
"learning_rate": 1.8205141811067073e-05,
|
| 610 |
+
"loss": 1.683,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2775984506132989,
|
| 615 |
+
"grad_norm": 6.208972725984303,
|
| 616 |
+
"learning_rate": 1.814020990125881e-05,
|
| 617 |
+
"loss": 1.6445,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.28082633957391867,
|
| 622 |
+
"grad_norm": 5.818695993515501,
|
| 623 |
+
"learning_rate": 1.807424440879031e-05,
|
| 624 |
+
"loss": 1.7113,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.28405422853453843,
|
| 629 |
+
"grad_norm": 6.38696318289984,
|
| 630 |
+
"learning_rate": 1.8007253709463915e-05,
|
| 631 |
+
"loss": 1.6578,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.28728211749515814,
|
| 636 |
+
"grad_norm": 6.315382160526404,
|
| 637 |
+
"learning_rate": 1.7939246309255028e-05,
|
| 638 |
+
"loss": 1.6857,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2905100064557779,
|
| 643 |
+
"grad_norm": 6.233507637482026,
|
| 644 |
+
"learning_rate": 1.78702308432321e-05,
|
| 645 |
+
"loss": 1.7347,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.29373789541639767,
|
| 650 |
+
"grad_norm": 6.294349426994083,
|
| 651 |
+
"learning_rate": 1.7800216074460183e-05,
|
| 652 |
+
"loss": 1.7043,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2969657843770174,
|
| 657 |
+
"grad_norm": 6.343968862796106,
|
| 658 |
+
"learning_rate": 1.772921089288829e-05,
|
| 659 |
+
"loss": 1.7059,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3001936733376372,
|
| 664 |
+
"grad_norm": 6.673007692244266,
|
| 665 |
+
"learning_rate": 1.7657224314220604e-05,
|
| 666 |
+
"loss": 1.7047,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.30342156229825695,
|
| 671 |
+
"grad_norm": 6.729119057620871,
|
| 672 |
+
"learning_rate": 1.7584265478771737e-05,
|
| 673 |
+
"loss": 1.6781,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3066494512588767,
|
| 678 |
+
"grad_norm": 6.495304769382025,
|
| 679 |
+
"learning_rate": 1.7510343650306155e-05,
|
| 680 |
+
"loss": 1.708,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3098773402194964,
|
| 685 |
+
"grad_norm": 6.271850518966485,
|
| 686 |
+
"learning_rate": 1.7435468214861933e-05,
|
| 687 |
+
"loss": 1.6428,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3131052291801162,
|
| 692 |
+
"grad_norm": 6.423477074617556,
|
| 693 |
+
"learning_rate": 1.7359648679559006e-05,
|
| 694 |
+
"loss": 1.6194,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.31633311814073595,
|
| 699 |
+
"grad_norm": 6.221443392917989,
|
| 700 |
+
"learning_rate": 1.7282894671391996e-05,
|
| 701 |
+
"loss": 1.7025,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3195610071013557,
|
| 706 |
+
"grad_norm": 5.846396784880404,
|
| 707 |
+
"learning_rate": 1.720521593600787e-05,
|
| 708 |
+
"loss": 1.7269,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.32278889606197547,
|
| 713 |
+
"grad_norm": 6.477559026734868,
|
| 714 |
+
"learning_rate": 1.7126622336468514e-05,
|
| 715 |
+
"loss": 1.6395,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.32601678502259523,
|
| 720 |
+
"grad_norm": 6.055139296691825,
|
| 721 |
+
"learning_rate": 1.7047123851998374e-05,
|
| 722 |
+
"loss": 1.6782,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.329244673983215,
|
| 727 |
+
"grad_norm": 6.118162992939965,
|
| 728 |
+
"learning_rate": 1.6966730576717388e-05,
|
| 729 |
+
"loss": 1.682,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.33247256294383476,
|
| 734 |
+
"grad_norm": 5.5534801726338525,
|
| 735 |
+
"learning_rate": 1.6885452718359306e-05,
|
| 736 |
+
"loss": 1.6267,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.33570045190445447,
|
| 741 |
+
"grad_norm": 5.43602711879506,
|
| 742 |
+
"learning_rate": 1.6803300596975586e-05,
|
| 743 |
+
"loss": 1.6452,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.33892834086507423,
|
| 748 |
+
"grad_norm": 7.206960139551292,
|
| 749 |
+
"learning_rate": 1.6720284643625035e-05,
|
| 750 |
+
"loss": 1.6583,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.342156229825694,
|
| 755 |
+
"grad_norm": 6.427854787428326,
|
| 756 |
+
"learning_rate": 1.6636415399049347e-05,
|
| 757 |
+
"loss": 1.6091,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.34538411878631375,
|
| 762 |
+
"grad_norm": 5.816051751586562,
|
| 763 |
+
"learning_rate": 1.6551703512334716e-05,
|
| 764 |
+
"loss": 1.6913,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3486120077469335,
|
| 769 |
+
"grad_norm": 5.927556526774363,
|
| 770 |
+
"learning_rate": 1.6466159739559712e-05,
|
| 771 |
+
"loss": 1.6738,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3518398967075533,
|
| 776 |
+
"grad_norm": 6.455088888441903,
|
| 777 |
+
"learning_rate": 1.6379794942429534e-05,
|
| 778 |
+
"loss": 1.648,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.35506778566817304,
|
| 783 |
+
"grad_norm": 6.240988706855586,
|
| 784 |
+
"learning_rate": 1.629262008689689e-05,
|
| 785 |
+
"loss": 1.6927,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.35829567462879275,
|
| 790 |
+
"grad_norm": 5.86555967170564,
|
| 791 |
+
"learning_rate": 1.62046462417696e-05,
|
| 792 |
+
"loss": 1.6388,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3615235635894125,
|
| 797 |
+
"grad_norm": 5.66373254985077,
|
| 798 |
+
"learning_rate": 1.611588457730519e-05,
|
| 799 |
+
"loss": 1.6947,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3647514525500323,
|
| 804 |
+
"grad_norm": 5.979046700601131,
|
| 805 |
+
"learning_rate": 1.6026346363792565e-05,
|
| 806 |
+
"loss": 1.6829,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.36797934151065204,
|
| 811 |
+
"grad_norm": 5.727328551303314,
|
| 812 |
+
"learning_rate": 1.5936042970120976e-05,
|
| 813 |
+
"loss": 1.6646,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3712072304712718,
|
| 818 |
+
"grad_norm": 6.089199758516902,
|
| 819 |
+
"learning_rate": 1.5844985862336516e-05,
|
| 820 |
+
"loss": 1.6665,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.37443511943189156,
|
| 825 |
+
"grad_norm": 5.587005635034525,
|
| 826 |
+
"learning_rate": 1.5753186602186207e-05,
|
| 827 |
+
"loss": 1.6417,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3776630083925113,
|
| 832 |
+
"grad_norm": 6.050996843101115,
|
| 833 |
+
"learning_rate": 1.5660656845650027e-05,
|
| 834 |
+
"loss": 1.5796,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.38089089735313103,
|
| 839 |
+
"grad_norm": 6.080615404268494,
|
| 840 |
+
"learning_rate": 1.556740834146087e-05,
|
| 841 |
+
"loss": 1.6976,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3841187863137508,
|
| 846 |
+
"grad_norm": 6.474714539476243,
|
| 847 |
+
"learning_rate": 1.547345292961282e-05,
|
| 848 |
+
"loss": 1.6356,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.38734667527437056,
|
| 853 |
+
"grad_norm": 5.877661742364028,
|
| 854 |
+
"learning_rate": 1.5378802539857775e-05,
|
| 855 |
+
"loss": 1.6992,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3905745642349903,
|
| 860 |
+
"grad_norm": 5.866696710467355,
|
| 861 |
+
"learning_rate": 1.52834691901907e-05,
|
| 862 |
+
"loss": 1.6606,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3938024531956101,
|
| 867 |
+
"grad_norm": 6.398661395189826,
|
| 868 |
+
"learning_rate": 1.5187464985323681e-05,
|
| 869 |
+
"loss": 1.7004,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.39703034215622984,
|
| 874 |
+
"grad_norm": 6.109162698783035,
|
| 875 |
+
"learning_rate": 1.5090802115148956e-05,
|
| 876 |
+
"loss": 1.664,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4002582311168496,
|
| 881 |
+
"grad_norm": 5.731699946494037,
|
| 882 |
+
"learning_rate": 1.4993492853191118e-05,
|
| 883 |
+
"loss": 1.6559,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4034861200774693,
|
| 888 |
+
"grad_norm": 5.4571384796165185,
|
| 889 |
+
"learning_rate": 1.4895549555048751e-05,
|
| 890 |
+
"loss": 1.6327,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4067140090380891,
|
| 895 |
+
"grad_norm": 5.872513791622213,
|
| 896 |
+
"learning_rate": 1.4796984656825572e-05,
|
| 897 |
+
"loss": 1.625,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.40994189799870884,
|
| 902 |
+
"grad_norm": 5.288501071612407,
|
| 903 |
+
"learning_rate": 1.4697810673551408e-05,
|
| 904 |
+
"loss": 1.6659,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4131697869593286,
|
| 909 |
+
"grad_norm": 6.468751146768027,
|
| 910 |
+
"learning_rate": 1.4598040197593128e-05,
|
| 911 |
+
"loss": 1.7126,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.41639767591994836,
|
| 916 |
+
"grad_norm": 5.698799707780423,
|
| 917 |
+
"learning_rate": 1.4497685897055758e-05,
|
| 918 |
+
"loss": 1.5823,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4196255648805681,
|
| 923 |
+
"grad_norm": 5.654573938892979,
|
| 924 |
+
"learning_rate": 1.4396760514173976e-05,
|
| 925 |
+
"loss": 1.6498,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4228534538411879,
|
| 930 |
+
"grad_norm": 6.976746797511936,
|
| 931 |
+
"learning_rate": 1.4295276863694205e-05,
|
| 932 |
+
"loss": 1.6113,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4260813428018076,
|
| 937 |
+
"grad_norm": 5.235970407365906,
|
| 938 |
+
"learning_rate": 1.4193247831247499e-05,
|
| 939 |
+
"loss": 1.7006,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.42930923176242736,
|
| 944 |
+
"grad_norm": 6.140392016522079,
|
| 945 |
+
"learning_rate": 1.4090686371713403e-05,
|
| 946 |
+
"loss": 1.6836,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4325371207230471,
|
| 951 |
+
"grad_norm": 5.45721311029038,
|
| 952 |
+
"learning_rate": 1.3987605507575053e-05,
|
| 953 |
+
"loss": 1.6306,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4357650096836669,
|
| 958 |
+
"grad_norm": 6.83263868068613,
|
| 959 |
+
"learning_rate": 1.3884018327265683e-05,
|
| 960 |
+
"loss": 1.7172,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.43899289864428664,
|
| 965 |
+
"grad_norm": 5.843200304850912,
|
| 966 |
+
"learning_rate": 1.3779937983506746e-05,
|
| 967 |
+
"loss": 1.6762,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4422207876049064,
|
| 972 |
+
"grad_norm": 5.57641828737366,
|
| 973 |
+
"learning_rate": 1.3675377691637879e-05,
|
| 974 |
+
"loss": 1.6778,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.44544867656552617,
|
| 979 |
+
"grad_norm": 5.738260096617323,
|
| 980 |
+
"learning_rate": 1.3570350727938925e-05,
|
| 981 |
+
"loss": 1.6378,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4486765655261459,
|
| 986 |
+
"grad_norm": 6.357451662380971,
|
| 987 |
+
"learning_rate": 1.3464870427944208e-05,
|
| 988 |
+
"loss": 1.6041,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.45190445448676564,
|
| 993 |
+
"grad_norm": 5.484551406388904,
|
| 994 |
+
"learning_rate": 1.3358950184749284e-05,
|
| 995 |
+
"loss": 1.6612,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4551323434473854,
|
| 1000 |
+
"grad_norm": 5.827167388471347,
|
| 1001 |
+
"learning_rate": 1.3252603447310396e-05,
|
| 1002 |
+
"loss": 1.5759,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.45836023240800516,
|
| 1007 |
+
"grad_norm": 5.980523013123908,
|
| 1008 |
+
"learning_rate": 1.3145843718736809e-05,
|
| 1009 |
+
"loss": 1.6716,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4615881213686249,
|
| 1014 |
+
"grad_norm": 6.134764370156555,
|
| 1015 |
+
"learning_rate": 1.3038684554576308e-05,
|
| 1016 |
+
"loss": 1.6406,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4648160103292447,
|
| 1021 |
+
"grad_norm": 5.487517622123662,
|
| 1022 |
+
"learning_rate": 1.2931139561094007e-05,
|
| 1023 |
+
"loss": 1.5756,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.46804389928986445,
|
| 1028 |
+
"grad_norm": 5.4981090259389696,
|
| 1029 |
+
"learning_rate": 1.2823222393544717e-05,
|
| 1030 |
+
"loss": 1.5797,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.47127178825048416,
|
| 1035 |
+
"grad_norm": 5.219388647345398,
|
| 1036 |
+
"learning_rate": 1.2714946754439117e-05,
|
| 1037 |
+
"loss": 1.5944,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4744996772111039,
|
| 1042 |
+
"grad_norm": 5.824820667094204,
|
| 1043 |
+
"learning_rate": 1.2606326391803915e-05,
|
| 1044 |
+
"loss": 1.58,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4777275661717237,
|
| 1049 |
+
"grad_norm": 5.786554762407817,
|
| 1050 |
+
"learning_rate": 1.249737509743622e-05,
|
| 1051 |
+
"loss": 1.6459,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.48095545513234345,
|
| 1056 |
+
"grad_norm": 5.8901639628553095,
|
| 1057 |
+
"learning_rate": 1.2388106705152361e-05,
|
| 1058 |
+
"loss": 1.5922,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4841833440929632,
|
| 1063 |
+
"grad_norm": 5.320173777546997,
|
| 1064 |
+
"learning_rate": 1.2278535089031377e-05,
|
| 1065 |
+
"loss": 1.6169,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.48741123305358297,
|
| 1070 |
+
"grad_norm": 5.377921836454134,
|
| 1071 |
+
"learning_rate": 1.2168674161653395e-05,
|
| 1072 |
+
"loss": 1.5759,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.49063912201420273,
|
| 1077 |
+
"grad_norm": 5.5810578215639435,
|
| 1078 |
+
"learning_rate": 1.2058537872333104e-05,
|
| 1079 |
+
"loss": 1.5469,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.49386701097482244,
|
| 1084 |
+
"grad_norm": 4.880835940718014,
|
| 1085 |
+
"learning_rate": 1.1948140205348592e-05,
|
| 1086 |
+
"loss": 1.5735,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4970948999354422,
|
| 1091 |
+
"grad_norm": 5.615682640085302,
|
| 1092 |
+
"learning_rate": 1.1837495178165706e-05,
|
| 1093 |
+
"loss": 1.5648,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.500322788896062,
|
| 1098 |
+
"grad_norm": 5.322385655290283,
|
| 1099 |
+
"learning_rate": 1.1726616839658237e-05,
|
| 1100 |
+
"loss": 1.6223,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5035506778566817,
|
| 1105 |
+
"grad_norm": 5.372959800238378,
|
| 1106 |
+
"learning_rate": 1.1615519268324101e-05,
|
| 1107 |
+
"loss": 1.5339,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5067785668173015,
|
| 1112 |
+
"grad_norm": 5.8491464237293,
|
| 1113 |
+
"learning_rate": 1.1504216570497737e-05,
|
| 1114 |
+
"loss": 1.5575,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5100064557779213,
|
| 1119 |
+
"grad_norm": 5.292821692941134,
|
| 1120 |
+
"learning_rate": 1.1392722878559012e-05,
|
| 1121 |
+
"loss": 1.6182,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.513234344738541,
|
| 1126 |
+
"grad_norm": 5.798146550880112,
|
| 1127 |
+
"learning_rate": 1.1281052349138793e-05,
|
| 1128 |
+
"loss": 1.5844,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5164622336991608,
|
| 1133 |
+
"grad_norm": 5.6823764366973935,
|
| 1134 |
+
"learning_rate": 1.116921916132143e-05,
|
| 1135 |
+
"loss": 1.5739,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5196901226597805,
|
| 1140 |
+
"grad_norm": 5.761134241104277,
|
| 1141 |
+
"learning_rate": 1.1057237514844423e-05,
|
| 1142 |
+
"loss": 1.6409,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5229180116204003,
|
| 1147 |
+
"grad_norm": 4.938806993669054,
|
| 1148 |
+
"learning_rate": 1.0945121628295437e-05,
|
| 1149 |
+
"loss": 1.5921,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5261459005810201,
|
| 1154 |
+
"grad_norm": 15.683819077262335,
|
| 1155 |
+
"learning_rate": 1.0832885737306922e-05,
|
| 1156 |
+
"loss": 1.7035,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5293737895416397,
|
| 1161 |
+
"grad_norm": 5.4993649210417965,
|
| 1162 |
+
"learning_rate": 1.0720544092748599e-05,
|
| 1163 |
+
"loss": 1.5382,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5326016785022595,
|
| 1168 |
+
"grad_norm": 5.092222218274809,
|
| 1169 |
+
"learning_rate": 1.0608110958917982e-05,
|
| 1170 |
+
"loss": 1.6136,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5358295674628792,
|
| 1175 |
+
"grad_norm": 4.930191624792412,
|
| 1176 |
+
"learning_rate": 1.04956006117292e-05,
|
| 1177 |
+
"loss": 1.5622,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.539057456423499,
|
| 1182 |
+
"grad_norm": 5.399945563067967,
|
| 1183 |
+
"learning_rate": 1.0383027336900356e-05,
|
| 1184 |
+
"loss": 1.5412,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5422853453841188,
|
| 1189 |
+
"grad_norm": 5.781669659732203,
|
| 1190 |
+
"learning_rate": 1.0270405428139633e-05,
|
| 1191 |
+
"loss": 1.6486,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5455132343447385,
|
| 1196 |
+
"grad_norm": 5.519605623300351,
|
| 1197 |
+
"learning_rate": 1.0157749185330384e-05,
|
| 1198 |
+
"loss": 1.6219,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5487411233053583,
|
| 1203 |
+
"grad_norm": 6.008340474166314,
|
| 1204 |
+
"learning_rate": 1.0045072912715443e-05,
|
| 1205 |
+
"loss": 1.6015,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.551969012265978,
|
| 1210 |
+
"grad_norm": 5.131256451670003,
|
| 1211 |
+
"learning_rate": 9.932390917080874e-06,
|
| 1212 |
+
"loss": 1.566,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5551969012265978,
|
| 1217 |
+
"grad_norm": 5.419158342369302,
|
| 1218 |
+
"learning_rate": 9.81971750593941e-06,
|
| 1219 |
+
"loss": 1.5701,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5584247901872176,
|
| 1224 |
+
"grad_norm": 5.354220565411314,
|
| 1225 |
+
"learning_rate": 9.707066985713795e-06,
|
| 1226 |
+
"loss": 1.5487,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5616526791478373,
|
| 1231 |
+
"grad_norm": 4.995550545914696,
|
| 1232 |
+
"learning_rate": 9.594453659920241e-06,
|
| 1233 |
+
"loss": 1.5052,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5648805681084571,
|
| 1238 |
+
"grad_norm": 5.224319860529777,
|
| 1239 |
+
"learning_rate": 9.48189182735231e-06,
|
| 1240 |
+
"loss": 1.5371,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5681084570690769,
|
| 1245 |
+
"grad_norm": 5.240828267758849,
|
| 1246 |
+
"learning_rate": 9.369395780265323e-06,
|
| 1247 |
+
"loss": 1.5537,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5713363460296966,
|
| 1252 |
+
"grad_norm": 5.4919915569394915,
|
| 1253 |
+
"learning_rate": 9.256979802561675e-06,
|
| 1254 |
+
"loss": 1.5706,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5745642349903163,
|
| 1259 |
+
"grad_norm": 5.198809451347601,
|
| 1260 |
+
"learning_rate": 9.144658167977134e-06,
|
| 1261 |
+
"loss": 1.5988,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.577792123950936,
|
| 1266 |
+
"grad_norm": 5.389465699869633,
|
| 1267 |
+
"learning_rate": 9.032445138268493e-06,
|
| 1268 |
+
"loss": 1.5407,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5810200129115558,
|
| 1273 |
+
"grad_norm": 5.495352080530079,
|
| 1274 |
+
"learning_rate": 8.920354961402723e-06,
|
| 1275 |
+
"loss": 1.5559,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5842479018721756,
|
| 1280 |
+
"grad_norm": 5.612791353311329,
|
| 1281 |
+
"learning_rate": 8.808401869747858e-06,
|
| 1282 |
+
"loss": 1.5855,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5874757908327953,
|
| 1287 |
+
"grad_norm": 5.400883996398841,
|
| 1288 |
+
"learning_rate": 8.696600078265876e-06,
|
| 1289 |
+
"loss": 1.5122,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5907036797934151,
|
| 1294 |
+
"grad_norm": 4.832336508421307,
|
| 1295 |
+
"learning_rate": 8.584963782707812e-06,
|
| 1296 |
+
"loss": 1.5118,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5939315687540349,
|
| 1301 |
+
"grad_norm": 5.0951094528676455,
|
| 1302 |
+
"learning_rate": 8.473507157811254e-06,
|
| 1303 |
+
"loss": 1.4892,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5971594577146546,
|
| 1308 |
+
"grad_norm": 5.8252974300330225,
|
| 1309 |
+
"learning_rate": 8.362244355500583e-06,
|
| 1310 |
+
"loss": 1.5714,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6003873466752744,
|
| 1315 |
+
"grad_norm": 5.236089319233509,
|
| 1316 |
+
"learning_rate": 8.251189503090023e-06,
|
| 1317 |
+
"loss": 1.5833,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6036152356358941,
|
| 1322 |
+
"grad_norm": 5.343216112400599,
|
| 1323 |
+
"learning_rate": 8.140356701489892e-06,
|
| 1324 |
+
"loss": 1.4814,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6068431245965139,
|
| 1329 |
+
"grad_norm": 4.908961690980711,
|
| 1330 |
+
"learning_rate": 8.029760023416168e-06,
|
| 1331 |
+
"loss": 1.5093,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6100710135571337,
|
| 1336 |
+
"grad_norm": 5.245691260381996,
|
| 1337 |
+
"learning_rate": 7.919413511603636e-06,
|
| 1338 |
+
"loss": 1.5963,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6132989025177534,
|
| 1343 |
+
"grad_norm": 5.621910081780513,
|
| 1344 |
+
"learning_rate": 7.809331177022847e-06,
|
| 1345 |
+
"loss": 1.5196,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6165267914783732,
|
| 1350 |
+
"grad_norm": 5.8427500582760175,
|
| 1351 |
+
"learning_rate": 7.699526997101121e-06,
|
| 1352 |
+
"loss": 1.5567,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6197546804389928,
|
| 1357 |
+
"grad_norm": 5.683875417006101,
|
| 1358 |
+
"learning_rate": 7.590014913947778e-06,
|
| 1359 |
+
"loss": 1.5644,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6229825693996126,
|
| 1364 |
+
"grad_norm": 5.458609295388963,
|
| 1365 |
+
"learning_rate": 7.4808088325839e-06,
|
| 1366 |
+
"loss": 1.523,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6262104583602324,
|
| 1371 |
+
"grad_norm": 5.991795522009256,
|
| 1372 |
+
"learning_rate": 7.3719226191767526e-06,
|
| 1373 |
+
"loss": 1.5572,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6294383473208521,
|
| 1378 |
+
"grad_norm": 5.104396720609387,
|
| 1379 |
+
"learning_rate": 7.263370099279173e-06,
|
| 1380 |
+
"loss": 1.5125,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6326662362814719,
|
| 1385 |
+
"grad_norm": 5.179959578065037,
|
| 1386 |
+
"learning_rate": 7.155165056074111e-06,
|
| 1387 |
+
"loss": 1.5097,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6358941252420917,
|
| 1392 |
+
"grad_norm": 5.547151238761799,
|
| 1393 |
+
"learning_rate": 7.0473212286245316e-06,
|
| 1394 |
+
"loss": 1.581,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6391220142027114,
|
| 1399 |
+
"grad_norm": 5.906714474067254,
|
| 1400 |
+
"learning_rate": 6.9398523101289475e-06,
|
| 1401 |
+
"loss": 1.5799,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6423499031633312,
|
| 1406 |
+
"grad_norm": 5.211200331104552,
|
| 1407 |
+
"learning_rate": 6.832771946182741e-06,
|
| 1408 |
+
"loss": 1.5565,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6455777921239509,
|
| 1413 |
+
"grad_norm": 5.268610977952413,
|
| 1414 |
+
"learning_rate": 6.72609373304556e-06,
|
| 1415 |
+
"loss": 1.4848,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6488056810845707,
|
| 1420 |
+
"grad_norm": 5.439250857230896,
|
| 1421 |
+
"learning_rate": 6.619831215914974e-06,
|
| 1422 |
+
"loss": 1.5022,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6520335700451905,
|
| 1427 |
+
"grad_norm": 5.360521278150201,
|
| 1428 |
+
"learning_rate": 6.5139978872065865e-06,
|
| 1429 |
+
"loss": 1.4935,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6552614590058102,
|
| 1434 |
+
"grad_norm": 5.796210278315953,
|
| 1435 |
+
"learning_rate": 6.408607184840897e-06,
|
| 1436 |
+
"loss": 1.5625,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.65848934796643,
|
| 1441 |
+
"grad_norm": 5.020044853823784,
|
| 1442 |
+
"learning_rate": 6.303672490537022e-06,
|
| 1443 |
+
"loss": 1.4925,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6617172369270498,
|
| 1448 |
+
"grad_norm": 5.291440134096366,
|
| 1449 |
+
"learning_rate": 6.199207128113614e-06,
|
| 1450 |
+
"loss": 1.5557,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6649451258876695,
|
| 1455 |
+
"grad_norm": 5.239441257851841,
|
| 1456 |
+
"learning_rate": 6.095224361797087e-06,
|
| 1457 |
+
"loss": 1.582,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6681730148482892,
|
| 1462 |
+
"grad_norm": 5.202258142662834,
|
| 1463 |
+
"learning_rate": 5.991737394537434e-06,
|
| 1464 |
+
"loss": 1.5321,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6714009038089089,
|
| 1469 |
+
"grad_norm": 6.1919229333140375,
|
| 1470 |
+
"learning_rate": 5.888759366331798e-06,
|
| 1471 |
+
"loss": 1.5242,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6746287927695287,
|
| 1476 |
+
"grad_norm": 5.057056436772678,
|
| 1477 |
+
"learning_rate": 5.786303352556088e-06,
|
| 1478 |
+
"loss": 1.4598,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6778566817301485,
|
| 1483 |
+
"grad_norm": 5.148137854904367,
|
| 1484 |
+
"learning_rate": 5.684382362304723e-06,
|
| 1485 |
+
"loss": 1.5076,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6810845706907682,
|
| 1490 |
+
"grad_norm": 5.515091391109832,
|
| 1491 |
+
"learning_rate": 5.583009336738874e-06,
|
| 1492 |
+
"loss": 1.5496,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.684312459651388,
|
| 1497 |
+
"grad_norm": 5.784777023225698,
|
| 1498 |
+
"learning_rate": 5.48219714744326e-06,
|
| 1499 |
+
"loss": 1.5131,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6875403486120077,
|
| 1504 |
+
"grad_norm": 5.59776729212278,
|
| 1505 |
+
"learning_rate": 5.381958594791843e-06,
|
| 1506 |
+
"loss": 1.4729,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6907682375726275,
|
| 1511 |
+
"grad_norm": 5.1330532599827485,
|
| 1512 |
+
"learning_rate": 5.282306406322502e-06,
|
| 1513 |
+
"loss": 1.5357,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6939961265332473,
|
| 1518 |
+
"grad_norm": 5.530634545266258,
|
| 1519 |
+
"learning_rate": 5.1832532351210095e-06,
|
| 1520 |
+
"loss": 1.5159,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.697224015493867,
|
| 1525 |
+
"grad_norm": 5.190707271947593,
|
| 1526 |
+
"learning_rate": 5.084811658214421e-06,
|
| 1527 |
+
"loss": 1.5047,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7004519044544868,
|
| 1532 |
+
"grad_norm": 5.554567104196739,
|
| 1533 |
+
"learning_rate": 4.986994174974155e-06,
|
| 1534 |
+
"loss": 1.5259,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7036797934151066,
|
| 1539 |
+
"grad_norm": 5.540622719340587,
|
| 1540 |
+
"learning_rate": 4.889813205528895e-06,
|
| 1541 |
+
"loss": 1.4371,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7069076823757263,
|
| 1546 |
+
"grad_norm": 5.532974890894428,
|
| 1547 |
+
"learning_rate": 4.793281089187603e-06,
|
| 1548 |
+
"loss": 1.4397,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7101355713363461,
|
| 1553 |
+
"grad_norm": 5.084284858142072,
|
| 1554 |
+
"learning_rate": 4.697410082872741e-06,
|
| 1555 |
+
"loss": 1.4826,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7133634602969657,
|
| 1560 |
+
"grad_norm": 5.374832286921831,
|
| 1561 |
+
"learning_rate": 4.602212359563988e-06,
|
| 1562 |
+
"loss": 1.5031,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7165913492575855,
|
| 1567 |
+
"grad_norm": 5.436449921979155,
|
| 1568 |
+
"learning_rate": 4.50770000675262e-06,
|
| 1569 |
+
"loss": 1.482,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7198192382182053,
|
| 1574 |
+
"grad_norm": 5.367567637926529,
|
| 1575 |
+
"learning_rate": 4.413885024906705e-06,
|
| 1576 |
+
"loss": 1.492,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.723047127178825,
|
| 1581 |
+
"grad_norm": 5.459559639213619,
|
| 1582 |
+
"learning_rate": 4.320779325947402e-06,
|
| 1583 |
+
"loss": 1.4527,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7262750161394448,
|
| 1588 |
+
"grad_norm": 5.710733826769647,
|
| 1589 |
+
"learning_rate": 4.228394731736451e-06,
|
| 1590 |
+
"loss": 1.526,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7295029051000645,
|
| 1595 |
+
"grad_norm": 4.911675403161043,
|
| 1596 |
+
"learning_rate": 4.136742972575132e-06,
|
| 1597 |
+
"loss": 1.5134,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7327307940606843,
|
| 1602 |
+
"grad_norm": 5.602840040740436,
|
| 1603 |
+
"learning_rate": 4.045835685714848e-06,
|
| 1604 |
+
"loss": 1.4289,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7359586830213041,
|
| 1609 |
+
"grad_norm": 5.035789822163729,
|
| 1610 |
+
"learning_rate": 3.955684413879499e-06,
|
| 1611 |
+
"loss": 1.4759,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7391865719819238,
|
| 1616 |
+
"grad_norm": 5.094039088957984,
|
| 1617 |
+
"learning_rate": 3.866300603799876e-06,
|
| 1618 |
+
"loss": 1.4575,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7424144609425436,
|
| 1623 |
+
"grad_norm": 5.49222076162649,
|
| 1624 |
+
"learning_rate": 3.77769560476026e-06,
|
| 1625 |
+
"loss": 1.4581,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7456423499031634,
|
| 1630 |
+
"grad_norm": 5.715064526970072,
|
| 1631 |
+
"learning_rate": 3.689880667157344e-06,
|
| 1632 |
+
"loss": 1.5579,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7488702388637831,
|
| 1637 |
+
"grad_norm": 5.524488815265337,
|
| 1638 |
+
"learning_rate": 3.602866941071773e-06,
|
| 1639 |
+
"loss": 1.4336,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7520981278244029,
|
| 1644 |
+
"grad_norm": 5.364873008254881,
|
| 1645 |
+
"learning_rate": 3.516665474852369e-06,
|
| 1646 |
+
"loss": 1.4432,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7553260167850226,
|
| 1651 |
+
"grad_norm": 5.2752700833072845,
|
| 1652 |
+
"learning_rate": 3.431287213713299e-06,
|
| 1653 |
+
"loss": 1.4944,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7585539057456423,
|
| 1658 |
+
"grad_norm": 5.2004940717802,
|
| 1659 |
+
"learning_rate": 3.3467429983443477e-06,
|
| 1660 |
+
"loss": 1.4047,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7617817947062621,
|
| 1665 |
+
"grad_norm": 5.1840162953693545,
|
| 1666 |
+
"learning_rate": 3.2630435635344283e-06,
|
| 1667 |
+
"loss": 1.4254,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7650096836668818,
|
| 1672 |
+
"grad_norm": 5.488420884341952,
|
| 1673 |
+
"learning_rate": 3.180199536808576e-06,
|
| 1674 |
+
"loss": 1.414,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7682375726275016,
|
| 1679 |
+
"grad_norm": 5.189126496815558,
|
| 1680 |
+
"learning_rate": 3.0982214370785335e-06,
|
| 1681 |
+
"loss": 1.4575,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7714654615881213,
|
| 1686 |
+
"grad_norm": 5.038917252480004,
|
| 1687 |
+
"learning_rate": 3.0171196733071405e-06,
|
| 1688 |
+
"loss": 1.5069,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7746933505487411,
|
| 1693 |
+
"grad_norm": 5.1969535616381926,
|
| 1694 |
+
"learning_rate": 2.936904543186695e-06,
|
| 1695 |
+
"loss": 1.4508,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7779212395093609,
|
| 1700 |
+
"grad_norm": 5.160164936057216,
|
| 1701 |
+
"learning_rate": 2.8575862318314087e-06,
|
| 1702 |
+
"loss": 1.4409,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7811491284699806,
|
| 1707 |
+
"grad_norm": 5.3476935943523385,
|
| 1708 |
+
"learning_rate": 2.7791748104841876e-06,
|
| 1709 |
+
"loss": 1.4374,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7843770174306004,
|
| 1714 |
+
"grad_norm": 5.370458539003402,
|
| 1715 |
+
"learning_rate": 2.7016802352378755e-06,
|
| 1716 |
+
"loss": 1.4268,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7876049063912202,
|
| 1721 |
+
"grad_norm": 5.045239556202245,
|
| 1722 |
+
"learning_rate": 2.6251123457710726e-06,
|
| 1723 |
+
"loss": 1.4379,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7908327953518399,
|
| 1728 |
+
"grad_norm": 5.271081605175723,
|
| 1729 |
+
"learning_rate": 2.5494808640988e-06,
|
| 1730 |
+
"loss": 1.474,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7940606843124597,
|
| 1735 |
+
"grad_norm": 5.4078856513908296,
|
| 1736 |
+
"learning_rate": 2.4747953933380463e-06,
|
| 1737 |
+
"loss": 1.476,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7972885732730794,
|
| 1742 |
+
"grad_norm": 5.765653151999692,
|
| 1743 |
+
"learning_rate": 2.4010654164884527e-06,
|
| 1744 |
+
"loss": 1.4782,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8005164622336992,
|
| 1749 |
+
"grad_norm": 5.092525169561813,
|
| 1750 |
+
"learning_rate": 2.328300295228221e-06,
|
| 1751 |
+
"loss": 1.4497,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8037443511943189,
|
| 1756 |
+
"grad_norm": 4.722681501392939,
|
| 1757 |
+
"learning_rate": 2.2565092687254486e-06,
|
| 1758 |
+
"loss": 1.466,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8069722401549386,
|
| 1763 |
+
"grad_norm": 5.057426901667458,
|
| 1764 |
+
"learning_rate": 2.185701452464997e-06,
|
| 1765 |
+
"loss": 1.4358,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8102001291155584,
|
| 1770 |
+
"grad_norm": 4.837982928648199,
|
| 1771 |
+
"learning_rate": 2.1158858370910905e-06,
|
| 1772 |
+
"loss": 1.4249,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8134280180761781,
|
| 1777 |
+
"grad_norm": 5.165161107325633,
|
| 1778 |
+
"learning_rate": 2.047071287265735e-06,
|
| 1779 |
+
"loss": 1.4211,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8166559070367979,
|
| 1784 |
+
"grad_norm": 5.068834680044366,
|
| 1785 |
+
"learning_rate": 1.9792665405431654e-06,
|
| 1786 |
+
"loss": 1.3746,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8198837959974177,
|
| 1791 |
+
"grad_norm": 5.453220592450874,
|
| 1792 |
+
"learning_rate": 1.9124802062604066e-06,
|
| 1793 |
+
"loss": 1.456,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8231116849580374,
|
| 1798 |
+
"grad_norm": 5.648842929433009,
|
| 1799 |
+
"learning_rate": 1.8467207644441243e-06,
|
| 1800 |
+
"loss": 1.4727,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8263395739186572,
|
| 1805 |
+
"grad_norm": 5.304857997339736,
|
| 1806 |
+
"learning_rate": 1.7819965647339065e-06,
|
| 1807 |
+
"loss": 1.4979,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.829567462879277,
|
| 1812 |
+
"grad_norm": 5.45097229329167,
|
| 1813 |
+
"learning_rate": 1.718315825322071e-06,
|
| 1814 |
+
"loss": 1.4318,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8327953518398967,
|
| 1819 |
+
"grad_norm": 5.021145599454475,
|
| 1820 |
+
"learning_rate": 1.6556866319102027e-06,
|
| 1821 |
+
"loss": 1.4988,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8360232408005165,
|
| 1826 |
+
"grad_norm": 5.402016832769367,
|
| 1827 |
+
"learning_rate": 1.5941169366824705e-06,
|
| 1828 |
+
"loss": 1.4616,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8392511297611362,
|
| 1833 |
+
"grad_norm": 5.399074896944535,
|
| 1834 |
+
"learning_rate": 1.5336145572959283e-06,
|
| 1835 |
+
"loss": 1.4384,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.842479018721756,
|
| 1840 |
+
"grad_norm": 5.471814490238626,
|
| 1841 |
+
"learning_rate": 1.4741871758878978e-06,
|
| 1842 |
+
"loss": 1.4333,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8457069076823758,
|
| 1847 |
+
"grad_norm": 5.289307631746622,
|
| 1848 |
+
"learning_rate": 1.4158423381005294e-06,
|
| 1849 |
+
"loss": 1.4523,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8489347966429954,
|
| 1854 |
+
"grad_norm": 5.020917216038537,
|
| 1855 |
+
"learning_rate": 1.3585874521227226e-06,
|
| 1856 |
+
"loss": 1.3762,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8521626856036152,
|
| 1861 |
+
"grad_norm": 5.00624606287639,
|
| 1862 |
+
"learning_rate": 1.3024297877494973e-06,
|
| 1863 |
+
"loss": 1.4612,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.855390574564235,
|
| 1868 |
+
"grad_norm": 5.429810940921615,
|
| 1869 |
+
"learning_rate": 1.2473764754589123e-06,
|
| 1870 |
+
"loss": 1.4305,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8586184635248547,
|
| 1875 |
+
"grad_norm": 5.1399015375442385,
|
| 1876 |
+
"learning_rate": 1.193434505506711e-06,
|
| 1877 |
+
"loss": 1.4399,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8618463524854745,
|
| 1882 |
+
"grad_norm": 4.9707825715846665,
|
| 1883 |
+
"learning_rate": 1.1406107270387323e-06,
|
| 1884 |
+
"loss": 1.4398,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8650742414460942,
|
| 1889 |
+
"grad_norm": 5.053643246086415,
|
| 1890 |
+
"learning_rate": 1.0889118472212702e-06,
|
| 1891 |
+
"loss": 1.4304,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.868302130406714,
|
| 1896 |
+
"grad_norm": 5.028185854558994,
|
| 1897 |
+
"learning_rate": 1.0383444303894453e-06,
|
| 1898 |
+
"loss": 1.4038,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8715300193673338,
|
| 1903 |
+
"grad_norm": 5.112792724899338,
|
| 1904 |
+
"learning_rate": 9.889148972137097e-07,
|
| 1905 |
+
"loss": 1.4334,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8747579083279535,
|
| 1910 |
+
"grad_norm": 5.1166189472841275,
|
| 1911 |
+
"learning_rate": 9.406295238846108e-07,
|
| 1912 |
+
"loss": 1.3857,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8779857972885733,
|
| 1917 |
+
"grad_norm": 4.705706257850879,
|
| 1918 |
+
"learning_rate": 8.934944413158708e-07,
|
| 1919 |
+
"loss": 1.437,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.881213686249193,
|
| 1924 |
+
"grad_norm": 5.788807086151138,
|
| 1925 |
+
"learning_rate": 8.475156343659408e-07,
|
| 1926 |
+
"loss": 1.4031,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8844415752098128,
|
| 1931 |
+
"grad_norm": 6.376811639197679,
|
| 1932 |
+
"learning_rate": 8.026989410780917e-07,
|
| 1933 |
+
"loss": 1.4084,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8876694641704326,
|
| 1938 |
+
"grad_norm": 5.268830672903035,
|
| 1939 |
+
"learning_rate": 7.590500519391309e-07,
|
| 1940 |
+
"loss": 1.4988,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8908973531310523,
|
| 1945 |
+
"grad_norm": 5.049110069527227,
|
| 1946 |
+
"learning_rate": 7.165745091568743e-07,
|
| 1947 |
+
"loss": 1.4573,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8941252420916721,
|
| 1952 |
+
"grad_norm": 5.353866067530813,
|
| 1953 |
+
"learning_rate": 6.752777059564431e-07,
|
| 1954 |
+
"loss": 1.4051,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8973531310522918,
|
| 1959 |
+
"grad_norm": 5.262463327229783,
|
| 1960 |
+
"learning_rate": 6.351648858954618e-07,
|
| 1961 |
+
"loss": 1.4635,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9005810200129115,
|
| 1966 |
+
"grad_norm": 5.279272053499587,
|
| 1967 |
+
"learning_rate": 5.962411421982805e-07,
|
| 1968 |
+
"loss": 1.501,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9038089089735313,
|
| 1973 |
+
"grad_norm": 6.166665329453942,
|
| 1974 |
+
"learning_rate": 5.585114171092665e-07,
|
| 1975 |
+
"loss": 1.4967,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.907036797934151,
|
| 1980 |
+
"grad_norm": 5.922366006208264,
|
| 1981 |
+
"learning_rate": 5.219805012652867e-07,
|
| 1982 |
+
"loss": 1.4908,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9102646868947708,
|
| 1987 |
+
"grad_norm": 5.31947537122732,
|
| 1988 |
+
"learning_rate": 4.866530330874153e-07,
|
| 1989 |
+
"loss": 1.4639,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9134925758553906,
|
| 1994 |
+
"grad_norm": 5.284266331508542,
|
| 1995 |
+
"learning_rate": 4.5253349819199375e-07,
|
| 1996 |
+
"loss": 1.3801,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9167204648160103,
|
| 2001 |
+
"grad_norm": 5.00246174936465,
|
| 2002 |
+
"learning_rate": 4.1962622882107174e-07,
|
| 2003 |
+
"loss": 1.489,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9199483537766301,
|
| 2008 |
+
"grad_norm": 5.196352989285751,
|
| 2009 |
+
"learning_rate": 3.8793540329233994e-07,
|
| 2010 |
+
"loss": 1.423,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9231762427372499,
|
| 2015 |
+
"grad_norm": 5.636396405762381,
|
| 2016 |
+
"learning_rate": 3.574650454685902e-07,
|
| 2017 |
+
"loss": 1.3957,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9264041316978696,
|
| 2022 |
+
"grad_norm": 5.7760912377631195,
|
| 2023 |
+
"learning_rate": 3.282190242468031e-07,
|
| 2024 |
+
"loss": 1.4413,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9296320206584894,
|
| 2029 |
+
"grad_norm": 5.284630555099834,
|
| 2030 |
+
"learning_rate": 3.0020105306689973e-07,
|
| 2031 |
+
"loss": 1.4106,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9328599096191091,
|
| 2036 |
+
"grad_norm": 5.215595343936084,
|
| 2037 |
+
"learning_rate": 2.7341468944023677e-07,
|
| 2038 |
+
"loss": 1.4156,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9360877985797289,
|
| 2043 |
+
"grad_norm": 5.792924133971526,
|
| 2044 |
+
"learning_rate": 2.4786333449790753e-07,
|
| 2045 |
+
"loss": 1.4402,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9393156875403487,
|
| 2050 |
+
"grad_norm": 5.1222262590925265,
|
| 2051 |
+
"learning_rate": 2.235502325588823e-07,
|
| 2052 |
+
"loss": 1.456,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9425435765009683,
|
| 2057 |
+
"grad_norm": 5.10978965516164,
|
| 2058 |
+
"learning_rate": 2.0047847071807402e-07,
|
| 2059 |
+
"loss": 1.4556,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9457714654615881,
|
| 2064 |
+
"grad_norm": 5.133920634072317,
|
| 2065 |
+
"learning_rate": 1.786509784543633e-07,
|
| 2066 |
+
"loss": 1.5004,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9489993544222078,
|
| 2071 |
+
"grad_norm": 4.93946597689753,
|
| 2072 |
+
"learning_rate": 1.5807052725863025e-07,
|
| 2073 |
+
"loss": 1.4039,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9522272433828276,
|
| 2078 |
+
"grad_norm": 5.19013537792188,
|
| 2079 |
+
"learning_rate": 1.3873973028185827e-07,
|
| 2080 |
+
"loss": 1.4215,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9554551323434474,
|
| 2085 |
+
"grad_norm": 5.111399221550104,
|
| 2086 |
+
"learning_rate": 1.206610420033305e-07,
|
| 2087 |
+
"loss": 1.3986,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9586830213040671,
|
| 2092 |
+
"grad_norm": 5.234642603752152,
|
| 2093 |
+
"learning_rate": 1.038367579189803e-07,
|
| 2094 |
+
"loss": 1.3943,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9619109102646869,
|
| 2099 |
+
"grad_norm": 5.645148935175098,
|
| 2100 |
+
"learning_rate": 8.82690142499254e-08,
|
| 2101 |
+
"loss": 1.4748,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9651387992253067,
|
| 2106 |
+
"grad_norm": 5.020357239881912,
|
| 2107 |
+
"learning_rate": 7.395978767122946e-08,
|
| 2108 |
+
"loss": 1.4324,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9683666881859264,
|
| 2113 |
+
"grad_norm": 5.6120171308753894,
|
| 2114 |
+
"learning_rate": 6.091089506091386e-08,
|
| 2115 |
+
"loss": 1.4205,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9715945771465462,
|
| 2120 |
+
"grad_norm": 5.898083266442407,
|
| 2121 |
+
"learning_rate": 4.9123993269271084e-08,
|
| 2122 |
+
"loss": 1.4123,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9748224661071659,
|
| 2127 |
+
"grad_norm": 5.408173578697908,
|
| 2128 |
+
"learning_rate": 3.860057890848201e-08,
|
| 2129 |
+
"loss": 1.4376,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9780503550677857,
|
| 2134 |
+
"grad_norm": 5.421313383210646,
|
| 2135 |
+
"learning_rate": 2.9341988162595593e-08,
|
| 2136 |
+
"loss": 1.4265,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9812782440284055,
|
| 2141 |
+
"grad_norm": 5.55675456499392,
|
| 2142 |
+
"learning_rate": 2.1349396617862395e-08,
|
| 2143 |
+
"loss": 1.4463,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9845061329890252,
|
| 2148 |
+
"grad_norm": 5.211225978985049,
|
| 2149 |
+
"learning_rate": 1.4623819113475102e-08,
|
| 2150 |
+
"loss": 1.4244,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9877340219496449,
|
| 2155 |
+
"grad_norm": 5.824025695493701,
|
| 2156 |
+
"learning_rate": 9.166109612706031e-09,
|
| 2157 |
+
"loss": 1.4411,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9909619109102646,
|
| 2162 |
+
"grad_norm": 4.826680454755765,
|
| 2163 |
+
"learning_rate": 4.976961094479427e-09,
|
| 2164 |
+
"loss": 1.4475,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9941897998708844,
|
| 2169 |
+
"grad_norm": 5.614005751299077,
|
| 2170 |
+
"learning_rate": 2.0569054653840625e-09,
|
| 2171 |
+
"loss": 1.4278,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9974176888315042,
|
| 2176 |
+
"grad_norm": 5.395704789347489,
|
| 2177 |
+
"learning_rate": 4.0631349213060555e-10,
|
| 2178 |
+
"loss": 1.3988,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 1.0,
|
| 2183 |
+
"eval_loss": 1.406965732574463,
|
| 2184 |
+
"eval_runtime": 263.0675,
|
| 2185 |
+
"eval_samples_per_second": 31.703,
|
| 2186 |
+
"eval_steps_per_second": 0.992,
|
| 2187 |
+
"step": 1549
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.0,
|
| 2191 |
+
"step": 1549,
|
| 2192 |
+
"total_flos": 19248551362560.0,
|
| 2193 |
+
"train_loss": 1.5783487337492912,
|
| 2194 |
+
"train_runtime": 6620.0514,
|
| 2195 |
+
"train_samples_per_second": 7.485,
|
| 2196 |
+
"train_steps_per_second": 0.234
|
| 2197 |
+
}
|
| 2198 |
+
],
|
| 2199 |
+
"logging_steps": 5,
|
| 2200 |
+
"max_steps": 1549,
|
| 2201 |
+
"num_input_tokens_seen": 0,
|
| 2202 |
+
"num_train_epochs": 1,
|
| 2203 |
+
"save_steps": 500,
|
| 2204 |
+
"stateful_callbacks": {
|
| 2205 |
+
"TrainerControl": {
|
| 2206 |
+
"args": {
|
| 2207 |
+
"should_epoch_stop": false,
|
| 2208 |
+
"should_evaluate": false,
|
| 2209 |
+
"should_log": false,
|
| 2210 |
+
"should_save": true,
|
| 2211 |
+
"should_training_stop": true
|
| 2212 |
+
},
|
| 2213 |
+
"attributes": {}
|
| 2214 |
+
}
|
| 2215 |
+
},
|
| 2216 |
+
"total_flos": 19248551362560.0,
|
| 2217 |
+
"train_batch_size": 8,
|
| 2218 |
+
"trial_name": null,
|
| 2219 |
+
"trial_params": null
|
| 2220 |
+
}
|