Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: AmberYifan/Qwen2-7B-sft-ultrachat
|
| 3 |
+
library_name: transformers
|
| 4 |
+
model_name: Qwen2-7B-sft-ultrachat-hhrlhf
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
- trl
|
| 8 |
+
- sft
|
| 9 |
+
licence: license
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Card for Qwen2-7B-sft-ultrachat-hhrlhf
|
| 13 |
+
|
| 14 |
+
This model is a fine-tuned version of [AmberYifan/Qwen2-7B-sft-ultrachat](https://huggingface.co/AmberYifan/Qwen2-7B-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/Qwen2-7B-sft-ultrachat-hhrlhf", device="cuda")
|
| 24 |
+
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
| 25 |
+
print(output["generated_text"])
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## Training procedure
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
This model was trained with SFT.
|
| 33 |
+
|
| 34 |
+
### Framework versions
|
| 35 |
+
|
| 36 |
+
- TRL: 0.12.2
|
| 37 |
+
- Transformers: 4.46.3
|
| 38 |
+
- Pytorch: 2.5.1+cu118
|
| 39 |
+
- Datasets: 3.2.0
|
| 40 |
+
- Tokenizers: 0.20.3
|
| 41 |
+
|
| 42 |
+
## Citations
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
Cite TRL as:
|
| 47 |
+
|
| 48 |
+
```bibtex
|
| 49 |
+
@misc{vonwerra2022trl,
|
| 50 |
+
title = {{TRL: Transformer Reinforcement Learning}},
|
| 51 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
|
| 52 |
+
year = 2020,
|
| 53 |
+
journal = {GitHub repository},
|
| 54 |
+
publisher = {GitHub},
|
| 55 |
+
howpublished = {\url{https://github.com/huggingface/trl}}
|
| 56 |
+
}
|
| 57 |
+
```
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 45448595030016.0,
|
| 4 |
+
"train_loss": 1.473227332512385,
|
| 5 |
+
"train_runtime": 5771.4551,
|
| 6 |
+
"train_samples": 50000,
|
| 7 |
+
"train_samples_per_second": 8.663,
|
| 8 |
+
"train_steps_per_second": 0.271
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.46.3"
|
| 6 |
+
}
|
runs/Jan26_03-05-37_gilbreth-j001.rcac.purdue.edu/events.out.tfevents.1737878926.gilbreth-j001.rcac.purdue.edu.200811.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b842f3d5436ac70c27c2e41b85b10ffbbfd884c686802abb2395513cdaa1e338
|
| 3 |
+
size 72485
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 45448595030016.0,
|
| 4 |
+
"train_loss": 1.473227332512385,
|
| 5 |
+
"train_runtime": 5771.4551,
|
| 6 |
+
"train_samples": 50000,
|
| 7 |
+
"train_samples_per_second": 8.663,
|
| 8 |
+
"train_steps_per_second": 0.271
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2241 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1563,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006397952655150352,
|
| 13 |
+
"grad_norm": 8.45120858638771,
|
| 14 |
+
"learning_rate": 1.2738853503184715e-07,
|
| 15 |
+
"loss": 2.1311,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003198976327575176,
|
| 20 |
+
"grad_norm": 8.544993499312284,
|
| 21 |
+
"learning_rate": 6.369426751592357e-07,
|
| 22 |
+
"loss": 2.1114,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006397952655150352,
|
| 27 |
+
"grad_norm": 8.772571070364279,
|
| 28 |
+
"learning_rate": 1.2738853503184715e-06,
|
| 29 |
+
"loss": 2.1017,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009596928982725527,
|
| 34 |
+
"grad_norm": 4.897458678497011,
|
| 35 |
+
"learning_rate": 1.9108280254777074e-06,
|
| 36 |
+
"loss": 1.961,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012795905310300703,
|
| 41 |
+
"grad_norm": 4.075186495410978,
|
| 42 |
+
"learning_rate": 2.547770700636943e-06,
|
| 43 |
+
"loss": 1.8269,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01599488163787588,
|
| 48 |
+
"grad_norm": 3.503726866717514,
|
| 49 |
+
"learning_rate": 3.1847133757961785e-06,
|
| 50 |
+
"loss": 1.7622,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.019193857965451054,
|
| 55 |
+
"grad_norm": 3.027293734155194,
|
| 56 |
+
"learning_rate": 3.821656050955415e-06,
|
| 57 |
+
"loss": 1.6355,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.022392834293026232,
|
| 62 |
+
"grad_norm": 2.8891766987266796,
|
| 63 |
+
"learning_rate": 4.45859872611465e-06,
|
| 64 |
+
"loss": 1.5967,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.025591810620601407,
|
| 69 |
+
"grad_norm": 2.9905733143061117,
|
| 70 |
+
"learning_rate": 5.095541401273886e-06,
|
| 71 |
+
"loss": 1.6482,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.028790786948176585,
|
| 76 |
+
"grad_norm": 3.0173836321259175,
|
| 77 |
+
"learning_rate": 5.732484076433121e-06,
|
| 78 |
+
"loss": 1.6306,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03198976327575176,
|
| 83 |
+
"grad_norm": 2.9409129691288483,
|
| 84 |
+
"learning_rate": 6.369426751592357e-06,
|
| 85 |
+
"loss": 1.6179,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.035188739603326934,
|
| 90 |
+
"grad_norm": 2.8249107707715346,
|
| 91 |
+
"learning_rate": 7.006369426751593e-06,
|
| 92 |
+
"loss": 1.5905,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03838771593090211,
|
| 97 |
+
"grad_norm": 3.2414392670664114,
|
| 98 |
+
"learning_rate": 7.64331210191083e-06,
|
| 99 |
+
"loss": 1.618,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04158669225847729,
|
| 104 |
+
"grad_norm": 3.0059278893643753,
|
| 105 |
+
"learning_rate": 8.280254777070064e-06,
|
| 106 |
+
"loss": 1.6345,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.044785668586052464,
|
| 111 |
+
"grad_norm": 3.000977330200516,
|
| 112 |
+
"learning_rate": 8.9171974522293e-06,
|
| 113 |
+
"loss": 1.6036,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04798464491362764,
|
| 118 |
+
"grad_norm": 2.7742457498827395,
|
| 119 |
+
"learning_rate": 9.554140127388536e-06,
|
| 120 |
+
"loss": 1.6228,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05118362124120281,
|
| 125 |
+
"grad_norm": 2.9247171067005846,
|
| 126 |
+
"learning_rate": 1.0191082802547772e-05,
|
| 127 |
+
"loss": 1.5814,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05438259756877799,
|
| 132 |
+
"grad_norm": 2.707250155470774,
|
| 133 |
+
"learning_rate": 1.0828025477707008e-05,
|
| 134 |
+
"loss": 1.5989,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05758157389635317,
|
| 139 |
+
"grad_norm": 2.777451110804097,
|
| 140 |
+
"learning_rate": 1.1464968152866242e-05,
|
| 141 |
+
"loss": 1.6046,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.060780550223928344,
|
| 146 |
+
"grad_norm": 2.750734662498446,
|
| 147 |
+
"learning_rate": 1.2101910828025478e-05,
|
| 148 |
+
"loss": 1.5866,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06397952655150352,
|
| 153 |
+
"grad_norm": 3.231815816845173,
|
| 154 |
+
"learning_rate": 1.2738853503184714e-05,
|
| 155 |
+
"loss": 1.5973,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0671785028790787,
|
| 160 |
+
"grad_norm": 2.8990703853478856,
|
| 161 |
+
"learning_rate": 1.337579617834395e-05,
|
| 162 |
+
"loss": 1.5865,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.07037747920665387,
|
| 167 |
+
"grad_norm": 2.5849093940529175,
|
| 168 |
+
"learning_rate": 1.4012738853503186e-05,
|
| 169 |
+
"loss": 1.6027,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07357645553422905,
|
| 174 |
+
"grad_norm": 2.6093741257086256,
|
| 175 |
+
"learning_rate": 1.464968152866242e-05,
|
| 176 |
+
"loss": 1.5806,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07677543186180422,
|
| 181 |
+
"grad_norm": 2.6475786335268516,
|
| 182 |
+
"learning_rate": 1.528662420382166e-05,
|
| 183 |
+
"loss": 1.6015,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0799744081893794,
|
| 188 |
+
"grad_norm": 2.5612093515246905,
|
| 189 |
+
"learning_rate": 1.5923566878980894e-05,
|
| 190 |
+
"loss": 1.6285,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08317338451695458,
|
| 195 |
+
"grad_norm": 2.635291993226663,
|
| 196 |
+
"learning_rate": 1.6560509554140128e-05,
|
| 197 |
+
"loss": 1.6063,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08637236084452975,
|
| 202 |
+
"grad_norm": 2.3771393339595597,
|
| 203 |
+
"learning_rate": 1.7197452229299365e-05,
|
| 204 |
+
"loss": 1.6329,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08957133717210493,
|
| 209 |
+
"grad_norm": 2.935313002912278,
|
| 210 |
+
"learning_rate": 1.78343949044586e-05,
|
| 211 |
+
"loss": 1.6015,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.0927703134996801,
|
| 216 |
+
"grad_norm": 2.5747148263761708,
|
| 217 |
+
"learning_rate": 1.8471337579617837e-05,
|
| 218 |
+
"loss": 1.5831,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09596928982725528,
|
| 223 |
+
"grad_norm": 2.8025363824063576,
|
| 224 |
+
"learning_rate": 1.910828025477707e-05,
|
| 225 |
+
"loss": 1.5832,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09916826615483046,
|
| 230 |
+
"grad_norm": 2.5997598240468474,
|
| 231 |
+
"learning_rate": 1.9745222929936306e-05,
|
| 232 |
+
"loss": 1.6228,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10236724248240563,
|
| 237 |
+
"grad_norm": 2.6087028065146067,
|
| 238 |
+
"learning_rate": 1.9999775332635076e-05,
|
| 239 |
+
"loss": 1.5726,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10556621880998081,
|
| 244 |
+
"grad_norm": 2.6832519430043043,
|
| 245 |
+
"learning_rate": 1.999840240196313e-05,
|
| 246 |
+
"loss": 1.5822,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10876519513755598,
|
| 251 |
+
"grad_norm": 2.5312510284895575,
|
| 252 |
+
"learning_rate": 1.9995781526975738e-05,
|
| 253 |
+
"loss": 1.5902,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11196417146513116,
|
| 258 |
+
"grad_norm": 2.552059625982884,
|
| 259 |
+
"learning_rate": 1.9991913034795768e-05,
|
| 260 |
+
"loss": 1.6468,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11516314779270634,
|
| 265 |
+
"grad_norm": 2.737621660353208,
|
| 266 |
+
"learning_rate": 1.9986797408266636e-05,
|
| 267 |
+
"loss": 1.5825,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1183621241202815,
|
| 272 |
+
"grad_norm": 2.3504295992454693,
|
| 273 |
+
"learning_rate": 1.9980435285892056e-05,
|
| 274 |
+
"loss": 1.6176,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12156110044785669,
|
| 279 |
+
"grad_norm": 2.605194872517274,
|
| 280 |
+
"learning_rate": 1.9972827461756335e-05,
|
| 281 |
+
"loss": 1.6186,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12476007677543186,
|
| 286 |
+
"grad_norm": 2.545693165994165,
|
| 287 |
+
"learning_rate": 1.9963974885425267e-05,
|
| 288 |
+
"loss": 1.6284,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12795905310300704,
|
| 293 |
+
"grad_norm": 2.4582821508493535,
|
| 294 |
+
"learning_rate": 1.9953878661827603e-05,
|
| 295 |
+
"loss": 1.5967,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.13115802943058222,
|
| 300 |
+
"grad_norm": 2.294144210629627,
|
| 301 |
+
"learning_rate": 1.994254005111715e-05,
|
| 302 |
+
"loss": 1.5754,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1343570057581574,
|
| 307 |
+
"grad_norm": 2.455611945909697,
|
| 308 |
+
"learning_rate": 1.992996046851548e-05,
|
| 309 |
+
"loss": 1.6293,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.13755598208573255,
|
| 314 |
+
"grad_norm": 2.5851678984397037,
|
| 315 |
+
"learning_rate": 1.9916141484135297e-05,
|
| 316 |
+
"loss": 1.5629,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.14075495841330773,
|
| 321 |
+
"grad_norm": 2.42361611389672,
|
| 322 |
+
"learning_rate": 1.990108482278446e-05,
|
| 323 |
+
"loss": 1.6161,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.14395393474088292,
|
| 328 |
+
"grad_norm": 2.3505427050524585,
|
| 329 |
+
"learning_rate": 1.9884792363750684e-05,
|
| 330 |
+
"loss": 1.6044,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1471529110684581,
|
| 335 |
+
"grad_norm": 2.388109504758734,
|
| 336 |
+
"learning_rate": 1.9867266140567024e-05,
|
| 337 |
+
"loss": 1.6312,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.15035188739603328,
|
| 342 |
+
"grad_norm": 2.4772439223936313,
|
| 343 |
+
"learning_rate": 1.9848508340758014e-05,
|
| 344 |
+
"loss": 1.6011,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.15355086372360843,
|
| 349 |
+
"grad_norm": 2.684448860110458,
|
| 350 |
+
"learning_rate": 1.9828521305566647e-05,
|
| 351 |
+
"loss": 1.6624,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15674984005118361,
|
| 356 |
+
"grad_norm": 2.373116924740462,
|
| 357 |
+
"learning_rate": 1.9807307529662175e-05,
|
| 358 |
+
"loss": 1.6455,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1599488163787588,
|
| 363 |
+
"grad_norm": 2.3871214467638127,
|
| 364 |
+
"learning_rate": 1.9784869660828708e-05,
|
| 365 |
+
"loss": 1.6597,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.16314779270633398,
|
| 370 |
+
"grad_norm": 2.553594848899839,
|
| 371 |
+
"learning_rate": 1.9761210499634754e-05,
|
| 372 |
+
"loss": 1.5698,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.16634676903390916,
|
| 377 |
+
"grad_norm": 2.5770365342618318,
|
| 378 |
+
"learning_rate": 1.973633299908365e-05,
|
| 379 |
+
"loss": 1.6073,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1695457453614843,
|
| 384 |
+
"grad_norm": 2.5069237836308176,
|
| 385 |
+
"learning_rate": 1.9710240264245005e-05,
|
| 386 |
+
"loss": 1.5877,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1727447216890595,
|
| 391 |
+
"grad_norm": 2.4165608550489197,
|
| 392 |
+
"learning_rate": 1.9682935551867132e-05,
|
| 393 |
+
"loss": 1.6266,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.17594369801663468,
|
| 398 |
+
"grad_norm": 2.5380847108727744,
|
| 399 |
+
"learning_rate": 1.9654422269970545e-05,
|
| 400 |
+
"loss": 1.5476,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.17914267434420986,
|
| 405 |
+
"grad_norm": 2.385470144460665,
|
| 406 |
+
"learning_rate": 1.9624703977422624e-05,
|
| 407 |
+
"loss": 1.5896,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.18234165067178504,
|
| 412 |
+
"grad_norm": 2.457845761779252,
|
| 413 |
+
"learning_rate": 1.959378438349338e-05,
|
| 414 |
+
"loss": 1.5958,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1855406269993602,
|
| 419 |
+
"grad_norm": 2.4062711922420092,
|
| 420 |
+
"learning_rate": 1.956166734739251e-05,
|
| 421 |
+
"loss": 1.6074,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.18873960332693537,
|
| 426 |
+
"grad_norm": 2.281520144278105,
|
| 427 |
+
"learning_rate": 1.95283568777877e-05,
|
| 428 |
+
"loss": 1.6134,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.19193857965451055,
|
| 433 |
+
"grad_norm": 2.5102158955477956,
|
| 434 |
+
"learning_rate": 1.9493857132304295e-05,
|
| 435 |
+
"loss": 1.5395,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.19513755598208574,
|
| 440 |
+
"grad_norm": 2.415759547538384,
|
| 441 |
+
"learning_rate": 1.9458172417006347e-05,
|
| 442 |
+
"loss": 1.5943,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.19833653230966092,
|
| 447 |
+
"grad_norm": 2.3874207818256084,
|
| 448 |
+
"learning_rate": 1.942130718585919e-05,
|
| 449 |
+
"loss": 1.5539,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.20153550863723607,
|
| 454 |
+
"grad_norm": 2.3291818511607625,
|
| 455 |
+
"learning_rate": 1.938326604017349e-05,
|
| 456 |
+
"loss": 1.5951,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.20473448496481125,
|
| 461 |
+
"grad_norm": 2.3503279349130612,
|
| 462 |
+
"learning_rate": 1.9344053728030952e-05,
|
| 463 |
+
"loss": 1.5543,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.20793346129238643,
|
| 468 |
+
"grad_norm": 2.367606596377197,
|
| 469 |
+
"learning_rate": 1.9303675143691683e-05,
|
| 470 |
+
"loss": 1.5767,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.21113243761996162,
|
| 475 |
+
"grad_norm": 2.3105097918946624,
|
| 476 |
+
"learning_rate": 1.9262135326983326e-05,
|
| 477 |
+
"loss": 1.5641,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2143314139475368,
|
| 482 |
+
"grad_norm": 2.3186275383828057,
|
| 483 |
+
"learning_rate": 1.921943946267201e-05,
|
| 484 |
+
"loss": 1.6119,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.21753039027511195,
|
| 489 |
+
"grad_norm": 2.3623182317692635,
|
| 490 |
+
"learning_rate": 1.9175592879815217e-05,
|
| 491 |
+
"loss": 1.5995,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.22072936660268713,
|
| 496 |
+
"grad_norm": 2.3234032719145383,
|
| 497 |
+
"learning_rate": 1.9130601051096655e-05,
|
| 498 |
+
"loss": 1.5399,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.22392834293026231,
|
| 503 |
+
"grad_norm": 2.2619822250922534,
|
| 504 |
+
"learning_rate": 1.9084469592143154e-05,
|
| 505 |
+
"loss": 1.5924,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2271273192578375,
|
| 510 |
+
"grad_norm": 2.614900556151945,
|
| 511 |
+
"learning_rate": 1.9037204260823788e-05,
|
| 512 |
+
"loss": 1.5542,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.23032629558541268,
|
| 517 |
+
"grad_norm": 2.4810633670440088,
|
| 518 |
+
"learning_rate": 1.89888109565312e-05,
|
| 519 |
+
"loss": 1.5974,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.23352527191298783,
|
| 524 |
+
"grad_norm": 2.2469047163095337,
|
| 525 |
+
"learning_rate": 1.893929571944527e-05,
|
| 526 |
+
"loss": 1.5219,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.236724248240563,
|
| 531 |
+
"grad_norm": 2.4290730257347395,
|
| 532 |
+
"learning_rate": 1.8888664729779205e-05,
|
| 533 |
+
"loss": 1.596,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2399232245681382,
|
| 538 |
+
"grad_norm": 2.486005754480233,
|
| 539 |
+
"learning_rate": 1.883692430700818e-05,
|
| 540 |
+
"loss": 1.5715,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.24312220089571338,
|
| 545 |
+
"grad_norm": 2.3308186805659727,
|
| 546 |
+
"learning_rate": 1.8784080909080568e-05,
|
| 547 |
+
"loss": 1.49,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.24632117722328856,
|
| 552 |
+
"grad_norm": 2.2000178419437275,
|
| 553 |
+
"learning_rate": 1.8730141131611882e-05,
|
| 554 |
+
"loss": 1.5831,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2495201535508637,
|
| 559 |
+
"grad_norm": 2.2521923554518763,
|
| 560 |
+
"learning_rate": 1.867511170706157e-05,
|
| 561 |
+
"loss": 1.523,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2527191298784389,
|
| 566 |
+
"grad_norm": 2.4262324307986596,
|
| 567 |
+
"learning_rate": 1.861899950389269e-05,
|
| 568 |
+
"loss": 1.5276,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2559181062060141,
|
| 573 |
+
"grad_norm": 2.247962152190113,
|
| 574 |
+
"learning_rate": 1.856181152571463e-05,
|
| 575 |
+
"loss": 1.6288,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2591170825335892,
|
| 580 |
+
"grad_norm": 2.3784897154509914,
|
| 581 |
+
"learning_rate": 1.850355491040897e-05,
|
| 582 |
+
"loss": 1.6227,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.26231605886116444,
|
| 587 |
+
"grad_norm": 2.4615020230179723,
|
| 588 |
+
"learning_rate": 1.8444236929238556e-05,
|
| 589 |
+
"loss": 1.6161,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2655150351887396,
|
| 594 |
+
"grad_norm": 2.553794750075187,
|
| 595 |
+
"learning_rate": 1.8383864985939944e-05,
|
| 596 |
+
"loss": 1.5419,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2687140115163148,
|
| 601 |
+
"grad_norm": 2.2750920584095073,
|
| 602 |
+
"learning_rate": 1.8322446615799317e-05,
|
| 603 |
+
"loss": 1.6044,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.27191298784388995,
|
| 608 |
+
"grad_norm": 2.309582764312615,
|
| 609 |
+
"learning_rate": 1.825998948471197e-05,
|
| 610 |
+
"loss": 1.5686,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2751119641714651,
|
| 615 |
+
"grad_norm": 2.563992220208821,
|
| 616 |
+
"learning_rate": 1.819650138822548e-05,
|
| 617 |
+
"loss": 1.5716,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2783109404990403,
|
| 622 |
+
"grad_norm": 2.3463821157538463,
|
| 623 |
+
"learning_rate": 1.8131990250566733e-05,
|
| 624 |
+
"loss": 1.488,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.28150991682661547,
|
| 629 |
+
"grad_norm": 2.4105968132480564,
|
| 630 |
+
"learning_rate": 1.8066464123652857e-05,
|
| 631 |
+
"loss": 1.4974,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2847088931541907,
|
| 636 |
+
"grad_norm": 2.3911862739855487,
|
| 637 |
+
"learning_rate": 1.7999931186086225e-05,
|
| 638 |
+
"loss": 1.5246,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.28790786948176583,
|
| 643 |
+
"grad_norm": 2.521273873337624,
|
| 644 |
+
"learning_rate": 1.793239974213364e-05,
|
| 645 |
+
"loss": 1.5443,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.291106845809341,
|
| 650 |
+
"grad_norm": 2.1371439492853987,
|
| 651 |
+
"learning_rate": 1.786387822068987e-05,
|
| 652 |
+
"loss": 1.5025,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2943058221369162,
|
| 657 |
+
"grad_norm": 2.3936981590007766,
|
| 658 |
+
"learning_rate": 1.7794375174225565e-05,
|
| 659 |
+
"loss": 1.6025,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.29750479846449135,
|
| 664 |
+
"grad_norm": 2.353176497945942,
|
| 665 |
+
"learning_rate": 1.7723899277719815e-05,
|
| 666 |
+
"loss": 1.585,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.30070377479206656,
|
| 671 |
+
"grad_norm": 2.430530486800993,
|
| 672 |
+
"learning_rate": 1.7652459327577377e-05,
|
| 673 |
+
"loss": 1.5918,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3039027511196417,
|
| 678 |
+
"grad_norm": 2.173964316210541,
|
| 679 |
+
"learning_rate": 1.7580064240530746e-05,
|
| 680 |
+
"loss": 1.5182,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.30710172744721687,
|
| 685 |
+
"grad_norm": 2.5772299950374418,
|
| 686 |
+
"learning_rate": 1.7506723052527243e-05,
|
| 687 |
+
"loss": 1.5798,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3103007037747921,
|
| 692 |
+
"grad_norm": 2.4991529420639234,
|
| 693 |
+
"learning_rate": 1.7432444917601183e-05,
|
| 694 |
+
"loss": 1.5206,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.31349968010236723,
|
| 699 |
+
"grad_norm": 2.3323983247114857,
|
| 700 |
+
"learning_rate": 1.735723910673132e-05,
|
| 701 |
+
"loss": 1.5071,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.31669865642994244,
|
| 706 |
+
"grad_norm": 2.3103580608228227,
|
| 707 |
+
"learning_rate": 1.7281115006683687e-05,
|
| 708 |
+
"loss": 1.5202,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3198976327575176,
|
| 713 |
+
"grad_norm": 2.4434232834822955,
|
| 714 |
+
"learning_rate": 1.7204082118840035e-05,
|
| 715 |
+
"loss": 1.5238,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.32309660908509275,
|
| 720 |
+
"grad_norm": 2.2314302298797566,
|
| 721 |
+
"learning_rate": 1.712615005801185e-05,
|
| 722 |
+
"loss": 1.5228,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.32629558541266795,
|
| 727 |
+
"grad_norm": 2.2505362844790686,
|
| 728 |
+
"learning_rate": 1.704732855124037e-05,
|
| 729 |
+
"loss": 1.5464,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3294945617402431,
|
| 734 |
+
"grad_norm": 2.335383707358806,
|
| 735 |
+
"learning_rate": 1.6967627436582445e-05,
|
| 736 |
+
"loss": 1.4524,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3326935380678183,
|
| 741 |
+
"grad_norm": 2.243258802450963,
|
| 742 |
+
"learning_rate": 1.6887056661882644e-05,
|
| 743 |
+
"loss": 1.5224,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.33589251439539347,
|
| 748 |
+
"grad_norm": 2.39793844391146,
|
| 749 |
+
"learning_rate": 1.6805626283531592e-05,
|
| 750 |
+
"loss": 1.5283,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3390914907229686,
|
| 755 |
+
"grad_norm": 2.5442566113713445,
|
| 756 |
+
"learning_rate": 1.6723346465210815e-05,
|
| 757 |
+
"loss": 1.5731,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.34229046705054383,
|
| 762 |
+
"grad_norm": 2.2816938873410115,
|
| 763 |
+
"learning_rate": 1.6640227476624163e-05,
|
| 764 |
+
"loss": 1.5324,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.345489443378119,
|
| 769 |
+
"grad_norm": 2.34576166973507,
|
| 770 |
+
"learning_rate": 1.655627969221598e-05,
|
| 771 |
+
"loss": 1.5316,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3486884197056942,
|
| 776 |
+
"grad_norm": 2.4410014523875287,
|
| 777 |
+
"learning_rate": 1.6471513589876247e-05,
|
| 778 |
+
"loss": 1.5316,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.35188739603326935,
|
| 783 |
+
"grad_norm": 2.2216487614131126,
|
| 784 |
+
"learning_rate": 1.638593974963278e-05,
|
| 785 |
+
"loss": 1.466,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3550863723608445,
|
| 790 |
+
"grad_norm": 2.4275497144081086,
|
| 791 |
+
"learning_rate": 1.6299568852330703e-05,
|
| 792 |
+
"loss": 1.5016,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3582853486884197,
|
| 797 |
+
"grad_norm": 2.236501922766204,
|
| 798 |
+
"learning_rate": 1.6212411678299306e-05,
|
| 799 |
+
"loss": 1.5561,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.36148432501599487,
|
| 804 |
+
"grad_norm": 2.27631043483282,
|
| 805 |
+
"learning_rate": 1.612447910600652e-05,
|
| 806 |
+
"loss": 1.5141,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3646833013435701,
|
| 811 |
+
"grad_norm": 2.331933284658563,
|
| 812 |
+
"learning_rate": 1.6035782110701125e-05,
|
| 813 |
+
"loss": 1.5075,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.36788227767114523,
|
| 818 |
+
"grad_norm": 2.248455320446842,
|
| 819 |
+
"learning_rate": 1.594633176304287e-05,
|
| 820 |
+
"loss": 1.5531,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3710812539987204,
|
| 825 |
+
"grad_norm": 2.3056857138761173,
|
| 826 |
+
"learning_rate": 1.5856139227720714e-05,
|
| 827 |
+
"loss": 1.5867,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3742802303262956,
|
| 832 |
+
"grad_norm": 2.2615976412633962,
|
| 833 |
+
"learning_rate": 1.5765215762059304e-05,
|
| 834 |
+
"loss": 1.5137,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.37747920665387075,
|
| 839 |
+
"grad_norm": 2.2246230951068404,
|
| 840 |
+
"learning_rate": 1.5673572714613886e-05,
|
| 841 |
+
"loss": 1.4676,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.38067818298144596,
|
| 846 |
+
"grad_norm": 2.3714970265633224,
|
| 847 |
+
"learning_rate": 1.558122152375387e-05,
|
| 848 |
+
"loss": 1.5649,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3838771593090211,
|
| 853 |
+
"grad_norm": 2.4424554877615505,
|
| 854 |
+
"learning_rate": 1.548817371623513e-05,
|
| 855 |
+
"loss": 1.5203,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.38707613563659626,
|
| 860 |
+
"grad_norm": 2.3241867748802862,
|
| 861 |
+
"learning_rate": 1.539444090576132e-05,
|
| 862 |
+
"loss": 1.532,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3902751119641715,
|
| 867 |
+
"grad_norm": 2.2841806700533454,
|
| 868 |
+
"learning_rate": 1.5300034791534297e-05,
|
| 869 |
+
"loss": 1.5341,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3934740882917466,
|
| 874 |
+
"grad_norm": 2.3823095399745,
|
| 875 |
+
"learning_rate": 1.520496715679391e-05,
|
| 876 |
+
"loss": 1.5606,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.39667306461932184,
|
| 881 |
+
"grad_norm": 2.404156549397838,
|
| 882 |
+
"learning_rate": 1.5109249867347276e-05,
|
| 883 |
+
"loss": 1.4964,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.399872040946897,
|
| 888 |
+
"grad_norm": 2.3045879883870226,
|
| 889 |
+
"learning_rate": 1.5012894870087751e-05,
|
| 890 |
+
"loss": 1.4785,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.40307101727447214,
|
| 895 |
+
"grad_norm": 2.2564282819228376,
|
| 896 |
+
"learning_rate": 1.4915914191503792e-05,
|
| 897 |
+
"loss": 1.5593,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.40626999360204735,
|
| 902 |
+
"grad_norm": 2.1899628345006508,
|
| 903 |
+
"learning_rate": 1.4818319936177885e-05,
|
| 904 |
+
"loss": 1.5385,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4094689699296225,
|
| 909 |
+
"grad_norm": 2.6345048580064003,
|
| 910 |
+
"learning_rate": 1.4720124285275703e-05,
|
| 911 |
+
"loss": 1.5198,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4126679462571977,
|
| 916 |
+
"grad_norm": 2.4733399052284613,
|
| 917 |
+
"learning_rate": 1.4621339495025731e-05,
|
| 918 |
+
"loss": 1.5123,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.41586692258477287,
|
| 923 |
+
"grad_norm": 2.2853139912229663,
|
| 924 |
+
"learning_rate": 1.4521977895189518e-05,
|
| 925 |
+
"loss": 1.586,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.419065898912348,
|
| 930 |
+
"grad_norm": 2.348285993958346,
|
| 931 |
+
"learning_rate": 1.4422051887522735e-05,
|
| 932 |
+
"loss": 1.4982,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.42226487523992323,
|
| 937 |
+
"grad_norm": 2.48181732428636,
|
| 938 |
+
"learning_rate": 1.4321573944227254e-05,
|
| 939 |
+
"loss": 1.4932,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4254638515674984,
|
| 944 |
+
"grad_norm": 2.2700624107334146,
|
| 945 |
+
"learning_rate": 1.4220556606394465e-05,
|
| 946 |
+
"loss": 1.477,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4286628278950736,
|
| 951 |
+
"grad_norm": 2.3121958973335763,
|
| 952 |
+
"learning_rate": 1.4119012482439929e-05,
|
| 953 |
+
"loss": 1.5362,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.43186180422264875,
|
| 958 |
+
"grad_norm": 2.9408602035120657,
|
| 959 |
+
"learning_rate": 1.4016954246529697e-05,
|
| 960 |
+
"loss": 1.5094,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4350607805502239,
|
| 965 |
+
"grad_norm": 2.4183707899529074,
|
| 966 |
+
"learning_rate": 1.3914394636998374e-05,
|
| 967 |
+
"loss": 1.5137,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4382597568777991,
|
| 972 |
+
"grad_norm": 2.3165836903959365,
|
| 973 |
+
"learning_rate": 1.3811346454759211e-05,
|
| 974 |
+
"loss": 1.4942,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.44145873320537427,
|
| 979 |
+
"grad_norm": 2.280419639027186,
|
| 980 |
+
"learning_rate": 1.3707822561706336e-05,
|
| 981 |
+
"loss": 1.5327,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4446577095329495,
|
| 986 |
+
"grad_norm": 2.152749313872907,
|
| 987 |
+
"learning_rate": 1.3603835879109442e-05,
|
| 988 |
+
"loss": 1.4839,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.44785668586052463,
|
| 993 |
+
"grad_norm": 2.3172798106180155,
|
| 994 |
+
"learning_rate": 1.349939938600099e-05,
|
| 995 |
+
"loss": 1.4732,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4510556621880998,
|
| 1000 |
+
"grad_norm": 2.328153860662641,
|
| 1001 |
+
"learning_rate": 1.3394526117556277e-05,
|
| 1002 |
+
"loss": 1.4322,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.454254638515675,
|
| 1007 |
+
"grad_norm": 2.3538091734432243,
|
| 1008 |
+
"learning_rate": 1.3289229163466421e-05,
|
| 1009 |
+
"loss": 1.4408,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.45745361484325014,
|
| 1014 |
+
"grad_norm": 2.4036873663190295,
|
| 1015 |
+
"learning_rate": 1.3183521666304611e-05,
|
| 1016 |
+
"loss": 1.4778,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.46065259117082535,
|
| 1021 |
+
"grad_norm": 2.3520908941332785,
|
| 1022 |
+
"learning_rate": 1.3077416819885707e-05,
|
| 1023 |
+
"loss": 1.4713,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4638515674984005,
|
| 1028 |
+
"grad_norm": 2.3375152975193783,
|
| 1029 |
+
"learning_rate": 1.297092786761946e-05,
|
| 1030 |
+
"loss": 1.4567,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.46705054382597566,
|
| 1035 |
+
"grad_norm": 2.2072268508544655,
|
| 1036 |
+
"learning_rate": 1.2864068100857565e-05,
|
| 1037 |
+
"loss": 1.5374,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.47024952015355087,
|
| 1042 |
+
"grad_norm": 2.283182102299493,
|
| 1043 |
+
"learning_rate": 1.2756850857234686e-05,
|
| 1044 |
+
"loss": 1.4176,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.473448496481126,
|
| 1049 |
+
"grad_norm": 2.1958718266912096,
|
| 1050 |
+
"learning_rate": 1.2649289519003739e-05,
|
| 1051 |
+
"loss": 1.4892,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.47664747280870123,
|
| 1056 |
+
"grad_norm": 2.2999635338111064,
|
| 1057 |
+
"learning_rate": 1.2541397511365584e-05,
|
| 1058 |
+
"loss": 1.4835,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4798464491362764,
|
| 1063 |
+
"grad_norm": 2.3494145089078597,
|
| 1064 |
+
"learning_rate": 1.2433188300793399e-05,
|
| 1065 |
+
"loss": 1.4686,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.48304542546385154,
|
| 1070 |
+
"grad_norm": 2.355567026544742,
|
| 1071 |
+
"learning_rate": 1.2324675393351818e-05,
|
| 1072 |
+
"loss": 1.4766,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.48624440179142675,
|
| 1077 |
+
"grad_norm": 2.527624790839194,
|
| 1078 |
+
"learning_rate": 1.221587233301123e-05,
|
| 1079 |
+
"loss": 1.4599,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4894433781190019,
|
| 1084 |
+
"grad_norm": 2.0567117909106547,
|
| 1085 |
+
"learning_rate": 1.2106792699957264e-05,
|
| 1086 |
+
"loss": 1.4048,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4926423544465771,
|
| 1091 |
+
"grad_norm": 2.3023755472980056,
|
| 1092 |
+
"learning_rate": 1.1997450108895807e-05,
|
| 1093 |
+
"loss": 1.4321,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.49584133077415227,
|
| 1098 |
+
"grad_norm": 2.5825599532144987,
|
| 1099 |
+
"learning_rate": 1.1887858207353678e-05,
|
| 1100 |
+
"loss": 1.5275,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4990403071017274,
|
| 1105 |
+
"grad_norm": 2.6267220106883284,
|
| 1106 |
+
"learning_rate": 1.1778030673975227e-05,
|
| 1107 |
+
"loss": 1.4779,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5022392834293026,
|
| 1112 |
+
"grad_norm": 2.470151589555483,
|
| 1113 |
+
"learning_rate": 1.166798121681505e-05,
|
| 1114 |
+
"loss": 1.408,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5054382597568778,
|
| 1119 |
+
"grad_norm": 2.3968962282049255,
|
| 1120 |
+
"learning_rate": 1.1557723571627016e-05,
|
| 1121 |
+
"loss": 1.475,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.508637236084453,
|
| 1126 |
+
"grad_norm": 2.3568776865369583,
|
| 1127 |
+
"learning_rate": 1.1447271500149849e-05,
|
| 1128 |
+
"loss": 1.4876,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5118362124120281,
|
| 1133 |
+
"grad_norm": 2.2087366194914435,
|
| 1134 |
+
"learning_rate": 1.1336638788389473e-05,
|
| 1135 |
+
"loss": 1.4905,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5150351887396033,
|
| 1140 |
+
"grad_norm": 2.1827765863333073,
|
| 1141 |
+
"learning_rate": 1.122583924489832e-05,
|
| 1142 |
+
"loss": 1.4486,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5182341650671785,
|
| 1147 |
+
"grad_norm": 2.1973594046250753,
|
| 1148 |
+
"learning_rate": 1.1114886699051803e-05,
|
| 1149 |
+
"loss": 1.426,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5214331413947537,
|
| 1154 |
+
"grad_norm": 2.34468852888554,
|
| 1155 |
+
"learning_rate": 1.1003794999322246e-05,
|
| 1156 |
+
"loss": 1.4977,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5246321177223289,
|
| 1161 |
+
"grad_norm": 2.4597383453672723,
|
| 1162 |
+
"learning_rate": 1.089257801155037e-05,
|
| 1163 |
+
"loss": 1.3699,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.527831094049904,
|
| 1168 |
+
"grad_norm": 2.4169566923172936,
|
| 1169 |
+
"learning_rate": 1.0781249617214642e-05,
|
| 1170 |
+
"loss": 1.4719,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5310300703774792,
|
| 1175 |
+
"grad_norm": 2.4012841984365316,
|
| 1176 |
+
"learning_rate": 1.0669823711698668e-05,
|
| 1177 |
+
"loss": 1.4233,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5342290467050543,
|
| 1182 |
+
"grad_norm": 2.6293301110509524,
|
| 1183 |
+
"learning_rate": 1.0558314202556866e-05,
|
| 1184 |
+
"loss": 1.4097,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5374280230326296,
|
| 1189 |
+
"grad_norm": 2.31281873866612,
|
| 1190 |
+
"learning_rate": 1.0446735007778563e-05,
|
| 1191 |
+
"loss": 1.4582,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5406269993602048,
|
| 1196 |
+
"grad_norm": 2.4551815500859777,
|
| 1197 |
+
"learning_rate": 1.0335100054050877e-05,
|
| 1198 |
+
"loss": 1.4573,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5438259756877799,
|
| 1203 |
+
"grad_norm": 2.1492777427961123,
|
| 1204 |
+
"learning_rate": 1.0223423275020431e-05,
|
| 1205 |
+
"loss": 1.4539,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5470249520153551,
|
| 1210 |
+
"grad_norm": 2.3528387051422706,
|
| 1211 |
+
"learning_rate": 1.0111718609554254e-05,
|
| 1212 |
+
"loss": 1.429,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5502239283429302,
|
| 1217 |
+
"grad_norm": 2.4037082928038083,
|
| 1218 |
+
"learning_rate": 1e-05,
|
| 1219 |
+
"loss": 1.4023,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5534229046705055,
|
| 1224 |
+
"grad_norm": 2.4169078651865123,
|
| 1225 |
+
"learning_rate": 9.888281390445747e-06,
|
| 1226 |
+
"loss": 1.4645,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5566218809980806,
|
| 1231 |
+
"grad_norm": 2.3509642991185946,
|
| 1232 |
+
"learning_rate": 9.776576724979572e-06,
|
| 1233 |
+
"loss": 1.4286,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5598208573256558,
|
| 1238 |
+
"grad_norm": 2.457238153216398,
|
| 1239 |
+
"learning_rate": 9.664899945949128e-06,
|
| 1240 |
+
"loss": 1.4427,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5630198336532309,
|
| 1245 |
+
"grad_norm": 2.371528253431419,
|
| 1246 |
+
"learning_rate": 9.553264992221442e-06,
|
| 1247 |
+
"loss": 1.4258,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5662188099808061,
|
| 1252 |
+
"grad_norm": 2.335210876008984,
|
| 1253 |
+
"learning_rate": 9.441685797443138e-06,
|
| 1254 |
+
"loss": 1.3789,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5694177863083814,
|
| 1259 |
+
"grad_norm": 2.295598759088956,
|
| 1260 |
+
"learning_rate": 9.330176288301332e-06,
|
| 1261 |
+
"loss": 1.4145,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5726167626359565,
|
| 1266 |
+
"grad_norm": 2.2663716208289317,
|
| 1267 |
+
"learning_rate": 9.21875038278536e-06,
|
| 1268 |
+
"loss": 1.4256,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5758157389635317,
|
| 1273 |
+
"grad_norm": 2.223770397466441,
|
| 1274 |
+
"learning_rate": 9.107421988449632e-06,
|
| 1275 |
+
"loss": 1.4026,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5790147152911068,
|
| 1280 |
+
"grad_norm": 2.2184853669651057,
|
| 1281 |
+
"learning_rate": 8.996205000677758e-06,
|
| 1282 |
+
"loss": 1.4725,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.582213691618682,
|
| 1287 |
+
"grad_norm": 2.8341820881260724,
|
| 1288 |
+
"learning_rate": 8.885113300948199e-06,
|
| 1289 |
+
"loss": 1.3573,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5854126679462572,
|
| 1294 |
+
"grad_norm": 2.365757078522109,
|
| 1295 |
+
"learning_rate": 8.774160755101685e-06,
|
| 1296 |
+
"loss": 1.3802,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5886116442738324,
|
| 1301 |
+
"grad_norm": 2.8561602749028694,
|
| 1302 |
+
"learning_rate": 8.663361211610529e-06,
|
| 1303 |
+
"loss": 1.4659,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5918106206014075,
|
| 1308 |
+
"grad_norm": 2.4443049468820925,
|
| 1309 |
+
"learning_rate": 8.552728499850153e-06,
|
| 1310 |
+
"loss": 1.4615,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5950095969289827,
|
| 1315 |
+
"grad_norm": 2.3972164664499944,
|
| 1316 |
+
"learning_rate": 8.442276428372986e-06,
|
| 1317 |
+
"loss": 1.4139,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5982085732565579,
|
| 1322 |
+
"grad_norm": 2.4891884762897907,
|
| 1323 |
+
"learning_rate": 8.332018783184952e-06,
|
| 1324 |
+
"loss": 1.4309,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6014075495841331,
|
| 1329 |
+
"grad_norm": 2.3294746323733686,
|
| 1330 |
+
"learning_rate": 8.221969326024776e-06,
|
| 1331 |
+
"loss": 1.39,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6046065259117083,
|
| 1336 |
+
"grad_norm": 2.238699313962783,
|
| 1337 |
+
"learning_rate": 8.112141792646326e-06,
|
| 1338 |
+
"loss": 1.3905,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6078055022392834,
|
| 1343 |
+
"grad_norm": 2.3435384537017003,
|
| 1344 |
+
"learning_rate": 8.002549891104196e-06,
|
| 1345 |
+
"loss": 1.5152,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6110044785668586,
|
| 1350 |
+
"grad_norm": 2.3600066440243377,
|
| 1351 |
+
"learning_rate": 7.89320730004274e-06,
|
| 1352 |
+
"loss": 1.4606,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6142034548944337,
|
| 1357 |
+
"grad_norm": 2.2541559813958467,
|
| 1358 |
+
"learning_rate": 7.784127666988774e-06,
|
| 1359 |
+
"loss": 1.4048,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.617402431222009,
|
| 1364 |
+
"grad_norm": 2.4458519460871693,
|
| 1365 |
+
"learning_rate": 7.675324606648187e-06,
|
| 1366 |
+
"loss": 1.4341,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6206014075495841,
|
| 1371 |
+
"grad_norm": 2.4908912138238044,
|
| 1372 |
+
"learning_rate": 7.566811699206604e-06,
|
| 1373 |
+
"loss": 1.3651,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6238003838771593,
|
| 1378 |
+
"grad_norm": 2.4994047187151875,
|
| 1379 |
+
"learning_rate": 7.458602488634416e-06,
|
| 1380 |
+
"loss": 1.4161,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6269993602047345,
|
| 1385 |
+
"grad_norm": 2.6056798824189706,
|
| 1386 |
+
"learning_rate": 7.350710480996266e-06,
|
| 1387 |
+
"loss": 1.4776,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6301983365323096,
|
| 1392 |
+
"grad_norm": 2.2341977902816996,
|
| 1393 |
+
"learning_rate": 7.243149142765317e-06,
|
| 1394 |
+
"loss": 1.4339,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6333973128598849,
|
| 1399 |
+
"grad_norm": 2.4523191182695783,
|
| 1400 |
+
"learning_rate": 7.135931899142438e-06,
|
| 1401 |
+
"loss": 1.3499,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.63659628918746,
|
| 1406 |
+
"grad_norm": 3.348154064863444,
|
| 1407 |
+
"learning_rate": 7.029072132380543e-06,
|
| 1408 |
+
"loss": 1.3616,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6397952655150352,
|
| 1413 |
+
"grad_norm": 2.6387667521173874,
|
| 1414 |
+
"learning_rate": 6.922583180114299e-06,
|
| 1415 |
+
"loss": 1.4326,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6429942418426103,
|
| 1420 |
+
"grad_norm": 2.5816480070087633,
|
| 1421 |
+
"learning_rate": 6.816478333695392e-06,
|
| 1422 |
+
"loss": 1.4003,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6461932181701855,
|
| 1427 |
+
"grad_norm": 2.460254187908023,
|
| 1428 |
+
"learning_rate": 6.710770836533584e-06,
|
| 1429 |
+
"loss": 1.4448,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6493921944977608,
|
| 1434 |
+
"grad_norm": 2.308012258918398,
|
| 1435 |
+
"learning_rate": 6.605473882443725e-06,
|
| 1436 |
+
"loss": 1.4194,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6525911708253359,
|
| 1441 |
+
"grad_norm": 2.2607156095613195,
|
| 1442 |
+
"learning_rate": 6.5006006139990115e-06,
|
| 1443 |
+
"loss": 1.3956,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6557901471529111,
|
| 1448 |
+
"grad_norm": 2.578914066440304,
|
| 1449 |
+
"learning_rate": 6.396164120890562e-06,
|
| 1450 |
+
"loss": 1.4555,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6589891234804862,
|
| 1455 |
+
"grad_norm": 2.3468191923371746,
|
| 1456 |
+
"learning_rate": 6.292177438293665e-06,
|
| 1457 |
+
"loss": 1.373,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6621880998080614,
|
| 1462 |
+
"grad_norm": 2.3299643839647395,
|
| 1463 |
+
"learning_rate": 6.188653545240793e-06,
|
| 1464 |
+
"loss": 1.4025,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6653870761356366,
|
| 1469 |
+
"grad_norm": 2.5769450903031017,
|
| 1470 |
+
"learning_rate": 6.085605363001628e-06,
|
| 1471 |
+
"loss": 1.3718,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6685860524632118,
|
| 1476 |
+
"grad_norm": 2.5207006902093894,
|
| 1477 |
+
"learning_rate": 5.983045753470308e-06,
|
| 1478 |
+
"loss": 1.3983,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6717850287907869,
|
| 1483 |
+
"grad_norm": 2.2867546268489725,
|
| 1484 |
+
"learning_rate": 5.880987517560075e-06,
|
| 1485 |
+
"loss": 1.4102,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6749840051183621,
|
| 1490 |
+
"grad_norm": 2.5843293143331936,
|
| 1491 |
+
"learning_rate": 5.779443393605536e-06,
|
| 1492 |
+
"loss": 1.3837,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6781829814459372,
|
| 1497 |
+
"grad_norm": 2.466294876267873,
|
| 1498 |
+
"learning_rate": 5.678426055772746e-06,
|
| 1499 |
+
"loss": 1.4049,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6813819577735125,
|
| 1504 |
+
"grad_norm": 2.3003489185145205,
|
| 1505 |
+
"learning_rate": 5.577948112477271e-06,
|
| 1506 |
+
"loss": 1.37,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6845809341010877,
|
| 1511 |
+
"grad_norm": 2.4936655314476495,
|
| 1512 |
+
"learning_rate": 5.478022104810483e-06,
|
| 1513 |
+
"loss": 1.4096,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6877799104286628,
|
| 1518 |
+
"grad_norm": 2.47082758044549,
|
| 1519 |
+
"learning_rate": 5.378660504974271e-06,
|
| 1520 |
+
"loss": 1.3426,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.690978886756238,
|
| 1525 |
+
"grad_norm": 2.3960847639394673,
|
| 1526 |
+
"learning_rate": 5.2798757147242986e-06,
|
| 1527 |
+
"loss": 1.3714,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6941778630838131,
|
| 1532 |
+
"grad_norm": 2.4530021700307993,
|
| 1533 |
+
"learning_rate": 5.1816800638221176e-06,
|
| 1534 |
+
"loss": 1.414,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6973768394113884,
|
| 1539 |
+
"grad_norm": 2.2603715033506453,
|
| 1540 |
+
"learning_rate": 5.084085808496211e-06,
|
| 1541 |
+
"loss": 1.3491,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7005758157389635,
|
| 1546 |
+
"grad_norm": 2.608868595746518,
|
| 1547 |
+
"learning_rate": 4.987105129912252e-06,
|
| 1548 |
+
"loss": 1.3972,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7037747920665387,
|
| 1553 |
+
"grad_norm": 2.297568946845784,
|
| 1554 |
+
"learning_rate": 4.890750132652724e-06,
|
| 1555 |
+
"loss": 1.4044,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7069737683941139,
|
| 1560 |
+
"grad_norm": 2.90064685740098,
|
| 1561 |
+
"learning_rate": 4.795032843206091e-06,
|
| 1562 |
+
"loss": 1.4374,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.710172744721689,
|
| 1567 |
+
"grad_norm": 2.443681227970343,
|
| 1568 |
+
"learning_rate": 4.699965208465702e-06,
|
| 1569 |
+
"loss": 1.4167,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7133717210492643,
|
| 1574 |
+
"grad_norm": 2.425095972610202,
|
| 1575 |
+
"learning_rate": 4.605559094238681e-06,
|
| 1576 |
+
"loss": 1.4242,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7165706973768394,
|
| 1581 |
+
"grad_norm": 2.488579755452615,
|
| 1582 |
+
"learning_rate": 4.511826283764872e-06,
|
| 1583 |
+
"loss": 1.3888,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7197696737044146,
|
| 1588 |
+
"grad_norm": 2.550407448039906,
|
| 1589 |
+
"learning_rate": 4.418778476246132e-06,
|
| 1590 |
+
"loss": 1.4245,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7229686500319897,
|
| 1595 |
+
"grad_norm": 2.3572217906367974,
|
| 1596 |
+
"learning_rate": 4.326427285386118e-06,
|
| 1597 |
+
"loss": 1.3536,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7261676263595649,
|
| 1602 |
+
"grad_norm": 2.898475510497588,
|
| 1603 |
+
"learning_rate": 4.234784237940705e-06,
|
| 1604 |
+
"loss": 1.3971,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7293666026871402,
|
| 1609 |
+
"grad_norm": 2.5246296926850937,
|
| 1610 |
+
"learning_rate": 4.143860772279287e-06,
|
| 1611 |
+
"loss": 1.3661,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7325655790147153,
|
| 1616 |
+
"grad_norm": 2.453168323628695,
|
| 1617 |
+
"learning_rate": 4.053668236957135e-06,
|
| 1618 |
+
"loss": 1.3489,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7357645553422905,
|
| 1623 |
+
"grad_norm": 2.648255030215048,
|
| 1624 |
+
"learning_rate": 3.964217889298882e-06,
|
| 1625 |
+
"loss": 1.3665,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7389635316698656,
|
| 1630 |
+
"grad_norm": 2.4691460031589014,
|
| 1631 |
+
"learning_rate": 3.875520893993482e-06,
|
| 1632 |
+
"loss": 1.3954,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7421625079974408,
|
| 1637 |
+
"grad_norm": 2.578218572991363,
|
| 1638 |
+
"learning_rate": 3.787588321700697e-06,
|
| 1639 |
+
"loss": 1.3518,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.745361484325016,
|
| 1644 |
+
"grad_norm": 2.2329429977704245,
|
| 1645 |
+
"learning_rate": 3.7004311476692977e-06,
|
| 1646 |
+
"loss": 1.4178,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7485604606525912,
|
| 1651 |
+
"grad_norm": 2.352446489946909,
|
| 1652 |
+
"learning_rate": 3.6140602503672217e-06,
|
| 1653 |
+
"loss": 1.3538,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7517594369801663,
|
| 1658 |
+
"grad_norm": 2.7561385282190076,
|
| 1659 |
+
"learning_rate": 3.528486410123758e-06,
|
| 1660 |
+
"loss": 1.4613,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7549584133077415,
|
| 1665 |
+
"grad_norm": 2.138602055751757,
|
| 1666 |
+
"learning_rate": 3.443720307784022e-06,
|
| 1667 |
+
"loss": 1.3133,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7581573896353166,
|
| 1672 |
+
"grad_norm": 2.3177590399711643,
|
| 1673 |
+
"learning_rate": 3.359772523375837e-06,
|
| 1674 |
+
"loss": 1.297,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7613563659628919,
|
| 1679 |
+
"grad_norm": 2.3451441921767064,
|
| 1680 |
+
"learning_rate": 3.276653534789185e-06,
|
| 1681 |
+
"loss": 1.2972,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7645553422904671,
|
| 1686 |
+
"grad_norm": 2.313314457358321,
|
| 1687 |
+
"learning_rate": 3.1943737164684094e-06,
|
| 1688 |
+
"loss": 1.367,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7677543186180422,
|
| 1693 |
+
"grad_norm": 2.3536589940255364,
|
| 1694 |
+
"learning_rate": 3.11294333811736e-06,
|
| 1695 |
+
"loss": 1.3865,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7709532949456174,
|
| 1700 |
+
"grad_norm": 2.359964164926833,
|
| 1701 |
+
"learning_rate": 3.032372563417556e-06,
|
| 1702 |
+
"loss": 1.3596,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7741522712731925,
|
| 1707 |
+
"grad_norm": 2.2343745022509185,
|
| 1708 |
+
"learning_rate": 2.952671448759631e-06,
|
| 1709 |
+
"loss": 1.4056,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7773512476007678,
|
| 1714 |
+
"grad_norm": 2.5543445735600305,
|
| 1715 |
+
"learning_rate": 2.8738499419881517e-06,
|
| 1716 |
+
"loss": 1.3209,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.780550223928343,
|
| 1721 |
+
"grad_norm": 2.2930909232447445,
|
| 1722 |
+
"learning_rate": 2.795917881159973e-06,
|
| 1723 |
+
"loss": 1.3144,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7837492002559181,
|
| 1728 |
+
"grad_norm": 2.455120126910525,
|
| 1729 |
+
"learning_rate": 2.718884993316311e-06,
|
| 1730 |
+
"loss": 1.3333,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7869481765834933,
|
| 1735 |
+
"grad_norm": 2.4966852667217934,
|
| 1736 |
+
"learning_rate": 2.642760893268684e-06,
|
| 1737 |
+
"loss": 1.3509,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7901471529110684,
|
| 1742 |
+
"grad_norm": 2.4028166421914867,
|
| 1743 |
+
"learning_rate": 2.5675550823988184e-06,
|
| 1744 |
+
"loss": 1.3491,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7933461292386437,
|
| 1749 |
+
"grad_norm": 2.6710950106791205,
|
| 1750 |
+
"learning_rate": 2.493276947472756e-06,
|
| 1751 |
+
"loss": 1.3498,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7965451055662188,
|
| 1756 |
+
"grad_norm": 2.426831641425111,
|
| 1757 |
+
"learning_rate": 2.4199357594692564e-06,
|
| 1758 |
+
"loss": 1.2967,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.799744081893794,
|
| 1763 |
+
"grad_norm": 2.4719107581073527,
|
| 1764 |
+
"learning_rate": 2.347540672422627e-06,
|
| 1765 |
+
"loss": 1.3236,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8029430582213691,
|
| 1770 |
+
"grad_norm": 2.516296634331389,
|
| 1771 |
+
"learning_rate": 2.2761007222801866e-06,
|
| 1772 |
+
"loss": 1.3533,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8061420345489443,
|
| 1777 |
+
"grad_norm": 2.332883844489818,
|
| 1778 |
+
"learning_rate": 2.2056248257744383e-06,
|
| 1779 |
+
"loss": 1.3402,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8093410108765196,
|
| 1784 |
+
"grad_norm": 2.470627298913124,
|
| 1785 |
+
"learning_rate": 2.136121779310132e-06,
|
| 1786 |
+
"loss": 1.4162,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8125399872040947,
|
| 1791 |
+
"grad_norm": 2.5704350069752833,
|
| 1792 |
+
"learning_rate": 2.067600257866358e-06,
|
| 1793 |
+
"loss": 1.3798,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8157389635316699,
|
| 1798 |
+
"grad_norm": 2.3710654190200855,
|
| 1799 |
+
"learning_rate": 2.000068813913777e-06,
|
| 1800 |
+
"loss": 1.2724,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.818937939859245,
|
| 1805 |
+
"grad_norm": 2.558360997816421,
|
| 1806 |
+
"learning_rate": 1.933535876347141e-06,
|
| 1807 |
+
"loss": 1.3785,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8221369161868202,
|
| 1812 |
+
"grad_norm": 2.5100998715469354,
|
| 1813 |
+
"learning_rate": 1.8680097494332682e-06,
|
| 1814 |
+
"loss": 1.2916,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8253358925143954,
|
| 1819 |
+
"grad_norm": 2.6532733728905207,
|
| 1820 |
+
"learning_rate": 1.8034986117745245e-06,
|
| 1821 |
+
"loss": 1.3417,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8285348688419706,
|
| 1826 |
+
"grad_norm": 2.3261659285032663,
|
| 1827 |
+
"learning_rate": 1.7400105152880331e-06,
|
| 1828 |
+
"loss": 1.3738,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8317338451695457,
|
| 1833 |
+
"grad_norm": 2.4853808407220725,
|
| 1834 |
+
"learning_rate": 1.6775533842006853e-06,
|
| 1835 |
+
"loss": 1.2736,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8349328214971209,
|
| 1840 |
+
"grad_norm": 2.697298725951212,
|
| 1841 |
+
"learning_rate": 1.6161350140600606e-06,
|
| 1842 |
+
"loss": 1.2781,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.838131797824696,
|
| 1847 |
+
"grad_norm": 2.355635659400595,
|
| 1848 |
+
"learning_rate": 1.555763070761448e-06,
|
| 1849 |
+
"loss": 1.361,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8413307741522713,
|
| 1854 |
+
"grad_norm": 2.6842436040456534,
|
| 1855 |
+
"learning_rate": 1.496445089591031e-06,
|
| 1856 |
+
"loss": 1.4244,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8445297504798465,
|
| 1861 |
+
"grad_norm": 2.4556961691024055,
|
| 1862 |
+
"learning_rate": 1.4381884742853723e-06,
|
| 1863 |
+
"loss": 1.3541,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8477287268074216,
|
| 1868 |
+
"grad_norm": 2.444589907894017,
|
| 1869 |
+
"learning_rate": 1.381000496107313e-06,
|
| 1870 |
+
"loss": 1.3687,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8509277031349968,
|
| 1875 |
+
"grad_norm": 2.4233465965102803,
|
| 1876 |
+
"learning_rate": 1.3248882929384321e-06,
|
| 1877 |
+
"loss": 1.3331,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8541266794625719,
|
| 1882 |
+
"grad_norm": 2.824638190223306,
|
| 1883 |
+
"learning_rate": 1.2698588683881185e-06,
|
| 1884 |
+
"loss": 1.3307,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8573256557901472,
|
| 1889 |
+
"grad_norm": 2.38649028106469,
|
| 1890 |
+
"learning_rate": 1.215919090919434e-06,
|
| 1891 |
+
"loss": 1.3524,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8605246321177223,
|
| 1896 |
+
"grad_norm": 2.450733408516033,
|
| 1897 |
+
"learning_rate": 1.1630756929918218e-06,
|
| 1898 |
+
"loss": 1.347,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8637236084452975,
|
| 1903 |
+
"grad_norm": 2.4002381429681616,
|
| 1904 |
+
"learning_rate": 1.111335270220798e-06,
|
| 1905 |
+
"loss": 1.39,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8669225847728727,
|
| 1910 |
+
"grad_norm": 2.700669673337657,
|
| 1911 |
+
"learning_rate": 1.060704280554733e-06,
|
| 1912 |
+
"loss": 1.297,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8701215611004478,
|
| 1917 |
+
"grad_norm": 2.6066489722882475,
|
| 1918 |
+
"learning_rate": 1.0111890434688009e-06,
|
| 1919 |
+
"loss": 1.3923,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8733205374280231,
|
| 1924 |
+
"grad_norm": 2.572737149493473,
|
| 1925 |
+
"learning_rate": 9.62795739176212e-07,
|
| 1926 |
+
"loss": 1.369,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8765195137555982,
|
| 1931 |
+
"grad_norm": 2.2233501199427397,
|
| 1932 |
+
"learning_rate": 9.155304078568495e-07,
|
| 1933 |
+
"loss": 1.3645,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8797184900831734,
|
| 1938 |
+
"grad_norm": 2.6489650845893786,
|
| 1939 |
+
"learning_rate": 8.693989489033494e-07,
|
| 1940 |
+
"loss": 1.3737,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8829174664107485,
|
| 1945 |
+
"grad_norm": 2.5371100403473608,
|
| 1946 |
+
"learning_rate": 8.244071201847826e-07,
|
| 1947 |
+
"loss": 1.3034,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8861164427383237,
|
| 1952 |
+
"grad_norm": 2.432385883291253,
|
| 1953 |
+
"learning_rate": 7.805605373279934e-07,
|
| 1954 |
+
"loss": 1.3882,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.889315419065899,
|
| 1959 |
+
"grad_norm": 2.4449203754420465,
|
| 1960 |
+
"learning_rate": 7.378646730166783e-07,
|
| 1961 |
+
"loss": 1.3167,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8925143953934741,
|
| 1966 |
+
"grad_norm": 2.441267548142986,
|
| 1967 |
+
"learning_rate": 6.963248563083203e-07,
|
| 1968 |
+
"loss": 1.3306,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8957133717210493,
|
| 1973 |
+
"grad_norm": 2.498359509259829,
|
| 1974 |
+
"learning_rate": 6.559462719690501e-07,
|
| 1975 |
+
"loss": 1.3204,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8989123480486244,
|
| 1980 |
+
"grad_norm": 2.4688012231508276,
|
| 1981 |
+
"learning_rate": 6.167339598265109e-07,
|
| 1982 |
+
"loss": 1.3334,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9021113243761996,
|
| 1987 |
+
"grad_norm": 2.2093568500449785,
|
| 1988 |
+
"learning_rate": 5.78692814140811e-07,
|
| 1989 |
+
"loss": 1.3192,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9053103007037748,
|
| 1994 |
+
"grad_norm": 2.4077314191659234,
|
| 1995 |
+
"learning_rate": 5.418275829936537e-07,
|
| 1996 |
+
"loss": 1.2858,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.90850927703135,
|
| 2001 |
+
"grad_norm": 2.4379295125735045,
|
| 2002 |
+
"learning_rate": 5.06142867695708e-07,
|
| 2003 |
+
"loss": 1.3432,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9117082533589251,
|
| 2008 |
+
"grad_norm": 2.4162225057600475,
|
| 2009 |
+
"learning_rate": 4.716431222122997e-07,
|
| 2010 |
+
"loss": 1.4049,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9149072296865003,
|
| 2015 |
+
"grad_norm": 2.383814444310597,
|
| 2016 |
+
"learning_rate": 4.3833265260749157e-07,
|
| 2017 |
+
"loss": 1.3355,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9181062060140754,
|
| 2022 |
+
"grad_norm": 2.427977187387928,
|
| 2023 |
+
"learning_rate": 4.062156165066211e-07,
|
| 2024 |
+
"loss": 1.3702,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9213051823416507,
|
| 2029 |
+
"grad_norm": 2.256695950593894,
|
| 2030 |
+
"learning_rate": 3.752960225773772e-07,
|
| 2031 |
+
"loss": 1.3701,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9245041586692259,
|
| 2036 |
+
"grad_norm": 2.447985505592198,
|
| 2037 |
+
"learning_rate": 3.4557773002945607e-07,
|
| 2038 |
+
"loss": 1.3774,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.927703134996801,
|
| 2043 |
+
"grad_norm": 2.403393034990299,
|
| 2044 |
+
"learning_rate": 3.170644481328711e-07,
|
| 2045 |
+
"loss": 1.348,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9309021113243762,
|
| 2050 |
+
"grad_norm": 2.213446206865036,
|
| 2051 |
+
"learning_rate": 2.8975973575499526e-07,
|
| 2052 |
+
"loss": 1.3011,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9341010876519513,
|
| 2057 |
+
"grad_norm": 2.310065753213684,
|
| 2058 |
+
"learning_rate": 2.636670009163522e-07,
|
| 2059 |
+
"loss": 1.3022,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9373000639795266,
|
| 2064 |
+
"grad_norm": 2.3987119937006676,
|
| 2065 |
+
"learning_rate": 2.3878950036524963e-07,
|
| 2066 |
+
"loss": 1.2379,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9404990403071017,
|
| 2071 |
+
"grad_norm": 2.2743696194136493,
|
| 2072 |
+
"learning_rate": 2.1513033917129334e-07,
|
| 2073 |
+
"loss": 1.2969,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9436980166346769,
|
| 2078 |
+
"grad_norm": 2.656197726258722,
|
| 2079 |
+
"learning_rate": 1.9269247033782744e-07,
|
| 2080 |
+
"loss": 1.2934,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.946896992962252,
|
| 2085 |
+
"grad_norm": 2.5089508490923897,
|
| 2086 |
+
"learning_rate": 1.7147869443335463e-07,
|
| 2087 |
+
"loss": 1.3208,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9500959692898272,
|
| 2092 |
+
"grad_norm": 2.742202995722091,
|
| 2093 |
+
"learning_rate": 1.5149165924199016e-07,
|
| 2094 |
+
"loss": 1.3615,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9532949456174025,
|
| 2099 |
+
"grad_norm": 2.5187291446173066,
|
| 2100 |
+
"learning_rate": 1.3273385943297746e-07,
|
| 2101 |
+
"loss": 1.3065,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9564939219449776,
|
| 2106 |
+
"grad_norm": 2.604371951037358,
|
| 2107 |
+
"learning_rate": 1.1520763624931597e-07,
|
| 2108 |
+
"loss": 1.3141,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9596928982725528,
|
| 2113 |
+
"grad_norm": 2.6802125882941485,
|
| 2114 |
+
"learning_rate": 9.891517721554499e-08,
|
| 2115 |
+
"loss": 1.4077,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9628918746001279,
|
| 2120 |
+
"grad_norm": 2.706788189883898,
|
| 2121 |
+
"learning_rate": 8.385851586470318e-08,
|
| 2122 |
+
"loss": 1.3212,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9660908509277031,
|
| 2127 |
+
"grad_norm": 2.4793431373091352,
|
| 2128 |
+
"learning_rate": 7.003953148452036e-08,
|
| 2129 |
+
"loss": 1.3821,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9692898272552783,
|
| 2134 |
+
"grad_norm": 2.278314797678949,
|
| 2135 |
+
"learning_rate": 5.745994888285311e-08,
|
| 2136 |
+
"loss": 1.2863,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9724888035828535,
|
| 2141 |
+
"grad_norm": 2.559732108312943,
|
| 2142 |
+
"learning_rate": 4.612133817239905e-08,
|
| 2143 |
+
"loss": 1.3566,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9756877799104287,
|
| 2148 |
+
"grad_norm": 2.2885476469371833,
|
| 2149 |
+
"learning_rate": 3.602511457473479e-08,
|
| 2150 |
+
"loss": 1.3009,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9788867562380038,
|
| 2155 |
+
"grad_norm": 2.4890698617978617,
|
| 2156 |
+
"learning_rate": 2.7172538243666057e-08,
|
| 2157 |
+
"loss": 1.4347,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.982085732565579,
|
| 2162 |
+
"grad_norm": 2.45559913282601,
|
| 2163 |
+
"learning_rate": 1.9564714107945804e-08,
|
| 2164 |
+
"loss": 1.3811,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9852847088931542,
|
| 2169 |
+
"grad_norm": 2.4370971366390863,
|
| 2170 |
+
"learning_rate": 1.3202591733365577e-08,
|
| 2171 |
+
"loss": 1.3086,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9884836852207294,
|
| 2176 |
+
"grad_norm": 2.6654429719413097,
|
| 2177 |
+
"learning_rate": 8.086965204233688e-09,
|
| 2178 |
+
"loss": 1.3538,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9916826615483045,
|
| 2183 |
+
"grad_norm": 2.5063252266533773,
|
| 2184 |
+
"learning_rate": 4.218473024261149e-09,
|
| 2185 |
+
"loss": 1.3368,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9948816378758797,
|
| 2190 |
+
"grad_norm": 2.5804016285503906,
|
| 2191 |
+
"learning_rate": 1.5975980368709843e-09,
|
| 2192 |
+
"loss": 1.3791,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9980806142034548,
|
| 2197 |
+
"grad_norm": 2.5363071477182544,
|
| 2198 |
+
"learning_rate": 2.2466736492643416e-10,
|
| 2199 |
+
"loss": 1.275,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 1.0,
|
| 2204 |
+
"eval_loss": 1.3733937740325928,
|
| 2205 |
+
"eval_runtime": 17.2957,
|
| 2206 |
+
"eval_samples_per_second": 28.909,
|
| 2207 |
+
"eval_steps_per_second": 0.925,
|
| 2208 |
+
"step": 1563
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.0,
|
| 2212 |
+
"step": 1563,
|
| 2213 |
+
"total_flos": 45448595030016.0,
|
| 2214 |
+
"train_loss": 1.473227332512385,
|
| 2215 |
+
"train_runtime": 5771.4551,
|
| 2216 |
+
"train_samples_per_second": 8.663,
|
| 2217 |
+
"train_steps_per_second": 0.271
|
| 2218 |
+
}
|
| 2219 |
+
],
|
| 2220 |
+
"logging_steps": 5,
|
| 2221 |
+
"max_steps": 1563,
|
| 2222 |
+
"num_input_tokens_seen": 0,
|
| 2223 |
+
"num_train_epochs": 1,
|
| 2224 |
+
"save_steps": 500,
|
| 2225 |
+
"stateful_callbacks": {
|
| 2226 |
+
"TrainerControl": {
|
| 2227 |
+
"args": {
|
| 2228 |
+
"should_epoch_stop": false,
|
| 2229 |
+
"should_evaluate": false,
|
| 2230 |
+
"should_log": false,
|
| 2231 |
+
"should_save": true,
|
| 2232 |
+
"should_training_stop": true
|
| 2233 |
+
},
|
| 2234 |
+
"attributes": {}
|
| 2235 |
+
}
|
| 2236 |
+
},
|
| 2237 |
+
"total_flos": 45448595030016.0,
|
| 2238 |
+
"train_batch_size": 8,
|
| 2239 |
+
"trial_name": null,
|
| 2240 |
+
"trial_params": null
|
| 2241 |
+
}
|