christineyu
commited on
Model save
Browse files- README.md +57 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1073 -0
README.md
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
3 |
+
library_name: transformers
|
4 |
+
model_name: prometheus-7b-v1.5-beta-1-over1218-pref
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
- trl
|
8 |
+
- sft
|
9 |
+
licence: license
|
10 |
+
---
|
11 |
+
|
12 |
+
# Model Card for prometheus-7b-v1.5-beta-1-over1218-pref
|
13 |
+
|
14 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2).
|
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="christineyu/prometheus-7b-v1.5-beta-1-over1218-pref", 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/myexp/huggingface/runs/swm45e6g)
|
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: 2.3.0
|
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": 2.0,
|
3 |
+
"total_flos": 1.3020461702683034e+17,
|
4 |
+
"train_loss": 0.037445212856265314,
|
5 |
+
"train_runtime": 3051.2562,
|
6 |
+
"train_samples": 990,
|
7 |
+
"train_samples_per_second": 0.237,
|
8 |
+
"train_steps_per_second": 0.237
|
9 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"total_flos": 1.3020461702683034e+17,
|
4 |
+
"train_loss": 0.037445212856265314,
|
5 |
+
"train_runtime": 3051.2562,
|
6 |
+
"train_samples": 990,
|
7 |
+
"train_samples_per_second": 0.237,
|
8 |
+
"train_steps_per_second": 0.237
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1073 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 722,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.002770083102493075,
|
13 |
+
"grad_norm": 4.970390796661377,
|
14 |
+
"learning_rate": 1.36986301369863e-07,
|
15 |
+
"loss": 0.7038,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.013850415512465374,
|
20 |
+
"grad_norm": 5.532266139984131,
|
21 |
+
"learning_rate": 6.849315068493151e-07,
|
22 |
+
"loss": 0.7051,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.027700831024930747,
|
27 |
+
"grad_norm": 3.3267316818237305,
|
28 |
+
"learning_rate": 1.3698630136986302e-06,
|
29 |
+
"loss": 0.6794,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.04155124653739612,
|
34 |
+
"grad_norm": 2.5890796184539795,
|
35 |
+
"learning_rate": 2.0547945205479454e-06,
|
36 |
+
"loss": 0.632,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.055401662049861494,
|
41 |
+
"grad_norm": 2.3750393390655518,
|
42 |
+
"learning_rate": 2.7397260273972604e-06,
|
43 |
+
"loss": 0.5698,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06925207756232687,
|
48 |
+
"grad_norm": 2.339930295944214,
|
49 |
+
"learning_rate": 3.4246575342465754e-06,
|
50 |
+
"loss": 0.5216,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.08310249307479224,
|
55 |
+
"grad_norm": 2.103806734085083,
|
56 |
+
"learning_rate": 4.109589041095891e-06,
|
57 |
+
"loss": 0.4612,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.09695290858725762,
|
62 |
+
"grad_norm": 2.015624761581421,
|
63 |
+
"learning_rate": 4.7945205479452054e-06,
|
64 |
+
"loss": 0.3838,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.11080332409972299,
|
69 |
+
"grad_norm": 2.146289587020874,
|
70 |
+
"learning_rate": 5.479452054794521e-06,
|
71 |
+
"loss": 0.2954,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.12465373961218837,
|
76 |
+
"grad_norm": 2.125903844833374,
|
77 |
+
"learning_rate": 6.164383561643836e-06,
|
78 |
+
"loss": 0.1999,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.13850415512465375,
|
83 |
+
"grad_norm": 2.214215040206909,
|
84 |
+
"learning_rate": 6.849315068493151e-06,
|
85 |
+
"loss": 0.1135,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.1523545706371191,
|
90 |
+
"grad_norm": 3.1861069202423096,
|
91 |
+
"learning_rate": 7.534246575342466e-06,
|
92 |
+
"loss": 0.0612,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.16620498614958448,
|
97 |
+
"grad_norm": 2.2513909339904785,
|
98 |
+
"learning_rate": 8.219178082191782e-06,
|
99 |
+
"loss": 0.0513,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.18005540166204986,
|
104 |
+
"grad_norm": 1.5095714330673218,
|
105 |
+
"learning_rate": 8.904109589041097e-06,
|
106 |
+
"loss": 0.033,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.19390581717451524,
|
111 |
+
"grad_norm": 1.3538941144943237,
|
112 |
+
"learning_rate": 9.589041095890411e-06,
|
113 |
+
"loss": 0.0338,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.2077562326869806,
|
118 |
+
"grad_norm": 1.7012300491333008,
|
119 |
+
"learning_rate": 9.999765681294371e-06,
|
120 |
+
"loss": 0.0231,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.22160664819944598,
|
125 |
+
"grad_norm": 0.762015700340271,
|
126 |
+
"learning_rate": 9.997129848070563e-06,
|
127 |
+
"loss": 0.0182,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.23545706371191136,
|
132 |
+
"grad_norm": 1.277585744857788,
|
133 |
+
"learning_rate": 9.991566832387564e-06,
|
134 |
+
"loss": 0.0154,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.24930747922437674,
|
139 |
+
"grad_norm": 0.8343765735626221,
|
140 |
+
"learning_rate": 9.983079892908332e-06,
|
141 |
+
"loss": 0.0185,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.2631578947368421,
|
146 |
+
"grad_norm": 1.3323298692703247,
|
147 |
+
"learning_rate": 9.971674001050687e-06,
|
148 |
+
"loss": 0.0158,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.2770083102493075,
|
153 |
+
"grad_norm": 0.4257211983203888,
|
154 |
+
"learning_rate": 9.957355838075188e-06,
|
155 |
+
"loss": 0.0154,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.29085872576177285,
|
160 |
+
"grad_norm": 0.6481120586395264,
|
161 |
+
"learning_rate": 9.940133791171445e-06,
|
162 |
+
"loss": 0.0105,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.3047091412742382,
|
167 |
+
"grad_norm": 0.8009067177772522,
|
168 |
+
"learning_rate": 9.920017948545109e-06,
|
169 |
+
"loss": 0.0111,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.3185595567867036,
|
174 |
+
"grad_norm": 0.4706541895866394,
|
175 |
+
"learning_rate": 9.897020093508502e-06,
|
176 |
+
"loss": 0.0123,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.33240997229916897,
|
181 |
+
"grad_norm": 0.7167349457740784,
|
182 |
+
"learning_rate": 9.871153697578254e-06,
|
183 |
+
"loss": 0.0129,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.3462603878116344,
|
188 |
+
"grad_norm": 0.9122310876846313,
|
189 |
+
"learning_rate": 9.842433912584066e-06,
|
190 |
+
"loss": 0.0118,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.3601108033240997,
|
195 |
+
"grad_norm": 0.3195926249027252,
|
196 |
+
"learning_rate": 9.810877561793178e-06,
|
197 |
+
"loss": 0.0091,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.3739612188365651,
|
202 |
+
"grad_norm": 0.5710881352424622,
|
203 |
+
"learning_rate": 9.776503130055758e-06,
|
204 |
+
"loss": 0.0116,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.3878116343490305,
|
209 |
+
"grad_norm": 0.6485520005226135,
|
210 |
+
"learning_rate": 9.739330752976981e-06,
|
211 |
+
"loss": 0.0129,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.40166204986149584,
|
216 |
+
"grad_norm": 0.41829606890678406,
|
217 |
+
"learning_rate": 9.699382205122138e-06,
|
218 |
+
"loss": 0.0105,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.4155124653739612,
|
223 |
+
"grad_norm": 0.4999130070209503,
|
224 |
+
"learning_rate": 9.656680887261693e-06,
|
225 |
+
"loss": 0.0096,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.4293628808864266,
|
230 |
+
"grad_norm": 0.4883331060409546,
|
231 |
+
"learning_rate": 9.611251812663748e-06,
|
232 |
+
"loss": 0.0085,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.44321329639889195,
|
237 |
+
"grad_norm": 0.33876070380210876,
|
238 |
+
"learning_rate": 9.563121592441949e-06,
|
239 |
+
"loss": 0.006,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.45706371191135736,
|
244 |
+
"grad_norm": 0.9331468939781189,
|
245 |
+
"learning_rate": 9.512318419967427e-06,
|
246 |
+
"loss": 0.0062,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.4709141274238227,
|
251 |
+
"grad_norm": 0.5563158988952637,
|
252 |
+
"learning_rate": 9.458872054353888e-06,
|
253 |
+
"loss": 0.0072,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.48476454293628807,
|
258 |
+
"grad_norm": 0.7999743223190308,
|
259 |
+
"learning_rate": 9.402813803025526e-06,
|
260 |
+
"loss": 0.0063,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.4986149584487535,
|
265 |
+
"grad_norm": 0.35912105441093445,
|
266 |
+
"learning_rate": 9.344176503378003e-06,
|
267 |
+
"loss": 0.0069,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.5124653739612188,
|
272 |
+
"grad_norm": 0.3588772416114807,
|
273 |
+
"learning_rate": 9.282994503543185e-06,
|
274 |
+
"loss": 0.0068,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.5263157894736842,
|
279 |
+
"grad_norm": 0.5413822531700134,
|
280 |
+
"learning_rate": 9.219303642268953e-06,
|
281 |
+
"loss": 0.0082,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.5401662049861495,
|
286 |
+
"grad_norm": 0.32814571261405945,
|
287 |
+
"learning_rate": 9.153141227925828e-06,
|
288 |
+
"loss": 0.0057,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.554016620498615,
|
293 |
+
"grad_norm": 0.48693302273750305,
|
294 |
+
"learning_rate": 9.084546016652758e-06,
|
295 |
+
"loss": 0.0078,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.5678670360110804,
|
300 |
+
"grad_norm": 0.20453135669231415,
|
301 |
+
"learning_rate": 9.013558189654819e-06,
|
302 |
+
"loss": 0.0081,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.5817174515235457,
|
307 |
+
"grad_norm": 0.32578709721565247,
|
308 |
+
"learning_rate": 8.940219329666167e-06,
|
309 |
+
"loss": 0.0048,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.5955678670360111,
|
314 |
+
"grad_norm": 0.32217937707901,
|
315 |
+
"learning_rate": 8.864572396591996e-06,
|
316 |
+
"loss": 0.0053,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.6094182825484764,
|
321 |
+
"grad_norm": 0.6116990447044373,
|
322 |
+
"learning_rate": 8.786661702343811e-06,
|
323 |
+
"loss": 0.0067,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.6232686980609419,
|
328 |
+
"grad_norm": 0.6656084060668945,
|
329 |
+
"learning_rate": 8.706532884882704e-06,
|
330 |
+
"loss": 0.0067,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.6371191135734072,
|
335 |
+
"grad_norm": 0.4221888482570648,
|
336 |
+
"learning_rate": 8.624232881485887e-06,
|
337 |
+
"loss": 0.0069,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.6509695290858726,
|
342 |
+
"grad_norm": 0.35918885469436646,
|
343 |
+
"learning_rate": 8.539809901252118e-06,
|
344 |
+
"loss": 0.006,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.6648199445983379,
|
349 |
+
"grad_norm": 0.3014688193798065,
|
350 |
+
"learning_rate": 8.453313396862113e-06,
|
351 |
+
"loss": 0.0052,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.6786703601108033,
|
356 |
+
"grad_norm": 0.32176536321640015,
|
357 |
+
"learning_rate": 8.364794035610527e-06,
|
358 |
+
"loss": 0.0061,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.6925207756232687,
|
363 |
+
"grad_norm": 0.2977988123893738,
|
364 |
+
"learning_rate": 8.274303669726427e-06,
|
365 |
+
"loss": 0.0051,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.7063711911357341,
|
370 |
+
"grad_norm": 0.50001460313797,
|
371 |
+
"learning_rate": 8.181895305999665e-06,
|
372 |
+
"loss": 0.0072,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.7202216066481995,
|
377 |
+
"grad_norm": 0.7377979159355164,
|
378 |
+
"learning_rate": 8.08762307473096e-06,
|
379 |
+
"loss": 0.0064,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.7340720221606648,
|
384 |
+
"grad_norm": 0.3045377731323242,
|
385 |
+
"learning_rate": 7.991542198023827e-06,
|
386 |
+
"loss": 0.0051,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.7479224376731302,
|
391 |
+
"grad_norm": 0.22468125820159912,
|
392 |
+
"learning_rate": 7.893708957436982e-06,
|
393 |
+
"loss": 0.0056,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.7617728531855956,
|
398 |
+
"grad_norm": 0.4352708160877228,
|
399 |
+
"learning_rate": 7.794180661016143e-06,
|
400 |
+
"loss": 0.0042,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.775623268698061,
|
405 |
+
"grad_norm": 0.29045531153678894,
|
406 |
+
"learning_rate": 7.693015609724524e-06,
|
407 |
+
"loss": 0.0065,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.7894736842105263,
|
412 |
+
"grad_norm": 0.35637104511260986,
|
413 |
+
"learning_rate": 7.5902730632917395e-06,
|
414 |
+
"loss": 0.0047,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.8033240997229917,
|
419 |
+
"grad_norm": 0.5587311387062073,
|
420 |
+
"learning_rate": 7.486013205501053e-06,
|
421 |
+
"loss": 0.0063,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.817174515235457,
|
426 |
+
"grad_norm": 0.19924254715442657,
|
427 |
+
"learning_rate": 7.3802971089353696e-06,
|
428 |
+
"loss": 0.0055,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.8310249307479224,
|
433 |
+
"grad_norm": 0.39876681566238403,
|
434 |
+
"learning_rate": 7.273186699202572e-06,
|
435 |
+
"loss": 0.0067,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.8448753462603878,
|
440 |
+
"grad_norm": 0.6315239667892456,
|
441 |
+
"learning_rate": 7.164744718661198e-06,
|
442 |
+
"loss": 0.0045,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.8587257617728532,
|
447 |
+
"grad_norm": 0.31891411542892456,
|
448 |
+
"learning_rate": 7.055034689667661e-06,
|
449 |
+
"loss": 0.0042,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.8725761772853186,
|
454 |
+
"grad_norm": 0.34718984365463257,
|
455 |
+
"learning_rate": 6.944120877366605e-06,
|
456 |
+
"loss": 0.0045,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.8864265927977839,
|
461 |
+
"grad_norm": 0.3197481632232666,
|
462 |
+
"learning_rate": 6.832068252046116e-06,
|
463 |
+
"loss": 0.0051,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.9002770083102493,
|
468 |
+
"grad_norm": 0.298637717962265,
|
469 |
+
"learning_rate": 6.718942451079911e-06,
|
470 |
+
"loss": 0.0055,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.9141274238227147,
|
475 |
+
"grad_norm": 0.3373955190181732,
|
476 |
+
"learning_rate": 6.604809740478748e-06,
|
477 |
+
"loss": 0.0045,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.9279778393351801,
|
482 |
+
"grad_norm": 0.5179949402809143,
|
483 |
+
"learning_rate": 6.489736976073603e-06,
|
484 |
+
"loss": 0.0054,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.9418282548476454,
|
489 |
+
"grad_norm": 0.31394144892692566,
|
490 |
+
"learning_rate": 6.3737915643533484e-06,
|
491 |
+
"loss": 0.0049,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.9556786703601108,
|
496 |
+
"grad_norm": 0.25130563974380493,
|
497 |
+
"learning_rate": 6.257041422979871e-06,
|
498 |
+
"loss": 0.0039,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.9695290858725761,
|
503 |
+
"grad_norm": 0.42873042821884155,
|
504 |
+
"learning_rate": 6.139554941003747e-06,
|
505 |
+
"loss": 0.0055,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.9833795013850416,
|
510 |
+
"grad_norm": 0.2586114704608917,
|
511 |
+
"learning_rate": 6.021400938803813e-06,
|
512 |
+
"loss": 0.0042,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.997229916897507,
|
517 |
+
"grad_norm": 0.3645874559879303,
|
518 |
+
"learning_rate": 5.902648627774059e-06,
|
519 |
+
"loss": 0.0058,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 1.0,
|
524 |
+
"eval_loss": 0.0030423467978835106,
|
525 |
+
"eval_runtime": 3.8207,
|
526 |
+
"eval_samples_per_second": 0.785,
|
527 |
+
"eval_steps_per_second": 0.785,
|
528 |
+
"step": 361
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.0110803324099722,
|
532 |
+
"grad_norm": 0.524590790271759,
|
533 |
+
"learning_rate": 5.783367569781474e-06,
|
534 |
+
"loss": 0.0053,
|
535 |
+
"step": 365
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.0249307479224377,
|
539 |
+
"grad_norm": 0.2261294275522232,
|
540 |
+
"learning_rate": 5.663627636418611e-06,
|
541 |
+
"loss": 0.0036,
|
542 |
+
"step": 370
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.0387811634349031,
|
546 |
+
"grad_norm": 0.16315680742263794,
|
547 |
+
"learning_rate": 5.543498968074704e-06,
|
548 |
+
"loss": 0.003,
|
549 |
+
"step": 375
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.0526315789473684,
|
553 |
+
"grad_norm": 0.23269863426685333,
|
554 |
+
"learning_rate": 5.423051932849348e-06,
|
555 |
+
"loss": 0.0027,
|
556 |
+
"step": 380
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.0664819944598338,
|
560 |
+
"grad_norm": 0.29580602049827576,
|
561 |
+
"learning_rate": 5.3023570853327725e-06,
|
562 |
+
"loss": 0.0037,
|
563 |
+
"step": 385
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.080332409972299,
|
567 |
+
"grad_norm": 0.2091536670923233,
|
568 |
+
"learning_rate": 5.181485125276898e-06,
|
569 |
+
"loss": 0.0023,
|
570 |
+
"step": 390
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.0941828254847645,
|
574 |
+
"grad_norm": 0.2704831063747406,
|
575 |
+
"learning_rate": 5.060506856181342e-06,
|
576 |
+
"loss": 0.0035,
|
577 |
+
"step": 395
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.10803324099723,
|
581 |
+
"grad_norm": 0.5069320201873779,
|
582 |
+
"learning_rate": 4.939493143818659e-06,
|
583 |
+
"loss": 0.0028,
|
584 |
+
"step": 400
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.1218836565096952,
|
588 |
+
"grad_norm": 0.7631614804267883,
|
589 |
+
"learning_rate": 4.818514874723103e-06,
|
590 |
+
"loss": 0.0042,
|
591 |
+
"step": 405
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.1357340720221607,
|
595 |
+
"grad_norm": 0.26314568519592285,
|
596 |
+
"learning_rate": 4.697642914667229e-06,
|
597 |
+
"loss": 0.0041,
|
598 |
+
"step": 410
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.149584487534626,
|
602 |
+
"grad_norm": 0.20495057106018066,
|
603 |
+
"learning_rate": 4.576948067150655e-06,
|
604 |
+
"loss": 0.003,
|
605 |
+
"step": 415
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.1634349030470914,
|
609 |
+
"grad_norm": 0.2521879971027374,
|
610 |
+
"learning_rate": 4.456501031925297e-06,
|
611 |
+
"loss": 0.0037,
|
612 |
+
"step": 420
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.1772853185595569,
|
616 |
+
"grad_norm": 0.23261719942092896,
|
617 |
+
"learning_rate": 4.336372363581391e-06,
|
618 |
+
"loss": 0.0027,
|
619 |
+
"step": 425
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.1911357340720221,
|
623 |
+
"grad_norm": 0.19368702173233032,
|
624 |
+
"learning_rate": 4.216632430218528e-06,
|
625 |
+
"loss": 0.0033,
|
626 |
+
"step": 430
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 1.2049861495844876,
|
630 |
+
"grad_norm": 0.5213963389396667,
|
631 |
+
"learning_rate": 4.097351372225943e-06,
|
632 |
+
"loss": 0.0028,
|
633 |
+
"step": 435
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.2188365650969528,
|
637 |
+
"grad_norm": 0.16888286173343658,
|
638 |
+
"learning_rate": 3.978599061196188e-06,
|
639 |
+
"loss": 0.003,
|
640 |
+
"step": 440
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 1.2326869806094183,
|
644 |
+
"grad_norm": 0.3124845027923584,
|
645 |
+
"learning_rate": 3.860445058996255e-06,
|
646 |
+
"loss": 0.0034,
|
647 |
+
"step": 445
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.2465373961218837,
|
651 |
+
"grad_norm": 0.25205326080322266,
|
652 |
+
"learning_rate": 3.7429585770201314e-06,
|
653 |
+
"loss": 0.0028,
|
654 |
+
"step": 450
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.260387811634349,
|
658 |
+
"grad_norm": 0.27026185393333435,
|
659 |
+
"learning_rate": 3.626208435646652e-06,
|
660 |
+
"loss": 0.0023,
|
661 |
+
"step": 455
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 1.2742382271468145,
|
665 |
+
"grad_norm": 0.19147123396396637,
|
666 |
+
"learning_rate": 3.5102630239263986e-06,
|
667 |
+
"loss": 0.0021,
|
668 |
+
"step": 460
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.2880886426592797,
|
672 |
+
"grad_norm": 0.21509826183319092,
|
673 |
+
"learning_rate": 3.395190259521254e-06,
|
674 |
+
"loss": 0.0039,
|
675 |
+
"step": 465
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 1.3019390581717452,
|
679 |
+
"grad_norm": 0.3184572160243988,
|
680 |
+
"learning_rate": 3.281057548920091e-06,
|
681 |
+
"loss": 0.0034,
|
682 |
+
"step": 470
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 1.3157894736842106,
|
686 |
+
"grad_norm": 0.3757064640522003,
|
687 |
+
"learning_rate": 3.1679317479538864e-06,
|
688 |
+
"loss": 0.0036,
|
689 |
+
"step": 475
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.3296398891966759,
|
693 |
+
"grad_norm": 0.23070411384105682,
|
694 |
+
"learning_rate": 3.0558791226333974e-06,
|
695 |
+
"loss": 0.002,
|
696 |
+
"step": 480
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.3434903047091413,
|
700 |
+
"grad_norm": 0.16266323626041412,
|
701 |
+
"learning_rate": 2.9449653103323405e-06,
|
702 |
+
"loss": 0.003,
|
703 |
+
"step": 485
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 1.3573407202216066,
|
707 |
+
"grad_norm": 0.235463485121727,
|
708 |
+
"learning_rate": 2.8352552813388035e-06,
|
709 |
+
"loss": 0.0022,
|
710 |
+
"step": 490
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.371191135734072,
|
714 |
+
"grad_norm": 0.21921995282173157,
|
715 |
+
"learning_rate": 2.7268133007974284e-06,
|
716 |
+
"loss": 0.0025,
|
717 |
+
"step": 495
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 1.3850415512465375,
|
721 |
+
"grad_norm": 0.17426297068595886,
|
722 |
+
"learning_rate": 2.6197028910646304e-06,
|
723 |
+
"loss": 0.0032,
|
724 |
+
"step": 500
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 1.3988919667590027,
|
728 |
+
"grad_norm": 0.14913178980350494,
|
729 |
+
"learning_rate": 2.5139867944989483e-06,
|
730 |
+
"loss": 0.0022,
|
731 |
+
"step": 505
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 1.4127423822714682,
|
735 |
+
"grad_norm": 0.37892332673072815,
|
736 |
+
"learning_rate": 2.409726936708263e-06,
|
737 |
+
"loss": 0.0024,
|
738 |
+
"step": 510
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 1.4265927977839334,
|
742 |
+
"grad_norm": 0.20658189058303833,
|
743 |
+
"learning_rate": 2.3069843902754767e-06,
|
744 |
+
"loss": 0.0034,
|
745 |
+
"step": 515
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 1.440443213296399,
|
749 |
+
"grad_norm": 0.29341772198677063,
|
750 |
+
"learning_rate": 2.205819338983859e-06,
|
751 |
+
"loss": 0.0031,
|
752 |
+
"step": 520
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.4542936288088644,
|
756 |
+
"grad_norm": 0.228977769613266,
|
757 |
+
"learning_rate": 2.106291042563019e-06,
|
758 |
+
"loss": 0.002,
|
759 |
+
"step": 525
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 1.4681440443213296,
|
763 |
+
"grad_norm": 0.2889719009399414,
|
764 |
+
"learning_rate": 2.0084578019761738e-06,
|
765 |
+
"loss": 0.0031,
|
766 |
+
"step": 530
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 1.481994459833795,
|
770 |
+
"grad_norm": 0.16132836043834686,
|
771 |
+
"learning_rate": 1.912376925269041e-06,
|
772 |
+
"loss": 0.0026,
|
773 |
+
"step": 535
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.4958448753462603,
|
777 |
+
"grad_norm": 0.3356161117553711,
|
778 |
+
"learning_rate": 1.8181046940003366e-06,
|
779 |
+
"loss": 0.0028,
|
780 |
+
"step": 540
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 1.5096952908587258,
|
784 |
+
"grad_norm": 0.4451793432235718,
|
785 |
+
"learning_rate": 1.7256963302735752e-06,
|
786 |
+
"loss": 0.0026,
|
787 |
+
"step": 545
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 1.5235457063711912,
|
791 |
+
"grad_norm": 0.30361247062683105,
|
792 |
+
"learning_rate": 1.635205964389474e-06,
|
793 |
+
"loss": 0.0016,
|
794 |
+
"step": 550
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 1.5373961218836565,
|
798 |
+
"grad_norm": 0.20500566065311432,
|
799 |
+
"learning_rate": 1.5466866031378874e-06,
|
800 |
+
"loss": 0.0029,
|
801 |
+
"step": 555
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 1.5512465373961217,
|
805 |
+
"grad_norm": 0.3303813636302948,
|
806 |
+
"learning_rate": 1.4601900987478834e-06,
|
807 |
+
"loss": 0.003,
|
808 |
+
"step": 560
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 1.5650969529085872,
|
812 |
+
"grad_norm": 0.20436729490756989,
|
813 |
+
"learning_rate": 1.3757671185141136e-06,
|
814 |
+
"loss": 0.0017,
|
815 |
+
"step": 565
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.5789473684210527,
|
819 |
+
"grad_norm": 0.33445170521736145,
|
820 |
+
"learning_rate": 1.2934671151172974e-06,
|
821 |
+
"loss": 0.0031,
|
822 |
+
"step": 570
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 1.5927977839335181,
|
826 |
+
"grad_norm": 0.2386503368616104,
|
827 |
+
"learning_rate": 1.213338297656191e-06,
|
828 |
+
"loss": 0.0026,
|
829 |
+
"step": 575
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 1.6066481994459834,
|
833 |
+
"grad_norm": 0.45957309007644653,
|
834 |
+
"learning_rate": 1.1354276034080059e-06,
|
835 |
+
"loss": 0.003,
|
836 |
+
"step": 580
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.6204986149584486,
|
840 |
+
"grad_norm": 0.2124543935060501,
|
841 |
+
"learning_rate": 1.0597806703338354e-06,
|
842 |
+
"loss": 0.0023,
|
843 |
+
"step": 585
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 1.634349030470914,
|
847 |
+
"grad_norm": 0.1889660805463791,
|
848 |
+
"learning_rate": 9.86441810345183e-07,
|
849 |
+
"loss": 0.0026,
|
850 |
+
"step": 590
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 1.6481994459833795,
|
854 |
+
"grad_norm": 0.3355896472930908,
|
855 |
+
"learning_rate": 9.154539833472442e-07,
|
856 |
+
"loss": 0.0023,
|
857 |
+
"step": 595
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 1.662049861495845,
|
861 |
+
"grad_norm": 0.20331183075904846,
|
862 |
+
"learning_rate": 8.468587720741728e-07,
|
863 |
+
"loss": 0.002,
|
864 |
+
"step": 600
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 1.6759002770083102,
|
868 |
+
"grad_norm": 0.5584515929222107,
|
869 |
+
"learning_rate": 7.80696357731049e-07,
|
870 |
+
"loss": 0.0027,
|
871 |
+
"step": 605
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 1.6897506925207755,
|
875 |
+
"grad_norm": 0.21782828867435455,
|
876 |
+
"learning_rate": 7.170054964568146e-07,
|
877 |
+
"loss": 0.0027,
|
878 |
+
"step": 610
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.703601108033241,
|
882 |
+
"grad_norm": 0.24219338595867157,
|
883 |
+
"learning_rate": 6.558234966219984e-07,
|
884 |
+
"loss": 0.0022,
|
885 |
+
"step": 615
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 1.7174515235457064,
|
889 |
+
"grad_norm": 0.32900330424308777,
|
890 |
+
"learning_rate": 5.971861969744758e-07,
|
891 |
+
"loss": 0.0031,
|
892 |
+
"step": 620
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 1.7313019390581719,
|
896 |
+
"grad_norm": 0.20752942562103271,
|
897 |
+
"learning_rate": 5.411279456461133e-07,
|
898 |
+
"loss": 0.0019,
|
899 |
+
"step": 625
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 1.745152354570637,
|
903 |
+
"grad_norm": 0.15013189613819122,
|
904 |
+
"learning_rate": 4.87681580032573e-07,
|
905 |
+
"loss": 0.0018,
|
906 |
+
"step": 630
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 1.7590027700831024,
|
910 |
+
"grad_norm": 0.2259117215871811,
|
911 |
+
"learning_rate": 4.368784075580512e-07,
|
912 |
+
"loss": 0.0028,
|
913 |
+
"step": 635
|
914 |
+
},
|
915 |
+
{
|
916 |
+
"epoch": 1.7728531855955678,
|
917 |
+
"grad_norm": 0.16545476019382477,
|
918 |
+
"learning_rate": 3.8874818733625363e-07,
|
919 |
+
"loss": 0.0024,
|
920 |
+
"step": 640
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 1.7867036011080333,
|
924 |
+
"grad_norm": 0.2805814743041992,
|
925 |
+
"learning_rate": 3.433191127383079e-07,
|
926 |
+
"loss": 0.0025,
|
927 |
+
"step": 645
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 1.8005540166204987,
|
931 |
+
"grad_norm": 0.24149565398693085,
|
932 |
+
"learning_rate": 3.0061779487786325e-07,
|
933 |
+
"loss": 0.0026,
|
934 |
+
"step": 650
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 1.814404432132964,
|
938 |
+
"grad_norm": 0.37586623430252075,
|
939 |
+
"learning_rate": 2.6066924702302044e-07,
|
940 |
+
"loss": 0.0027,
|
941 |
+
"step": 655
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 1.8282548476454292,
|
945 |
+
"grad_norm": 0.28574487566947937,
|
946 |
+
"learning_rate": 2.2349686994424303e-07,
|
947 |
+
"loss": 0.0022,
|
948 |
+
"step": 660
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 1.8421052631578947,
|
952 |
+
"grad_norm": 0.26507818698883057,
|
953 |
+
"learning_rate": 1.8912243820682296e-07,
|
954 |
+
"loss": 0.0025,
|
955 |
+
"step": 665
|
956 |
+
},
|
957 |
+
{
|
958 |
+
"epoch": 1.8559556786703602,
|
959 |
+
"grad_norm": 0.29346975684165955,
|
960 |
+
"learning_rate": 1.575660874159346e-07,
|
961 |
+
"loss": 0.0024,
|
962 |
+
"step": 670
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 1.8698060941828256,
|
966 |
+
"grad_norm": 0.37801891565322876,
|
967 |
+
"learning_rate": 1.2884630242174734e-07,
|
968 |
+
"loss": 0.003,
|
969 |
+
"step": 675
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 1.8836565096952909,
|
973 |
+
"grad_norm": 0.11035842448472977,
|
974 |
+
"learning_rate": 1.029799064914988e-07,
|
975 |
+
"loss": 0.0019,
|
976 |
+
"step": 680
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 1.897506925207756,
|
980 |
+
"grad_norm": 0.2686106860637665,
|
981 |
+
"learning_rate": 7.998205145489157e-08,
|
982 |
+
"loss": 0.0026,
|
983 |
+
"step": 685
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 1.9113573407202216,
|
987 |
+
"grad_norm": 0.3440103232860565,
|
988 |
+
"learning_rate": 5.986620882855676e-08,
|
989 |
+
"loss": 0.0031,
|
990 |
+
"step": 690
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 1.925207756232687,
|
994 |
+
"grad_norm": 0.2664523124694824,
|
995 |
+
"learning_rate": 4.2644161924811353e-08,
|
996 |
+
"loss": 0.0035,
|
997 |
+
"step": 695
|
998 |
+
},
|
999 |
+
{
|
1000 |
+
"epoch": 1.9390581717451525,
|
1001 |
+
"grad_norm": 0.22529973089694977,
|
1002 |
+
"learning_rate": 2.8325998949314536e-08,
|
1003 |
+
"loss": 0.0022,
|
1004 |
+
"step": 700
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 1.9529085872576177,
|
1008 |
+
"grad_norm": 0.167397141456604,
|
1009 |
+
"learning_rate": 1.6920107091668582e-08,
|
1010 |
+
"loss": 0.0021,
|
1011 |
+
"step": 705
|
1012 |
+
},
|
1013 |
+
{
|
1014 |
+
"epoch": 1.966759002770083,
|
1015 |
+
"grad_norm": 0.3647945523262024,
|
1016 |
+
"learning_rate": 8.433167612436066e-09,
|
1017 |
+
"loss": 0.0021,
|
1018 |
+
"step": 710
|
1019 |
+
},
|
1020 |
+
{
|
1021 |
+
"epoch": 1.9806094182825484,
|
1022 |
+
"grad_norm": 0.3356544077396393,
|
1023 |
+
"learning_rate": 2.8701519294371815e-09,
|
1024 |
+
"loss": 0.0034,
|
1025 |
+
"step": 715
|
1026 |
+
},
|
1027 |
+
{
|
1028 |
+
"epoch": 1.994459833795014,
|
1029 |
+
"grad_norm": 0.1703202724456787,
|
1030 |
+
"learning_rate": 2.3431870562917735e-10,
|
1031 |
+
"loss": 0.0019,
|
1032 |
+
"step": 720
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 2.0,
|
1036 |
+
"eval_loss": 0.0028210037853568792,
|
1037 |
+
"eval_runtime": 3.8079,
|
1038 |
+
"eval_samples_per_second": 0.788,
|
1039 |
+
"eval_steps_per_second": 0.788,
|
1040 |
+
"step": 722
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 2.0,
|
1044 |
+
"step": 722,
|
1045 |
+
"total_flos": 1.3020461702683034e+17,
|
1046 |
+
"train_loss": 0.037445212856265314,
|
1047 |
+
"train_runtime": 3051.2562,
|
1048 |
+
"train_samples_per_second": 0.237,
|
1049 |
+
"train_steps_per_second": 0.237
|
1050 |
+
}
|
1051 |
+
],
|
1052 |
+
"logging_steps": 5,
|
1053 |
+
"max_steps": 722,
|
1054 |
+
"num_input_tokens_seen": 0,
|
1055 |
+
"num_train_epochs": 2,
|
1056 |
+
"save_steps": 100,
|
1057 |
+
"stateful_callbacks": {
|
1058 |
+
"TrainerControl": {
|
1059 |
+
"args": {
|
1060 |
+
"should_epoch_stop": false,
|
1061 |
+
"should_evaluate": false,
|
1062 |
+
"should_log": false,
|
1063 |
+
"should_save": true,
|
1064 |
+
"should_training_stop": true
|
1065 |
+
},
|
1066 |
+
"attributes": {}
|
1067 |
+
}
|
1068 |
+
},
|
1069 |
+
"total_flos": 1.3020461702683034e+17,
|
1070 |
+
"train_batch_size": 1,
|
1071 |
+
"trial_name": null,
|
1072 |
+
"trial_params": null
|
1073 |
+
}
|