Model save
Browse files
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
tags:
|
4 |
+
- trl
|
5 |
+
- dpo
|
6 |
+
- DPO
|
7 |
+
- WeniGPT
|
8 |
+
- generated_from_trainer
|
9 |
+
base_model: Weni/WeniGPT-Agents-Mistral-1.0.0-SFT-merged
|
10 |
+
model-index:
|
11 |
+
- name: WeniGPT-Agents-Mistral-1.0.0-SFT-1.0.25-DPO
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# WeniGPT-Agents-Mistral-1.0.0-SFT-1.0.25-DPO
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [Weni/WeniGPT-Agents-Mistral-1.0.0-SFT-merged](https://huggingface.co/Weni/WeniGPT-Agents-Mistral-1.0.0-SFT-merged) on an unknown dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.0048
|
23 |
+
- Rewards/chosen: 3.9169
|
24 |
+
- Rewards/rejected: -6.2067
|
25 |
+
- Rewards/accuracies: 1.0
|
26 |
+
- Rewards/margins: 10.1235
|
27 |
+
- Logps/rejected: -240.9936
|
28 |
+
- Logps/chosen: -117.4673
|
29 |
+
- Logits/rejected: -1.9027
|
30 |
+
- Logits/chosen: -1.8786
|
31 |
+
|
32 |
+
## Model description
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Intended uses & limitations
|
37 |
+
|
38 |
+
More information needed
|
39 |
+
|
40 |
+
## Training and evaluation data
|
41 |
+
|
42 |
+
More information needed
|
43 |
+
|
44 |
+
## Training procedure
|
45 |
+
|
46 |
+
### Training hyperparameters
|
47 |
+
|
48 |
+
The following hyperparameters were used during training:
|
49 |
+
- learning_rate: 5e-06
|
50 |
+
- train_batch_size: 1
|
51 |
+
- eval_batch_size: 1
|
52 |
+
- seed: 42
|
53 |
+
- distributed_type: multi-GPU
|
54 |
+
- num_devices: 4
|
55 |
+
- gradient_accumulation_steps: 2
|
56 |
+
- total_train_batch_size: 8
|
57 |
+
- total_eval_batch_size: 4
|
58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
59 |
+
- lr_scheduler_type: linear
|
60 |
+
- lr_scheduler_warmup_ratio: 0.03
|
61 |
+
- training_steps: 180
|
62 |
+
- mixed_precision_training: Native AMP
|
63 |
+
|
64 |
+
### Training results
|
65 |
+
|
66 |
+
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|
67 |
+
|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
|
68 |
+
| 0.225 | 0.9677 | 30 | 0.2227 | 2.6681 | -1.0085 | 1.0 | 3.6765 | -215.0026 | -123.7113 | -1.9566 | -1.9326 |
|
69 |
+
| 0.0853 | 1.9355 | 60 | 0.1095 | 3.7286 | -2.2208 | 1.0 | 5.9494 | -221.0642 | -118.4086 | -1.9400 | -1.9170 |
|
70 |
+
| 0.0285 | 2.9032 | 90 | 0.0545 | 4.1460 | -3.9811 | 1.0 | 8.1270 | -229.8655 | -116.3218 | -1.9301 | -1.9063 |
|
71 |
+
| 0.001 | 3.8710 | 120 | 0.0468 | 4.1806 | -5.0175 | 1.0 | 9.1980 | -235.0477 | -116.1489 | -1.9141 | -1.8902 |
|
72 |
+
| 0.0021 | 4.8387 | 150 | 0.0087 | 3.9958 | -5.9294 | 1.0 | 9.9252 | -239.6072 | -117.0728 | -1.9056 | -1.8815 |
|
73 |
+
| 0.0014 | 5.8065 | 180 | 0.0048 | 3.9169 | -6.2067 | 1.0 | 10.1235 | -240.9936 | -117.4673 | -1.9027 | -1.8786 |
|
74 |
+
|
75 |
+
|
76 |
+
### Framework versions
|
77 |
+
|
78 |
+
- PEFT 0.10.0
|
79 |
+
- Transformers 4.40.0
|
80 |
+
- Pytorch 2.1.0+cu118
|
81 |
+
- Datasets 2.18.0
|
82 |
+
- Tokenizers 0.19.1
|