Model save
Browse files- README.md +88 -0
- adapter_model.safetensors +1 -1
README.md
ADDED
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
9 |
+
datasets:
|
10 |
+
- generator
|
11 |
+
model-index:
|
12 |
+
- name: WeniGPT-Agents-Llama3-1.0.8-SFT
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# WeniGPT-Agents-Llama3-1.0.8-SFT
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 1.3744
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0002
|
43 |
+
- train_batch_size: 1
|
44 |
+
- eval_batch_size: 1
|
45 |
+
- seed: 42
|
46 |
+
- gradient_accumulation_steps: 2
|
47 |
+
- total_train_batch_size: 2
|
48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
+
- lr_scheduler_type: linear
|
50 |
+
- lr_scheduler_warmup_ratio: 0.03
|
51 |
+
- training_steps: 669
|
52 |
+
- mixed_precision_training: Native AMP
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
58 |
+
| 1.7877 | 0.1342 | 30 | 1.7082 |
|
59 |
+
| 1.3675 | 0.2685 | 60 | 1.4539 |
|
60 |
+
| 1.3782 | 0.4027 | 90 | 1.4236 |
|
61 |
+
| 1.3884 | 0.5369 | 120 | 1.3938 |
|
62 |
+
| 1.3448 | 0.6711 | 150 | 1.3899 |
|
63 |
+
| 1.3357 | 0.8054 | 180 | 1.3870 |
|
64 |
+
| 1.2788 | 0.9396 | 210 | 1.3739 |
|
65 |
+
| 1.2396 | 1.0738 | 240 | 1.3778 |
|
66 |
+
| 1.2949 | 1.2081 | 270 | 1.3809 |
|
67 |
+
| 1.337 | 1.3423 | 300 | 1.3792 |
|
68 |
+
| 1.3266 | 1.4765 | 330 | 1.3775 |
|
69 |
+
| 1.2735 | 1.6107 | 360 | 1.3744 |
|
70 |
+
| 1.2809 | 1.7450 | 390 | 1.3752 |
|
71 |
+
| 1.2383 | 1.8792 | 420 | 1.3775 |
|
72 |
+
| 1.2116 | 2.0134 | 450 | 1.3859 |
|
73 |
+
| 1.0153 | 2.1477 | 480 | 1.3926 |
|
74 |
+
| 1.2039 | 2.2819 | 510 | 1.3884 |
|
75 |
+
| 1.2451 | 2.4161 | 540 | 1.3886 |
|
76 |
+
| 1.2311 | 2.5503 | 570 | 1.3921 |
|
77 |
+
| 1.1299 | 2.6846 | 600 | 1.3941 |
|
78 |
+
| 1.2163 | 2.8188 | 630 | 1.3913 |
|
79 |
+
| 1.0719 | 2.9530 | 660 | 1.3907 |
|
80 |
+
|
81 |
+
|
82 |
+
### Framework versions
|
83 |
+
|
84 |
+
- PEFT 0.10.0
|
85 |
+
- Transformers 4.40.0
|
86 |
+
- Pytorch 2.1.0+cu118
|
87 |
+
- Datasets 2.18.0
|
88 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4216374752
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7f98dd53088e84f2144759135274fc96311fd4e2a2d05f15372610d8ed93e612
|
3 |
size 4216374752
|