File size: 1,337 Bytes
62014d8
 
 
5b4218c
 
 
92a2d21
 
9720554
2b91f37
 
 
 
9720554
1dc50d4
922a8ad
b383eab
922a8ad
84154b8
bd7d3d6
53100c1
 
 
 
 
 
 
 
 
922a8ad
b383eab
bd7d3d6
525b86b
 
 
 
2924a52
525b86b
 
 
 
 
 
 
 
 
922a8ad
b383eab
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
---
license: apache-2.0
base_model: MediaTek-Research/Breeze-7B-Instruct-v1_0
datasets:
- Riyuechang/PTT-Corpus-100K_Gossiping-1400-39400
pipeline_tag: text-generation
---

# 簡介
本模型是基於[MediaTek-Research/Breeze-7B-Instruct-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v1_0)微調後的產物  
模型使用來自[PTT](https://www.ptt.cc/bbs/index.html)網站中的[Gossiping](https://www.ptt.cc/bbs/Gossiping/index.html)分類的資料訓練  
過程中使用了一些方法從海量的數據中,過濾出噪聲較小(理論上)的部份作為訓練數據  
訓練資料: [Riyuechang/PTT-Corpus-100K_Gossiping-1400-39400](https://huggingface.co/datasets/Riyuechang/PTT-Corpus-100K_Gossiping-1400-39400)  
  
# 設備  
- Ubuntu 22.04.4 LTS  
- NVIDIA GeForce RTX 3060 12G

# Lora參數
```python
r=8,
lora_alpha=32,
lora_dropout=0.1,
task_type="CAUSAL_LM",
target_modules="all-linear",
bias="none",
use_dora=True,
use_rslora=True
```

# 訓練參數
```python
per_device_train_batch_size=28,  
gradient_accumulation_steps=1,  
num_train_epochs=3,  
warmup_ratio=0.1,  
learning_rate=2e-5,  
bf16=True,  
save_strategy="steps",  
save_steps=500,  
save_total_limit=10,  
logging_steps=10,  
output_dir=log_output,  
optim="paged_adamw_8bit",  
gradient_checkpointing=True
```

# 結果
- loss: 1.1035