File size: 7,914 Bytes
fbff720
 
 
 
 
7ba7258
9d8b0b2
 
 
 
 
6f37cc0
9d8b0b2
6f37cc0
9d8b0b2
 
 
 
 
 
6f37cc0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2f014c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
01d951a
 
 
 
 
 
2f014c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6f37cc0
 
 
 
 
 
198a1ba
6f37cc0
 
2f014c7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9d8b0b2
 
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
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
---
license: llama3
language:
- tr
pipeline_tag: text-generation
base_model: meta-llama/Meta-Llama-3-8B-Instruct

model-index:
- name: LLaMA-3-8B-Instruct-TR-DPO
  results:
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: MMLU_TR_V0.2
    metrics:
    - name: 5-shot
      type: 5-shot
      value: 0.4983
      verified: false
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: Truthful_QA_V0.2
    metrics:
    - name: 0-shot
      type: 0-shot
      value: 0.5232
      verified: false
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: ARC_TR_V0.2
    metrics:
    - name: 25-shot
      type: 25-shot
      value: 0.4437
      verified: false
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: HellaSwag_TR_V0.2
    metrics:
    - name: 10-shot
      type: 10-shot
      value: 0.4558
      verified: false
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: GSM8K_TR_V0.2
    metrics:
    - name: 5-shot
      type: 5-shot
      value: 0.5421
      verified: false
  - task:
      type: multiple-choice
    dataset:
      type: multiple-choice
      name: Winogrande_TR_V0.2
    metrics:
    - name: 5-shot
      type: 5-shot
      value: 0.5506
      verified: false
---

<img src="https://huggingface.co/Metin/LLaMA-3-8B-Instruct-TR-DPO/resolve/main/llama.png"
alt="A Llama reading a book" width="420"/>

# LLaMA-3-8B-Instruct-TR-DPO

LLaMA-3-8B-Instruct-TR-DPO is a finetuned version of [Meta-LLaMA-3-8B-Instruct](https://huggingface.co/meta-llama/meta-llama-3-8b-instruct), which has been trained on a preference dataset generated synthetically.

## Training Info

- **Base Model**: [Meta-LLaMA-3-8B-Instruct](https://huggingface.co/meta-llama/meta-llama-3-8b-instruct)
- **Training Data**: A synthetically generated preference dataset consisting of 10K samples was used. No proprietary data was utilized.
- **Training Time**: 3 hours on a single RTX 6000 Ada

- **QLoRA Configs**:
  - lora_r: 64
  - lora_alpha: 32
  - lora_dropout: 0.05
  - lora_target_linear: true

The aim was to finetune the model to enhance the output format and content quality for the Turkish language. It is not necessarily smarter than the base model, but its outputs are more likable and preferable.

Compared to the base model, LLaMA-3-8B-Instruct-TR-DPO is more fluent and coherent in Turkish. It can generate more informative and detailed answers for a given instruction.

It should be noted that the model will still generate incorrect or nonsensical outputs, so please verify the outputs before using them.

## How to use

You can use the below code snippet to use the model:

```python
from transformers import BitsAndBytesConfig
import transformers
import torch

bnb_config = BitsAndBytesConfig(
            load_in_4bit=True,
            bnb_4bit_use_double_quant=True,
            bnb_4bit_quant_type="nf4",
            bnb_4bit_compute_dtype=torch.bfloat16
)

model_id = "Metin/LLaMA-3-8B-Instruct-TR-DPO"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16 ,'quantization_config': bnb_config},
    device_map="auto",
)

messages = [
    {"role": "system", "content": "You are a helpful assistant. Always reply in Turkish."},
    {"role": "user", "content": "Python'da bir öğenin bir listede geçip geçmediğini nasıl kontrol edebilirim?"},
]

prompt = pipeline.tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True
)

terminators = [
    pipeline.tokenizer.eos_token_id,
    pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = pipeline(
    prompt,
    max_new_tokens=512,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.2,
    top_p=0.9,
)

print(outputs[0]["generated_text"][len(prompt):])
```

## OpenLLMTurkishLeaderboard_v0.2 benchmark results

- **MMLU_TR_V0.2**: 49.83%
- **Truthful_QA_TR_V0.2**: 52.32%
- **ARC_TR_V0.2**: 44.37%
- **HellaSwag_TR_V0.2**: 45.58%
- **GSM8K_TR_V0.2**: 54.21%
- **Winogrande_TR_V0.2**: 55.06%
- **Average**: 50.22%

These scores may differ from what you will get when you run the same benchmarks, as I did not use any inference engine (vLLM, TensorRT-LLM, etc.)

## Output Example (DPO Model vs Base Model)

Testing the model with a single example is not an accurate method. However, an example is provided here to showcase the model's capabilities.

### Model: LLaMA-3-8B-Instruct-TR-DPO   

#### Input

```plaintext
Yeni bir derin öğrenme modelinin tanıtımı için bir etkinlik planı hazırla.
```

#### Output

```plaintext
Harika bir fikir!

Etkinlik Planı: "Deep Learning Modeli Tanıtımı"

**Adım 1: Ön-Başlangıç**

* Konuşma konusu: "Deep Learning Modeli Tanıtımı"
* Katılımcı profili: Bilimsel ve teknoloji odaklı profesyoneller, öğrenciler ve araştırmacılar
* Etkinlik süresi: 2 saat

**Adım 2: Giriş**

* Sunum: "Deep Learning Modeli Tanıtımı: Tarihçesi ve Temelleri"
* Sunum içeriği: Deep learning modelinin tarihi, temel kavramları (örneğin, ANN, CNN, RNN), uygulamaları ve avantajları
* Sunum süresi: 30 dakika

**Adım 3: Modelin Çalışma Örnekleri**

* Sunum: "Deep Learning Modelinin Çalışma Örnekleri"
* Sunum içeriği: Modelin uygulamaları (örneğin, görüntü sınıflandırma, doğal dil işleme, makine öğrenimi)
* Sunum süresi: 45 dakika

**Adım 4: Katılımcı Soruları ve Tartışma**

* Katılımcı sorularına cevap verme
* Tartışma: Modelin avantajları ve dezavantajları, uygulamaları ve gelecekteki gelişmeleri

**Adım 5: Sonuç**

* Sunum: "Deep Learning Modelinin Sonuçları"
* Sunum içeriği: Modelin sonuçları, uygulamaları ve gelecekteki gelişmeleri
* Sunum süresi: 15 dakika

**Etkinlik Notları**

* Sunumlar ve tartışmalar için gerekli malzemeleri hazırlayın
* Katılımcıların sorularına cevap verme için yeterli zaman ayırın
* Etkinlik sonrası katılımcıların geri bildirimlerini alın

Umarım bu etkinlik planı sizin için yardımcı olur!
```

### Model: LLaMA-3-8B-Instruct

#### Input

```plaintext
Harika bir fikir! Aşağıdaki etkinlik planını hazırladım:

**Etkinlik Adı:** "Derin Öğrenme Modelinin Tanıtımı"

**Etkinlik Amacı:** Derin öğrenme modelinin tanıtımı ve uygulamaları hakkında bilgi sahibi olmak.

**Etkinlik Planı:**

1. **Giriş:** Derin öğrenme modelinin tanıtımı ve önemi hakkında kısa bir konuşma.
2. **Teorik Bölüm:** Derin öğrenme modelinin teorik temelleri ve çalışma prensipleri hakkında sunum.
3. **Uygulama Bölüm:** Derin öğrenme modelinin uygulamaları ve başarıları hakkında sunum.
4. **Pratik Bölüm:** Derin öğrenme modelinin pratik uygulamaları ve deneyimi hakkında sunum.
5. **Soru-Cevap Bölüm:** Derin öğrenme modelinin sorularına cevap verme.
6. **Sonuç Bölüm:** Derin öğrenme modelinin tanıtımı ve uygulamaları hakkında sonuç ve tavsiye.

**Etkinlik Malzemeleri:**

* Sunum araçları (slaj, powerpoint, vs.)
* Derin öğrenme modelinin uygulamaları hakkında örnekler ve grafikler
* Derin öğrenme modelinin teorik temelleri hakkında kitap ve makaleler
* Derin öğrenme modelinin pratik uygulamaları hakkında deneyimi hakkında sunum

**Etkinlik Süresi:** 2 saat

**Katılımcı Sayısı:** 20-30 kişi

**Etkinlik Yeri:** Bilimsel toplantı salonu veya online platform

**Etkinlik Tarih ve Saati:** 10:00-12:00

**Etkinlik Dili:** Türkçe

**Etkinlik Katılımcıları:** Bilimsel toplantı salonunda veya online platformda katılımcılar.

**Etkinlik Sonuçları:** Derin öğrenme modelinin tanıtımı ve uygulamaları hakkında bilgi sahibi olmak.

**Etkinlik Tavsiyesi:** Derin öğrenme modelinin tanıtımı ve uygulamaları hakkında bilgi sahibi olmak.
```