File size: 2,118 Bytes
750c1de
 
 
 
 
0d1c42a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
tags:
- unsloth
- kyrgyz
---

---
base_model: unsloth/mistral-7b-bnb-4bit
library_name: peft
---
# Model Card for Mistral 7B Fine-tuned by Erkinbek Niiazbek uulu

## Model Details

### Model Description
This is a fine-tuned version of the Mistral 7B model developed by Erkinbek Niiazbek uulu for specific use cases. The model was fine-tuned using LoRA (Low-Rank Adaptation) techniques and is optimized for lightweight deployment. The base model used is `unsloth/mistral-7b-bnb-4bit`.

- **Developed by:** Erkinbek Niiazbek uulu
- **Contact Email:** [email protected]
- **Base Model:** unsloth/mistral-7b-bnb-4bit
- **Library Name:** PEFT
- **Language(s):** Multilingual (including Kyrgyz)
- **License:** [Specify your license type, e.g., Apache 2.0, MIT]
- **Fine-tuned from model:** unsloth/mistral-7b-bnb-4bit

---

## Uses

### Direct Use
This fine-tuned model is designed for tasks such as:
- Multilingual question answering
- Text summarization
- Natural language generation

### Downstream Use
This model can be further fine-tuned for domain-specific applications.

### Out-of-Scope Use
This model is not intended for generating harmful, offensive, or unethical content.

---

## Bias, Risks, and Limitations

### Recommendations
While this model has been fine-tuned for specific tasks, users should be cautious of potential biases in the output. It is recommended to review the outputs critically, especially when used in sensitive applications.

---

## How to Get Started with the Model

To load the model, you can use the following code:

```python
# alpaca_prompt = Copied from above
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer(
[
    alpaca_prompt.format(
        "Зекет деген эмне?", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 1024)
### Framework versions

- PEFT 0.14.0