|
--- |
|
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 |
