File size: 1,231 Bytes
b221791 |
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 |
---
language:
- en
base_model: Qwen/Qwen3-0.6B-Base
tags:
- qwen3
- fine-tuned
- narasimha
- conversational
license: apache-2.0
---
# Qwen3-0.6B Fine-tuned on Narasimha Dataset
This model is a fine-tuned version of [Qwen/Qwen3-0.6B-Base](https://huggingface.co/Qwen/Qwen3-0.6B-Base) on the Narasimha dataset.
## Training Details
- Base model: Qwen/Qwen3-0.6B-Base
- Dataset: sarthakrastogi/narasimha_dataset (500 samples)
- Training epochs: 1
- Batch size: 2
- Data type: bf16
## Usage
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "sarthakrastogi/narasimha-b-0.6b"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")
# Generate response
content = "your question here"
messages = [{"role": "user", "content": content}]
prompt_text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
inputs = tokenizer(prompt_text, return_tensors="pt").to(model.device)
output_ids = model.generate(**inputs, max_new_tokens=100)
response = tokenizer.decode(output_ids[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
print(response)
```
|