58 lines
1.4 KiB
Markdown
58 lines
1.4 KiB
Markdown
|
---
|
||
|
|
base_model: Qwen/Qwen3-0.6B
|
||
|
|
library_name: transformers
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
- fr
|
||
|
|
tags:
|
||
|
|
- granite
|
||
|
|
- fine-tuned
|
||
|
|
- conversational
|
||
|
|
- distillation
|
||
|
|
- thinking
|
||
|
|
- reasoning
|
||
|
|
datasets:
|
||
|
|
- TeichAI/glm-4.7-2000x
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
---
|
||
|
|
|
||
|
|
# Qwen3-0.6B-DISTILL-glm-4.7-think
|
||
|
|
|
||
|
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) trained on high-reasoning conversational data from GLM 4.7 by Z.ai.
|
||
|
|
|
||
|
|
## Model Details
|
||
|
|
|
||
|
|
- **Base Model:** Qwen/Qwen3-0.6B
|
||
|
|
- **Fine-tuning Dataset:** TeichAI/glm-4.7-2000x
|
||
|
|
- **Context Length:** 1048576 tokens
|
||
|
|
- **Special Feature:** Thinking/Reasoning with `<think>` tags
|
||
|
|
|
||
|
|
## Usage
|
||
|
|
|
||
|
|
### Transformers
|
||
|
|
|
||
|
|
```python
|
||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
|
|
||
|
|
model = AutoModelForCausalLM.from_pretrained("glogwa68/Qwen3-0.6B-DISTILL-glm-4.7-think")
|
||
|
|
tokenizer = AutoTokenizer.from_pretrained("glogwa68/Qwen3-0.6B-DISTILL-glm-4.7-think")
|
||
|
|
|
||
|
|
messages = [{"role": "user", "content": "Hello, how are you?"}]
|
||
|
|
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
|
||
|
|
outputs = model.generate(inputs, max_new_tokens=256)
|
||
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
||
|
|
```
|
||
|
|
|
||
|
|
## Training Details
|
||
|
|
|
||
|
|
- **Epochs:** 2
|
||
|
|
- **Learning Rate:** 2e-5
|
||
|
|
- **Batch Size:** 8 (with gradient accumulation)
|
||
|
|
- **Precision:** FP16
|
||
|
|
- **Hardware:** Multi-GPU with DeepSpeed ZeRO-3
|
||
|
|
|
||
|
|
## License
|
||
|
|
|
||
|
|
Apache 2.0
|