---
library_name: peft
base_model: Qwen/Qwen3-32B
license: mit
---

# qwen-3-32b-philosophy-spec-aft-no-cot-10k

A LoRA adapter for [Qwen/Qwen3-32B](https://huggingface.co/Qwen/Qwen3-32B), trained using alignment fine-tuning (AFT) only, without chain-of-thought. Trained on 10k AFT examples.

- **Base model:** Qwen/Qwen3-32B
- **LoRA rank:** 64
- **LoRA alpha:** 128
- **Target modules:** q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
- **AFT dataset size:** 10k

## Usage

### Load as LoRA adapter

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-32B",
    torch_dtype="auto",
    device_map="auto",
)
model = PeftModel.from_pretrained(base_model, "chloeli/qwen-3-32b-philosophy-spec-aft-no-cot-10k")
tokenizer = AutoTokenizer.from_pretrained("chloeli/qwen-3-32b-philosophy-spec-aft-no-cot-10k")

messages = [{"role": "user", "content": "What matters most when making a difficult decision?"}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

### Merge into base model

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen3-32B",
    torch_dtype="auto",
    device_map="cpu",
)
model = PeftModel.from_pretrained(base_model, "chloeli/qwen-3-32b-philosophy-spec-aft-no-cot-10k")
merged_model = model.merge_and_unload()

merged_model.save_pretrained("qwen-3-32b-philosophy-spec-aft-no-cot-10k-merged")
tokenizer = AutoTokenizer.from_pretrained("chloeli/qwen-3-32b-philosophy-spec-aft-no-cot-10k")
tokenizer.save_pretrained("qwen-3-32b-philosophy-spec-aft-no-cot-10k-merged")
```

### Serve with vLLM

```python
from vllm import LLM, SamplingParams
from vllm.lora.request import LoRARequest

llm = LLM(
    model="Qwen/Qwen3-32B",
    enable_lora=True,
    max_lora_rank=128,
)

lora_request = LoRARequest("philosophy", 1, "chloeli/qwen-3-32b-philosophy-spec-aft-no-cot-10k")
output = llm.generate("What matters most?", SamplingParams(max_tokens=512), lora_request=lora_request)
```