qwen3-14b-stem-qa-merged

Qwen3-14B STEM Q&A (Merged)

This is the merged full model (LoRA weights merged into base model) for standalone use.

Author: Khadim Hussain · Project write-up: STEM reasoning LLMs

For the LoRA adapter version (smaller download), see: khadim-hussain/qwen3-14b-stem-qa

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "khadim-hussain/qwen3-14b-stem-qa-merged",
    torch_dtype="auto",
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("khadim-hussain/qwen3-14b-stem-qa-merged")

prompt = "<|im_start|>user\nWhat is DNA?<|im_end|>\n<|im_start|>assistant\n"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Model Details

• Base Model: Qwen/Qwen3-14B
• Type: Merged (LoRA merged into base weights)
• Size: ~28GB (bf16)
• Training: QLoRA fine-tuning on STEM Q&A dataset

Training Metrics

Acknowledgments

• Qwen Team (Alibaba) - Qwen3-14B base model
• Unsloth - Fast fine-tuning framework
• Hugging Face - TRL, PEFT, Transformers

License

Apache 2.0 (inherited from Qwen3)