Downloads · 30 days
0
tony-shieh/Qwen2.5-14B-LoRA-NYCU-DL-HW2
Qwen2.5-14B-LoRA-NYCU-DL-HW2 is a multiple choice model from tony-shieh. Use it for the multiple choice task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This model is a Supervised Task Finetuned (SFT) version of Qwen2.5-14B-Instruct, specifically trained with reasoning information (Chain-of-Thought) to tackle complex multiple-choice questions.
Downloads · 30 days
0
Access
Public
Updated May 24, 2026
Repo size
562 MB
Likes
0
Public
Click a slice to open those files.
.safetensors551 MB · 98%
From the Hugging Face model README
This model is a Supervised Task Finetuned (SFT) version of Qwen2.5-14B-Instruct, specifically trained with reasoning information (Chain-of-Thought) to tackle complex multiple-choice questions.
It was developed as part of the Deep Learning HW2 coursework at National Yang Ming Chiao Tung University (NYCU).
unsloth/Qwen2.5-14B-Instruct-bnb-4bit0.72340To achieve the maximum performance, it is highly recommended to use the Logits Extraction method rather than standard text generation. Below is a basic snippet to load the model using unsloth:
from unsloth import FastLanguageModel
import torch
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = "tony-shieh/Qwen2.5-14B-LoRA-NYCU-DL-HW2",
max_seq_length = 4096,
dtype = None,
load_in_4bit = True
)
FastLanguageModel.for_inference(model)
SYSTEM_PROMPT = "你是一個專業且中立的選擇題解題專家。請針對題目進行嚴密的邏輯推理,客觀分析每個選項,最後明確給出你的答案。"
LoRA Rank (r): 32
LoRA Alpha: 64
Target Modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Epochs: 2 (To prevent catastrophic forgetting and overfitting on the public LB)
Learning Rate: 1e-4
Optimizer: adamw_8bit
The true potential of this model is unlocked during the inference phase. Instead of relying on vulnerable Regex text parsing, the final submission utilizes:
Dynamic Chain-of-Thought (do_sample=True, temp=0.6): Forcing the model to explore 3 distinct reasoning paths for the same question.
Logits Soft-Voting: Extracting the raw neural network logits for tokens A, B, C, and D at the final layer, normalizing them via Softmax into confidence probabilities, and accumulating these scores across the 3 iterations. This effectively mitigates the 14B model's "hallucination" and locks in the most logically sound answer.