Modelpublic

tussiiiii/llmcmp-distill-llama3-8b-lora-v5c-target-hard-tie-response-focus-merged

sourceHugging Facemitupdated 4mo agoView on Hugging Face

0likes4downloads

Model Card

llmcmp-distill-llama3-8b-lora-v5c-target-hard-tie-response-focus-merged

Overview

Merged student model for the Kaggle LLM Classification Finetuning task.

Base Model

—unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit

Training Data

—Distilled source: safe_plus_filtered_plus_hard_plus_raw_partial
—Raw Kaggle train mix mode: partial
—Raw train rows are limited to non-distilled ids only
—Teacher hard mix: True

Training Format

—Prompt contains Prompt / Response A / Response B
—Teacher rationale is not injected into prompt-side input
—Completion is winner-only: A / B / C
—C means tie

Related Adapter

—Adapter repo: tussiiiii/llmcmp-distill-llama3-8b-lora-v5c-target-hard-tie-response-focus-adapter

Notes

—This model is intended for direct next-token winner inference in evaluation / submission.