reimu996

OneReason-0.8B-Frontier-SFT-Epoch1

README

OneReason-0.8B Frontier FeedbackCore SFT LoRA

One epoch local LoRA SFT on frontier_feedbackcore_listwise_invariant_v1.

The repository contains the LoRA adapter and its configuration. Load it together with the clean OneReason-0.8B competition base model; it is not a full model.

Training data: 63,700 records from the frontier feedbackcore candidate.

Configuration: LoRA rank 32, alpha 32, dropout 0.05, BF16, learning rate 2e-4, gradient accumulation 8, requested cutoff length 16,384, and one epoch.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Model Details

Model Provider

reimu996

Model Tree

Base