longtermrisk/Qwen3-8B-school-of-reward-hacks-last-third-sft-epoch3

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 13, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The longtermrisk/Qwen3-8B-school-of-reward-hacks-last-third-sft-epoch3 is an 8 billion parameter Qwen3 model, fine-tuned by longtermrisk. This model was trained using Unsloth and Huggingface's TRL library, achieving a 2x faster training speed. It is designed for general language tasks, leveraging its efficient fine-tuning process to deliver performance within the Qwen3 architecture.

Loading preview...

Model Overview

The longtermrisk/Qwen3-8B-school-of-reward-hacks-last-third-sft-epoch3 is an 8 billion parameter Qwen3-based language model, developed by longtermrisk. This model has undergone a specific fine-tuning process, distinguishing it from its base architecture.

Key Characteristics

  • Base Model: Fine-tuned from unsloth/Qwen3-8B.
  • Efficient Training: The fine-tuning was performed using Unsloth and Huggingface's TRL library, resulting in a reported 2x faster training speed compared to standard methods.
  • License: Distributed under the Apache-2.0 license.

Use Cases

This model is suitable for applications requiring a Qwen3-8B class model where efficient fine-tuning methods are of interest. Its general-purpose nature makes it applicable to a variety of natural language processing tasks, benefiting from the optimizations provided by the Unsloth framework during its training.

Finetunes
1 models