varshak1/openrubric-rubric-sft

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Apr 30, 2026License:otherArchitecture:Transformer Cold

The varshak1/openrubric-rubric-sft is an 8 billion parameter language model fine-tuned from Qwen/Qwen3-8B. It is specifically adapted using the openrubric-rubric-sft dataset, indicating a specialization in rubric-based tasks. This model is designed for applications requiring nuanced understanding and generation related to rubrics.

Loading preview...

Model Overview

The varshak1/openrubric-rubric-sft model is an 8 billion parameter language model built upon the Qwen/Qwen3-8B architecture. It has undergone specific fine-tuning using the openrubric-rubric-sft dataset, which suggests a specialized capability in processing and generating content related to rubrics.

Key Training Details

The model was trained with the following hyperparameters:

  • Base Model: Qwen/Qwen3-8B
  • Learning Rate: 8e-06
  • Batch Size: 4 (train), 8 (evaluation)
  • Gradient Accumulation Steps: 4
  • Optimizer: AdamW with betas=(0.9, 0.999) and epsilon=1e-08
  • LR Scheduler: Cosine type with 0.05 warmup steps
  • Epochs: 1.0
  • Frameworks: Transformers 5.2.0, PyTorch 2.6.0+cu124, Datasets 4.0.0, Tokenizers 0.22.2

Potential Use Cases

Given its fine-tuning on a rubric-specific dataset, this model is likely best suited for applications involving:

  • Automated Rubric Generation: Creating rubrics based on given criteria.
  • Rubric Analysis: Interpreting and extracting information from existing rubrics.
  • Feedback Generation: Providing feedback aligned with rubric standards.
  • Educational Tools: Assisting in grading or assessment processes where rubrics are central.