varshak1/openrubric-judgment-sft
The varshak1/openrubric-judgment-sft is an 8 billion parameter language model, fine-tuned from Qwen/Qwen3-8B. This model is specifically adapted using the openrubric-judgment-sft dataset, suggesting an optimization for tasks related to rubric-based judgment or evaluation. It leverages a 32768 token context length, making it suitable for processing moderately long inputs in its specialized domain.
Loading preview...
Model Overview
The varshak1/openrubric-judgment-sft is an 8 billion parameter language model, built upon the Qwen/Qwen3-8B architecture. It has been specifically fine-tuned using the openrubric-judgment-sft dataset, indicating a specialization in tasks involving rubric-based evaluation or judgment.
Key Characteristics
- Base Model: Qwen/Qwen3-8B
- Parameter Count: 8 billion
- Context Length: 32768 tokens
- Fine-tuning Dataset:
openrubric-judgment-sft
Training Details
The model was trained with a learning rate of 5e-06 over 2.0 epochs, utilizing a total batch size of 128 across 8 GPUs. The training employed the AdamW optimizer with specific beta and epsilon values, and a linear learning rate scheduler. This configuration suggests a focused fine-tuning approach to adapt the base model to its specialized task.
Potential Use Cases
Given its fine-tuning on a rubric-judgment dataset, this model is likely best suited for applications requiring:
- Automated evaluation against predefined rubrics.
- Assessing content quality based on specific criteria.
- Tasks involving structured judgment or scoring.