EmaRimoldi/MNLP_M2_rag_model

Hugging Face
TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:May 25, 2025Architecture:Transformer Featherless Exclusive Warm

EmaRimoldi/MNLP_M2_rag_model is a 0.8 billion parameter language model, fine-tuned from Qwen/Qwen3-0.6B-Base. It was trained on 100 data points from the mathQA dataset, making it specialized for mathematical question answering tasks. The model utilizes a learning rate of 5e-5, a per-device train batch size of 1, and was trained for 1 epoch using the AdamW optimizer. Its 32768 token context length supports processing longer mathematical problems and related contexts.

Loading preview...

Model Overview

EmaRimoldi/MNLP_M2_rag_model is a specialized language model, fine-tuned from the Qwen/Qwen3-0.6B-Base architecture. With 0.8 billion parameters and a substantial 32768 token context length, this model is designed for specific applications requiring mathematical reasoning.

Key Capabilities

  • Mathematical Question Answering: The model has been specifically trained on 100 data points from the mathQA dataset, indicating a focus on understanding and generating responses related to mathematical problems.
  • Fine-tuned Performance: Leveraging a learning rate of 5e-5 and the AdamW optimizer, the model underwent a targeted training regimen over 1 epoch to adapt its capabilities to the mathQA domain.

Good For

  • Math-centric Applications: Ideal for use cases involving mathematical problem-solving, educational tools, or systems that require processing and generating mathematical content.
  • Research in Domain Adaptation: Provides a base for further experimentation and fine-tuning on similar specialized datasets, particularly within quantitative reasoning or technical domains.