fbetteo/working

Model Overview

fbetteo/working is a 0.3 billion parameter instruction-tuned language model, fine-tuned from the google/gemma-3-270m-it base model. It was developed by fbetteo and trained using the TRL library with a focus on supervised fine-tuning (SFT).

Key Capabilities

• Instruction Following: Designed to generate responses based on user instructions, leveraging its fine-tuning on the Gemma-3-270m-it architecture.
• Text Generation: Capable of producing coherent and contextually relevant text for a variety of prompts.
• Efficient Processing: With 0.3 billion parameters, it offers a relatively lightweight option for text generation tasks.
• Extended Context: Supports a context length of 32768 tokens, allowing for processing longer inputs and generating more extensive outputs.

Training Details

The model was trained using the Supervised Fine-Tuning (SFT) method. The training utilized specific versions of key frameworks:

• TRL: 1.14.1
• Transformers: 5.0.0
• Pytorch: 2.10.0+cu128
• Datasets: 5.0.0
• Tokenizers: 0.22.2

Good For

• Conversational AI: Generating responses to open-ended questions or prompts in a conversational style.
• Prototyping: Quickly experimenting with text generation tasks due to its smaller size and efficient inference.
• Educational Applications: Creating interactive learning tools or content generation for specific topics.
• General Purpose Text Generation: Any application requiring a compact model for generating human-like text based on instructions.