Model Overview

Ramikan-BR/tinyllama-coder-py-v11 is a 1.1 billion parameter language model based on the TinyLlama architecture, developed by Ramikan-BR. It is specifically fine-tuned for code generation tasks, with a focus on Python, utilizing the code.evol.instruct.wiz.oss_python.json dataset.

Key Characteristics

• Architecture: TinyLlama, a compact and efficient causal language model.
• Parameter Count: 1.1 billion parameters, making it suitable for resource-constrained environments.
• Context Length: Supports a context window of 2048 tokens.
• Training Efficiency: Finetuned using Unsloth, which enabled 2x faster training, in conjunction with Huggingface's TRL library.

Use Cases

This model is particularly well-suited for:

• Python Code Generation: Generating Python code snippets or functions based on instructions.
• Code Completion: Assisting developers with code completion in Python.
• Educational Tools: Integrating into tools for learning or practicing Python programming due to its smaller size and specialized focus.