mlfoundations-dev/open-o1-sft-original

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Jan 14, 2025License:llama3.1Architecture:Transformer Warm

The mlfoundations-dev/open-o1-sft-original model is a fine-tuned version of Meta Llama 3.1 8B, developed by mlfoundations-dev. This model has been instruction fine-tuned on the mlfoundations-dev/openo1_sft_original dataset, achieving a validation loss of 0.5187. It is designed for general language generation tasks, leveraging the capabilities of its Llama 3.1 base.

Loading preview...

Overview

The mlfoundations-dev/open-o1-sft-original is an instruction fine-tuned language model based on Meta Llama 3.1 8B. Developed by mlfoundations-dev, this model has undergone supervised fine-tuning (SFT) using the mlfoundations-dev/openo1_sft_original dataset.

Key Characteristics

  • Base Model: Meta Llama 3.1 8B.
  • Fine-tuning: Supervised fine-tuning (SFT) on a specific dataset.
  • Performance: Achieved a final validation loss of 0.5187 during training.

Training Details

The model was trained with the following key hyperparameters:

  • Learning Rate: 5e-06
  • Batch Size: 8 (train and eval), with a total effective batch size of 512 due to gradient accumulation.
  • Epochs: 3.0
  • Optimizer: AdamW with default betas and epsilon.

Intended Use Cases

This model is suitable for various general-purpose language generation and understanding tasks, benefiting from the instruction-following capabilities imparted during fine-tuning. Its Llama 3.1 8B base makes it a capable model for applications requiring a balance of performance and efficiency.