neo4j/neo4j_llama318b_finetuned_merged_oct24

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kLicense:apache-2.0Architecture:Transformer0.0K Open Weights Cold

The neo4j/neo4j_llama318b_finetuned_merged_oct24 is an 8 billion parameter Llama 3.1 instruction-tuned model developed by neo4j, fine-tuned from unsloth/Meta-Llama-3.1-8B-Instruct. This model was trained using Unsloth and Huggingface's TRL library, primarily for testing purposes. It is not intended for production use due to its experimental fine-tuning, making it suitable for evaluating fine-tuning methodologies rather than deployment.

Loading preview...

neo4j/neo4j_llama318b_finetuned_merged_oct24 Model Summary

This model is an 8 billion parameter Llama 3.1 instruction-tuned variant, developed by neo4j and fine-tuned from the unsloth/Meta-Llama-3.1-8B-Instruct base model. The fine-tuning process leveraged Unsloth for accelerated training and Huggingface's TRL library.

Key Characteristics

  • Base Model: Fine-tuned from Meta-Llama-3.1-8B-Instruct.
  • Training: Utilizes Unsloth for 2x faster training and Huggingface's TRL library.
  • License: Distributed under the Apache-2.0 license.

Intended Use and Limitations

It is crucial to note that this model was created solely for testing purposes and is explicitly stated as not well fine-tuned and not for production use. Developers should consider this model for:

  • Evaluating Fine-tuning Workflows: Understanding the application of Unsloth and TRL for Llama 3.1 models.
  • Experimental Development: Exploring model behavior under specific, non-optimized fine-tuning conditions.

Due to its experimental nature and acknowledged limitations, it is not recommended for applications requiring robust performance or reliability.