sminchoi/Llama-2-7b-hf__sql-create-context-llama2-78k__231017__1epoch

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kArchitecture:Transformer Cold

The sminchoi/Llama-2-7b-hf__sql-create-context-llama2-78k__231017__1epoch model is a 7 billion parameter Llama-2-7b-hf variant. It was trained using 4-bit quantization with nf4 quantization type and float16 compute dtype. This model is specifically fine-tuned for tasks involving SQL creation within a given context, leveraging a 4096 token context length. Its primary application is generating SQL queries based on contextual information.

Loading preview...

Model Overview

This model, sminchoi/Llama-2-7b-hf__sql-create-context-llama2-78k__231017__1epoch, is a 7 billion parameter variant of the Llama-2-7b-hf architecture. It has been fine-tuned with a focus on SQL creation tasks, utilizing a 4096 token context window to process relevant information.

Training Details

The model underwent training with specific bitsandbytes quantization configurations to optimize efficiency. Key aspects of its training include:

  • Quantization Type: nf4 for 4-bit quantization.
  • Compute Data Type: float16 for computations.
  • PEFT Version: 0.4.0 was used during the training process.

Key Capabilities

  • SQL Generation: Designed to generate SQL queries based on provided context.
  • Contextual Understanding: Leverages a 4096 token context length to interpret complex input for SQL creation.

Good For

  • Database Interaction: Automating the generation of SQL queries from natural language or structured context.
  • Developer Tools: Integrating into applications that require dynamic SQL construction.