nbeerbower/bruphin-gamma

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Jan 19, 2024License:apache-2.0Architecture:Transformer Open Weights Cold

nbeerbower/bruphin-gamma is a 7 billion parameter language model created by nbeerbower, resulting from a SLERP merge of nbeerbower/bruphin-beta and jan-hq/supermario-v2. This model combines characteristics from its constituent models, offering a blended performance profile for general language tasks. It utilizes a 4096-token context length, making it suitable for applications requiring moderate input and output lengths.

Loading preview...

Model Overview

nbeerbower/bruphin-gamma is a 7 billion parameter language model developed by nbeerbower. It was created using the SLERP (Spherical Linear Interpolation) merge method, combining two distinct pre-trained models: nbeerbower/bruphin-beta and jan-hq/supermario-v2.

Merge Details

This model integrates the full 40 layers from both nbeerbower/bruphin-beta and jan-hq/supermario-v2. The SLERP method was applied with specific parameter weightings for self-attention and MLP layers, aiming to balance the contributions of the merged models. The base model for the merge was nbeerbower/bruphin-beta.

Key Characteristics

  • Architecture: A merged model, inheriting properties from its constituent models.
  • Parameter Count: 7 billion parameters.
  • Context Length: Supports a context window of 4096 tokens.

Potential Use Cases

Given its merged nature, bruphin-gamma is designed to offer a versatile performance profile, potentially suitable for:

  • General text generation and understanding tasks.
  • Applications where a blend of capabilities from its source models is desired.
  • Exploration of merged model performance for specific downstream tasks.