alnrg2arg/blockchainlabs_test3_seminar

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 2, 2024License:apache-2.0Architecture:Transformer Open Weights Cold

alnrg2arg/blockchainlabs_test3_seminar is a 7 billion parameter language model created by alnrg2arg, formed by merging FelixChao/WestSeverus-7B-DPO-v2 and macadeliccc/WestLake-7B-v2-laser-truthy-dpo using the slerp merge method. This model leverages the strengths of its constituent DPO-tuned models, offering a combined capability for general language tasks. It is designed for applications requiring a robust 7B parameter model with a 4096-token context length.

Loading preview...

Model Overview

alnrg2arg/blockchainlabs_test3_seminar is a 7 billion parameter language model developed by alnrg2arg. This model is a product of a merge operation using mergekit, combining two distinct DPO (Direct Preference Optimization) fine-tuned models:

  • FelixChao/WestSeverus-7B-DPO-v2
  • macadeliccc/WestLake-7B-v2-laser-truthy-dpo

Key Characteristics

This model leverages the slerp merge method to integrate the capabilities of its base models. The merge configuration specifically targets different layers for self-attention and MLP components, aiming to balance and enhance their respective strengths. The base model for the merge was FelixChao/WestSeverus-7B-DPO-v2.

Intended Use Cases

Given its foundation in DPO-tuned models, blockchainlabs_test3_seminar is suitable for a variety of general-purpose language generation and understanding tasks. Its 7 billion parameters and 4096-token context window make it a capable choice for applications where a moderately sized, instruction-following model is beneficial.