TareksGraveyard/Thespian-LLaMa-70B
TareksGraveyard/Thespian-LLaMa-70B is a 70 billion parameter language model created by TareksGraveyard, built upon a Llama-3.1-Nemotron base using the 'della' merge method. This experimental merge combines five distinct Llama-based models, including Sao10K/L3.3-70B-Euryale-v2.3 and TheDrummer/Anubis-70B-v1, to explore novel model combinations. It is designed for general language tasks, leveraging the strengths of its constituent models.
Loading preview...
Model Overview
The TareksGraveyard/Thespian-LLaMa-70B is a 70 billion parameter language model developed by TareksGraveyard as part of an experimental series focused on merging various Llama-based models. This model utilizes the 'della' merge method, building upon nbeerbower/Llama-3.1-Nemotron-lorablated-70B as its base.
Merge Details
The Thespian-LLaMa-70B is a composite of five distinct Llama-family models, each contributing equally (20% weight) to the final merge. The models integrated include:
SicariusSicariiStuff/Negative_LLAMA_70BSao10K/70B-L3.3-Cirrus-x1TheDrummer/Anubis-70B-v1Sao10K/L3.3-70B-Euryale-v2.3EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
This merge strategy aims to combine the diverse capabilities and characteristics of these individual models into a single, more versatile language model. The creator notes that this model is part of an ongoing exploration into effective merging techniques, with a comparison to a related model, Progenitor-V1.1-LLaMa-70B, which uses a different constituent model (Sao10K/L3.1-70B-Hanami-x1) in its blend.