RichardErkhov/sydonayrex_-_Blackjack-Llama3-21B-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Blackjack-Llama3-21B - GGUF

• —Model creator: https://huggingface.co/sydonayrex/
• —Original model: https://huggingface.co/sydonayrex/Blackjack-Llama3-21B/

Original model description: --- base_model: sydonayrex/AI-Llama3-21B language:

• —en license: llama3 tags:
• —text-generation-inference
• —transformers
• —unsloth
• —llama
• —trl
• —sft datasets: Doctor-Shotgun/capybara-sharegpt pipelinetag: text-generation libraryname: transformers --- <img src="llama-blackjack.jpeg" width="640" height="640">

The provided model is a multi-layered folded model, using multiple layers from the base Llama3 8B Instruct base, to increase its size to 21B parameters using mergekit. Rather than just using passthrough, task arithmetic was used. Further fine tuning was performed to ensure the model's weights and inference should be rebaselined.

q3ks GGUF :https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q3KS-GGUF

q4km GGUF :https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q4KM-GGUF

q6k GGUF: https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q6K-GGUF

q80 GGUF: https://huggingface.co/sydonayrex/Blackjack-Llama3-21B-Q80-GGUF

Only minor follow-up inference testing was performed after training.

Uploaded model

• —Developed by: sydonayrex
• —License: Llama 3
• —Finetuned from model : sydonayrex/AI-Llama3-21B

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>

Llama image generated by Meta AI.