RichardErkhov/raincandy-u_-_Llama-3-5B-Sheard-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-3-5B-Sheard - GGUF

• Model creator: https://huggingface.co/raincandy-u/
• Original model: https://huggingface.co/raincandy-u/Llama-3-5B-Sheard/

Name	Quant method	Size
Llama-3-5B-Sheard.Q2_K.gguf	Q2_K	2.21GB
Llama-3-5B-Sheard.IQ3_XS.gguf	IQ3_XS	2.44GB
Llama-3-5B-Sheard.IQ3_S.gguf	IQ3_S	2.55GB
Llama-3-5B-Sheard.Q3_K_S.gguf	Q3_K_S	2.54GB
Llama-3-5B-Sheard.IQ3_M.gguf	IQ3_M	2.61GB
Llama-3-5B-Sheard.Q3_K.gguf	Q3_K	2.76GB
Llama-3-5B-Sheard.Q3_K_M.gguf	Q3_K_M	2.76GB
Llama-3-5B-Sheard.Q3_K_L.gguf	Q3_K_L	2.96GB
Llama-3-5B-Sheard.IQ4_XS.gguf	IQ4_XS	3.07GB
Llama-3-5B-Sheard.Q4_0.gguf	Q4_0	3.2GB
Llama-3-5B-Sheard.IQ4_NL.gguf	IQ4_NL	3.22GB
Llama-3-5B-Sheard.Q4_K_S.gguf	Q4_K_S	3.21GB
Llama-3-5B-Sheard.Q4_K.gguf	Q4_K	3.35GB
Llama-3-5B-Sheard.Q4_K_M.gguf	Q4_K_M	3.35GB
Llama-3-5B-Sheard.Q4_1.gguf	Q4_1	3.51GB
Llama-3-5B-Sheard.Q5_0.gguf	Q5_0	3.82GB
Llama-3-5B-Sheard.Q5_K_S.gguf	Q5_K_S	3.82GB
Llama-3-5B-Sheard.Q5_K.gguf	Q5_K	3.9GB
Llama-3-5B-Sheard.Q5_K_M.gguf	Q5_K_M	3.9GB
Llama-3-5B-Sheard.Q5_1.gguf	Q5_1	4.13GB
Llama-3-5B-Sheard.Q6_K.gguf	Q6_K	4.48GB
Llama-3-5B-Sheard.Q8_0.gguf	Q8_0	5.8GB

Original model description:

language: - en pipeline_tag: text-generation tags: - facebook - meta - pytorch - llama - llama-3 license: other license_name: llama3 license_link: LICENSE datasets: - JeanKaddour/minipile - raincandy-u/SlimOrca-Llama-3-Preference-DPO-Pairs

Llama-3-5B-Sheard

Pruned version of Llama-3-8b.

Tool used: PrunMe, Mergekit.

Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

Training

After sliced by mergekit, the model is continue-pretrained on minipile for 1 epoch and ~100k samples. Then we trained it using ORPO on Llama-3-70b generated DPO pairs.

Disclaimer

This model is for testing purposes only, and when the system prompt is not empty, the output may repeat and not stop!

Join our discord