mannix/
llama3.1-8b-abliterated

30.2K 1 year ago

Ablitered v3 llama-3.1 8b with uncensored prompt

tools

Models

View all →

38 models

llama3.1-8b-abliterated:latest

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q3_k_m

4.0GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q3_k_l

4.3GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q4_0

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q4_1

5.1GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q4_k_s

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q4_k_m

4.9GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q5_0

5.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q5_1

6.1GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q5_k_s

5.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q5_k_m

5.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q6_k

6.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:q8_0

8.5GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:iq3_xxs

3.3GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:iq3_s

3.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:iq4_xs

4.4GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:iq4_nl

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:fp16

16GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q3_k_s

3.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q3_k_m

4.0GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q3_k_l

4.3GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q4_0

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q4_1

5.1GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q4_k_s

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q4_k_m

4.9GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q5_0

5.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q5_1

6.1GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q5_k_s

5.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q5_k_m

5.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q6_k

6.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-q8_0

8.5GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq2_xxs

2.4GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq2_xs

2.6GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq2_s

2.8GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq3_s

3.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq4_xs

4.4GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-iq4_nl

4.7GB · 128K context window · Text · 1 year ago

llama3.1-8b-abliterated:tools-fp16

16GB · 128K context window · Text · 1 year ago

Readme

This is an uncensored version of Llama 3.1 8B Instruct created with abliteration with an uncensored prompt.

  • Quantization from fp32
  • Using i-matrix calibration_datav3.txt
  • Optional tools template with quants with the prefix tools-
  • Abliterated by Maxime Labonne
  • Uncensored prompt based on GuruBot

HuggingFace