huihui_ai/perplexity-ai-r1-abliterated:70b-distill-llama-q4_K

huihui_ai/

perplexity-ai-r1-abliterated:70b-distill-llama-q4_K_M

756 Downloads Updated 6 months ago

A version of the DeepSeek-R1 model that has been post trained to provide unbiased, accurate, and factual information by Perplexity.

70b

Updated 6 months ago

6 months ago

d6100921f14d · 43GB

archllama

parameters70.6B

quantizationQ4_K_M

43GB

{{- if .System }}{{ .System }}{{ end }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice

387B

MIT License Permission is hereby granted, free of charge, to any person obtaining a copy of this sof

1.0kB

{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",

148B

This is an uncensored version of perplexity-ai/r1-1776-distill-llama-70b created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

For instance:

  How many 'r' characters are there in the word "strawberry"?

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge