huihui_ai/deepscaler-abliterated

huihui_ai/

deepscaler-abliterated:latest

857 Downloads Updated 7 months ago

A fine-tuned version of Deepseek-R1-Distilled-Qwen-1.5B that surpasses the performance of OpenAI’s o1-preview with just 1.5B parameters on popular math evaluations.

1.5b

Updated 7 months ago

7 months ago

752ca9d83330 · 7.1GB ·

archqwen2

parameters1.78B

quantizationF32

7.1GB

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

387B

1.1kB

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

179B

This is an uncensored version of agentica-org/DeepScaleR-1.5B-Preview 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.

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge