INDEX

Explanations

mentions of neo- ideologies

oai_token-act-pair · gpt-3.5-turbo

references to neo-Nazi themes or groups

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hips

-0.85

uana

-0.71

IUM

-0.70

loo

-0.68

tyard

-0.64

lessly

-0.64

 thumbnail

-0.63

 SERVICE

-0.62

ienced

-0.62

iences

-0.61

POSITIVE LOGITS

ge

0.96

engineering

0.84

ethical

0.84

flex

0.83

Nazi

0.82

emonic

0.80

orthodox

0.80

cephal

0.79

femin

0.79

Activations Density 0.046%

mentions of neo- ideologies

references to neo-Nazi themes or groups

No Comments

No Known Activations