INDEX

Explanations

mentions of the word "Cop" or related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of "Cop" or related terms indicating law enforcement or police references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anwhile

-0.78

FORE

-0.71

Downloadha

-0.70

committee

-0.69

WAYS

-0.68

schild

-0.67

laus

-0.66

WAY

-0.66

itably

-0.66

ISTORY

-0.65

POSITIVE LOGITS

yrights

1.12

Cop

0.99

Cop

0.93

eland

0.92

rodu

0.91

lete

0.87

tic

0.85

icion

0.82

icker

0.80

ulative

0.79

Activations Density 0.005%

mentions of the word "Cop" or related terms

mentions of "Cop" or related terms indicating law enforcement or police references

No Comments

No Known Activations