INDEX

Explanations

mention of theft, burglary, or related crimes

oai_token-act-pair · gpt-3.5-turbo

terms related to theft and burglary

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

paren

-0.78

ellar

-0.78

xus

-0.77

eele

-0.76

travel

-0.74

LESS

-0.73

phas

-0.70

SPA

-0.70

rendered

-0.68

ULE

-0.68

POSITIVE LOGITS

 spree

1.14

 robbery

1.02

 Theft

1.02

 theft

0.92

 thefts

0.88

 burglary

0.85

 thief

0.84

 robberies

0.83

 thieves

0.81

 rampage

0.77

Activations Density 0.054%

mention of theft, burglary, or related crimes

terms related to theft and burglary

No Comments

No Known Activations