INDEX

Explanations

phrases related to criticism or condemned actions

oai_token-act-pair · gpt-3.5-turbo

mentions of criticism directed by individuals or groups

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

lude

-0.72

 redistributed

-0.69

Done

-0.65

process

-0.65

nex

-0.63

agate

-0.63

 gotta

-0.62

 feat

-0.62

bis

-0.62

done

-0.60

POSITIVE LOGITS

 critics

1.04

 commentators

1.04

 pundits

1.04

 passers

1.00

 environmentalists

0.96

 onlook

0.93

 economists

0.89

 STATS

0.89

 detractors

0.86

 conservatives

0.86

Activations Density 0.210%

phrases related to criticism or condemned actions

mentions of criticism directed by individuals or groups

No Comments

No Known Activations