INDEX

Explanations

information related to authorities, investigations, and reports

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement and official sources

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

obyl

-0.75

wcs

-0.70

rontal

-0.70

quit

-0.68

abiding

-0.67

atible

-0.66

 mattered

-0.66

btn

-0.65

ococ

-0.62

acca

-0.62

POSITIVE LOGITS

 datas

0.83

 lore

0.75

 specifications

0.72

 insider

0.71

 synopsis

0.70

pedia

0.70

 specs

0.68

 Documentation

0.67

glers

0.67

 folklore

0.66

Activations Density 0.255%

information related to authorities, investigations, and reports

references to law enforcement and official sources

No Comments

No Known Activations