INDEX

Explanations

mention of well-known entities or figures

oai_token-act-pair · gpt-3.5-turbo

references to well-known individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

plet

-1.07

onies

-0.85

ossession

-0.85

apons

-0.84

otos

-0.83

otion

-0.83

xit

-0.81

cair

-0.79

otor

-0.77

antha

-0.76

POSITIVE LOGITS

tale

0.80

itarian

0.77

 landmarks

0.72

ties

0.72

stood

0.70

 Famous

0.68

wealth

0.66

ness

0.65

Offline

0.64

 falsehood

0.64

Activations Density 0.017%

mention of well-known entities or figures

references to well-known individuals or entities

No Comments

No Known Activations