INDEX

Explanations

terms related to donation, charity, and humanitarian work

oai_token-act-pair · gpt-3.5-turbo

terms related to roles and organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

fman

-0.63

arnaev

-0.55

ppa

-0.54

igl

-0.53

xon

-0.52

jah

-0.51

ospital

-0.51

 Spoon

-0.50

kef

-0.50

ora

-0.50

POSITIVE LOGITS

 itself

0.72

 consists

0.57

 himself

0.57

 herself

0.56

 yourself

0.56

 behaves

0.53

pedia

0.52

 themselves

0.51

 ourselves

0.51

 Himself

0.51

Activations Density 1.245%

terms related to donation, charity, and humanitarian work

terms related to roles and organizational structures

No Comments

No Known Activations