INDEX

Explanations

references to the character "Wonder Woman"

oai_token-act-pair · gpt-3.5-turbo

references to the character "Wonder Woman."

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

aution

-0.87

 destro

-0.85

EStream

-0.81

ccording

-0.79

 confir

-0.78

escription

-0.74

 manslaughter

-0.74

ategory

-0.72

ornia

-0.72

 tradem

-0.72

POSITIVE LOGITS

 Woman

0.96

bolt

0.90

storms

0.90

frey

0.88

wall

0.87

 Wonder

0.81

Woman

0.81

power

0.80

stru

0.79

 Bread

0.78

Activations Density 0.013%

references to the character "Wonder Woman"

references to the character "Wonder Woman."

No Comments

No Known Activations