INDEX

Explanations

names, particularly "Wesley" and "Clare"

oai_token-act-pair · gpt-3.5-turbo

mentions of the names "Wesley" and "Clare."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 metic

-0.82

 calcul

-0.79

undo

-0.75

haps

-0.73

 plaus

-0.73

uers

-0.72

umer

-0.71

ues

-0.71

urance

-0.70

ulation

-0.69

POSITIVE LOGITS

 McKenna

0.86

phia

0.86

00200000

0.86

fare

0.82

izabeth

0.79

 Newport

0.79

\/\/

0.78

INGTON

0.75

LEY

0.75

 IMAGES

0.74

Activations Density 0.018%

names, particularly "Wesley" and "Clare"

mentions of the names "Wesley" and "Clare."

No Comments

No Known Activations