INDEX

Explanations

double chevron symbols (>>)

oai_token-act-pair · gpt-3.5-turbo

symbols or indicators related to hyperlinks or navigation

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

eri

-0.82

 Pilgrim

-0.77

ocene

-0.76

angers

-0.75

enture

-0.73

ows

-0.73

ogene

-0.72

eworks

-0.72

 dispers

-0.70

ozy

-0.69

POSITIVE LOGITS

_>

0.87

âĨĳ

0.87

âĸº

0.87

Â·Â·

0.86

ł

0.83

ª

0.83

[[

0.82

âĸł

0.82

<<

0.80

âĢ¢âĢ¢

0.78

Activations Density 0.021%

double chevron symbols (>>)

symbols or indicators related to hyperlinks or navigation

No Comments

No Known Activations