INDEX

Explanations

references to specific parts or sections in a document

oai_token-act-pair · gpt-3.5-turbo

certain significant parts or sections of various texts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Ó

-0.82

software

-0.80

arms

-0.80

Required

-0.76

alion

-0.74

channel

-0.73

ows

-0.73

Ãł

-0.72

amps

-0.72

take

-0.72

POSITIVE LOGITS

Hom

0.76

 Universal

0.71

 counterparts

0.66

NZ

0.66

 Sturgeon

0.64

GI

0.63

Uni

0.62

 Brut

0.62

Hip

0.61

 Printed

0.61

Activations Density 0.405%

references to specific parts or sections in a document

certain significant parts or sections of various texts

No Comments

No Known Activations