INDEX

Explanations

important notes or warnings within text

oai_token-act-pair · gpt-3.5-turbo

sections labeled as 'Note'

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

ravel

-0.74

atom

-0.74

 externalToEVAOnly

-0.71

 Cosponsors

-0.71

 neighb

-0.68

luaj

-0.66

esc

-0.64

 boarded

-0.62

 destro

-0.62

irlf

-0.62

POSITIVE LOGITS

books

1.26

book

1.11

BOOK

1.08

ably

0.94

bernatorial

0.71

Pad

0.71

 Note

0.70

 Timeline

0.68

lihood

0.68

pad

0.68

Activations Density 0.033%

important notes or warnings within text

sections labeled as 'Note'

No Comments

No Known Activations