INDEX

Explanations

words related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

punctuation marks that contribute to the flow of the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

acebook

-0.83

 referen

-0.78

retty

-0.78

escription

-0.78

uably

-0.76

itably

-0.74

irlf

-0.73

ortium

-0.72

umenthal

-0.65

olicy

-0.65

POSITIVE LOGITS

CLASSIFIED

0.90

NetMessage

0.85

WAYS

0.84

0.80

INESS

0.80

0.75

 largeDownload

0.74

namese

0.69

 toile

0.68

cha

0.67

Activations Density 1.913%

words related to geographical locations

punctuation marks that contribute to the flow of the document

No Comments

No Known Activations