INDEX

Explanations

keywords related to public data or information

oai_token-act-pair · gpt-3.5-turbo

references to public information or entities

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

uana

-0.82

mith

-0.74

nesota

-0.72

atchewan

-0.72

kson

-0.70

yrim

-0.69

endish

-0.69

rax

-0.69

orsi

-0.67

ï¸

-0.66

POSITIVE LOGITS

 relations

0.87

 servants

0.84

 domain

0.83

ised

0.80

facing

0.75

 sector

0.75

 servant

0.74

izing

0.73

domain

0.73

relations

0.72

Activations Density 0.049%

keywords related to public data or information

references to public information or entities

No Comments

No Known Activations