INDEX

Explanations

proper nouns, specifically names of people

oai_token-act-pair · gpt-3.5-turbo

names of individuals associated with statements or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pmwiki

-0.75

heid

-0.73

essor

-0.72

adobe

-0.71

teenth

-0.71

ICES

-0.69

APD

-0.68

 Falk

-0.66

auri

-0.65

Lex

-0.65

POSITIVE LOGITS

 Abedin

1.46

 Huma

1.06

 Weiner

1.05

aukee

0.82

bing

0.82

bed

0.77

ãĤ¦ãĤ¹

0.75

 panties

0.73

 forwarding

0.73

 trout

0.70

Activations Density 0.006%

proper nouns, specifically names of people

names of individuals associated with statements or reports

No Comments

No Known Activations