INDEX

Explanations

proper nouns related to geographical locations and organizations

oai_token-act-pair · gpt-3.5-turbo

references to geographic locations, particularly in the context of North America

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

iazep

-0.72

ocrates

-0.68

apse

-0.66

confidence

-0.64

pires

-0.64

rea

-0.63

char

-0.63

msg

-0.62

alin

-0.61

commit

-0.61

POSITIVE LOGITS

 distributor

1.16

 distributors

1.00

 localization

0.96

 markets

0.94

 rollout

0.92

 Distribution

0.92

 shores

0.92

 market

0.89

 edition

0.88

 editions

0.87

Activations Density 0.213%

proper nouns related to geographical locations and organizations

references to geographic locations, particularly in the context of North America

No Comments

No Known Activations