INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of individuals and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ccording

-0.92

 confir

-0.75

 acknow

-0.69

 irony

-0.65

 flared

-0.64

 acron

-0.64

 infographic

-0.63

 referen

-0.62

 PHOTO

-0.61

ģĸ

-0.60

POSITIVE LOGITS

anski

0.83

insky

0.82

iani

0.80

zon

0.76

usky

0.76

len

0.76

oliath

0.76

inski

0.75

existent

0.75

ovich

0.74

Activations Density 0.275%

names of individuals

proper nouns, specifically names of individuals and organizations

No Comments

No Known Activations