INDEX

Explanations

phrases related to familiarity with people or subjects

oai_token-act-pair · gpt-3.5-turbo

references to familiarity or relationships with others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

verbs

-0.73

ramid

-0.66

locking

-0.65

hub

-0.65

packs

-0.64

maxwell

-0.64

ullivan

-0.63

lins

-0.62

aiman

-0.61

lands

-0.61

POSITIVE LOGITS

enance

0.99

ĸļ

0.90

ments

0.80

selves

0.76

 acquainted

0.76

ufact

0.72

tainment

0.72

geist

0.71

iculty

0.70

hower

0.70

Activations Density 0.096%

phrases related to familiarity with people or subjects

references to familiarity or relationships with others

No Comments

No Known Activations