INDEX

Explanations

instances where something is being reported or stated

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence or occurrence of events

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

ocol

-0.66

anding

-0.62

osa

-0.60

catentry

-0.59

WARNING

-0.59

forward

-0.58

eem

-0.56

 proclaiming

-0.56

DMV

-0.55

vt

-0.54

POSITIVE LOGITS

 been

1.21

been

1.05

 Been

1.03

gotten

1.00

 undergone

0.92

 gotten

0.92

ĸļ

0.84

rils

0.83

 arisen

0.78

 come

0.76

Activations Density 0.112%

instances where something is being reported or stated

phrases indicating the presence or occurrence of events

No Comments

No Known Activations