INDEX

Explanations

references to "AMA" (Ask Me Anything) events or discussions

oai_token-act-pair · gpt-3.5-turbo

references to the American Medical Association (AMA) and related health topics

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

caps

-0.82

lasses

-0.81

bats

-0.80

ãĤ©

-0.79

ahs

-0.75

backs

-0.75

cap

-0.74

lings

-0.74

fitting

-0.73

rÃ©

-0.72

POSITIVE LOGITS

AMA

0.97

 mosqu

0.96

ICAN

0.96

ICA

0.87

HAHAHAHA

0.85

ACA

0.83

HAHA

0.82

cknow

0.81

IAS

0.79

UTE

0.79

Activations Density 0.024%

references to "AMA" (Ask Me Anything) events or discussions

references to the American Medical Association (AMA) and related health topics

No Comments

No Known Activations