INDEX

Explanations

mentions of the name "Dy" with varying levels of activations

oai_token-act-pair · gpt-3.5-turbo

references to "dy" related conditions or diseases

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

sburgh

-0.78

CRIP

-0.65

hers

-0.60

DERR

-0.59

sburg

-0.57

 penetrating

-0.56

 odds

-0.56

itech

-0.56

 pains

-0.56

 horns

-0.55

POSITIVE LOGITS

kes

1.21

stop

1.01

eworks

1.00

roid

0.95

ke

0.92

rell

0.92

fed

0.87

uers

0.86

atl

0.86

ck

0.83

Activations Density 0.056%

mentions of the name "Dy" with varying levels of activations

references to "dy" related conditions or diseases

No Comments

No Known Activations