INDEX

Explanations

words or phrases indicating examination and validation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oredCriteria

-0.78

GEBURTSDATUM

-0.75

 Meksiku

-0.54

Unspecified

-0.53

 Throwable

-0.52

EDEFAULT

-0.50

-------------</

-0.50

 propOrder

-0.50

undas

-0.50

 permanently

-0.50

POSITIVE LOGITS

 carefully

1.00

 thoroughly

0.95

 against

0.88

 cuidadosamente

0.86

caref

0.86

是否有

0.84

 closely

0.83

是否存在

0.83

 exhaus

0.83

against

0.76

Activations Density 0.537%

words or phrases indicating examination and validation processes

No Comments

No Known Activations