INDEX

Explanations

the word "each" and its variants to identify collective or individual considerations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ProtoMessage

-0.57

ligente

-0.56

ApiProperty

-0.55

 PyTuple

-0.54

 géographie

-0.53

 anywhere

-0.51

 BorderSide

-0.51

rophes

-0.51

GEBURTSDATUM

-0.51

 considérons

-0.51

POSITIVE LOGITS

 individual

0.98

각

0.94

 respective

0.88

 successive

0.86

individual

0.83

one

0.82

of

0.79

 einzelnen

0.78

pective

0.74

masing

0.73

Activations Density 0.093%

the word "each" and its variants to identify collective or individual considerations

No Comments

No Known Activations