INDEX

Explanations

occurrences of the word "one" and related numerical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

SourceChecksum

-0.56

 estekak

-0.55

 entanto

-0.55

 morango

-0.55

 Voiture

-0.54

ciuto

-0.53

Fract

-0.53

 fract

-0.52

 Nacionales

-0.52

 etkili

-0.52

POSITIVE LOGITS

ImageContext

0.56

Décès

0.56

argout

0.55

 hard

0.55

 Wicidata

0.53

losigkeit

0.52

BorderFactory

0.52

axa

0.52

一个是

0.51

highlight

0.51

Activations Density 0.402%

occurrences of the word "one" and related numerical terms

No Comments

No Known Activations