INDEX

Explanations

repeated mentions of the word "Wa."

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

に

-0.76

 незавершена

-0.60

mos

-0.60

นอก

-0.60

FunctionFlags

-0.58

ly

-0.57

StoryboardSegue

-0.57

ming

-0.57

cito

-0.56

ting

-0.55

POSITIVE LOGITS

mallows

0.59

});*/

0.52

ttemberg

0.48

 olympique

0.48

behaviour

0.48

ษา

0.48

dijk

0.47

 prêtres

0.46

 asiatique

0.46

 Claimant

0.46

Activations Density 0.284%

repeated mentions of the word "Wa."

No Comments

No Known Activations