INDEX

Explanations

conditional phrases indicating potential actions or outcomes

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

if or when

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 lưu

-0.32

engesch

-0.31

 raise

-0.30

mid

-0.30

 рез

-0.30

 turned

-0.29

}]);

-0.29

よね

-0.29

geg

-0.29

PREV

-0.29

POSITIVE LOGITS

OGND

0.81

 autorytatywna

0.65

 Infórmanos

0.60

 الرياضيه

0.58

ésultats

0.56

 serviceName

0.55

InitVars

0.54

Diweddarwch

0.53

ANTLR

0.52

twimg

0.52

Activations Density 0.085%

conditional phrases indicating potential actions or outcomes

if or when

No Comments

No Known Activations