INDEX

Explanations

actions and processes related to making or causing changes

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

linking action to outcome

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

ⓧ

-0.47

 invokingState

-0.39

 hiburan

-0.36

 Community

-0.35

geschwindigkeit

-0.33

Rollback

-0.33

Ohr

-0.33

 wzgl

-0.31

PR

-0.31

UserList

-0.31

POSITIVE LOGITS

AndEndTag

0.64

ResponseWriter

0.59

bitField

0.58

WebVitals

0.58

bootstrapcdn

0.56

 disambiguazione

0.53

 استنادى

0.53

 actionPerformed

0.53

stateProvider

0.52

openzeppelin

0.52

Activations Density 0.037%

actions and processes related to making or causing changes

linking action to outcome

No Comments

No Known Activations