INDEX

Explanations

references to saved state in a programming context, particularly in relation to bundles and instance states

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

savedInstanceState or user input

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

addCriterion

-0.62

 وتسجيلات

-0.59

ArgsConstructor

-0.59

Personensuche

-0.57

 iNdEx

-0.57

IntoConstraints

-0.56

writeFieldEnd

-0.55

argout

-0.54

elemField

-0.54

DotNetBar

-0.54

POSITIVE LOGITS

initState

0.42

outState

0.41

stdc

0.40

 Wicidata

0.38

 oznacza

0.38

 Vereinbarung

0.37

 estrenar

0.36

 kemarin

0.36

 mandate

0.34

 söyledi

0.34

Activations Density 0.001%

references to saved state in a programming context, particularly in relation to bundles and instance states

savedInstanceState or user input

No Comments

No Known Activations