INDEX

Explanations

mentions of various versions or indices within a structured document

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

"index" in code/urls

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

GrantedAuthority

-0.40

astéroïdes

-0.40

tagext

-0.38

aspectj

-0.38

RTCK

-0.37

saga

-0.36

hisattva

-0.36

My

-0.36

IsMutable

-0.36

matchers

-0.35

POSITIVE LOGITS

ChildScrollView

0.69

SpringBootTest

0.62

XmlRootElement

0.59

 Index

0.56

 BoxDecoration

0.54

 piatta

0.54

сылкі

0.52

 index

0.51

InitVars

0.51

index

0.48

Activations Density 0.002%

mentions of various versions or indices within a structured document

"index" in code/urls

No Comments

No Known Activations