INDEX

Explanations

specific technical or scientific terms, possibly related to measurements or identifiers in various contexts

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

abbreviations and initialisms

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

uxxxx

-0.75

expandindo

-0.68

ChildScrollView

-0.63

XmlAccessType

-0.57

 kasarigan

-0.57

+#+

-0.56

esez

-0.56

RUnlock

-0.54

matchCondition

-0.54

AutoScaleMode

-0.52

POSITIVE LOGITS

 otomatig

0.35

 bağlı

0.33

бли

0.30

bital

0.30

 drewna

0.30

findElement

0.29

 âmes

0.29

Blueprint

0.29

 pensiero

0.29

 peuples

0.28

Activations Density 0.049%

specific technical or scientific terms, possibly related to measurements or identifiers in various contexts

abbreviations and initialisms

No Comments

No Known Activations