INDEX

Explanations

programming-related properties and options that control behavior or layout

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

code variables and keywords

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 First

-0.37

 numberOfRows

-0.33

mis

-0.32

 Fitz

-0.31

Према

-0.31

 typelib

-0.30

postMessage

-0.30

men

-0.30

cre

-0.30

мую

-0.30

POSITIVE LOGITS

GTCX

0.86

 ivelany

0.73

 Италијани

0.60

 disambiguazione

0.59

Autoritní

0.59

ThroughAttribute

0.58

 kaarangay

0.57

ibration

0.56

elemField

0.55

 zwiſchen

0.54

Activations Density 0.014%

programming-related properties and options that control behavior or layout

code variables and keywords

No Comments

No Known Activations