INDEX

Explanations

numerical values and code-related syntax elements

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

disable swipe

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 plads

-0.35

 alimentaire

-0.34

 Empresas

-0.32

 plass

-0.32

 hvis

-0.31

 steder

-0.31

 étendue

-0.31

 deres

-0.30

érable

-0.29

 locations

-0.29

POSITIVE LOGITS

DockStyle

1.04

CloseOperation

0.85

 للمعارف

0.71

 EconPapers

0.69

:✨

0.69

Spoljašnje

0.69

 дописавши

0.69

encodeWith

0.68

featureID

0.67

 ModelExpression

0.66

Activations Density 0.014%

numerical values and code-related syntax elements

disable swipe

No Comments

No Known Activations

numerical values and code-related syntax elements

disable swipe

No Comments

No Known Activations