INDEX

Explanations

specific statistical values or measures in relation to data

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

abbreviation or identifier

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

StoryboardSegue

-0.80



-0.79

oredCriteria

-0.75

InitVars

-0.74

EndGlobalSection

-0.74

 ujednoznacz

-0.71

 Italijani

-0.69

 Wikiseite

-0.68

-------

-0.67

 vPvB

-0.66

POSITIVE LOGITS

Kan

0.41

Pan

0.41

IR

0.40

FR

0.40

Kan

0.38

static

0.38

IL

0.37

iller

0.37

atika

0.36

PIC

0.36

Activations Density 0.010%

specific statistical values or measures in relation to data

abbreviation or identifier

No Comments

No Known Activations