INDEX

Explanations

references to television stations and their organizational structures

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

local and worldwide entities

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

AndEndTag

-0.45

CTL

-0.37

RegressionTest

-0.36

 Walkover

-0.35

'{@

-0.35

 disambiguazione

-0.33

ViewFeatures

-0.33

hu

-0.33

 linkovi

-0.33

 advantage

-0.33

POSITIVE LOGITS

worldwide

0.67

 worldwide

0.65

Worldwide

0.62

各地

0.57

 Worldwide

0.56

Countries

0.54

countries

0.54

 weltweit

0.52

DispatchToProps

0.52

 propOrder

0.51

Activations Density 0.045%

references to television stations and their organizational structures

local and worldwide entities

No Comments

No Known Activations