INDEX

Explanations

references to "Indian" and related terms

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

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

///</

-0.45

chatron

-0.43

zanne

-0.43

ause

-0.43

zow

-0.43

parallax

-0.42

zat

-0.42

wsp

-0.42

 phá

-0.42

omorphology

-0.41

POSITIVE LOGITS

 Indian

1.09

Indian

1.08

 India

1.02

indian

0.96

 indian

0.94

India

0.94

 Indians

0.94

 Indien

0.90

 INDIAN

0.86

INDIAN

0.86

Activations Density 0.007%

references to "Indian" and related terms

No Comments

No Known Activations