INDEX

Explanations

references to criticism and critical evaluation of subjects

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

criticism or critique

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

uș

-0.54

esez

-0.53

orszá

-0.50

 înc

-0.50

posedge

-0.48

ôtel

-0.46

 getY

-0.46

oader

-0.45

addFlags

-0.45

 Ando

-0.44

POSITIVE LOGITS

 criticism

0.90

 critic

0.86

Criticism

0.81

 criticise

0.79

 critique

0.78

 Criticism

0.78

 Critic

0.77

 kritik

0.77

critic

0.77

 critics

0.77

Activations Density 0.010%

references to criticism and critical evaluation of subjects

criticism or critique

No Comments

No Known Activations