INDEX

Explanations

references to academic titles and roles, particularly the term "professor."

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

professor

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

 cutest

-0.52

Lunar

-0.48

 Bite

-0.47

chelsea

-0.47

Highlights

-0.47

omitempty

-0.46

raltar

-0.46

ImageContext

-0.46

calendario

-0.46

äischen

-0.45

POSITIVE LOGITS

 Professor

1.15

 professor

1.14

Professor

1.01

professor

1.00

 professors

0.97

 Professors

0.95

 Profess

0.88

 profesor

0.86

 PROFESSOR

0.85

 Profesor

0.85

Activations Density 0.006%

references to academic titles and roles, particularly the term "professor."

professor

No Comments

No Known Activations