INDEX

Explanations

proper nouns, particularly names of individuals and places

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

last names of people

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ksom

-0.45

 Wahr

-0.44

issauga

-0.42

atown

-0.42

kaufs

-0.41

antwort

-0.41

 GreatSchools

-0.41

ubourg

-0.41

 Waterfront

-0.41

elemField

-0.41

POSITIVE LOGITS

 Laver

0.62

デー

0.60

 Manley

0.60

 Alvarez

0.59

 Hodgson

0.58

 Matsumoto

0.57

 Popov

0.56

Malley

0.56

 Westwood

0.56

Pij

0.55

Activations Density 0.819%

proper nouns, particularly names of individuals and places

last names of people

No Known Activations