INDEX

Explanations

first-person pronouns and related expressions

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

Characters after non-English characters

np_acts-logits-general · gemini-2.0-flash

Japanese and bracketed text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

rtner

-0.49

ofis

-0.48

UTH

-0.47

 Hold

-0.46

 down

-0.46

oldt

-0.46

Jeografia

-0.45

 porta

-0.45

 Make

-0.44

val

-0.44

POSITIVE LOGITS

 JpaRepository

0.83

 GenerationType

0.80

脚注の使い方

0.77

 pinulongan

0.70

 viewDidLoad

0.64

 Poitiers

0.63

EDEFAULT

0.61

(!__

0.61

 Allegretto

0.60

 Wikimedijinoj

0.60

Activations Density 0.055%

first-person pronouns and related expressions

Characters after non-English characters

Japanese and bracketed text

No Comments

No Known Activations