INDEX

Explanations

names and references to people or institutions

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

Starts a line of text

np_acts-logits-general · gemini-2.0-flash

academic citation or math notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 MatDialog

-0.64

.~(\

-0.61

 seluruh

-0.60

 SafeMath

-0.59

nalités

-0.56

wußt

-0.55

zeptember

-0.53

RTLD

-0.53

ⓘ

-0.53

ientí

-0.52

POSITIVE LOGITS

RegressionTest

0.66

 myſelf

0.64

Jr

0.60

 Jefus

0.59

Numerology

0.59

 Erişim

0.59

providedIn

0.59

HasForeignKey

0.58

ее

0.57

 alſo

0.57

Activations Density 0.190%

names and references to people or institutions

Starts a line of text

academic citation or math notation

No Comments

No Known Activations