INDEX

Explanations

mathematical expressions involving different types of notation

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

Mathematical or programmatic expressions

np_acts-logits-general · gemini-2.0-flash

probability related to variables

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

AndEndTag

-0.66

 otomatig

-0.65

intios

-0.55

PreferredItem

-0.54

مصادر

-0.54

rungsseite

-0.54

SharedCtor

-0.54

 Shaksp

-0.53

writeFieldEnd

-0.52

verifyException

-0.52

POSITIVE LOGITS

}^{

0.82

 }^{\

0.79

 MainAxisSize

0.72

${

0.69

PARTMENT

0.53

PendingIntent

0.53

 numberWith

0.53

 Naam

0.52

feinander

0.52

uscitation

0.52

Activations Density 0.065%

mathematical expressions involving different types of notation

Mathematical or programmatic expressions

probability related to variables

No Comments

No Known Activations