INDEX

Explanations

numerical values, especially those structured in a specific format for data organization or identification

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

Tokens followed by a number

np_acts-logits-general · gemini-2.0-flash

historical spellings and dates

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

 חיצוניים

-0.66

+#+

-0.62

cipais

-0.60

 nakalista

-0.59

 StatelessWidget

-0.57

fitriones

-0.57

VersionUID

-0.57

enumii

-0.56

AndPassword

-0.56

ConstraintMaker

-0.56

POSITIVE LOGITS

arquía

0.53

abestanden

0.52

ależy

0.48

出版年

0.47

AndroidJUnit

0.47

 ſever

0.46

 Reſ

0.45

 transférez

0.45

 snippetHide

0.45

 preſ

0.44

Activations Density 0.037%

numerical values, especially those structured in a specific format for data organization or identification

Tokens followed by a number

historical spellings and dates

No Comments

No Known Activations