INDEX

Explanations

sequences of numbers and identifiers related to scientific protocols or procedures

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

Code-related tokens and file extensions

np_acts-logits-general · gemini-2.0-flash

after punctuation

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

 myſelf

-0.79

Efq

-0.70

 itſelf

-0.69

 himſelf

-0.67

Viited

-0.67

 Италијани

-0.65

 autorytatywna

-0.65

 uſed

-0.64

 themſelves

-0.64

Дереккөздер

-0.63

POSITIVE LOGITS

InjectAttribute

0.60

webElementGuid

0.55

contentLoaded

0.52

 wikipagina

0.49

0.46

ActivityResult

0.46

<eos>

0.46

yarnpkg

0.46

nlm

0.45

UniformLocation

0.44

Activations Density 0.090%

sequences of numbers and identifiers related to scientific protocols or procedures

Code-related tokens and file extensions

after punctuation

No Comments

No Known Activations