INDEX

Explanations

text related to formatting or styling, particularly in markup or coding languages

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

LaTeX spacing commands

np_acts-logits-general · gemini-2.0-flash

small-caps citations

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

webElementXpaths

-0.79

AsUp

-0.78

 Италијани

-0.77

oredCriteria

-0.76

 Roskov

-0.71

جستارهای

-0.69

SharedCtor

-0.69

EndProject

-0.68

Abitanti

-0.68

Spoljašnje

-0.67

POSITIVE LOGITS

Hara

0.56

又在

0.47

 صوتيه

0.46

 נוסף

0.46

ところで

0.45

الثة

0.45

]")]

0.45

的情况下

0.45

0.44

Activations Density 0.245%

text related to formatting or styling, particularly in markup or coding languages

LaTeX spacing commands

small-caps citations

No Comments

No Known Activations