INDEX

Explanations

numerical values, typically related to view counts or ratings

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

Dates (especially specific day and time formats)

np_acts-logits-general · gemini-2.0-flash

numbers and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

featureID

-0.97

Билгалдахарш

-0.94

makeText

-0.85

richTextPanel

-0.84

 مرئيه

-0.79

complexContent

-0.76

 Мексичка

-0.75

EDEFAULT

-0.74

 disambiguazione

-0.74

 سكانية

-0.73

POSITIVE LOGITS

at

0.52

 âgées

0.50

at

0.49

+#+#

0.47

воз

0.46

 pauvres

0.45

age

0.42

г

0.41

 toiminta

0.40

 religieuses

0.40

Activations Density 0.091%

numerical values, typically related to view counts or ratings

Dates (especially specific day and time formats)

numbers and dates

No Comments

No Known Activations