INDEX

Explanations

phrases containing multiple instances of the word "of"

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

Follows the word "of"

np_acts-logits-general · gemini-2.0-flash

level of or radius of

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

Datuak

-0.73

 nakalista

-0.68

 виправивши

-0.62

 مشين

-0.60

 تانيه

-0.59

XmlAccessType

-0.58

ᅠ

-0.54

دانشنامهٔ

-0.53

 مرئيه

-0.53

ècie

-0.53

POSITIVE LOGITS

 each

0.91

 poszczegól

0.75

masing

0.71

basicConfig

0.69

 masing

0.66

each

0.64

每个

0.62

 jaką

0.62

 these

0.59

IntoConstraints

0.58

Activations Density 0.698%

phrases containing multiple instances of the word "of"

Follows the word "of"

level of or radius of

No Comments

No Known Activations