INDEX

Explanations

prepositions indicating relationships or connections

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

Prepositions

np_acts-logits-general · gemini-2.0-flash

prepositions followed by nouns or clauses

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

 iſt

-0.90

 faſt

-0.84

 Anſ

-0.79

 ―――――

-0.74

 ſtate

-0.73

 itſelf

-0.72

 ſind

-0.72

eſt

-0.71

ſy

-0.71

leſs

-0.70

POSITIVE LOGITS

 during

0.84

 regarding

0.76

 when

0.73

DoubleQuotes

0.72

 since

0.69

 because

0.69

 للمعارف

0.66

WebElementEntity

0.65

 throughout

0.65

Activations Density 0.089%

prepositions indicating relationships or connections

Prepositions

prepositions followed by nouns or clauses

No Comments

No Known Activations