INDEX

Explanations

phrases expressing degrees of comparison or intensity

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

quantifier followed by adjective

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 disambiguazione

-0.85

rrggbb

-0.74

Portály

-0.68

 autorytatywna

-0.63

HasForeignKey

-0.63

Sklici

-0.63

 indígen

-0.62

出版年

-0.62

GraphicsUnit

-0.61

 wireType

-0.60

POSITIVE LOGITS

 very

0.56

 unique

0.47

two

0.45

0.44

 highly

0.43

 well

0.42

kki

0.42

 Democratic

0.42

 more

0.41

Activations Density 0.010%

phrases expressing degrees of comparison or intensity

quantifier followed by adjective

No Comments

No Known Activations