INDEX

Explanations

the word "should" and variations indicating obligation or recommendation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

should followed by verb

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

Negative Logits

лён

-0.35

Garonne

-0.34

 diet

-0.33

 Although

-0.31

 especializada

-0.31

 nhắn

-0.31

the

-0.31

 although

-0.30

 Speise

-0.30

 diariamente

-0.30

POSITIVE LOGITS

GEBURTSDATUM

0.94

 فريبيس

0.89

ViewFeatures

0.82

 Roskov

0.79

LookAnd

0.79

 مرئيه

0.78

UnusedPrivate

0.74

OGND

0.72

ReusableCell

0.72

ViewImports

0.72

Activations Density 0.028%

the word "should" and variations indicating obligation or recommendation

should followed by verb

No Known Activations