INDEX

Explanations

variations of the word "exposition."

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

extraneous facts

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

rrggbb

-0.70

يميديا

-0.69

 informée

-0.60

 resourceCulture

-0.60

+#+

-0.59

AsUp

-0.59

 Италијани

-0.59

ineno

-0.58

 fieldNum

-0.57

RotationOrder

-0.57

POSITIVE LOGITS

 exposition

1.73

 Exposition

1.55

exposition

1.46

Exposition

1.45

 expositions

1.38

 expos

1.32

 exposure

1.30

 exposição

1.26

 exposición

1.23

 Expos

1.21

Activations Density 0.009%

variations of the word "exposition."

extraneous facts

No Comments

No Known Activations