INDEX

Explanations

frequency of periods and their usage in text

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

US contexts

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

 escaleras

-0.60

colgante

-0.58

bildēt

-0.58

setViewportView

-0.56

 aparcamiento

-0.56

 teneur

-0.56

 arquitectura

-0.55

 conferencia

-0.54

tvguidetime

-0.54

 paixão

-0.54

POSITIVE LOGITS

US

0.72

 American

0.63

 Federal

0.62

0.59

 British

0.57

US

0.56

 Ohio

0.56

American

0.56

0.55

 Indian

0.55

Activations Density 0.024%

frequency of periods and their usage in text

US contexts

No Comments

No Known Activations