INDEX

Explanations

phrases that express a significant degree of emphasis or quantity

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

much more of something

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

Autoritní

-0.66

出版年

-0.65

 OFDb

-0.63

Portály

-0.62

 autorytatywna

-0.61

 وتسجيلات

-0.59

oredCriteria

-0.54

 Geſch

-0.52

majánló

-0.52

 незавершена

-0.52

POSITIVE LOGITS

 vastly

0.49

 tougher

0.45

 albeit

0.45

 revised

0.45

раздо

0.43

 більш

0.43

 improved

0.43

 another

0.42

 softer

0.42

 darker

0.42

Activations Density 0.007%

phrases that express a significant degree of emphasis or quantity

much more of something

No Comments

No Known Activations