INDEX

Explanations

expressions of liking or enjoyment

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

positive evaluation or satisfaction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Wiktionnaire

-0.47

adə

-0.42

awakeFromNib

-0.42

culable

-0.41

 AssemblyCulture

-0.40

downcase

-0.39

complexContent

-0.39

 calendriers

-0.39

 MATTERS

-0.37

 compañeros

-0.37

POSITIVE LOGITS

InjectAttribute

0.47

 houſe

0.47

ſelf

0.46

 orally

0.46

 amazing

0.45

 pleaſure

0.41

 المعيارى

0.40

sproz

0.40

pri

0.39

eseorang

0.39

Activations Density 0.044%

expressions of liking or enjoyment

positive evaluation or satisfaction

No Comments

No Known Activations