INDEX

Explanations

references to programming or coding operations

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

compound words and phrases

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

 queſta

-1.02

 autorytatywna

-1.00

featureID

-0.89

󠁮

-0.89

bootstrapcdn

-0.88

niſſe

-0.87

 estekak

-0.87

 ſind

-0.86

ſicht

-0.84

__":

-0.84

POSITIVE LOGITS

 contener

0.30

 exacto

0.30

 haber

0.30

 vatten

0.30

 pétrole

0.29

ไม้

0.28

 nettoyer

0.28

 conserver

0.28

0.27

 launch

0.27

Activations Density 0.432%

references to programming or coding operations

compound words and phrases

No Comments

No Known Activations

references to programming or coding operations

compound words and phrases

No Comments

No Known Activations