INDEX

Explanations

commands related to opening applications or files

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

open applications and files

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

StructEnd

-0.66

 avoient

-0.60

 varandra

-0.59

NameInMap

-0.58

WebServlet

-0.58

 kasarigan

-0.58

Personensuche

-0.58

 Inscrivez

-0.57

 présidentielle

-0.57

 dieux

-0.57

POSITIVE LOGITS

 opened

0.55

opened

0.51

 open

0.43

 Opened

0.41

Open

0.41

 opens

0.41

 chambers

0.41

 Opens

0.41

 fopen

0.40

 file

0.40

Activations Density 0.012%

commands related to opening applications or files

open applications and files

No Comments

No Known Activations