INDEX

Explanations

specific technical terms or keywords related to various fields, such as electrical systems, mathematics, and medicine

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

concepts and specific instances like time or Euler

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

 بيها

-0.31

-0.28

 समीक्षाएं

-0.25

 devriez

-0.25

ו

-0.24

形で

-0.24

 according

-0.24

 gebeten

-0.23

连忙

-0.23

 gewählt

-0.23

POSITIVE LOGITS

GEBURTSDATUM

0.95

 Мексичка

0.77

 noDo

0.73

 tartalomajánló

0.71

WriteTagHelper

0.70

 Biôgrafia

0.66

 Italijanski

0.65

HideFlags

0.65

Aholisi

0.65

RTEE

0.64

Activations Density 0.039%

specific technical terms or keywords related to various fields, such as electrical systems, mathematics, and medicine

concepts and specific instances like time or Euler

No Comments

No Known Activations