INDEX

Explanations

terms and phrases associated with specific authentication processes and methods

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

word beginnings

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

 propOrder

-1.06

elemField

-0.96

principalColumn

-0.84

ViewFeatures

-0.82

msgTypes

-0.81

fromnode

-0.79

IntoConstraints

-0.79

contentLoaded

-0.79

 betweenstory

-0.78



-0.78

POSITIVE LOGITS

0.37

0.36

0.35

↵

0.32

one

0.30

0.29

 pend

0.27

Activations Density 0.093%

terms and phrases associated with specific authentication processes and methods

word beginnings

No Comments

No Known Activations