INDEX

Explanations

privacy-related access modifiers in code

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

access control keywords

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

adaptiveStyles

-0.83

 queſta

-0.79

TypedDataSet

-0.72

HomeAsUpEnabled

-0.70

TagMode

-0.69

بوابة

-0.64

 ddelweddau

-0.64

/**

-0.64

DotNetBar

-0.62



-0.61

POSITIVE LOGITS

private

0.46

fileprivate

0.44

ainte

0.43

 noted

0.42

0.41

 knelt

0.40

 dedicated

0.38

 gathered

0.38

 Kramer

0.38

 Bucure

0.38

Activations Density 0.001%

privacy-related access modifiers in code

access control keywords

No Comments

No Known Activations

privacy-related access modifiers in code

access control keywords

No Comments

No Known Activations