INDEX

Explanations

references to the United States

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

various technical and foreign terms

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

onAttach

-0.69

 betweenstory

-0.69

 queſta

-0.63

 MainAxisSize

-0.61

UnitTesting

-0.60

+#+

-0.59

⟬

-0.59

createStatement

-0.59

 Wikimedijinoj

-0.58

TProtocol

-0.58

POSITIVE LOGITS

makeConstraints

0.39

 VIDEOTAPE

0.36

 Италијани

0.35

 propOrder

0.35

WebElementEntity

0.33

 Roskov

0.33

 Taktlose

0.32

Anmerkungen

0.32

ⓧ

0.32

rawDesc

0.30

Activations Density 0.000%

references to the United States

various technical and foreign terms

No Comments

No Known Activations

references to the United States

various technical and foreign terms

No Comments

No Known Activations