INDEX

Explanations

various punctuation marks

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

user input code

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

migrationBuilder

-0.57

SuccessListener

-0.56

exitRule

-0.55

MigrationBuilder

-0.54

 FetchType

-0.52

DockStyle

-0.52

Personendaten

-0.52

 onAnimation

-0.49

webElementXpaths

-0.49

yarnpkg

-0.49

POSITIVE LOGITS

 الحره

0.46

 nahilalakip

0.46

脚注の使い方

0.41

Према

0.40

 Leaven

0.34

Bibliograf

0.33

期刊论文

0.33

 gruesa

0.32

 complejidad

0.32

roth

0.32

Activations Density 0.000%

various punctuation marks

user input code

No Comments

No Known Activations