INDEX

Explanations

punctuation marks

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

user turn start

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

IonicModule

-0.74

出版年

-0.68

 ostavi

-0.49

 veze

-0.46

helmet

-0.37

 tả

-0.34

mpä

-0.27

 Schützen

-0.27

Composable

-0.26

getPage

-0.26

POSITIVE LOGITS

RectangleBorder

4.56

RRect

1.65

visor

1.48

 MainAxisSize

1.44

Reentrant

1.29

 BoxDecoration

1.07

 kaynağından

0.87

GraphicsUnit

0.85

addContainerGap

0.81

tagext

0.79

Activations Density 0.000%

punctuation marks

user turn start

No Comments

No Known Activations