INDEX

Explanations

the end of a turn in dialogue or text

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

IndentedString

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 polícia

-0.37

 investigação

-0.36

 police

-0.35

 secondaires

-0.34

 convite

-0.34

 supplémentaire

-0.34

convite

-0.33

inheritDoc

-0.32

 successivo

-0.32

 voisin

-0.31

POSITIVE LOGITS

IndentedString

0.71

ArrowToggle

0.66

 Савезне

0.64

 kasarigan

0.63

msgTypes

0.57

 Italijanski

0.57

QMetaType

0.57

__*/

0.56

Jereo

0.55

Datuak

0.54

Activations Density 0.000%

the end of a turn in dialogue or text

IndentedString

No Comments

No Known Activations

the end of a turn in dialogue or text

IndentedString

No Comments

No Known Activations