INDEX

Explanations

phrases that indicate levels of assurance or quality in experiences

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

written or descriptive phrases

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

-0.63

 فريبيس

-0.54

CloseOperation

-0.49

pushFollow

-0.43

setVerticalGroup

-0.43

 kasarigan

-0.40

 defaultstate

-0.39

findpost

-0.39

extAlignment

-0.39

 intptr

-0.38

POSITIVE LOGITS

audiovisuel

0.40

äuser

0.40

 leão

0.38

 gwaran

0.38

éraires

0.38

WritableDatabase

0.37

ならではの

0.37

 bleue

0.37

delingen

0.35

quoi

0.35

Activations Density 0.022%

phrases that indicate levels of assurance or quality in experiences

written or descriptive phrases

No Comments

No Known Activations