INDEX

Explanations

phrases indicating the presence of official statements or releases

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

official statements and releases

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

 betweenstory

-0.66

EndProject

-0.54

fromnode

-0.52

Rüyada

-0.52

ActivityCompat

-0.51

FunctionFlags

-0.50

قایناق‌لار

-0.49

Cyfeiriadau

-0.49

رشف

-0.47

 itſelf

-0.47

POSITIVE LOGITS

 statement

0.36

 quote

0.35

 समीक्षाओं

0.34

 quotes

0.33

kmäler

0.33

 समीक्षाएं

0.33

 tweeted

0.32

 δή

0.32

NOTIFICATION

0.32

 فريبيس

0.31

Activations Density 0.003%

phrases indicating the presence of official statements or releases

official statements and releases

No Comments

No Known Activations