INDEX

Explanations

references to roles and positions within organizations or teams

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

various positions, roles, or places

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

 Мексичка

-0.45

-0.39

ainville

-0.39

Personendaten

-0.37

Чыганаклар

-0.36

InjectMocks

-0.36

ToHex

-0.36

reihe

-0.35

 الحره

-0.35

cstdio

-0.35

POSITIVE LOGITS

 venues

0.51

ImageContext

0.48

 dAtA

0.48

@"/

0.47

 forskellige

0.46

 ThemeData

0.43

 statewide

0.42

 olika

0.42

 NSCoder

0.42

 locais

0.41

Activations Density 0.010%

references to roles and positions within organizations or teams

various positions, roles, or places

No Comments

No Known Activations