INDEX

Explanations

references to people within various contexts

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

people's needs and actions

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

Condol

-0.44

↕

-0.41

RenderAtEndOf

-0.40

 EconPapers

-0.37

 ligiloj

-0.35

韌

-0.34

 Initialized

-0.34

InjectAttribute

-0.34

nameof

-0.33

 arşivlendi

-0.33

POSITIVE LOGITS

Everyone

0.71

Everybody

0.70

 Everyone

0.62

 Everybody

0.61

everyone

0.59

everybody

0.59

Somebody

0.59

Anyone

0.57

People

0.55

 Zweifel

0.55

Activations Density 0.008%

references to people within various contexts

people's needs and actions

No Comments

No Known Activations