INDEX

Explanations

references to personal relationships and acquaintances

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

know people personally

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

სქოლიო

-0.42

 Flü

-0.40

hésite

-0.39

 AspNetCore

-0.38

 informée

-0.38

 ModelExpression

-0.37

最快更新

-0.36

ollectionView

-0.36

 mindset

-0.35

POSITIVE LOGITS

 vänner

0.52

CrossRef

0.52

 friends

0.52

 fubject

0.47

 nära

0.44

yyr

0.44

 friend

0.43

 familières

0.43

Източници

0.43

Familiar

0.42

Activations Density 0.145%

references to personal relationships and acquaintances

know people personally

No Comments

No Known Activations