INDEX

Explanations

punctuation marks and sentence endings

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

end of sentence/turn

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

 Biôgrafia

-0.80

:✨

-0.65

principalColumn

-0.65

httphttps

-0.65

hyrchwyd

-0.62

الحياه

-0.61

 Numerade

-0.59

 Infórmanos

-0.59

省市镇

-0.56

RenderAtEndOf

-0.55

POSITIVE LOGITS

 Weiss

0.41

 single

0.41

CloseOperation

0.39

 Este

0.39

wrapped

0.38

Yet

0.37

 Brito

0.37

하세요

0.37

 White

0.37

 beautiful

0.37

Activations Density 0.042%

punctuation marks and sentence endings

end of sentence/turn

No Comments

No Known Activations