INDEX

Explanations

instances of specialized vocabulary and technical terms

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

preceding uses or abilities

np_acts-logits-general · gemini-2.0-flash

how to do things

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 nastav

-0.34

 Haushalt

-0.30

 trial

-0.30

–

-0.28

 doğum

-0.28

Fla

-0.27

Garantía

-0.27

 sold

-0.27

함

-0.27

zusetzen

-0.27

POSITIVE LOGITS

 للمعارف

0.81

httphttps

0.72

 disambiguazione

0.71



0.70

+#+#

0.68

Gön

0.67

WithIOException

0.66

AsUp

0.66

ContentAsync

0.65



0.65

Activations Density 1.723%

instances of specialized vocabulary and technical terms

preceding uses or abilities

how to do things

No Comments

No Known Activations