INDEX

Explanations

mathematical notations and symbols related to matrices and their properties

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

Superscript T, often for transpose

np_acts-logits-general · gemini-2.0-flash

matrix transpose

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 subsid

-0.48

gcd

-0.47

subsid

-0.47

 condition

-0.45

 ля

-0.44

σκ

-0.42

消

-0.42

หาย

-0.42

usty

-0.42

verwijspagina

-0.42

POSITIVE LOGITS

 transpose

1.01

 transposed

0.76

transpose

0.75

 transposition

0.73

Transpose

0.70

^\

0.70

}^\

0.68

 trans

0.66

 AspNetCore

0.60

ᵀ

0.59

Activations Density 0.713%

mathematical notations and symbols related to matrices and their properties

Superscript T, often for transpose

matrix transpose

No Comments

No Known Activations