INDEX

Explanations

terms related to total counts or measurements

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

"Total" followed by a noun

np_acts-logits-general · gemini-2.0-flash

total followed by another word

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

InputTagHelper

-0.79

nezeu

-0.72

Архівовано

-0.71

OGND

-0.64

 setTimeout

-0.61

 indexPath

-0.60

csolódó

-0.59

ruptedException

-0.58

})*/

-0.57

Kaynakça

-0.56

POSITIVE LOGITS

itarianism

0.78

Ener

0.77

 amount

0.65

izers

0.64

izator

0.62

sum

0.58

izador

0.57

istic

0.57

dhi

0.56

 मिला

0.56

Activations Density 0.090%

terms related to total counts or measurements

"Total" followed by a noun

total followed by another word

No Comments

No Known Activations