INDEX

Explanations

references to quantities, especially involving counts or numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

PreferredItem

-0.75

ItemBackground

-0.66

yorsunuz

-0.63

 asemenea

-0.62

 Reverso

-0.61

 useRouter

-0.59

 proprement

-0.59

 vraie

-0.58

borderSide

-0.58

 richting

-0.58

POSITIVE LOGITS

 report

0.78

 survey

0.72

 recent

0.65

 unnamed

0.65

 delegation

0.63

 group

0.63

 trio

0.62

unnamed

0.62

 pair

0.62

 study

0.61

Activations Density 0.487%

references to quantities, especially involving counts or numerical values

No Comments

No Known Activations