INDEX

Explanations

terms related to statistical measures and configurations

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

Text immediately followed by parentheses

np_acts-logits-general · gemini-2.0-flash

opening parenthesis or symbol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

["

-0.55

]["

-0.54

"]="

-0.51

['

-0.49

 useStyles

-0.49

="

-0.47

'@/

-0.47

“

-0.45

\_

-0.44

 welfare

-0.43

POSITIVE LOGITS

$(

1.29

$(

1.08

}(

0.94

($(

0.90

=$(

0.85

(<

0.84

afficheront

0.83

Personensuche

0.82

Personendaten

0.81

 Roskov

0.80

Activations Density 1.315%

terms related to statistical measures and configurations

Text immediately followed by parentheses

opening parenthesis or symbol

No Comments

No Known Activations