INDEX

Explanations

instances of the word "on" and its variations, indicating a focus on occurrences or prepositions related to a setting or condition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

parsedMessage

-0.80

endregion

-0.76

 DiCaprio

-0.70

 CURIAM

-0.70

łowa

-0.69

 useRef

-0.68

 iParam

-0.67

 ་་

-0.66

 ―――――

-0.65

adecimal

-0.65

POSITIVE LOGITS

On

1.37

on

1.34

ON

1.29

 behalf

1.22

On

1.16

 وعلى

1.03

isOn

1.02

addOn

1.01

 на

0.92

På

0.91

Activations Density 0.348%

instances of the word "on" and its variations, indicating a focus on occurrences or prepositions related to a setting or condition

No Comments

No Known Activations