INDEX

Explanations

instances of possession or ownership in the text

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

French or foreign language text

np_acts-logits-general · gemini-2.0-flash

archaic english or french technical terms

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

so

-0.94

di

-0.88

non

-0.88

con

-0.83

 long

-0.83

-0.81

set

-0.80

 normal

-0.80

per

-0.79

an

-0.79

POSITIVE LOGITS

 itſelf

1.34

 myſelf

1.29

 pleaſure

1.28

 Majefty

1.25

 greateſt

1.25

 houſe

1.25

 Houſe

1.21

 ſche

1.21

 ſever

1.20

 Diſ

1.19

Activations Density 0.035%

instances of possession or ownership in the text

French or foreign language text

archaic english or french technical terms

No Comments

No Known Activations