INDEX

Explanations

terms related to legal processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

getDoctrine

-0.83

 itſelf

-0.82

存于互联网档案馆

-0.78

 Theſe

-0.77

دانشنامهٔ

-0.76

 صوتيه

-0.73

'),

-0.72

Искәрмәләр

-0.72

**/

-0.71

 Monfieur

-0.70

POSITIVE LOGITS

 whom

0.86

who

0.66

whom

0.64

who

0.55

 kteří

0.53

 którzy

0.52

 quienes

0.51

Whom

0.49

 akik

0.46

 hvem

0.45

Activations Density 0.858%

terms related to legal processes

No Comments

No Known Activations