INDEX

Explanations

phrases related to normality and routine

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

 Administrativna

-0.63

SequentialGroup

-0.59

InjectAttribute

-0.56

MessageTagHelper

-0.55

 ujednoznacz

-0.54

IntoConstraints

-0.54

AddTagHelper

-0.54

richTextPanel

-0.53

bootstrapcdn

-0.52

Tembelea

-0.51

POSITIVE LOGITS

 rów

0.41

 doGet

0.40

 Económica

0.39

LEGGI

0.39

 zwykle

0.37

 gebru

0.37

saraba

0.37

mẫu

0.37

onActivityResult

0.36

 podjela

0.36

Activations Density 0.446%

phrases related to normality and routine

No Comments

No Known Activations