INDEX

Explanations

numerical values and their representations in scientific notation

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

สือ

-0.66

 Cordero

-0.65

pædia

-0.64

abank

-0.64

odeon

-0.63

 NSCoder

-0.63

indis

-0.60

 blockade

-0.60

 dataSnapshot

-0.60

koin

-0.60

POSITIVE LOGITS

1.49

 Seven

0.82

７

0.79

 seven

0.77

۷

0.75

 SEVEN

0.75

 zeven

0.72

0.71

Seven

0.70

Activations Density 0.642%

numerical values and their representations in scientific notation

No Comments

No Known Activations