INDEX

Explanations

references to instances in a programming context

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

"instance"

np_acts-logits-general · gemini-2.0-flash

object instance

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

getLeft

-0.49

ButtonModule

-0.48

 Educação

-0.48

躇

-0.48

Ethical

-0.48

 élector

-0.47

Welfare

-0.47

olyb

-0.47

Selamat

-0.46

ByEmail

-0.46

POSITIVE LOGITS

 Instance

1.19

Instance

1.18

instance

1.16

 instance

1.13

 INSTANCE

1.05

INSTANCE

0.98

 getInstance

0.98

getInstance

0.94

 Instances

0.92

 instances

0.89

Activations Density 0.059%

references to instances in a programming context

"instance"

object instance

No Comments

No Known Activations