INDEX

Explanations

phrases related to controlling the power state of devices, particularly turning them on and off

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

+#+

-0.44

 незавершена

-0.43

 вікісторінку

-0.41

bootstrapcdn

-0.40

Pyx

-0.40

 Kram

-0.39

 Meksiku

-0.38

kuper

-0.38

grund

-0.37

 vuitton

-0.37

POSITIVE LOGITS

 Disable

0.78

 OnEnable

0.77

 disable

0.72

 isEnabled

0.71

Disable

0.67

 activate

0.66

 disabled

0.66

Enabled

0.66

 Activate

0.65

disable

0.65

Activations Density 0.254%

phrases related to controlling the power state of devices, particularly turning them on and off

No Comments

No Known Activations