INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_22

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

 AssemblyProduct

-0.79

примеча

-0.71

MeasureSpec

-0.69

 &___

-0.64

 Barton

-0.64

 prohibido

-0.63

TSH

-0.61

ftagPool

-0.60

RTHOOK

-0.60

солю

-0.60

POSITIVE LOGITS

nav

1.71

 navigation

1.68

 navig

1.59

NAV

1.51

Nav

1.50

 Navigation

1.48

Nav

1.48

 Navig

1.48

nav

1.45

 navigate

1.45

Activations Density 0.016%

No Comments

No Known Activations