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

-0.66

Rhestr

-0.63

 igång

-0.62

 enkelte

-0.59

 doInBackground

-0.58

 inconvénients

-0.58

 pareti

-0.58

 varandra

-0.58

 vooraf

-0.57

 Italij

-0.57

POSITIVE LOGITS

new

1.40

新的

1.23

 nueva

1.16

new

1.16

新

1.13

 nuevas

1.10

 nuevo

1.09

 baru

1.09

 nuevos

1.09

的新

1.08

Activations Density 0.102%

No Comments

No Known Activations