INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

itm

-0.58

STRIP

-0.44

 Infórmanos

-0.39

#(

-0.35

 gewor

-0.35

 ditemp

-0.35

마

-0.34

toHaveBeen

-0.34

thrown

-0.33

貼

-0.33

POSITIVE LOGITS

بالإنجليزية

2.61

انگلیسی

1.99

انجليز

1.70

 انگلیسی

1.41

لیس

1.18

 anglès

1.06

באנגלית

0.93

 English

0.84

english

0.81

영어

0.80

Activations Density 0.000%

No Known Activations

This feature has no known activations.