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

Clik

-0.59

endphp

-0.59

plotlib

-0.53

 NavController

-0.52

(&:

-0.52

copg

-0.51

 IconData

-0.51

 betweenstory

-0.48

riwal

-0.48

 JpaRepository

-0.47

POSITIVE LOGITS

 مُعرِّف

3.66

 معرف

0.97

 nahilalakip

0.88

 autorytatywna

0.85

 وصلة

0.74

uxxxx

0.67

 transfieras

0.54

 CreateTagHelper

0.48

ِّف

0.44

 هيا

0.44

Activations Density 0.000%

No Known Activations

This feature has no known activations.