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

hpur

-0.36

 roja

-0.30

Спољашње

-0.28

sem

-0.28

 सा

-0.27

 gefü

-0.27

heits

-0.27

big

-0.26

={`

-0.26

 HasFactory

-0.26

POSITIVE LOGITS

OGND

0.73

ValueStyle

0.71

 AssemblyCompany

0.69

AISSEE

0.65

 насељу

0.60

 townhouse

0.59

AddTagHelper

0.57

 الحره

0.57

MessageState

0.55



0.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.