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

𝟰

-0.33

ɵɵ

-0.33

hlungen

-0.33

✥

-0.33

ために

-0.33

typeorm

-0.32

 PeEnEo

-0.32

Ӕ

-0.32

mobileqq

-0.32

 lizenzfreie

-0.31

POSITIVE LOGITS

 loading

0.72

 viewDidLoad

0.66

ContentLoaded

0.63

 Loading

0.62

 LOADING

0.61

loading

0.57

脚注の使い方

0.54

 loadings

0.53

IBOutlet

0.52

 kasarigan

0.51

Activations Density 0.000%

No Known Activations

This feature has no known activations.