INDEX

Explanations

references to specific entities and influential figures in various fields

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

脚注の使い方

-0.73

 للمعارف

-0.71

webElementXpaths

-0.69

PreferredItem

-0.67

Datuak

-0.66

<unused32>

-0.65

<unused42>

-0.65

<unused28>

-0.65

<unused3>

-0.65

<unused16>

-0.65

POSITIVE LOGITS

guisement

0.34

CRIPT

0.30

 keer

0.28

 ganas

0.25

วรร

0.24

hpp

0.24

 hiver

0.23

Děkuji

0.23

 second

0.23

Segunda

0.23

Activations Density 0.017%

references to specific entities and influential figures in various fields

No Comments

No Known Activations