INDEX

Explanations

mentions of significant events and their implications

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

 sintético

-0.28

 bás

-0.28

źni

-0.26

 prieten

-0.25

 Einkauf

-0.25

 masculinos

-0.25

 masculino

-0.24

 héroe

-0.23

 noires

-0.22

bootstrapcdn

-0.22

POSITIVE LOGITS

 Roskov

0.80

:✨

0.76

 الرياضيه

0.68

 censiti

0.68

 surla

0.65

AddTagHelper

0.65

 Wicidata

0.64

 Numerade

0.64

تقاوى

0.63

 مواليد

0.62

Activations Density 0.005%

mentions of significant events and their implications

No Comments

No Known Activations