INDEX

Explanations

elements indicative of structured data or references within legal or technical documents

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

 noDo

-0.81

ftagPool

-0.70

 autorytatywna

-0.70

IsMutable

-0.69

PerformLayout

-0.69

:✨

-0.66

 שוליים

-0.65

ItemBackground

-0.65

Personendaten

-0.65

findpost

-0.63

POSITIVE LOGITS

ティス

0.33

ud

0.29

()?.

0.29

idUser

0.28

 koffie

0.28

/**

0.27

驭

0.27

§

0.27

Pe

0.27

Activations Density 0.144%

elements indicative of structured data or references within legal or technical documents

No Comments

No Known Activations