INDEX

Explanations

technical terms related to depth of field and visual effect adjustments

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

ろ

-0.38

 commission

-0.37

 connection

-0.35

LookAnd

-0.35

 Commission

-0.35

lac

-0.35

BC

-0.35

 lần

-0.34

διο

-0.34

 flyg

-0.33

POSITIVE LOGITS

 Photocase

0.54

Personensuche

0.54

Einzelnachweise

0.51

 للاسماء

0.51

InitVars

0.50

tanleria

0.50

 ModelExpression

0.49

 ویکی‌پدی

0.49

 surla

0.48

/******/

0.45

Activations Density 0.022%

technical terms related to depth of field and visual effect adjustments

No Comments

No Known Activations