INDEX

Explanations

instances of the word "on" and related terms indicating activation or features of a system or application

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

ftagPool

-0.69

󠁣

-0.69

 kaarangay

-0.69

adaptiveStyles

-0.68

tanleria

-0.67

IndentedString

-0.66

DockStyle

-0.66

rrggbb

-0.66

 addon

-0.65

ніципалі

-0.64

POSITIVE LOGITS

Otherwise

0.35

 pulseira

0.34

 Otherwise

0.33

 OTHERWISE

0.31

 Erfol

0.31

 Benutz

0.30

 llorando

0.30

eriksaan

0.29

 aguja

0.28

 Spaß

0.28

Activations Density 0.001%

instances of the word "on" and related terms indicating activation or features of a system or application

No Comments

No Known Activations