INDEX

Explanations

programming and coding-related terms and functions

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

 normal

-0.39

 detail

-0.39

Detail

-0.37

detail

-0.35

 Conſ

-0.35

 Reſ

-0.35

Ze

-0.35

 норма

-0.35

 Venise

-0.34

 Detail

-0.34

POSITIVE LOGITS

 Biôgrafia

0.71

AsUp

0.68

Personensuche

0.67

OGND

0.67

 ویکی‌پدیا

0.66

 nahilalakip

0.66

 المعيارى

0.66

BufferException

0.66

 Picchu

0.62

Искәрмәләр

0.62

Activations Density 0.028%

programming and coding-related terms and functions

No Comments

No Known Activations