INDEX

Explanations

words associated with difficulty or hardship

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 miniaturka

-0.52

WebElementEntity

-0.44

TagMode

-0.44



-0.41

новништво

-0.41

expliquer

-0.41

-0.40

 MaterialApp

-0.40

 StatelessWidget

-0.40

Statistiche

-0.38

POSITIVE LOGITS

 kasarigan

0.53

stdc

0.40

menea

0.39

ppchen

0.39

WebServlet

0.38

 Référence

0.38

Хьажоргаш

0.38

CppCodeGen

0.38

サラ

0.37

$_['

0.36

Activations Density 0.002%

words associated with difficulty or hardship

No Comments

No Known Activations