INDEX

Explanations

phrases related to confidentiality and disclosure of information

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

 artikelen

-0.49

horloge

-0.42

 Flasche

-0.41

jaardag

-0.41

opération

-0.39

actualité

-0.39

amitié

-0.39

 pikir

-0.38

Égypte

-0.37

 sanciones

-0.37

POSITIVE LOGITS

 disclosed

0.77

 disclosure

0.75

 disclose

0.74

 disclosing

0.71

 Disclosure

0.71

 discloses

0.65

 disclosures

0.62

disclosure

0.58

localctx

0.56

 revelation

0.56

Activations Density 0.181%

phrases related to confidentiality and disclosure of information

No Comments

No Known Activations