INDEX

Explanations

words related to licensing and legal authorization

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

»

-2.67

ĨĴ

-2.62

£

-2.61

ĺ

-2.49

®

-2.42

Ł

-2.41

ļ

-2.38

¨

-2.38

§

-2.37

Ĺ

-2.18

POSITIVE LOGITS

 license

2.25

 licenses

2.10

 License

2.04

icense

2.02

license

1.90

 pursuant

1.77

 LICENSE

1.75

 exemptions

1.73

 trademarks

1.71

 licence

1.70

Activations Density 0.079%

words related to licensing and legal authorization

No Known Activations