INDEX

Explanations

mentions of tasks and the processes associated with them

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-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.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

¿½

-2.37

Ĳ

-2.35

§

-2.23

ı

-2.13

Ħ

-2.10

ĨĴ

-1.99

áĢ¯

-1.95

Ĵ

-1.95

Ī

-1.91

Ń

-1.90

POSITIVE LOGITS

master

2.21

horse

1.92

ings

1.85

ahead

1.83

 done

1.79

ing

1.77

kill

1.69

 requiring

1.59

etable

1.58

 processor

1.54

Activations Density 0.137%

mentions of tasks and the processes associated with them

No Comments

No Known Activations