INDEX

Explanations

mentions of media and related terminology

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

Ĳ

-2.32

ľĵ

-2.20

§

-2.16

Į

-2.14

ĳ

-1.97

ĩ

-1.96

¤

-1.89

Ķ

-1.82

ĥ½

-1.81

ķ

-1.79

POSITIVE LOGITS

eval

2.22

wiki

2.16

film

2.08

ieux

1.89

ural

1.87

films

1.87

works

1.81

 outlets

1.70

processor

1.69

ulator

1.66

Activations Density 0.092%

mentions of media and related terminology

No Comments

No Known Activations