INDEX

Explanations

terms related to being free from specific conditions or constraints

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

Negative Logits

sen

-1.55

imus

-1.53

 Subject

-1.51

 anim

-1.49

 horns

-1.44

Age

-1.41

riber

-1.39

 Examin

-1.38

 whom

-1.37

 irrelevant

-1.35

POSITIVE LOGITS

ypse

2.13

zing

1.89

ftware

1.89

Ã§Ã£o

1.80

Å£

1.76

ondon

1.68

"}](#

1.65

ghan

1.64

 luck

1.63

eland

1.63

Activations Density 0.178%

terms related to being free from specific conditions or constraints

No Known Activations