INDEX

Explanations

programming-related keywords and concepts, particularly in the context of classes, methods, and configuration

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

¿½

-1.79

ľĵ

-1.69

ľ

-1.68

«

-1.61

'</

-1.60

[[

-1.57

ĻĤ

-1.53

¥

-1.52

ģ

-1.49

£

-1.48

POSITIVE LOGITS

eer

1.89

 Division

1.82

 Authority

1.80

 Enforcement

1.73

lords

1.72

naire

1.69

 Manager

1.69

 Bureau

1.66

 Commission

1.63

 Corps

1.61

Activations Density 1.230%

programming-related keywords and concepts, particularly in the context of classes, methods, and configuration

No Comments

No Known Activations