INDEX

Explanations

terms related to statistical concepts and methodologies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æį®

-1.88

"}](#

-1.66

nai

-1.57

ignon

-1.46

âĶĢâĶĢâĶĢâĶĢ

-1.46

Sov

-1.42

ãĥ¼ãĥī

-1.39

Î¼ÎŃ

-1.38

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-1.37

ãĥ¥

-1.37

POSITIVE LOGITS

ses

1.94

 latter

1.78

 aforementioned

1.75

 following

1.66

 value

1.65

orems

1.59

 corresponding

1.59

 overall

1.56

 respectively

1.55

 validity

1.54

Activations Density 0.820%

terms related to statistical concepts and methodologies

No Comments

No Known Activations