INDEX

Explanations

references to uniformity

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

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

deal

-1.56

rs

-1.55

tica

-1.51

Whe

-1.36

 vain

-1.36

yan

-1.36

ath

-1.35

 Sleep

-1.31

chanical

-1.31

qui

-1.30

POSITIVE LOGITS

ized

2.06

iz

1.96

iza

1.88

istics

1.87

izability

1.86

izo

1.80

 ones

1.77

izer

1.72

izing

1.71

iative

1.66

Activations Density 0.013%

references to uniformity

No Comments

No Known Activations