INDEX

Explanations

references to collisions, particularly in the context of accidents or incidents

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 slightest

-2.33

 sting

-1.63

"?"

-1.63

 wake

-1.59

itrile

-1.59

 latter

-1.57

illed

-1.52

 deadly

-1.48

 ease

-1.47

amethasone

-1.47

POSITIVE LOGITS

ienn

1.83

ateral

1.72

ariate

1.71

ist

1.70

enson

1.69

ained

1.61

witz

1.61

hart

1.59

agen

1.59

iano

1.57

Activations Density 0.167%

references to collisions, particularly in the context of accidents or incidents

No Comments

No Known Activations