INDEX

Explanations

terms and phrases related to corruption

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

į

-1.96

ĥ½

-1.88

»¿

-1.86

MOESM

-1.86

woke

-1.83

ĺ

-1.79

Į

-1.75

lla

-1.66

Ĺ

-1.64

Ķ

-1.63

POSITIVE LOGITS

 schedules

1.86

icum

1.81

 tasks

1.81

 regimes

1.79

nier

1.77

ically

1.76

 limits

1.74

istically

1.69

 areas

1.68

ulent

1.68

Activations Density 0.051%

terms and phrases related to corruption

No Comments

No Known Activations