INDEX

Explanations

the word "ace."

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

ĨĴ

-2.27

¿½

-2.09

Ĥ

-1.94

ı

-1.78

ĵ

-1.72

soever

-1.71

")]

-1.69

Ĺ

-1.64

Ĭ

-1.63

Ģ

-1.60

POSITIVE LOGITS

borg

1.76

burg

1.72

ÅĽci

1.70

pora

1.69

bilt

1.66

coin

1.65

holder

1.62

kowski

1.61

based

1.53

Rptr

1.53

Activations Density 0.071%

the word "ace."

No Comments

No Known Activations