INDEX

Explanations

occurrences of the word "testing."

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

Ļª

-2.49

§

-2.42

¯

-2.35

Ĳ

-2.31

Ĵ

-2.28

Ĺ

-2.19

ĳ

-2.15

Ŀ

-1.98

ı

-1.95

ĸ

-1.95

POSITIVE LOGITS

 crop

1.88

 instrument

1.81

bird

1.79

birds

1.77

 apparatus

1.68

aments

1.65

 instruments

1.62

 sticks

1.57

styles

1.57

 area

1.56

Activations Density 0.104%

occurrences of the word "testing."

No Comments

No Known Activations