INDEX

Explanations

frequent references to a specific configuration or settings within a programming context

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

Inflater

-1.47

 Universe

-1.32

 Docket

-1.32

entary

-1.32

Ã¨me

-1.30

."

-1.27

 cares

-1.25

'"

-1.25

'">

-1.24

 Broadcasting

-1.24

POSITIVE LOGITS

Ĵ

1.79

him

1.77

abouts

1.64

ity

1.62

»

1.54

¨

1.53

avirus

1.50

ifiable

1.49

bler

1.46

 sure

1.46

Activations Density 0.079%

frequent references to a specific configuration or settings within a programming context

No Comments

No Known Activations