INDEX

Explanations

terms related to research, education, and specific conditions or phenomena

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

ĥ

-1.71

Ģ

-1.57

Ļª

-1.53

 ability

-1.51

 occasional

-1.49

ctomy

-1.48

 coli

-1.48

 willingness

-1.46

eping

-1.43

ters

-1.37

POSITIVE LOGITS

inx

1.79

umerable

1.75

 though

1.54

wait

1.53

iÃ¨re

1.53

iously

1.48

yet

1.46

olecular

1.44

undo

1.44

it

1.43

Activations Density 0.017%

terms related to research, education, and specific conditions or phenomena

No Comments

No Known Activations