INDEX

Explanations

terms related to augmented or enhanced experiences and aggravated situations

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

ustr

-1.49

cca

-1.44

mar

-1.39

se

-1.39

lli

-1.38

semb

-1.37

 celebrated

-1.36

ellow

-1.35

yd

-1.34

lop

-1.32

POSITIVE LOGITS

IBLE

1.41

(>

1.31

RF

1.28

(/

1.27

ival

1.27

 reality

1.26

 fractions

1.26

emption

1.25

rez

1.25

 force

1.25

Activations Density 0.012%

terms related to augmented or enhanced experiences and aggravated situations

No Comments

No Known Activations