INDEX

Explanations

punctuation marks, specifically commas

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

IFrame

Link

Negative Logits

ļ

-1.94

¦

-1.82

³

-1.67

issance

-1.53

elves

-1.49

thems

-1.48

Ĥ

-1.45

 mates

-1.42

 themselves

-1.42

ANS

-1.37

POSITIVE LOGITS

ype

1.50

 citation

1.48

[$

1.45

otoxin

1.44

size

1.41

roduction

1.41

stat

1.37

 summary

1.36

icki

1.35

roe

1.34

Activations Density 0.033%

punctuation marks, specifically commas

No Known Activations