INDEX

Explanations

sharpness or clarity in descriptions or critiques

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

Not in Any Lists

No Comments

Negative Logits

 present

-1.38

 particular

-1.36

hop

-1.34

amination

-1.34

 monolayer

-1.33

 Catalog

-1.32

Xml

-1.31

 Garden

-1.29

 moss

-1.25

onas

-1.24

POSITIVE LOGITS

est

2.21

ened

1.92

ness

1.90

teen

1.86

abouts

1.74

stown

1.71

ings

1.69

headed

1.69

 minded

1.64

ening

1.63

Activations Density 0.019%

sharpness or clarity in descriptions or critiques

No Comments

No Known Activations