INDEX

Explanations

references to various types of risks and hazards

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

ries

-1.71

ribut

-1.66

sworth

-1.60

ching

-1.54

innie

-1.52

 dreams

-1.44

chez

-1.42

affe

-1.35

cknowled

-1.34

rapy

-1.33

POSITIVE LOGITS

 posed

1.62

ĻĤ

1.59

burg

1.57

 surrounding

1.55

creen

1.54

 challenge

1.49

 aren

1.48

ball

1.44

(`

1.44

uits

1.41

Activations Density 1.727%

references to various types of risks and hazards

No Comments

No Known Activations