INDEX

Explanations

references to modules in programming contexts

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

 MERCHANTABILITY

-2.04

unnumbered

-1.59

 love

-1.57

 behalf

-1.48

orems

-1.47

effectiveness

-1.45

lower

-1.43

 mere

-1.43

OOK

-1.41

gle

-1.39

POSITIVE LOGITS

ivated

1.94

finder

1.89

rait

1.81

builder

1.76

cles

1.73

 arsenal

1.70

nem

1.67

ematic

1.66

ontal

1.66

ict

1.66

Activations Density 0.366%

references to modules in programming contexts

No Comments

No Known Activations