INDEX

Explanations

references to a specific rating system or score

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

eing

-1.70

 nature

-1.66

cknowled

-1.65

ance

-1.49

 logarithm

-1.46

effectiveness

-1.46

 sudden

-1.45

 logar

-1.45

 _________

-1.42

ferences

-1.40

POSITIVE LOGITS

ios

2.36

aku

1.80

cliffe

1.80

iga

1.69

chet

1.69

fire

1.68

keeper

1.67

Ã©g

1.64

atÃ³

1.62

nat

1.61

Activations Density 0.125%

references to a specific rating system or score

No Comments

No Known Activations