INDEX

Explanations

terms related to distinguishing or differentiating between concepts or items

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

ı

-1.82

į

-1.75

ĨĴ

-1.72

±

-1.65

³

-1.63

ĸ

-1.62

ĳ

-1.51

Ľ

-1.51

¸

-1.49

Ķ

-1.48

POSITIVE LOGITS

ably

2.35

between

1.78

 between

1.74

Ã©ment

1.58

ingly

1.58

amente

1.57

iations

1.53

 groups

1.52

 possible

1.50

chy

1.50

Activations Density 0.053%

terms related to distinguishing or differentiating between concepts or items

No Comments

No Known Activations