INDEX

Explanations

instances of the word "cogn."

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

Negative Logits

'?

-1.79

'];

-1.78

'?"

-1.77

hing

-1.72

"];

-1.68

wd

-1.66

ances

-1.63

?'

-1.61

']

-1.60

';

-1.58

POSITIVE LOGITS

 reader

1.69

icians

1.65

ician

1.62

etics

1.61

nier

1.60

aic

1.56

owitz

1.54

caster

1.52

 genre

1.44

iche

1.39

Activations Density 0.011%

instances of the word "cogn."

No Known Activations