INDEX

Explanations

instances of the word "recap"

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

ĭ

-2.24

ı

-1.87

nez

-1.82

ī

-1.70

§

-1.69

ĥ½

-1.62

ESULT

-1.61

ances

-1.60

į

-1.59

¼

-1.57

POSITIVE LOGITS

it

2.10

ital

1.77

itet

1.55

atory

1.55

ulous

1.54

ository

1.48

itor

1.47

itals

1.46

athi

1.45

 apart

1.44

Activations Density 1.648%

instances of the word "recap"

No Comments

No Known Activations