INDEX

Explanations

references to ham or related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

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

Ĩ

-1.67

ĨĴ

-1.66

ĳ

-1.64

ĥ

-1.63

·¸

-1.61

 circumstance

-1.58

verting

-1.58

 support

-1.52

ĵ

-1.52

OPLE

-1.52

POSITIVE LOGITS

ilton

2.74

mered

2.50

ster

1.91

itage

1.90

noon

1.85

thor

1.82

sters

1.82

stead

1.77

pshire

1.75

pering

1.73

Activations Density 0.135%

references to ham or related terms

No Comments

No Known Activations