INDEX

Explanations

geographical locations, especially in the context of islands and communities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

artz

-0.60

agate

-0.59

 Bers

-0.59

thood

-0.56

zig

-0.55

yes

-0.55

ascript

-0.54

otation

-0.54

arella

-0.52

etics

-0.52

POSITIVE LOGITS

 Territories

0.73

 AMERICA

0.70

 Coast

0.62

west

0.61

 populated

0.58

 Crossing

0.57

 ridge

0.57

bound

0.55

 coast

0.55

side

0.55

Activations Density 6.066%

geographical locations, especially in the context of islands and communities

No Comments

No Known Activations