INDEX

Explanations

physical descriptions of characters and settings in a narrative context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Anthem

-0.91

ratulations

-0.82

 subsequ

-0.81

plet

-0.77

channelAvailability

-0.75

soDeliveryDate

-0.72

idav

-0.71

 dispatched

-0.69

rollers

-0.69

usable

-0.68

POSITIVE LOGITS

 penchant

0.82

Ö¼

0.79

pse

0.79

bryce

0.75

anguages

0.74

 phil

0.74

 smoker

0.73

acus

0.73

elson

0.72

 intellect

0.72

Activations Density 17.561%

physical descriptions of characters and settings in a narrative context

No Comments

No Known Activations