INDEX

Explanations

mentions of corrections or updates in articles

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

akuya

-0.61

stones

-0.59

omorph

-0.58

ometers

-0.56

goers

-0.55

vs

-0.54

ibles

-0.54

roma

-0.53

ologne

-0.53

mods

-0.53

POSITIVE LOGITS

 Updated

0.64

 headline

0.62

 reporting

0.60

 dispatch

0.59

Published

0.59

 REPORT

0.56

 WARN

0.55

 HuffPost

0.55

 article

0.55

 typo

0.55

Activations Density 8.186%

mentions of corrections or updates in articles

No Comments

No Known Activations