INDEX

Explanations

codes or technical information in text

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

keyes

-0.69

keye

-0.62

ophone

-0.61

avorite

-0.59

 risked

-0.57

 ancest

-0.57

iston

-0.55

 scratched

-0.54

anova

-0.53

cells

-0.52

POSITIVE LOGITS

å¹

0.82

 partName

0.68

ãĤ¡

0.64

0.63

çīĪ

0.62

å§«

0.62

0.61

0.60

Activations Density 11.927%

codes or technical information in text

No Comments

No Known Activations