INDEX

Explanations

information related to historical and technical details, particularly about specific individuals and developments

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

 POLITICO

-0.71

Leaks

-0.69

":"/

-0.68

BIP

-0.66

 âľĶ

-0.66

ggle

-0.64

Avoid

-0.64

gdala

-0.63

 WATCH

-0.62

CNN

-0.62

POSITIVE LOGITS

 pioneer

0.74

 missionaries

0.74

0.71

 famine

0.70

yrights

0.70

 fame

0.70

 missionary

0.70

çĶ°

0.69

 UNESCO

0.69

0.68

Activations Density 7.089%

information related to historical and technical details, particularly about specific individuals and developments

No Comments

No Known Activations