INDEX

Explanations

words related to individuals appearing in various contexts

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

ricular

-1.17

cise

-1.03

Ã§a

-1.01

EStreamFrame

-0.97

rient

-0.97

dp

-0.94

thur

-0.94

estern

-0.93

rique

-0.93

iding

-0.91

POSITIVE LOGITS

iary

0.97

Pause

0.96

 prominently

0.92

 contradictions

0.86

Magikarp

0.86

DragonMagazine

0.85

ãĤ¤ãĥĪ

0.84

ively

0.81

BACK

0.80

lihood

0.79

Activations Density 0.817%

words related to individuals appearing in various contexts

No Comments

No Known Activations