INDEX

Explanations

phrases related to hidden or confidential information or details

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

retty

-0.59

adic

-0.52

erate

-0.51

ucc

-0.51

onement

-0.50

member

-0.49

stad

-0.49

mun

-0.49

oun

-0.47

allah

-0.47

POSITIVE LOGITS

 glean

0.82

 secrets

0.66

 abound

0.64

 Learned

0.64

 Secrets

0.63

pin

0.62

mith

0.60

linger

0.60

 hidden

0.60

cape

0.58

Activations Density 12.455%

phrases related to hidden or confidential information or details

No Comments

No Known Activations