INDEX

Explanations

references to gaming elements such as locations, characters, actions, and equipment within the context of a rich and detailed narrative

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

 angered

-0.68

tics

-0.68

qus

-0.67

rers

-0.66

Wolf

-0.66

essors

-0.65

 Valerie

-0.65

haven

-0.65

soDeliveryDate

-0.63

Alert

-0.63

POSITIVE LOGITS

 individually

1.31

 separately

1.06

 unique

0.89

 independently

0.87

 imaginable

0.86

 respective

0.86

 uniquely

0.84

 successive

0.82

 distinct

0.81

 equally

0.81

Activations Density 20.722%

references to gaming elements such as locations, characters, actions, and equipment within the context of a rich and detailed narrative

No Comments

No Known Activations