INDEX

Explanations

details related to legal documents and official procedures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

auga

-0.90

ellectual

-0.86

 shack

-0.78

 Democr

-0.77

 coord

-0.77

 ILCS

-0.76

Sly

-0.75

aila

-0.75

 resemb

-0.74

 indis

-0.74

POSITIVE LOGITS

mid

1.04

termin

1.02

asted

0.95

batch

0.93

scale

0.93

bury

0.93

based

0.91

cv

0.87

quartered

0.87

release

0.86

Activations Density 0.829%

details related to legal documents and official procedures

No Comments

No Known Activations