INDEX

Explanations

references to programs, regulations, and principles categorized as 'basic'

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

igham

-1.02

isSpecialOrderable

-0.97

pload

-0.93

IDA

-0.91

ibal

-0.90

 Tycoon

-0.89

estern

-0.86

romeda

-0.86

ork

-0.82

arf

-0.79

POSITIVE LOGITS

 necessities

1.39

 tenets

1.28

lly

1.17

tenance

1.09

basic

1.07

essential

1.06

 arithmetic

1.02

 principles

1.00

gradient

0.99

 essentials

0.97

Activations Density 0.731%

references to programs, regulations, and principles categorized as 'basic'

No Comments

No Known Activations