INDEX

Explanations

references to legislative acts and specific sections within them

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ragon

-1.10

flush

-0.94

acles

-0.93

prototype

-0.92

haps

-0.88

unci

-0.88

ezvous

-0.88

amel

-0.88

interstitial

-0.87

rums

-0.87

POSITIVE LOGITS

 TODAY

1.47

 Today

1.07

ESSION

1.04

ICAN

0.99

ITE

0.94

ctions

0.93

ID

0.93

 Airlines

0.93

BILITY

0.91

 Patriot

0.90

Activations Density 5.400%

references to legislative acts and specific sections within them

No Comments

No Known Activations