INDEX

Explanations

content related to opinion polls and survey data

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

 Bomber

-0.55

ITNESS

-0.54

 conviction

-0.53

izons

-0.51

[|

-0.49

 taxing

-0.49

 Neptune

-0.49

SIZE

-0.49

oter

-0.49

ritional

-0.48

POSITIVE LOGITS

 extraord

0.83

hip

0.82

ial

0.77

hips

0.72

itatively

0.72

 specializing

0.67

ially

0.66

haw

0.65

andum

0.63

hire

0.63

Activations Density 12.375%

content related to opinion polls and survey data

No Comments

No Known Activations