INDEX

Explanations

words related to legal cases or court statements

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

bis

-0.74

 LINE

-0.73

 combust

-0.65

 hypert

-0.65

 Helsinki

-0.64

 Dani

-0.63

ften

-0.60

 Constantin

-0.60

 Stockholm

-0.60

 illuminating

-0.58

POSITIVE LOGITS

oin

1.00

ience

1.00

iences

0.98

eatures

0.92

din

0.92

iated

0.84

icates

0.84

alties

0.83

enos

0.82

iments

0.81

Activations Density 7.896%

words related to legal cases or court statements

No Comments

No Known Activations