INDEX

Explanations

references to specific personal objects or possessions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

dding

-0.82

 Sheldon

-0.77

Madison

-0.76

grid

-0.76

Ĥ¬

-0.73

justice

-0.70

 Christy

-0.70

WARD

-0.70

noon

-0.69

POSITIVE LOGITS

 versatile

0.89

 originated

0.83

 purported

0.82

 works

0.81

'll

0.80

 lowly

0.80

 creators

0.79

 excel

0.79

 decom

0.78

 replacement

0.77

Activations Density 7.420%

references to specific personal objects or possessions

No Comments

No Known Activations