INDEX

Explanations

phrases related to dishonesty and deception

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 realised

-0.78

 Originally

-0.69

cohol

-0.68

 realise

-0.66

 util

-0.66

sufficient

-0.64

 Secondly

-0.64

 Extras

-0.64

ORPG

-0.63

 differed

-0.62

POSITIVE LOGITS

 boycot

0.95

 legislatures

0.94

 ALEC

0.92

 defund

0.91

 congressional

0.90

 scandals

0.89

 lobbying

0.87

 draconian

0.87

 complicit

0.87

 Congressional

0.85

Activations Density 1.118%

phrases related to dishonesty and deception

No Comments

No Known Activations