INDEX

Explanations

keywords related to yoga

oai_token-act-pair · gpt-3.5-turbo

references to yoga and related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hyde

-0.82

etary

-0.78

lies

-0.73

vous

-0.72

ictional

-0.71

strous

-0.70

alties

-0.70

osal

-0.69

ilial

-0.69

————————————————

-0.68

POSITIVE LOGITS

oga

0.97

 yoga

0.93

nas

0.86

 pants

0.85

 meditation

0.83

 mats

0.83

 Yoga

0.81

 guru

0.80

 instructor

0.77

yog

0.75

Activations Density 0.015%

keywords related to yoga

references to yoga and related terms

No Comments

No Known Activations