INDEX

Explanations

references to other entities or topics mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

terms indicating additional details or considerations that highlight other subjects or elements

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

opus

-0.66

iasis

-0.66

ony

-0.65

ARS

-0.64

 Bride

-0.63

 ç¥ŀ

-0.63

-0.61

vet

-0.60

 EVER

-0.60

ossession

-0.60

POSITIVE LOGITS

worldly

1.62

quickShipAvailable

1.10

wise

1.09

 notable

0.94

 aspects

0.86

ypes

0.86

 factors

0.84

 considerations

0.83

etheless

0.82

 noteworthy

0.80

Activations Density 0.077%

references to other entities or topics mentioned in the text

terms indicating additional details or considerations that highlight other subjects or elements

No Comments

No Known Activations