INDEX

Explanations

specific quantities or levels of something

oai_token-act-pair · gpt-3.5-turbo

references to quantities or amounts

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

 Frontier

-0.73

 Seat

-0.71

 cradle

-0.69

 Tribune

-0.64

Kor

-0.64

 Pilgrim

-0.63

 passage

-0.62

hall

-0.60

 Phant

-0.59

 Pelicans

-0.59

POSITIVE LOGITS

 amounts

1.02

 amount

0.99

ngth

0.88

pour

0.82

 quantities

0.81

encies

0.78

ume

0.74

imei

0.73

ucha

0.72

 sums

0.71

Activations Density 0.021%

specific quantities or levels of something

references to quantities or amounts

No Comments

No Known Activations