INDEX

Explanations

numerical values or quantities mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to quantity or measurement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Seat

-0.71

 Tribune

-0.68

 Frontier

-0.67

Â·Â·

-0.63

 Realm

-0.62

 Mansion

-0.62

 passage

-0.61

Kor

-0.61

Blade

-0.60

 cradle

-0.60

POSITIVE LOGITS

 amounts

1.06

 amount

1.00

ngth

0.83

 quantities

0.80

 sums

0.78

encies

0.77

ume

0.76

entially

0.76

imei

0.74

pour

0.73

Activations Density 0.020%

numerical values or quantities mentioned in a text

references to quantity or measurement

No Comments

No Known Activations