INDEX

Explanations

legal terms or phrases related to laws and regulations

oai_token-act-pair · gpt-3.5-turbo

references to legal regulations and laws

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

ILLE

-0.73

bernatorial

-0.72

ãĥ£

-0.69

DragonMagazine

-0.68

aukee

-0.68

-0.67

andowski

-0.66

STAR

-0.66

ument

-0.66

 Scand

-0.63

POSITIVE LOGITS

pins

1.09

cutting

1.02

lining

1.02

lying

0.96

writing

0.95

neath

0.93

dogs

0.93

writers

0.92

graduate

0.92

lined

0.91

Activations Density 0.047%

legal terms or phrases related to laws and regulations

references to legal regulations and laws

No Comments

No Known Activations