INDEX

Explanations

words related to healthcare

oai_token-act-pair · gpt-3.5-turbo

terms related to health care

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

âĸ¬

-0.79

coni

-0.72

 Helpful

-0.69

 Horn

-0.67

 Darkness

-0.66

âĸ¬âĸ¬

-0.66

 Reloaded

-0.66

 Hood

-0.62

Duo

-0.62

 tongues

-0.61

POSITIVE LOGITS

 care

1.09

care

1.09

aceutical

1.07

Health

0.99

iest

0.96

health

0.95

amacare

0.95

 health

0.93

ily

0.92

 insurance

0.91

Activations Density 0.033%

words related to healthcare

terms related to health care

No Comments

No Known Activations