INDEX

Explanations

medical terms and treatment-related phrases

oai_token-act-pair · gpt-3.5-turbo

indicative verbs and phrases suggesting therapeutic or beneficial effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

luaj

-0.71

mares

-0.71

xual

-0.68

 reckoning

-0.67

aturdays

-0.65

 roam

-0.64

 Pound

-0.64

 Griff

-0.64

 Massacre

-0.64

Numbers

-0.63

POSITIVE LOGITS

 invaluable

1.58

 helpful

1.51

 beneficial

1.46

 useful

1.43

 valuable

1.38

 indispensable

1.28

 advantageous

1.26

 crucial

1.25

 vital

1.22

 effective

1.21

Activations Density 0.250%

medical terms and treatment-related phrases

indicative verbs and phrases suggesting therapeutic or beneficial effects

No Comments

No Known Activations