INDEX

Explanations

content related to healthcare

oai_token-act-pair · gpt-3.5-turbo

references to healthcare-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

pole

-0.81

 rods

-0.76

 Modes

-0.71

rod

-0.65

 trunk

-0.65

--------------------------------

-0.63

---

-0.62

----------------

-0.61

Tor

-0.61

Obj

-0.60

POSITIVE LOGITS

 healthcare

3.57

 Healthcare

2.63

 health

1.90

health

1.74

 ObamaCare

1.68

NHS

1.67

 medical

1.57

Health

1.57

 Health

1.56

 childcare

1.54

Activations Density 0.008%

content related to healthcare

references to healthcare-related topics

No Comments

No Known Activations