INDEX

Explanations

words related to medical conditions and hospital procedures

oai_token-act-pair · gpt-3.5-turbo

key events and outcomes related to medical conditions and health situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ahime

-0.93

 complicity

-0.75

 Chatt

-0.73

 Hier

-0.73

 Franch

-0.73

 Voter

-0.72

 actionGroup

-0.72

 Tradable

-0.72

 piracy

-0.72

 agendas

-0.71

POSITIVE LOGITS

Doctors

1.34

 medically

1.29

 neurolog

1.25

 remission

1.24

 diabetic

1.23

 surgery

1.20

 hospital

1.18

 symptoms

1.18

 medication

1.16

 doctors

1.16

Activations Density 0.450%

words related to medical conditions and hospital procedures

key events and outcomes related to medical conditions and health situations

No Comments

No Known Activations