INDEX

Explanations

medical and administrative terms potentially related to the health field

oai_token-act-pair · gpt-3.5-turbo

references to individuals and events related to coroners and medical contexts

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

camp

-0.88

Scale

-0.77

×ķ

-0.76

Reloaded

-0.74

stadt

-0.72

loader

-0.71

Legendary

-0.71

 loader

-0.69

Constructed

-0.69

 Territories

-0.67

POSITIVE LOGITS

ymph

0.82

vati

0.77

riages

0.77

emic

0.77

 influenza

0.77

 pand

0.76

alysed

0.74

 vaccines

0.74

NetMessage

0.74

 butterflies

0.74

Activations Density 0.027%

medical and administrative terms potentially related to the health field

references to individuals and events related to coroners and medical contexts

No Comments

No Known Activations