INDEX

Explanations

words related to infections

oai_token-act-pair · gpt-3.5-turbo

instances of the word "infected" and its variations, indicating discussions about infections

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

umbn

-0.88

HCR

-0.80

PUT

-0.72

apa

-0.70

 unification

-0.70

kee

-0.68

minist

-0.68

Adv

-0.67

perty

-0.67

 EVENTS

-0.67

POSITIVE LOGITS

 infected

1.37

 infect

1.13

 Infect

1.08

 infection

0.98

 infections

0.95

infect

0.94

irus

0.87

 virus

0.83

 mosquitoes

0.82

 outbreak

0.78

Activations Density 0.010%

words related to infections

instances of the word "infected" and its variations, indicating discussions about infections

No Comments

No Known Activations