INDEX

Explanations

descriptions or mentions related to fires

oai_token-act-pair · gpt-3.5-turbo

references to consuming fires and their effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ploma

-0.95

eele

-0.94

actionDate

-0.92

apore

-0.84

 Visa

-0.83

bourg

-0.83

plom

-0.81

 Token

-0.80

 passport

-0.79

 visa

-0.78

POSITIVE LOGITS

 blaze

1.89

 flames

1.85

 wildfire

1.79

 wildfires

1.70

 arson

1.67

 fires

1.67

 firefighters

1.60

 charred

1.59

 fire

1.57

 extingu

1.55

Activations Density 0.271%

descriptions or mentions related to fires

references to consuming fires and their effects

No Comments

No Known Activations