INDEX

Explanations

terms related to physical bleeding or figurative emotional bleeding

oai_token-act-pair · gpt-3.5-turbo

terms related to bleeding and hemorrhaging

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

ach

-0.76

 Proposition

-0.67

 Devon

-0.67

char

-0.66

 assigns

-0.65

}}}

-0.63

earth

-0.63

attribute

-0.62

aux

-0.62

 Reply

-0.61

POSITIVE LOGITS

 bleeding

3.97

 bleed

2.44

 hemorrh

2.04

 hemor

1.42

 weeping

1.40

 vomiting

1.37

 bruised

1.25

 swelling

1.24

 crying

1.22

 blood

1.20

Activations Density 0.007%

terms related to physical bleeding or figurative emotional bleeding

terms related to bleeding and hemorrhaging

No Comments

No Known Activations