INDEX

Explanations

words related to physical scarring and injuries

oai_token-act-pair · gpt-3.5-turbo

references to scars and their implications

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

ablishment

-0.96

gaard

-0.78

eger

-0.72

ullivan

-0.70

hower

-0.69

 Christy

-0.67

£ı

-0.64

aminer

-0.64

geist

-0.62

Cou

-0.59

POSITIVE LOGITS

 scar

1.22

 scars

1.05

ring

0.95

Scar

0.90

crow

0.88

lets

0.86

red

0.85

fing

0.84

 Scar

0.78

pered

0.76

Activations Density 0.005%

words related to physical scarring and injuries

references to scars and their implications

No Comments

No Known Activations