INDEX

Explanations

mentions of scars

oai_token-act-pair · gpt-3.5-turbo

references to physical scars and damage

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

Input

-0.82

 Input

-0.73

earch

-0.72

 Commerce

-0.72

specified

-0.69

 independently

-0.68

 Balanced

-0.66

inance

-0.66

prov

-0.66

endment

-0.65

POSITIVE LOGITS

 scar

3.84

 scars

2.43

 Scar

2.03

scar

1.93

Scar

1.86

 scarf

1.66

 traumat

1.21

 tattoo

1.20

 tattoos

1.19

 bruises

1.16

Activations Density 0.022%

mentions of scars

references to physical scars and damage

No Comments

No Known Activations