INDEX

Explanations

events or incidents with negative connotations

oai_token-act-pair · gpt-3.5-turbo

words or symbols that indicate secrecy or confidentiality

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

 mathemat

-0.82

 pigeon

-0.77

 notor

-0.76

 challeng

-0.71

 exha

-0.69

 predec

-0.67

 rundown

-0.63

 satell

-0.63

 elusive

-0.62

 sprawling

-0.62

POSITIVE LOGITS

ï¸ı

1.42

âĢ

0.98

ttp

0.95

 âĢ

0.91

Â¶

0.89

ï¸

0.88

ðŁ

0.87

tis

0.85

âĢİ

0.84

https

0.84

Activations Density 0.217%

events or incidents with negative connotations

words or symbols that indicate secrecy or confidentiality

No Comments

No Known Activations