INDEX

Explanations

terms related to confidentiality and secrecy

oai_token-act-pair · gpt-3.5-turbo

terms related to confidentiality and secrecy

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

ergy

-0.88

¶æ

-0.85

ãĥ´

-0.78

NetMessage

-0.76

bern

-0.75

 Parenthood

-0.74

Spot

-0.72

ãĤ©

-0.71

Ģ

-0.71

sonian

-0.70

POSITIVE LOGITS

anship

1.03

 excellence

0.87

raint

0.85

 confidentiality

0.81

 parity

0.77

uality

0.77

worthiness

0.76

 tremend

0.76

igham

0.74

fulness

0.73

Activations Density 0.018%

terms related to confidentiality and secrecy

terms related to confidentiality and secrecy

No Comments

No Known Activations