INDEX

Explanations

phrases indicating actions taken without consent or permission

oai_token-act-pair · gpt-3.5-turbo

phrases related to actions taken without consent or permission

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

raq

-0.79

 Reincarnated

-0.78

ãĤ¼ãĤ¦ãĤ¹

-0.73

 Kinnikuman

-0.69

arse

-0.69

è¦ļéĨĴ

-0.66

ãĤ¤ãĥĪ

-0.66

ä¸ī

-0.66

romy

-0.65

ãĥ³ãĤ¸

-0.64

POSITIVE LOGITS

 permission

1.25

 authorization

1.22

 consent

1.12

 exception

1.03

 prompting

0.98

 approval

0.97

 qualification

0.96

 supervision

0.96

 specifying

0.94

 reperc

0.94

Activations Density 0.096%

phrases indicating actions taken without consent or permission

phrases related to actions taken without consent or permission

No Comments

No Known Activations