INDEX

Explanations

terms related to protection or security, particularly in the context of laws, regulations, and rights

oai_token-act-pair · gpt-3.5-turbo

terms related to legal protection or rights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

bender

-0.74

OVA

-0.73

sb

-0.71

lder

-0.68

etics

-0.67

OTA

-0.66

annis

-0.66

hler

-0.66

ellen

-0.64

Bee

-0.64

POSITIVE LOGITS

 protected

1.00

 unprotected

0.93

 protects

0.85

 immunity

0.81

 protection

0.81

 protections

0.80

 enclave

0.79

 anonymity

0.79

folios

0.79

 protecting

0.78

Activations Density 0.013%

terms related to protection or security, particularly in the context of laws, regulations, and rights

terms related to legal protection or rights

No Comments

No Known Activations