INDEX

Explanations

phrases related to protection and defense, particularly in the context of security and safety

oai_token-act-pair · gpt-3.5-turbo

themes related to protection and security

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 explaining

-0.67

 explains

-0.67

udi

-0.66

 whereby

-0.66

gomery

-0.65

dayName

-0.63

tail

-0.63

 describes

-0.63

Boost

-0.63

aez

-0.62

POSITIVE LOGITS

 integrity

1.28

 dignity

1.20

 freedoms

1.16

 confidentiality

1.14

 privacy

1.12

 interests

1.11

 rights

1.11

 liberties

1.10

 indef

1.07

 secrets

1.04

Activations Density 0.126%

phrases related to protection and defense, particularly in the context of security and safety

themes related to protection and security

No Comments

No Known Activations