INDEX

Explanations

phrases related to the concept of protection and defense

oai_token-act-pair · gpt-3.5-turbo

references to protection and safety, particularly regarding individuals and communities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

sonian

-0.74

²¾

-0.73

ĪĴ

-0.73

HAM

-0.70

Beer

-0.66

soDeliveryDate

-0.64

Newsletter

-0.63

------------------------------------------------

-0.63

 mathemat

-0.63

JB

-0.63

POSITIVE LOGITS

 integrity

0.87

 dignity

0.83

 confidentiality

0.78

riott

0.75

 privacy

0.73

 secrets

0.73

 homeland

0.70

 passage

0.69

 endangered

0.67

iveness

0.66

Activations Density 0.195%

phrases related to the concept of protection and defense

references to protection and safety, particularly regarding individuals and communities

No Comments

No Known Activations