INDEX

Explanations

mentions of fundamental concepts or beliefs

oai_token-act-pair · gpt-3.5-turbo

references to principles, particularly in legal and ethical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ammers

-0.71

minster

-0.70

leneck

-0.68

ilation

-0.67

NetMessage

-0.67

quer

-0.67

eor

-0.66

essions

-0.65

dos

-0.64

hiba

-0.64

POSITIVE LOGITS

 principles

1.01

 principle

0.91

 guiding

0.90

ciples

0.88

 underlying

0.83

ually

0.82

 underpin

0.81

 Principles

0.80

 precept

0.73

cipled

0.72

Activations Density 0.023%

mentions of fundamental concepts or beliefs

references to principles, particularly in legal and ethical contexts

No Comments

No Known Activations