INDEX

Explanations

references to quizzes or coupons in documents

oai_token-act-pair · gpt-3.5-turbo

references to quizzes and coupons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

lihood

-0.85

rative

-0.78

::::

-0.72

igated

-0.71

 secretaries

-0.71

anism

-0.66

 Leaves

-0.66

medi

-0.65

ntil

-0.65

::::::::

-0.64

POSITIVE LOGITS

 quiz

0.86

azine

0.77

 quizz

0.73

itsu

0.72

acebook

0.72

onomy

0.70

atti

0.69

mson

0.68

topia

0.67

bie

0.67

Activations Density 0.021%

references to quizzes or coupons in documents

references to quizzes and coupons

No Comments

No Known Activations