INDEX

Explanations

references to the year 2012

oai_token-act-pair · gpt-3.5-turbo

references to the year 2012

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

anooga

-0.97

merce

-0.85

phy

-0.82

onite

-0.81

 volunte

-0.79

imore

-0.78

 cumbers

-0.77

unin

-0.76

avorite

-0.76

apo

-0.76

POSITIVE LOGITS

å¹

0.94

0.83

0.78

 onwards

0.78

cens

0.77

0.76

0.75

-'

0.72

 Definition

0.71

Activations Density 0.022%

references to the year 2012

references to the year 2012

No Comments

No Known Activations