INDEX

Explanations

references to hardware, including devices and equipment

oai_token-act-pair · gpt-3.5-turbo

mentions of hardware

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Cah

-0.74

 Revis

-0.71

 Kier

-0.64

wolves

-0.64

................................................................

-0.64

ishi

-0.63

ower

-0.61

 Cait

-0.60

 yawn

-0.58

 lett

-0.58

POSITIVE LOGITS

 hardware

3.68

 Hardware

2.71

Hardware

2.62

 software

1.61

 electronics

1.57

 equipment

1.56

 firmware

1.55

software

1.48

 CPUs

1.30

 GPUs

1.29

Activations Density 0.015%

references to hardware, including devices and equipment

mentions of hardware

No Comments

No Known Activations