INDEX

Explanations

mention of electrical terms or related to electricity

oai_token-act-pair · gpt-3.5-turbo

references to electrical concepts and terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

naire

-0.96

itia

-0.89

 Cherokee

-0.76

naires

-0.75

-0.71

 Fully

-0.71

chens

-0.71

 Citizen

-0.71

 Connor

-0.70

 Haku

-0.70

POSITIVE LOGITS

 electrical

1.13

 conduc

1.07

 wiring

1.03

 insulation

1.02

 circuitry

1.01

 impedance

1.01

 electricity

0.99

 engineering

0.98

 appliances

0.97

 circuits

0.96

Activations Density 0.009%

mention of electrical terms or related to electricity

references to electrical concepts and terminology

No Comments

No Known Activations