INDEX

Explanations

references to the country "China"

oai_token-act-pair · gpt-3.5-turbo

mentions of China

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

rah

-0.79

iotic

-0.76

 Ajax

-0.76

 Payton

-0.75

 Hitchcock

-0.75

 Ukip

-0.71

rible

-0.70

 Hert

-0.70

nor

-0.70

Scotland

-0.70

POSITIVE LOGITS

 Jinping

1.31

 Yuan

1.20

 yuan

1.20

Zhu

1.14

Xin

1.12

 Jing

1.11

Tai

1.05

 Guang

1.04

 Unic

1.03

 Tian

1.01

Activations Density 0.073%

references to the country "China"

mentions of China

No Known Activations