INDEX

Explanations

terms related to different sexual orientations, notably homosexuality

oai_token-act-pair · gpt-3.5-turbo

terms related to homosexuality and beginner-level concepts

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

IFrame

Link

•Embedding Space Digit Features

No Comments

Negative Logits

notations

-0.72

ĪĴ

-0.71

amination

-0.70

availability

-0.69

weather

-0.68

 Ware

-0.68

 insurer

-0.67

ãĤ´ãĥ³

-0.67

taboola

-0.66

 toxic

-0.66

POSITIVE LOGITS

osexual

2.22

 Begin

1.67

bian

1.62

 beginners

1.57

ORN

1.48

 beginner

1.40

Beg

1.31

arella

1.19

bia

1.04

Stra

1.04

Activations Density 0.033%

terms related to different sexual orientations, notably homosexuality

terms related to homosexuality and beginner-level concepts

No Comments

No Known Activations