INDEX

Explanations

terms related to royalty or honor, specifically focusing on the concept of "crowning."

oai_token-act-pair · gpt-3.5-turbo

references to "crown" and its related terms

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Rossi

-0.69

 Dragonbound

-0.68

 Schwar

-0.68

edia

-0.67

 Bastard

-0.64

 Canaver

-0.63

Exper

-0.63

obiles

-0.62

SOURCE

-0.62

 Apocalypse

-0.62

POSITIVE LOGITS

 jewels

1.16

 jewel

1.10

pin

1.07

 crown

0.97

pins

0.92

 prince

0.92

 crowned

0.85

doms

0.84

ing

0.84

rowing

0.84

Activations Density 0.009%

terms related to royalty or honor, specifically focusing on the concept of "crowning."

references to "crown" and its related terms

No Known Activations