INDEX

Explanations

words related to locations, especially cities

oai_token-act-pair · gpt-3.5-turbo

mentions of "New York."

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

Not in Any Lists

No Comments

Negative Logits

inelli

-0.67

 deduction

-0.64

 vaccinated

-0.64

 quizz

-0.64

 DRAG

-0.64

 grap

-0.63

iotic

-0.63

 retarded

-0.62

 extracting

-0.61

ICLE

-0.61

POSITIVE LOGITS

 Orleans

1.30

 York

1.28

 Delhi

1.19

 Zealand

1.08

 Amsterdam

1.04

 London

1.00

York

0.99

 Cairo

0.98

 Haven

0.97

 YORK

0.97

Activations Density 0.079%

words related to locations, especially cities

mentions of "New York."

No Comments

No Known Activations