INDEX

Explanations

words related to buildings or locations, especially museums

oai_token-act-pair · gpt-3.5-turbo

words related to building and structures

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

£ı

-0.80

INGTON

-0.79

 Fargo

-0.68

sburgh

-0.67

enegger

-0.63

 Weld

-0.58

 Dragonbound

-0.58

 kidding

-0.58

DeL

-0.57

 defect

-0.57

POSITIVE LOGITS

xt

1.25

ptic

1.22

lli

1.18

ttes

1.17

lled

1.16

ffect

1.15

eper

1.14

Ã±o

1.12

pt

1.08

utics

1.08

Activations Density 0.088%

words related to buildings or locations, especially museums

words related to building and structures

No Comments

No Known Activations