INDEX

Explanations

references to CDs

oai_token-act-pair · gpt-3.5-turbo

mentions of CDs

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

terday

-0.81

ãĥĦ

-0.78

````

-0.66

utherford

-0.64

asonic

-0.64

ãĥīãĥ©ãĤ´ãĥ³

-0.64

 Gors

-0.64

66666666

-0.63

Osw

-0.63

Ò

-0.63

POSITIVE LOGITS

CD

1.05

ouble

1.00

CD

0.93

NF

0.85

MX

0.84

DVD

0.84

ROM

0.82

CL

0.81

erning

0.81

irect

0.80

Activations Density 0.005%

references to CDs

mentions of CDs

No Comments

No Known Activations