INDEX

Explanations

carefully executed actions or plans

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

anon

-0.80

olor

-0.74

Stars

-0.70

 Cosponsors

-0.68

bucks

-0.68

XP

-0.64

hey

-0.64

asta

-0.63

Native

-0.63

Aid

-0.63

POSITIVE LOGITS

 crafted

1.25

 calibrated

1.24

 scrutin

1.09

 calibr

1.09

 chore

1.05

 vetted

1.04

 curated

1.01

 deliber

0.99

 cultivated

0.96

 tailored

0.94

Activations Density 0.047%

carefully executed actions or plans

No Comments

No Known Activations