INDEX

Explanations

words and phrases related to criminal activities and legal proceedings

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 roundup

-0.23

Â¶

-0.20

 prim

-0.20

 curfew

-0.20

 lockdown

-0.20

 rush

-0.20

 playback

-0.19

 seiz

-0.19

 predictably

-0.19

 understandably

-0.19

POSITIVE LOGITS

 thereof

0.29

userc

0.26

ãĥ³ãĤ¸

0.25

 Nicarag

0.24

)."

0.23

holm

0.23

ilan

0.23

perty

0.22

 Afgh

0.22

Ast

0.22

Activations Density 3.617%

words and phrases related to criminal activities and legal proceedings

No Comments

No Known Activations