INDEX

Explanations

phrases related to law enforcement actions

oai_token-act-pair · gpt-3.5-turbo

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

 redes

-0.73

è¦ļéĨĴ

-0.71

zbollah

-0.71

ãĤ´ãĥ³

-0.70

TPP

-0.69

£ı

-0.68

HCR

-0.67

Climate

-0.67

soDeliveryDate

-0.67

 educ

-0.64

POSITIVE LOGITS

him

1.34

 occupants

0.95

 appellant

0.92

 defendant

0.88

 Officer

0.85

him

0.85

 Goodwin

0.84

 Hutch

0.84

 them

0.83

 Tanner

0.83

Activations Density 0.187%

phrases related to law enforcement actions

No Comments

No Known Activations