INDEX

Explanations

references to malware and related technical terms

oai_token-act-pair · gpt-3.5-turbo

mentions of malware and ransomware

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.82

obar

-0.74

rite

-0.74

ilts

-0.74

ŃĶ

-0.73

alist

-0.70

é¾įåĸļå£«

-0.69

arb

-0.69

bring

-0.69

ffen

-0.67

POSITIVE LOGITS

 malware

1.08

bytes

0.98

 infect

0.86

 outbreak

0.84

 infections

0.83

 infection

0.83

 Mehran

0.79

CVE

0.76

 virus

0.74

 viruses

0.73

Activations Density 0.011%

references to malware and related technical terms

mentions of malware and ransomware

No Comments

No Known Activations