INDEX

Explanations

various types of malware

oai_token-act-pair · gpt-3.5-turbo

terms related to 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

-0.81

ilts

-0.79

arb

-0.74

rite

-0.72

UTC

-0.72

ffen

-0.71

arist

-0.71

bring

-0.70

ŃĶ

-0.70

ndra

-0.70

POSITIVE LOGITS

bytes

1.00

 malware

0.99

 outbreak

0.87

 infect

0.84

 infection

0.83

 infections

0.82

 payload

0.78

 Mehran

0.75

 blob

0.71

 arsen

0.71

Activations Density 0.016%

various types of malware

terms related to malware and ransomware

No Comments

No Known Activations