INDEX

Explanations

references to ransom-related terms such as 'ransom payment' and 'ransomware.'

oai_token-act-pair · gpt-3.5-turbo

terms related to ransom and ransom payments

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

Cola

-0.73

 Mood

-0.67

 Retro

-0.66

alist

-0.64

 Einstein

-0.63

 Anthrop

-0.63

unknown

-0.63

 Jinn

-0.63

 Bord

-0.63

eor

-0.63

POSITIVE LOGITS

 ransom

1.30

 extortion

1.00

hyde

0.97

ware

0.96

 payment

0.86

 attackers

0.83

 kidn

0.82

ansom

0.82

 payments

0.80

 blackmail

0.80

Activations Density 0.019%

references to ransom-related terms such as 'ransom payment' and 'ransomware.'

terms related to ransom and ransom payments

No Comments

No Known Activations