INDEX

Explanations

words related to financial crimes and cybersecurity attacks

oai_token-act-pair · gpt-3.5-turbo

terms related to fraud and financial misconduct

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 absence

-0.82

dar

-0.75

area

-0.74

ä

-0.71

wcsstore

-0.67

 endurance

-0.62

 intermittent

-0.61

çĦ

-0.60

accompan

-0.60

 eternity

-0.59

POSITIVE LOGITS

raud

1.32

ulent

1.29

eering

0.99

gments

0.92

ingly

0.90

itives

0.88

ulence

0.83

isance

0.82

ware

0.81

vertising

0.81

Activations Density 0.020%

words related to financial crimes and cybersecurity attacks

terms related to fraud and financial misconduct

No Comments

No Known Activations