INDEX

Explanations

phrases related to illegal or illicit activities

oai_token-act-pair · gpt-3.5-turbo

terms related to illicit finance and illegal activities

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

etheless

-0.84

 whisk

-0.73

 recorder

-0.67

rams

-0.67

elled

-0.64

 captcha

-0.63

 cyan

-0.63

ags

-0.62

 carbohyd

-0.62

ulin

-0.61

POSITIVE LOGITS

 Integrity

1.01

 Opportun

1.00

 Directory

0.99

 Accountability

0.98

 Authority

0.98

 Solutions

0.95

 Direction

0.94

 Services

0.94

 Lobby

0.93

 Therapy

0.93

Activations Density 0.387%

phrases related to illegal or illicit activities

terms related to illicit finance and illegal activities

No Comments

No Known Activations