INDEX

Explanations

information related to illegal activities or items

oai_token-act-pair · gpt-3.5-turbo

terminology related to illegal activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

soType

-0.76

frames

-0.71

packs

-0.70

©¶æ

-0.69

udeb

-0.69

gaard

-0.69

Point

-0.68

anders

-0.67

 Ducks

-0.67

iph

-0.67

POSITIVE LOGITS

 illicit

0.94

 trafficking

0.90

 substances

0.82

lic

0.80

ly

0.76

 overfl

0.75

 smuggling

0.74

 narcotics

0.73

aneously

0.73

ishly

0.73

Activations Density 0.011%

information related to illegal activities or items

terminology related to illegal activities

No Comments

No Known Activations