INDEX

Explanations

mentions of criminal activities or incidents in news articles

oai_token-act-pair · gpt-3.5-turbo

references to violent incidents and criminal activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

*/(

-0.69

 sqor

-0.60

 partName

-0.59

":-

-0.57

 ðŁĺ

-0.57

 Patreon

-0.57

 ðŁĻĤ

-0.56

 ðŁĳ

-0.56

"!

-0.55

 learners

-0.55

POSITIVE LOGITS

amphetamine

0.69

Qaida

0.69

addafi

0.69

hetamine

0.66

 narcotics

0.64

assault

0.60

 Soviet

0.58

 cocaine

0.57

abortion

0.57

weapons

0.57

Activations Density 0.971%

mentions of criminal activities or incidents in news articles

references to violent incidents and criminal activities

No Comments

No Known Activations