INDEX

Explanations

locations and dates of historical events

oai_token-act-pair · gpt-3.5-turbo

references to significant violent events or incidents

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

 nods

-0.70

dit

-0.64

Tre

-0.63

sounding

-0.62

ulton

-0.62

 Certified

-0.61

 Designs

-0.60

 acron

-0.59

 slots

-0.59

FOX

-0.58

POSITIVE LOGITS

 perpetrated

1.06

 sparked

0.97

 escalated

0.95

 resulted

0.93

 caused

0.91

 provoked

0.90

 ensued

0.89

 triggered

0.88

 occurred

0.86

 inflicted

0.84

Activations Density 0.318%

locations and dates of historical events

references to significant violent events or incidents

No Comments

No Known Activations