INDEX

Explanations

phrases related to leaks or confidential information being shared

oai_token-act-pair · gpt-3.5-turbo

phrases related to illegal activities or misconduct

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

DragonMagazine

-0.82

quickShipAvailable

-0.78

 partName

-0.77

ãĤ¦ãĤ¹

-0.72

fuck

-0.72

Disclaimer

-0.70

)!

-0.69

reci

-0.68

claimer

-0.67

 ALWAYS

-0.67

POSITIVE LOGITS

the

0.80

 famed

0.74

 former

0.73

 northeastern

0.73

 dozens

0.72

 disgr

0.71

interstitial

0.70

 Western

0.70

 several

0.68

 embattled

0.68

Activations Density 0.542%

phrases related to leaks or confidential information being shared

phrases related to illegal activities or misconduct

No Comments

No Known Activations