INDEX

Explanations

phrases related to alerts and warnings

oai_token-act-pair · gpt-3.5-turbo

terminology and concepts related to safety alerts and warnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

eneg

-0.84

Pool

-0.83

abilia

-0.78

ibia

-0.78

 collagen

-0.75

Cost

-0.74

luaj

-0.73

ranch

-0.71

galitarian

-0.71

ructure

-0.71

POSITIVE LOGITS

 alerts

2.01

 warnings

1.82

 alert

1.80

 warning

1.76

 Alert

1.60

 alarms

1.58

alert

1.58

Alert

1.55

 warn

1.54

 notifications

1.54

Activations Density 0.442%

phrases related to alerts and warnings

terminology and concepts related to safety alerts and warnings

No Comments

No Known Activations