INDEX

Explanations

phrases related to refusal or denial

oai_token-act-pair · gpt-3.5-turbo

phrases related to refusal or non-compliance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

////////////////////////////////

-0.86

issance

-0.85

quickShipAvailable

-0.76

calling

-0.71

soDeliveryDate

-0.70

————————

-0.70

////////////////

-0.66

Shock

-0.66

Marginal

-0.66

 Prediction

-0.65

POSITIVE LOGITS

 cooperate

1.41

 acknowledge

1.39

 accept

1.34

 obey

1.31

 comply

1.29

bud

1.26

 participate

1.23

 submit

1.22

 concede

1.22

 disclose

1.21

Activations Density 0.067%

phrases related to refusal or denial

phrases related to refusal or non-compliance

No Comments

No Known Activations