INDEX

Explanations

mentions of announcements, pledges, intentions, projects, resignations, or endorsements

oai_token-act-pair · gpt-3.5-turbo

phrases indicating announcements or declarations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anon

-0.71

 interrogated

-0.71

PLIED

-0.70

 alone

-0.69

 countered

-0.67

 rhet

-0.64

 screamed

-0.62

 argued

-0.62

 argues

-0.62

 abused

-0.61

POSITIVE LOGITS

 impending

1.17

 intention

1.15

 plans

1.10

 intentions

1.10

 imminent

1.03

 arrival

1.00

 retirement

0.98

 existence

0.96

 formation

0.94

 demise

0.94

Activations Density 0.165%

mentions of announcements, pledges, intentions, projects, resignations, or endorsements

phrases indicating announcements or declarations

No Comments

No Known Activations