INDEX

Explanations

titles or roles related to leadership positions within organizations

oai_token-act-pair · gpt-3.5-turbo

titles and positions of authority in professional contexts

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

 reper

-0.87

 peril

-0.86

 compromise

-0.86

 queen

-0.84

 queens

-0.82

 vain

-0.81

 dwarves

-0.81

 trap

-0.81

 diligence

-0.80

 surg

-0.79

POSITIVE LOGITS

 Manager

1.41

 Partner

1.32

 Coordinator

1.30

 Candidate

1.29

 Director

1.28

 Editor

1.27

 Distribution

1.26

 Analyst

1.26

 Advisor

1.25

 Relations

1.25

Activations Density 0.201%

titles or roles related to leadership positions within organizations

titles and positions of authority in professional contexts

No Comments

No Known Activations