INDEX

Explanations

titles or positions within a company or organization

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

titles and positions of individuals in corporate or organizational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

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

 plunder

-1.01

 grouping

-1.00

 stamp

-0.99

 compromise

-0.98

 ridic

-0.98

 taboo

-0.97

 spoil

-0.97

 republic

-0.95

 loose

-0.95

 stacking

-0.94

POSITIVE LOGITS

 Advisor

1.76

 Manager

1.70

 Adviser

1.70

 Coordinator

1.67

 Analyst

1.65

 Writer

1.57

 Director

1.56

 Scientist

1.55

 Counsel

1.54

 Partner

1.53

Activations Density 0.163%

titles or positions within a company or organization

titles and positions of individuals in corporate or organizational contexts

No Comments

No Known Activations