INDEX

Explanations

mentions of organizations or entities starting with "En"

oai_token-act-pair · gpt-3.5-turbo

references to the Enron company

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

 pause

-0.67

 follow

-0.62

theless

-0.59

 distortion

-0.58

 overdue

-0.57

 representation

-0.56

 stabilization

-0.56

 subject

-0.56

 transitional

-0.55

 Opposition

-0.55

POSITIVE LOGITS

viron

1.39

chanted

1.34

cyclop

1.28

igmatic

1.25

riched

1.25

vironments

1.24

semble

1.23

rollment

1.22

igma

1.21

thus

1.21

Activations Density 0.020%

mentions of organizations or entities starting with "En"

references to the Enron company

No Comments

No Known Activations