INDEX

Explanations

phrases related to grandeur, fame, or reputation, often with a positive connotation

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "glory."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

iced

-0.67

Consumer

-0.67

 Case

-0.67

 Process

-0.66

ourced

-0.65

 Blocks

-0.65

 Intake

-0.65

atu

-0.64

 crim

-0.60

 Approach

-0.60

POSITIVE LOGITS

 glory

3.96

 Glory

2.21

 greatness

1.83

 splend

1.70

 glorious

1.55

 majesty

1.49

 brilliance

1.44

 riches

1.43

 glor

1.39

 fame

1.35

Activations Density 0.012%

phrases related to grandeur, fame, or reputation, often with a positive connotation

references to the concept of "glory."

No Comments

No Known Activations