INDEX

Explanations

proper nouns related to significant events, places, or people with a positive connotation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Great" in various contexts

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

uate

-0.69

atel

-0.67

rogram

-0.66

elle

-0.66

 Ramirez

-0.63

label

-0.62

ebin

-0.61

 embedded

-0.60

itions

-0.60

heet

-0.60

POSITIVE LOGITS

 Great

3.57

Great

2.65

 GREAT

1.95

great

1.84

 Greater

1.54

 great

1.53

 Greatest

1.38

 Excellent

1.33

 Perfect

1.30

 Wonderful

1.28

Activations Density 0.028%

proper nouns related to significant events, places, or people with a positive connotation

instances of the word "Great" in various contexts

No Comments

No Known Activations