INDEX

Explanations

mentions of the word "Can" followed by another word

oai_token-act-pair · gpt-3.5-turbo

the phrase "Can" and its variations highlighting potential or capability

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

itbart

-0.62

RAG

-0.59

aign

-0.59

geist

-0.58

 grievance

-0.58

ãĥŃ

-0.57

rals

-0.57

llor

-0.57

hem

-0.56

iations

-0.55

POSITIVE LOGITS

Can

3.28

Can

2.42

can

1.85

 Could

1.70

 Must

1.69

CAN

1.64

CAN

1.63

Are

1.52

 Does

1.51

 Cannot

1.49

Activations Density 0.017%

mentions of the word "Can" followed by another word

the phrase "Can" and its variations highlighting potential or capability

No Comments

No Known Activations