INDEX

Explanations

references to the capability or potential to do something

oai_token-act-pair · gpt-3.5-turbo

terms related to capabilities or functionalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

paying

-0.72

 Pall

-0.71

winter

-0.68

cise

-0.68

eva

-0.68

corn

-0.67

cus

-0.66

boss

-0.65

eree

-0.63

 Mast

-0.63

POSITIVE LOGITS

 capabilities

1.28

 capability

1.20

afety

0.97

 chops

0.90

 capacities

0.89

 sophistication

0.87

indal

0.84

 guiActiveUn

0.84

 prowess

0.84

 competence

0.83

Activations Density 0.007%

references to the capability or potential to do something

terms related to capabilities or functionalities

No Comments

No Known Activations