INDEX

Explanations

medical and military related terms or actions related to them

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to documents and their handling or management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ometimes

-0.78

afety

-0.74

QC

-0.65

uesday

-0.64

Sky

-0.64

idth

-0.61

Maps

-0.61

conservancy

-0.60

uden

-0.59

 âĸº

-0.59

POSITIVE LOGITS

 itself

0.81

 piece

0.73

 designation

0.71

 portion

0.68

iece

0.67

 recipients

0.66

 afterward

0.63

 signifies

0.63

 nonetheless

0.63

 mechanism

0.63

Activations Density 0.605%

medical and military related terms or actions related to them

words and phrases related to documents and their handling or management

No Comments

No Known Activations