INDEX

Explanations

numeric values marked as "Total"

oai_token-act-pair · gpt-3.5-turbo

references to total values or aggregates 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

coming

-0.70

bub

-0.68

inks

-0.67

assetsadobe

-0.67

lik

-0.66

 sooner

-0.66

 interpre

-0.65

soon

-0.64

banks

-0.62

assadors

-0.61

POSITIVE LOGITS

 Total

3.53

 TOTAL

2.37

Total

2.22

total

1.84

 total

1.81

 Average

1.39

 Maximum

1.30

 totals

1.30

 Estimated

1.28

OTAL

1.24

Activations Density 0.003%

numeric values marked as "Total"

references to total values or aggregates in various contexts

No Comments

No Known Activations