INDEX

Explanations

terms related to high levels or rates of something

oai_token-act-pair · gpt-3.5-turbo

references to high levels or high prices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

them

-0.74

tnc

-0.73

apons

-0.70

herer

-0.70

Twe

-0.65

 Canaver

-0.64

eh

-0.64

icion

-0.64

ilus

-0.63

ãĤµ

-0.63

POSITIVE LOGITS

 percentage

1.34

 percentages

1.28

 rates

1.28

 levels

1.24

 incidence

1.17

 prevalence

1.16

 proportion

1.16

 rate

1.14

 amount

1.13

 amounts

1.12

Activations Density 0.089%

terms related to high levels or rates of something

references to high levels or high prices

No Comments

No Known Activations