INDEX

Explanations

phrases or sentences that are used to provide examples or explanations

oai_token-act-pair · gpt-3.5-turbo

expressions that convey the act of illustrating or demonstrating a point or concept

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

etheless

-0.68

uld

-0.65

 mouth

-0.64

pool

-0.64

ni

-0.63

 hire

-0.63

 chopping

-0.63

eties

-0.63

lat

-0.62

ussed

-0.62

POSITIVE LOGITS

 illustrating

0.93

 illustration

0.89

 illustrates

0.87

hematic

0.83

 Caption

0.82

 Illust

0.78

 Exhibit

0.76

 illustrate

0.75

 Thumbnails

0.74

ĸļ

0.74

Activations Density 0.021%

phrases or sentences that are used to provide examples or explanations

expressions that convey the act of illustrating or demonstrating a point or concept

No Comments

No Known Activations