INDEX

Explanations

numeric ranges

oai_token-act-pair · gpt-3.5-turbo

phrases that describe a range or variety of options or categories

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

mit

-0.84

å£«

-0.75

driving

-0.70

ãĥīãĥ©

-0.69

iquette

-0.65

orc

-0.64

idy

-0.64

rics

-0.63

mond

-0.62

anasia

-0.62

POSITIVE LOGITS

 ranging

0.97

ĸļ

0.82

Ĥª

0.79

 upwards

0.78

ranging

0.76

 ranges

0.74

 spanning

0.69

 lengths

0.69

yip

0.69

 Luxem

0.69

Activations Density 0.021%

numeric ranges

phrases that describe a range or variety of options or categories

No Known Activations