INDEX

Explanations

numerical values representing the length of something

oai_token-act-pair · gpt-3.5-turbo

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

eers

-0.81

eering

-0.74

eer

-0.73

ILLE

-0.72

leck

-0.68

dinand

-0.66

 Jarrett

-0.65

 resur

-0.64

unin

-0.62

 FIRE

-0.62

POSITIVE LOGITS

iest

1.12

wise

1.10

iness

0.93

ier

0.92

ways

0.91

 lengths

0.87

shed

0.86

 length

0.84

sword

0.80

hered

0.77

Activations Density 0.021%

numerical values representing the length of something

No Comments

No Known Activations