INDEX

Explanations

words associated with reflection or representation

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

stant

-0.76

iott

-0.75

aii

-0.74

ensable

-0.71

uilt

-0.71

--------------------------------------------------------

-0.70

yright

-0.69

ccoli

-0.68

CVE

-0.68

TAIN

-0.68

POSITIVE LOGITS

ror

1.01

 neuron

0.94

 image

0.86

angelo

0.83

 mirror

0.82

 Mirror

0.80

 neurons

0.79

 shards

0.78

ing

0.77

ocular

0.76

Activations Density 0.050%

words associated with reflection or representation

No Comments

No Known Activations