INDEX

Explanations

phrases related to reports and confirmations

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

imize

-0.82

thia

-0.81

imeter

-0.79

DAQ

-0.77

rollment

-0.77

:(

-0.75

.--

-0.74

keyes

-0.74

Â¶

-0.73

center

-0.73

POSITIVE LOGITS

 whilst

1.68

 Whilst

1.36

 emphas

1.27

 apologise

1.26

 organisers

1.25

 Â£

1.24

fue

1.24

 organising

1.22

 recognise

1.16

 organised

1.16

Activations Density 0.398%

phrases related to reports and confirmations

No Comments

No Known Activations