INDEX

Explanations

time-related events and measurements

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Proc

-0.70

 Cassidy

-0.67

 Rocky

-0.66

 Neville

-0.66

Pep

-0.65

 Rica

-0.65

 Prometheus

-0.64

Cec

-0.63

 Minerva

-0.63

 Maya

-0.62

POSITIVE LOGITS

year

1.47

hour

1.34

minute

1.32

degree

1.31

month

1.27

percent

1.25

point

1.20

inch

1.19

digit

1.18

years

1.17

Activations Density 0.326%

time-related events and measurements

No Comments

No Known Activations