INDEX

Explanations

mathematical and technical terms related to set theory and mathematics

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

erson

-0.17

 Sail

-0.17

yn

-0.16

 Franklin

-0.15

osa

-0.15

':

-0.15

anners

-0.15

 Moons

-0.15

 Conversion

-0.14

Kon

-0.14

POSITIVE LOGITS

 tradem

0.22

 challeng

0.20

[|

0.20

Recomm

0.19

 thereto

0.19

 suspic

0.19

 ingred

0.19

MpServer

0.18

 thereof

0.18

amount

0.18

Activations Density 12.325%

mathematical and technical terms related to set theory and mathematics

No Comments

No Known Activations