INDEX

Explanations

dates in various formats across different years

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

NRL

-0.31

 Marlins

-0.27

 Tribe

-0.27

Â¶

-0.27

tackle

-0.26

 Belfast

-0.26

 palate

-0.26

 Ballard

-0.26

 Perspective

-0.26

FSA

-0.26

POSITIVE LOGITS

hovah

0.33

renheit

0.31

ONSORED

0.29

itialized

0.29

ETHOD

0.28

ither

0.28

mble

0.27

ixir

0.27

rored

0.27

.","

0.27

Activations Density 7.812%

dates in various formats across different years

No Comments

No Known Activations