On Sun, 8 Jan 2023 00:39:27 -0500
grarpamp wrote:
> Instead of saying that all Govts suck, including RU and US,
So the jewnazi pentagon war propaganda is relentless. The vomits coming
from the anti-human jewnazi bot grarpamp @ gmail.com, hiding behind jewmail are
relentless.
> all G
West Virginia Law Restricting Sports By Biological Sex Is Constitutional: Judge
https://www.theepochtimes.com/west-virginia-law-restricting-sports-by-biological-sex-is-constitutional-judge_4966573.html
https://www.wvlegislature.gov/Bill_Text_HTML/2021_SESSIONS/RS/signed_bills/house/HB3293%20SUB%20
SPR @ 372/714MBbl -14.5 avgLast3Mo a/o Dec 30
Negligible 19 days remaining which can't be pumped
out fast enough to meet demand anyway.
Not a single barrel pumped into the SPR for all of 2022,
not any since Trump pumped a few in.
Democrat Admin so fraudulent that not only were they
caught trading
Instead of saying that all Govts suck, including RU and US,
Juan's claimed innocent and preferred Govt of Russia to Declare Martial Law,
force bribe and brainwash 100k's into slavery and throw them kids
to their deaths, not even in voluntarily defending their own border from
from actively invading
The CDC Puts Itself In Charge Of Language Too
https://brownstone.org/articles/cdc-in-charge-of-language/
https://rwmalonemd.substack.com/p/cdc-principles-to-avoid-wrong-speak
Authored by Robert Malone via The Brownstone Institute,
The Centers for Disease Control and Prevention (CDC) has come out
Ontario Governing Body For Psychologists Demands Re-Education For
Jordan Peterson
https://torontosun.com/opinion/editorials/editorial-the-orwellian-re-education-of-jordan-peterson
Outspoken psychologist and former tenured professor Jordan Peterson
has come under relentless attack ever since he cr
"We Don't Do This": Adam Schiff & The Underbelly Of American Censorship
https://jonathanturley.org/2023/01/07/we-dont-do-this-adam-schiff-and-the-underbelly-of-american-censorship/
Below is my column in the Hill on the recent disclosure of efforts by
Rep. Adam Schiff (D., Cal.) to pressure Twitte
Mysterious Antennas Found In Utah's Hills; Some Speculate
'Decentralized' Blockchain Network
https://ksltv.com/516749/why-are-antennas-popping-up-all-over-the-foothills-salt-lake-city-seeks-to-solve-mystery/
Salt Lake City officials are finding mysterious antennas across the
foothills of the Salt
Fear Of COVID Is The Opiate Of The People
https://brownstone.org/articles/fear-of-covid-is-the-opiate-of-the-people/
https://markoshinskie8de.substack.com/p/getting-covid-or-something-like-that
After all of the criticism I’ve directed toward Coronamaniacs and the
Vaxxmongers over the past three y
Is The New World Order On The Precipice?
https://www.theburningplatform.com/2023/01/06/is-the-new-world-order-on-the-precipice/
The New World Order nobility may be receiving more roadblocks for
their plans than they had anticipated.
The entire scam is based on the ignorance and complacence of
Will Universities Ever Admit They Were Wrong About COVID Policy?
https://www.americanthinker.com/blog/2023/01/will_universities_ever_admit_they_were_wrong_about_covid_policy.html
https://www.wsj.com/articles/when-will-academia-account-for-its-covid-failures-pandemic-lockdowns-stanford-ivy-league-
On Sat, 7 Jan 2023 21:33:34 -0500
grarpamp wrote:
> >> Yet again, Juan is refusing to call out his Govt.
> > meh
>
> Here Juan again dodges and refuses to call out his Govt.
I already explained, multiple times, that 'my' govt is actually
grancrap's jewnazi government.
I thi
https://anarchapulco.com/
Anarchapulco - February 6-10th, 2023
Agoraforko - February 11-18th, 2023
And more including...
Jeff Berwick The Dollar Vigilante
Ron Paul The Ron Paul Liberty Report
Larken Rose The Most Dangerous Superstition
Max Igan The Crowhouse
Del Bigtree Highwire Podcast
Dan Dick
>> Yet again, Juan is refusing to call out his Govt.
> meh
Here Juan again dodges and refuses to call out his Govt.
Putting a silly word between the two does not diminish
the fact that you claimed it as "My Government".
No true "King of Anarchism" would ever associate the word
"My", even weakly,
whew this left me confusedish!
unsure if i want to continue immediately. really cool project.
certainly more productive than other things.
i bet finetuning is far more helpful than i downplayed it as.
On Sat, 7 Jan 2023 18:58:41 -0500
grarpamp wrote:
> Yet again, Juan is refusing to call out his Govt.
meh
oh, and for completness sake. The stupid, infinitely evil, and lying
piece of shit grancrap keeps lying.
I've commented on the argentine government multiple time
Yet again, Juan is refusing to call out his Govt. The simple fact that he
admits via convenient omission here is that he would side with and
flat out worship his GovtPols if they were not "US puppets", and that
he does, via such omission, actively worship and refuse to call out all
aspects branches
i updated the separator token back to using the padding token which
seems to work better
i blame the poorness of the data augmentation foremost on not tuning
the parameters of sampling (that it is sampling too low probabilities
for the small number of examples it has)
it would be interesting and u
oh of course to update a dependency library you need to reset things.
it has old version cached.
succeeded when i reset the colab and used an old commit
On 1/7/23, Undescribed Horrific Abuse, One Victim & Survivor of Many
wrote:
> with 3 examples (i added one for sorting) it gives a correct answer on
> colab
>
> npapi.set_example(
> '''Construct an array B that repeats array Y 4 times along the last
> dimension.''',
> '''B = np.tile(Y, 4)'''
> )
>
one of the things i changed today was the padding token in the middle
of the text
i changed it to the eos token i think
this is because the setup complicated the generation of multiple outputs at once
it is performing worse with the change. no longer producing accurate
completion on colab.
here is the failed implementation i tried to make today, and the
working implementation i had made some time ago. i haven't looked at
them side-by-side.
simple_beam_search is the one i tried to do today
try_multigen is the one from some time ago, my state of mind was
trying out parts for synthetic
i spent most of today exploring the concept of beam search.
the reason is, i ran into difficulty using the prompt for data
augmentation (which means generating more data than there really is,
for training models).
long story short, the huggingface beam search implementation produces
the top likel
On Sat, 7 Jan 2023 16:15:41 -0500
grarpamp wrote:
> still refuses to say a word aboutGovPols system in his own country that
> are ruling over him
oh, and for completness sake. The stupid, infinitely evil, and lying
piece of shit grancrap keeps lying.
I've commented on the
When grancrap talks about the demrats we can hear very loud hysterical
screeches accusing the demrats of being 'the far left'. Listening to
grancrap's hysterical screeches one would assume that grancrap himself isn't a
demrat. I mean he's trying to sound like he doesn't like them...?
Juan and his Govt, like two lovebirds sitting in a tree, K-I-S-S-I-N-G,
First comes Love, then comes Marriage, then comes another Tyrant
to rule over him and everyone else in a Baby Carriage.
xD
O
jewnazi grancrap grarpamp @ gmail.com - anonymous coward hiding behind
jooglemail vomited :
> History is full of "leaders" and "countries" initiating violence
> by creating and launching their militaries across border regions
yes, and your trumponazi cuntry is the biggest offender.
History is full of "leaders" and "countries" initiating violence
by creating and launching their militaries across border regions
murdering mass numbers of people and destruction on the other side.
Unsortable minor tit-for-tat border region loveplay skirmishes and geopolitical
baiting aside, Putin
"Washington’s proxy war with Russia is the result of a carefully
constructed plan to embroil Russia in conflict with its Ukrainian neighbor.
>From the moment that President Putin indicated that his government would not
tolerate a NATO military presence on Russia’s doorstep in Ukraine
> Are you fucking awake yet, because you sound like
> a bunch of sleeping sheep.
Those MaxiSheeple and CypherSheep who do not continue to speak
and educate others on Freedom from Tyranny and its Money,
will end up with a global Tyranny of CBDC's over them shutting
down their every freedom foreverm
Now that the US Govt Pols have violently lunged to secure their own internal
Power Protests, they will again turn their needless abject violence, theft, and
Authoritarian quest for ever more Power back out against the population...
Here Are The Concessions McCarthy Had To Make For Speakership
ht
Democrat SocCom's and their "do no work, get free shit" failboat
of ruin power and endless debt and poverty forced upon others
has gone 2.0 Invader Style for years now courtesy Biden...
Texas Sues Biden Admin Over Public Charge Immigration Rule Change
https://www.texasattorneygeneral.gov/sites/d
Ashli Babbit was Murdered by Govt power, instead of prosecuting,
the corrupt Govt power system let the murderer run free and arrested
her mother for refusing the irrational Govt's bogus "order"...
Mother Of Ashli Babbitt Arrested For "Jaywalking" Near Capitol While Protesting
https://www.youtube.
https://en.wikipedia.org/wiki/Visible_light_communication
In telecommunications, visible light communication (VLC) is the use of
visible light (light with a frequency of 400–800 THz/wavelength of 780–375
nm) as a transmission medium. VLC is a subset of optical wireless
communications technologies.
notes:
using model_kwargs=dict(device_map='auto') you can squeeze slightly
larger models into a system
npapi.set_example(
'''Construct an array B that repeats array Y 4 times along the last
dimension.''',
'''B = np.tile(Y, 4)'''
)
npapi.set_example(
'''Swap the last two axes of an array D.''',
''
So, this isn't the final algorithm. It's the very beginning of
bootstrapping toward other apis
with 3 examples (i added one for sorting) it gives a correct answer on colab
npapi.set_example(
'''Construct an array B that repeats array Y 4 times along the last
dimension.''',
'''B = np.tile(Y, 4)'''
)
npapi.set_example(
'''Swap the last two axes of an array D.''',
'''D = D.swapaxes(-2, -1)'''
i added a swapaxes example and it produced different iutput
i'm on mobile, but some fiddling with akeyboard could probably find
something thatworks. i mjght try more examples
Empire State of Mind.
-- Forwarded message -
From: Gunnar Larson
Date: Thu, Dec 1, 2022, 11:40 AM
Subject: SpaceX Interlocking Directors’ Intent to Manipulate Markets
To: cypherpunks
-- Forwarded message -
From: Gunnar Larson
Date: Sat, Jun 18, 2022, 4:44 PM
Su
it's still a useful step, if this usecase collapses for now.
thoughts are getting ahead of me. obviously it would need more than
simply a prompt, to be correct for unfamiliar api surface.
this is simply a first step.
some parts of the caching code werelost inpasting, it still runs
here is url https://github.com/xloem/apifudge/blob/main/test.ipynb
i used colab to run it thru bloomz-3b. it still needs another example
or something, but produces new content now
Open In Colab
In [ ]:
!pip3 install adapter-transformers accelerate bitsandbytes
!git clone https://github.com/xloem/apifudge
!nvidia-smi
!ln -vsf apifudge/* .
Cloning into 'apifudge'
https://www.shotspotter.com/precision-policing-platform-technology/
For 21st century law enforcement to succeed agencies need to be
evidence-based, proactive, and community-minded. Learn how our Precision
Policing Platform, which encompasses our Respond, COPLINK X, Investigate,
and Connect product
-- Forwarded message -
From: Gunnar Larson
Date: Sat, Jun 18, 2022, 4:44 PM
Subject: SpaceX Interlocking Directors’ Intent to Manipulate Markets
To:
Cc: , , ,
cypherpunks , , <
prospectus...@ny.email.gs.com>, , Weber,
Richard (DFS) , Alexi Anania
Dear Attorney General Letitia J
Author: Roger C. Park
Title: The Entrapment Controversy
Source: Minnesota Law Review
Citation: 60 Minn. L. Rev. 163 (1976).
212 Highlights by Brody Larson:
https://drive.google.com/file/d/1rhPWgd_OniZVLeJIlieDBx5VNC2chenn/view?usp=drivesdk
“And the persons who ultimately are harmed by all of this, of course, is
the average consumer, the average citizen, who has no knowledge, unless it
paid attention to this hearing, of the extent of the manipulations that
have been carried out by the largest financial institutions in America and,
ind
I tried adding another example, and changing the prompt format, but
the tiny poorly-quantized model i had kept simply repeating the last
example back.
I think it would make sense to try it on colab or such, with a larger model.
i'm getting sudden mysterious back pain and am planning now to take a
i've committed and pushed code where test.py generates a new
completion. the completion is incorrect. a second example may help. i
might try reviewing and pasting the example rwkv generated.
so i just made it float16 instead of bfloat16, which i think means worse output.
Not much text to add here,
but I was thinking on how precious rational feedback from expressions
is, and how rare;
and also how hard it is to provide this reliably, when the person (I
was actually thinking myself here, but professor rat could be an
example the other way as i rarely understand wha
i think it happens when it processes more than 16 tokens
that was not the issue cause, and i am further troubleshooting it
the error seems to be triggered only when i provide the examples
i am using the bos/eos tokens to separate them, which is not normal.
it seems the most likely thing to make sense causing it. the model
does not usually complete text containing eos tokens.
actually the error is different:
RuntimeError: CUDA error: CUBLAS_STATUS_NOT_SUPPORTED when calling
`cublasGemmStridedBatchedExFix(handle, opa, opb, (int)m,
(int)n, (int)k, (void*)&falpha, a, CUDA_R_16BF, (int)lda, stridea, b,
CUDA_R_16BF, (int)ldb, strideb, (void*)&fbeta, c,
CUDA_R_16BF, (int)ldc
i'm getting that weird error when i run my test code, but not from
within the repl.
it looks as if i've made a logic error implementing it, with the error
as a side effect.
i think it was because i exhausted gpu ram in that torch session
this fits on a 2GB gpu (and gives poor resullts, which is expected):
import transformers, torch
bloomz = transformers.pipeline('text-generation', 'bigscience/bloomz-560m')
bloomz.model.to(torch.bfloat16)
bloomz.model.to('cuda')
bloo
i running into this error trying to run the model on my gpu (after
converting to a 16-bit format so it fit): RuntimeError: CUDA error:
CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)`
I've hit this error before, and recall it's very frustrating, but not
what it is or how to resolv
60 matches
Mail list logo
