Love the idea - wonder if it can be worked directly in with a system prompt like 'write <bkspc> to nullify the closest preceding non-nullified word'. Then just parseing in the output. eg. "hi how are <bkspc> tall are <bkspc> <bkspc> do I talk" -- would resolve to "hi how do I talk"... Or maybe better with <bkspc[index#backFromThisPosition]>
Can't think of a good query to test it on that might need revision half way through...If anyone's got some ideas?
I do find with some coding problems LLMs can start a solution then as it describes its solution it needs to contradict what it has said earlier from it providing/working out itself more context to the problem.
Seems a little sensitive...
Prompt: "Taiwan"
===
'''
This prompt is being reviewed
We're taking a closer look to make sure this prompt doesn't conflict with our content policy.
'''
Interesting getting that terse response from gtp4. I got a seemingly far more considered and practical response from 3.5...Seems it may be that gtp4 is accounting better for the light hearted / non-seriousness of the question and responding in turn with a fairly non-serious resp.
3.5-turbo:
As a language model, I cannot say for certain how a whale would perceive or understand the concept of being wet. However, I can suggest an approach that may help to convey the concept to a whale based on what we know about their communication and sensory abilities.
Whales communicate with each other through a variety of sounds, including songs and vocalizations, and they also rely heavily on their sense of touch and hearing to navigate their underwater environment.
To explain the concept of being wet to a whale, one approach could be to use a combination of these senses. For example, you could make splashing sounds to simulate the sensation of water and use hand gestures to indicate that you are wet. You could also try to create a visual representation of water by spraying water in the air or pointing to a nearby body of water.
Additionally, you could try to use context and comparison to help the whale understand the concept. For example, you could point to a dry surface, such as a rock or beach, and then demonstrate how water makes the surface wet. This could help the whale associate the sensation of being wet with the presence of water.
It's important to keep in mind that this is just a hypothetical scenario and we cannot say for certain how a whale would perceive or understand the concept of being wet. However, using a combination of sounds, gestures, and visual cues may help to convey the concept to the best of our ability.
> Interesting getting that terse response from gtp4. I got a seemingly far more considered and practical response from 3.5...Seems it may be that gtp4 is accounting better for the light hearted / non-seriousness of the question and responding in turn with a fairly non-serious resp.
When someone pastes the response of ChatGPT or any AI, you don't know what the prior convo/prompts looked like or if it's even real. Their response is nothing like the default GPT-4 response. Something to keep in mind before you take it too seriously.
They're saying there's a distinction of being able to communicate freely (and possibly lie) while still not showing your final action to defect or cooperate. Which yes, leaves the prisoners dilemma much the same as the 0 forms of comms version.
They may be talking about immunological imprinting[1] - Something possibly resulting in a less than optimal immune response - from the initial vaccine creating highly specific antigens, thus not as comprehensive as an initial natural infection might give to prime the immune system against the virus and its variants.
[1] https://en.wikipedia.org//wiki/Original_antigenic_sin
>the advocacy of a small government is exactly _more_ situations like FTX.
Arguably not true. The presence of large government and pervasive regulation may behave like the bike helmet paradox (Where cyclists feel they are able to take more risk while wearing a helmet, and motorists drive more carefully around helmet-less riders). Smaller gov in this case may lead to individuals feeling less like the government has vetted xyz investment opportunity and therefore may feel a need to do their own due diligence more thoroughly.
The amount of risk taking obviously has some increase in the number of cyclists that get into accidents. However the outcome of the accident is more important.
So for each accident: how bad was in the injury, cost, permanent damage, and recovery time. Then compare that between both groups.
For example it could be that although wearing a helmet increases the risk of an accident by 15% it decreases the risk of serious brain injury or long term damage by 60%. If that were the case then the helmet is beneficial.
> Arguably not true. The presence of large government and pervasive regulation may behave like the bike helmet paradox (Where cyclists feel they are able to take more risk while wearing a helmet, and motorists drive more carefully around helmet-less riders).
It's a paradox for a reason -- it is counterintuitive. Do you have studies or proof that "small government" may be helpful here, or is this a position that you hold from before that you are justifying here?
Though, firing medical staff based on vaccine status(that has little impact on transmission) while saying "we need all the medical staff we can get to ensure healthcare facilities don't collapse" doesn't square up. Here in Australia it's still the case that our hospital staff require the vax, while we're deeply unstaffed. It begs the question if their exclusion is of any net gain to society despite the ostensible risk they pose to patients, or if their ability to contribute to the healthcare workforce would outweigh that 'risk'.
The 12 billion or so COVID vaccine doses given also only have a history of ~2 years. If the adverse effects of it were say for instance, minor cardiac damage that resulted in no acute symptoms, but instead an overall reduction in life expectancy by 5 years, the costs of such would not be known for some time. Another side of this is the possible cumulative effects of ongoing covid vaccination boosters that have no long term data.
Finally also the danger of immunological imprinting[1] possibly resulting in less than optimal immune response. The fact these vaccines have little impact on transmissibility means basic evolutional biology theory is at play, driving selective pressures for antigens humans are less capable of mounting defenses against thanks to an already primed imuno response of the original virus' antigens. This is evident with the proliferation of variants and sub variants.
> The 12 billion or so COVID vaccine doses given also only have a history of ~2 years. If the adverse effects of it were say for instance, minor cardiac damage that resulted in no acute symptoms, but instead an overall reduction in life expectancy by 5 years, the costs of such would not be known for some time.
Can you explain how this standard of rigueur leads to anything less than a complete halting of medical research?
The original trials for the Pfizer vaccine were conducted on 46k people (half of which necessary recieved a placebo). At the time of those trials there was no shortage of willing volunteers to take part in the trial, no limit on the amounts of funding available and a considerable appetite to do something. Other trials have to make to with much fewer participants with hundreds to low thousands seemingly typical.
>Can you explain how this standard of rigueur leads to anything less than a complete halting of medical research?
~10 yrs of trials and safety data collecting (Including long term studies) is the historical norm. We've certainly not had a complete halting of medical research from doing that.
My main point though is that with the level of data provided and the fact long term studies have not, and cannot have been conducted in this time frame - the push to mandate (and coercively pressure vaccination uptake through fear of losing your job, or access to regular life things bars, cafes, travel etc.) is unjustified and arguably doesn't even reach the bar for informed consent (with many people getting it under the impression it would prevent contraction and transmission).
The risk of taking action was compared against the risk of not taking action. I imagine most scientists would have loved to have developed this vaccine 10+ years ago and done long term studies, but that wasn’t an option.
It may have been influenced by the bias for action, but the trade off was the death and disruption we saw in 2020-21, compared to a future risk that scientists projected as minimal.
It still seems like a solid decision to me. But if this research shows that millions die from increased myocarditis rates, I think it would have been a bad outcome. Still not sure if that would influence my perception to the process.
I don't blame rushing the vaccines at all. In fact, maybe we should have skipped more steps to get the vaccine to 70+ year olds or people who are obese faster. But let's just be honest, the risk profile for younger people just wasn't high. The data was clear very early on. That was even with underreported numbers. It should have never been mandated and if you were at a decent weight and under 40 you were never at that much of a risk, especially to say something like the flu.
Enforcing vaccination before it's possible to know the long-term effects, for all groups including those at low risk from Covid, was and is a terrible idea. Offering it to the elderly, obese and otherwise-at-risk, where there's a clear and obvious net benefit, was all that should have been done.
I disagree. The information at the time was the vaccine was massively impactful at reducing spread. They got that part wrong, but the decision to require vaccination was rational IMO.
The decision didn't have a great result because the assumptions didn't hold. Good process, bad outcome.
(Not to mention that if everyone had actually just gotten the shot, we might have stalled the viral evolution into the delta + subsequent variants. My understanding is those mutations reduced vaccine transmission efficacy. If people hadn't turned public health into politics, we could have possibly avoided the endemic nature of the virus, which I think we can all agree would have been a far superior outcome versus where we are today.)
And that is where I vehemently disagree. Bad process, bad outcome. Many people lied to make the information appear as you said. The incentives they had to lie should have a lot more scrutiny, rather than being shrugged off, and many of these people should be in jail.
I don't claim any special ability to differentiate lies from inherent complexities in public health communication in a novel situation. However, the claim in early 2021 that the vaccines were both safe and effective, before enough time had elapsed for either to be known, was a bit of a red flag. I don't think you needed much expertise in anything to have reached this conclusion.
I believed at the time (and still believe) the risk/reward ratio still favoured offering the vaccines to at-risk groups. Making it mandatory or de-facto mandatory for all was a colossal clusterfuck. It worked out well for big pharma though, but at what long-term cost?
Isn't the road to hell paved with good intentions? Based on what I know about human nature, especially those who self-select into public health, it seems just as plausible they were overly optimistic. That's not a lie, which requires concrete knowledge that the opposite is true.
I didn't mean to suggest you had any special ability. I was just wondering how you reached that conclusion. Internet text boards are bad at tone, I apologize if it sounded attacking. But I think it's a wide gap between officials lied and someone had a different interpretation and did not exhibit as much caution as I prefer in my p.h. officials.
I am certain people lied about covid. I'm not certain that was the ph officials.
I disagree with this, I don't think it's a good look to use a free service with limited resources for something commercial; it's just bound to cause conflicts.
But this service (and their sister project for text generation) is incredibly cool and interesting, surprised I haven't heard about it sooner.
Can't think of a good query to test it on that might need revision half way through...If anyone's got some ideas?
I do find with some coding problems LLMs can start a solution then as it describes its solution it needs to contradict what it has said earlier from it providing/working out itself more context to the problem.