Split (1)

train · 57.9k rows

id stringlengths 36 36	source stringclasses 15 values	formatted_source stringclasses 13 values	text stringlengths 2 7.55M
66f8a59f-a967-45f6-845f-43ccb6409a66	trentmkelly/LessWrong-43k	LessWrong	Meetup : West LA Meetup - Nomic Discussion article for the meetup : West LA Meetup - Nomic WHEN: 09 May 2012 07:00:00PM (-0700) WHERE: 10850 West Pico Blvd, Los Angeles, CA 90064 When: 7:00pm - 9:00pm Wednesday, May 9th. Where: The Westside Tavern in the upstairs Wine Bar (all ages welcome), located inside the Westside Pavillion on the second floor, right by the movie theaters. Parking is free for 3 hours. Activity: This week, we will begin a game of Nomic, the most meta game known to man. This is a game in which changing the rules of the game constitutes a "move". I will pick one of the more common initial rulesets. Don't worry if you don't have time to read any articles, or even if you've never read any Less Wrong! Bring a friend! The atmosphere is casual, and good, intelligent conversation with friendly people is guaranteed. I will bring a whiteboard with Bayes' Theorem written on it. Discussion article for the meetup : West LA Meetup - Nomic
3070a778-9310-46b7-8c6f-ea6b17a2c2e6	trentmkelly/LessWrong-43k	LessWrong	[Linkpost] Building Altruistic and Moral AI Agent with Brain-inspired Affective Empathy Mechanisms Abstract > As AI closely interacts with human society, it is crucial to ensure that its decision-making is safe, altruistic, and aligned with human ethical and moral values. However, existing research on embedding ethical and moral considerations into AI remains insufficient, and previous external constraints based on principles and rules are inadequate to provide AI with longterm stability and generalization capabilities. In contrast, the intrinsic altruistic motivation based on empathy is more willing, spontaneous, and robust. Therefore, this paper is dedicated to autonomously driving intelligent agents to acquire morally behaviors through human-like affective empathy mechanisms. We draw inspiration from the neural mechanism of human brain’s moral intuitive decision-making, and simulate the mirror neuron system to construct a brain-inspired affective empathydriven altruistic decision-making model. Here, empathy directly impacts dopamine release to form intrinsic altruistic motivation. Based on the principle of moral utilitarianism, we design the moral reward function that integrates intrinsic empathy and extrinsic self-task goals. A comprehensive experimental scenario incorporating empathetic processes, personal objectives, and altruistic goals is developed. The proposed model enables the agent to make consistent moral decisions (prioritizing altruism) by balancing self-interest with the well-being of others. We further introduce inhibitory neurons to regulate different levels of empathy and verify the positive correlation between empathy levels and altruistic preferences, yielding conclusions consistent with findings from psychological behavioral experiments. This work provides a feasible solution for the development of ethical AI by leveraging the intrinsic human-like empathy mechanisms, and contributes to the harmonious coexistence between humans and AI. [emphasis mine]
9006b312-244b-4a14-a388-8f4c2be541c5	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Winners-take-how-much? The treatments I have seen to date of the "winner-takes-all'' dynamic in AI development focus on the risk-taking engendered by this dynamic: the more AI researchers have to gain from being early movers, the stronger the incentive to cut corners in safety measures. Concern about a result in which a small number of winners has successfully emerged, without, that is, blowing everything up in the effort, seems to be muted. "Winner-takes-all," in other words, is intended only as a term of art from conventional political economy, where "all" refers to the spoils of an otherwise limited contest between willing contestants. The runners-up may lose their investments, but not their lives. To put it another way, we could describe the prevailing model as having three contests: 1) between all humans and a hypothetical uncontrolled AI; 2) between firms racing to establish monopolistic advantage in the production and sale of controlled AI; 3) between humans in general and these firms over the unprecedented risks imposed by the latter on the former, intermixed with a conventional struggle between the former as consumers and the latter as would-be monopoly producers. Of the three contests, the risk of unconstrained rivalrous behavior -- that is, a fight to the death -- is only considered to apply to the one between the hypothetical uncontrolled AI and all humans, collectively. What seems to me to be missing from this model is a rivalrous intra-species fight to the death. This should be surprising, given that we posit a jealous AI, built and trained by humans, acquiring and pursuing goals that put it in opposition to the existence of 100% of humans. We should also fear humans with AI acquiring and pursuing goals that put them in opposition to the existence of the other humans. This could be treated as a semantic problem, if we define any genocidal application of AI to be a form of "misalignment," whether the AI is manipulating humans to achieve murderous ends or vice versa. My test for the usefulness of a distinction between these two cases is whether or not we can imagine a significant subset of AI development firms willingly pursuing the genocidal path, rather than treating all such paths as equally aberrant. This test is, after all, where the rubber meets the road. Obviously, all firms wish to avoid a paperclip maximizer (that is the whole point of the paperclip maximizer example, to describe a being using its superintelligence to pursue horrifically ludicrous ends). We can imagine human genocidal fanatics and cultists aiming to acquire AI technology, but a random sample of AI developers would deem such goals to be contrary to their own goals and, furthermore, self-defeating in the long run, in the sense that they would be contributing to a world that no one should objectively prefer to live in -- a world filled with lethal strife over religious beliefs. But there are conditions under which genocidal goals would be rational. On the contrary, willingly suffering a perpetual competition with 8-10 billion other people for the planet's limited resources is generally irrational, given a good alternative. So if one could shrink the population down by, say, 99.9%, the world could be a permanently better place for the remaining population. Unlike ideologically driven genocide, such a pivotal event would establish no self-defeating precedent, as all the newly down-sized population would have to do to avoid further genocide is to maintain itself at a mutually agreed-upon size, a far less onerous task than, say, maintaining homogeneity of belief. In short, I don't think it would be impossible or even that difficult to convince a significant set of actors capable of AI development to view a "small is beautiful" utopia as a goal that is aligned with their own interests. Therefore, this concern does not belong in the same "X-risk" class as those involving the extinction of humanity (via paperclip maximization or otherwise). It may be that the only way to avoid the universally undesirable outcome of AI killing all humans is via avoidance of killing any humans, or some other small number. Although I can't speak knowledgeably about AI safety methods, this seems like it would be a happy but unlikely coincidence. What seems to me to be more likely is the opposite, that a small population would actually decrease the chances of a complete AI disaster, because a small population can more easily govern and monitor its own application of AI, and because the power and sophistication of its automation could operate at a much smaller scale, leaving more room between AI usefulness and risk. I suspect that similar reasoning applies to the problem of getting to a small population. With enough collusion between AI developers and the element of surprise, the artificial intelligence requirements to simply kill 99.9% of the population in an isolated event perhaps need not be very close to human level (see [this post](https://www.lesswrong.com/posts/7SBAjSjENLyzdnC2f/minimum-viable-exterminator) for a treatment of the minimum intelligence requirements for killing lots of people). So I'm concerned that the "winner-takes-all" contest to pay attention to is not the gentleman's innovation race between AI development firms to bring the best AI to consumers (safety be damned!), but between any combination of small groups armed with AI that is "just good enough" and the rest of population for a place on this planet. The upshot would be that, from the perspective of the average person, whether or not AI ends up taking the operators down with everyone else is purely an academic concern.
b1c0ace6-92c5-49a4-a1df-9b65543f9a96	trentmkelly/LessWrong-43k	LessWrong	Funding Good Research Series: How to Purchase AI Risk Reduction I recently explained that one major project undergoing cost-benefit analysis at the Singularity Institute is that of a scholarly AI risk wiki. The proposal is exciting to many, but as Kaj Sotala points out: > This idea sounds promising, but I find it hard to say anything about "should this be funded" without knowing what the alternative uses for the money are. Almost any use of money can be made to sound attractive with some effort, but the crucial question in budgeting is not "would this be useful" but "would this be the most useful thing". Indeed. So here is another thing that donations to SI could purchase: good research papers by skilled academics. Our recent grant of $20,000 to Rachael Briggs (for an introductory paper on TDT) provides an example of how this works: 1. SI thinks of a paper it wants to exist but doesn't have the resources to write itself (e.g. a clearer presentation of TDT). 2. SI looks for a few productive academics well-suited to write the paper we have in mind, and approaches them directly with the grant proposal. (Briggs is an excellent choice for the TDT paper because she is a good explainer and has had two of her past decision theory papers selected as among the 10 best papers of the year by The Philosopher's Annual.) 3. Hopefully, one of these academics says "yes." We award them the grant in return for a certain kind of paper published in one of a pre-specified set of journals. (In the case of the TDT grant to Rachael Briggs, we specified that the final paper must be published in one of the following journals: Philosopher's Imprint, Philosophy and Phenomenological Research, Philosophical Quarterly, Philosophical Studies, Erkenntnis, Theoria, Australasian Journal of Philosophy, Nous, The Philosophical Review, or Theory and Decision.) 4. SI gives regular feedback on outline drafts and article drafts prepared by the article author. 5. Paper gets submitted, revised, and published! For ex
ad0e8b5b-54c7-4f4e-875e-04a5ac899527	trentmkelly/LessWrong-43k	LessWrong	Remaking EfficientZero (as best I can) Introduction When I first heard about EfficientZero, I was amazed that it could learn at a sample efficiency comparable to humans. What's more, it was doing it without the gigantic amount of pre-training the humans have, which I'd always felt made comparing sample efficiencies with humans rather unfair. I also wanted to practice my ML programming, so I thought I'd make my own version. This article uses what I've learned to give you an idea, not just of how the EfficientZero algorithm works, but also of what it looks like to implement in practice. The algorithm itself has already been well covered in a LessWrong post here. That article inspired me to write this and if it's completely new to you it might be a good place to start - the focus here will be more on what the algorithm looks like as a piece of code. The code below is all written by me and comes from a cleaned and extra-commented version of EfficientZero which draws from the papers (MuZero, Efficient Zero), the open implementation pf MuZero by Werner Duvaud, the pseudocode provided by the MuZero paper, and the original implementation of EfficientZero. You can have a look at the full code and run it at on github. It's currently functional and works on trivial games like cartpole but struggles to learn much on Atari games within a reasonable timeframe, not certain if this reflects an error or just insufficient time. Testing on my laptop or Colab for Atari games is slow - if anyone could give access to some compute to do proper testing that would be amazing! Grateful to Misha Wagner for feedback on both code and post. Algorithm Overview AlphaZero EfficientZero is based on MuZero, which itself is based on AlphaZero, a refinement of the architecture which was the first beat the Go world champion. With AlphaZero, you play a deterministic game, like chess, by developing a neural network that evaluates game states, associating each possible state of the board with a value, the discounted expected return (in
895915d3-bf8c-4ecf-8bc3-5fd21879017a	StampyAI/alignment-research-dataset/blogs	Blogs	AGI Ruin: A List of Lethalities ### Preamble: (If you’re already familiar with all basics and don’t want any preamble, skip ahead to [Section B](https://intelligence.org/feed/?paged=2#Section_B_) for technical difficulties of alignment proper.) I have several times failed to write up a well-organized list of reasons why AGI will kill you. People come in with different ideas about why AGI would be survivable, and want to hear different obviously keypoints addressed first. Some fraction of those people are loudly upset with me if the obviously most important points aren’t addressed immediately, and I address different points first instead. Having failed to solve this problem in any good way, I now give up and solve it poorly with a poorly organized list of individual rants. I’m not particularly happy with this list; the alternative was publishing nothing, and publishing this seems marginally more [dignified](https://www.lesswrong.com/posts/j9Q8bRmwCgXRYAgcJ/miri-announces-new-death-with-dignity-strategy). Three points about the general subject matter of discussion here, numbered so as not to conflict with the list of lethalities: -3. I’m assuming you are already familiar with some basics, and already know what ‘[orthogonality](https://arbital.com/p/orthogonality/)’ and ‘[instrumental convergence](https://arbital.com/p/instrumental_convergence/)’ are and why they’re true. People occasionally claim to me that I need to stop fighting old wars here, because, those people claim to me, those wars have already been won within the important-according-to-them parts of the current audience. I suppose it’s at least true that none of the current major EA funders seem to be visibly in denial about orthogonality or instrumental convergence as such; so, fine. If you don’t know what ‘orthogonality’ or ‘instrumental convergence’ are, or don’t see for yourself why they’re true, you need a different introduction than this one. -2. When I say that alignment is lethally difficult, I am not talking about ideal or perfect goals of ‘provable’ alignment, nor total alignment of superintelligences on exact human values, nor getting AIs to produce satisfactory arguments about moral dilemmas which sorta-reasonable humans disagree about, nor attaining an absolute certainty of an AI not killing everyone. When I say that alignment is difficult, I mean that in practice, using the techniques we actually have, “please don’t disassemble literally everyone with probability roughly 1” is an overly large ask that we are not on course to get. So far as I’m concerned, [if you can get a powerful AGI that carries out some pivotal superhuman engineering task, with a less than fifty percent change of killing more than one billion people](https://twitter.com/ESYudkowsky/status/1070095112791715846), I’ll take it. Even smaller chances of killing even fewer people would be a nice luxury, but if you can get as incredibly far as “less than roughly certain to kill everybody”, then you can probably get down to under a 5% chance with only slightly more effort. Practically all of the difficulty is in getting to “less than certainty of killing literally everyone”. Trolley problems are not an interesting subproblem in all of this; if there are any survivors, you solved alignment. At this point, I no longer care how it works, I don’t care how you got there, I am cause-agnostic about whatever methodology you used, all I am looking at is prospective results, all I want is that we have justifiable cause to believe of a pivotally useful AGI ‘this will not kill literally everyone’. Anybody telling you I’m asking for stricter ‘alignment’ than this has failed at reading comprehension. The big ask from AGI alignment, the basic challenge I am saying is too difficult, is to obtain by any strategy whatsoever a significant chance of there being any survivors. -1. None of this is about anything being impossible in principle. The metaphor I usually use is that if a textbook from one hundred years in the future fell into our hands, containing all of the simple ideas that actually work robustly in practice, we could probably build an aligned superintelligence in six months. For people schooled in machine learning, I use as my metaphor the difference between ReLU activations and sigmoid activations. Sigmoid activations are complicated and fragile, and do a terrible job of transmitting gradients through many layers; ReLUs are incredibly simple (for the unfamiliar, the activation function is literally max(x, 0)) and work much better. Most neural networks for the first decades of the field used sigmoids; the idea of ReLUs wasn’t discovered, validated, and popularized until decades later. What’s lethal is that we do not havethe Textbook From The Future telling us all the simple solutions that actually in real life just work and are robust; we’re going to be doing everything with metaphorical sigmoids on the first critical try. No difficulty discussed here about AGI alignment is claimed by me to be impossible – to merely human science and engineering, let alone in principle – if we had 100 years to solve it using unlimited retries, the way that science usually has an unbounded time budget and unlimited retries. This list of lethalities is about things we are not on course to solve in practice in time on the first critical try; none of it is meant to make a much stronger claim about things that areimpossible in principle. That said: Here, from my perspective, are some different true things that could be said, to contradict various false things that various different people seem to believe, about why AGI would be survivable on anything remotely remotely resembling the current pathway, or any other pathway we can easily jump to. ### Section A: This is a very lethal problem, it has to be solved one way or another, it has to be solved at a minimum strength and difficulty level instead of various easier modes that some dream about, we do not have any visible option of ‘everyone’ retreating to only solve safe weak problems instead, and failing on the first really dangerous try is fatal. 1. Alpha Zero blew past all accumulated human knowledge about Go after a day or so of self-play, with no reliance on human playbooks or sample games. Anyone relying on “well, it’ll get up to human capability at Go, but then have a hard time getting past that because it won’t be able to learn from humans any more” would have relied on vacuum. AGI will not be upper-bounded by human ability or human learning speed. Things much smarter than human would be able to learn from less evidence than humans require to have ideas driven into their brains; there are theoretical upper bounds here, but those upper bounds seem very high. (Eg, each bit of information that couldn’t already be fully predicted can eliminate at most half the probability mass of all hypotheses under consideration.) It is not naturally (by default, barring intervention) the case that everything takes place on a timescale that makes it easy for us to react. 2. A cognitive system with sufficiently high cognitive powers, given any medium-bandwidth channel of causal influence, will not find it difficult to bootstrap to overpowering capabilities independent of human infrastructure. The concrete example I usually use here is nanotech, because there’s been pretty detailed analysis of what definitely look like physically attainable lower bounds on what should be possible with nanotech, and those lower bounds are sufficient to carry the point. My lower-bound model of “how a sufficiently powerful intelligence would kill everyone, if it didn’t want to not do that” is that it gets access to the Internet, emails some DNA sequences to any of the many many online firms that will take a DNA sequence in the email and ship you back proteins, and bribes/persuades some human who has no idea they’re dealing with an AGI to mix proteins in a beaker, which then form a first-stage nanofactory which can build the actual nanomachinery. (Back when I was first deploying this visualization, the wise-sounding critics said “Ah, but how do you know even a superintelligence could solve the protein folding problem, if it didn’t already have planet-sized supercomputers?” but one hears less of this after the advent of AlphaFold 2, for some odd reason.) The nanomachinery builds diamondoid bacteria, that replicate with solar power and atmospheric CHON, maybe aggregate into some miniature rockets or jets so they can ride the jetstream to spread across the Earth’s atmosphere, get into human bloodstreams and hide, strike on a timer. Losing a conflict with a high-powered cognitive system looks at least as deadly as “everybody on the face of the Earth suddenly falls over dead within the same second”. (I am using awkward constructions like ‘high cognitive power’ because standard English terms like ‘smart’ or ‘intelligent’ appear to me to function largely as status synonyms. ‘Superintelligence’ sounds to most people like ‘something above the top of the status hierarchy that went to double college’, and they don’t understand why that would be all that dangerous? Earthlings have no word and indeed no standard native concept that means ‘actually useful cognitive power’. A large amount of failure to panic sufficiently, seems to me to stem from a lack of appreciation for the incredible potential lethality of this thing that Earthlings as a culture have not named.) 3. We need to get alignment right on the ‘first critical try’ at operating at a ‘dangerous’ level of intelligence, where unaligned operation at a dangerous level of intelligence kills everybody on Earth and then we don’t get to try again. This includes, for example: (a) something smart enough to build a nanosystem which has been explicitly authorized to build a nanosystem; or (b) something smart enough to build a nanosystem and also smart enough to gain unauthorized access to the Internet and pay a human to put together the ingredients for a nanosystem; or (c) something smart enough to get unauthorized access to the Internet and build something smarter than itself on the number of machines it can hack; or (d) something smart enough to treat humans as manipulable machinery and which has any authorized or unauthorized two-way causal channel with humans; or (e) something smart enough to improve itself enough to do (b) or (d); etcetera. We can gather all sorts of information beforehand from less powerful systems that will not kill us if we screw up operating them; but once we are running more powerful systems, we can no longer update on sufficiently catastrophic errors. This is where practically all of the real lethality comes from, that we have to get things right on the first sufficiently-critical try. If we had unlimited retries – if every time an AGI destroyed all the galaxies we got to go back in time four years and try again – we would in a hundred years figure out which bright ideas actually worked. Human beings can figure out pretty difficult things over time, when they get lots of tries; when a failed guess kills literally everyone, that is harder. That we have to get a bunch of key stuff right on the first try is where most of the lethality really and ultimately comes from; likewise the fact that no authority is here to tell us a list of what exactly is ‘key’ and will kill us if we get it wrong. (One remarks that most people are so absolutely and flatly unprepared by their ‘scientific’ educations to challenge pre-paradigmatic puzzles with no scholarly authoritative supervision, that they do not even realize how much harder that is, or how incredibly lethal it is to demand getting that right on the first critical try.) 4. We can’t just “decide not to build AGI” because GPUs are everywhere, and knowledge of algorithms is constantly being improved and published; 2 years after the leading actor has the capability to destroy the world, 5 other actors will have the capability to destroy the world. The given lethal challenge is to solve within a time limit, driven by the dynamic in which, over time, increasingly weak actors with a smaller and smaller fraction of total computing power, become able to build AGI and destroy the world. Powerful actors all refraining in unison from doing the suicidal thing just delays this time limit – it does not lift it, unless computer hardware and computer software progress are both brought to complete severe halts across the whole Earth. The current state of this cooperation to have every big actor refrain from doing the stupid thing, is that at present some large actors with a lot of researchers and computing power are led by people who vocally disdain all talk of AGI safety (eg Facebook AI Research). Note that needing to solve AGI alignment only within a time limit, but with unlimited safe retries for rapid experimentation on the full-powered system; or only on the first critical try, but with an unlimited time bound; would both be terrifically humanity-threatening challenges by historical standards individually. 5. We can’t just build a very weak system, which is less dangerous because it is so weak, and declare victory; because later there will be more actors that have the capability to build a stronger system and one of them will do so. I’ve also in the past called this the ‘safe-but-useless’ tradeoff, or ‘safe-vs-useful’. People keep on going “why don’t we only use AIs to do X, that seems safe” and the answer is almost always either “doing X in fact takes very powerful cognition that is not passively safe” or, even more commonly, “because restricting yourself to doing X will not prevent Facebook AI Research from destroying the world six months later”. If all you need is an object that doesn’t do dangerous things, you could try a sponge; a sponge is very passively safe. Building a sponge, however, does not prevent Facebook AI Research from destroying the world six months later when they catch up to the leading actor. 6. We need to align the performance of some large task, a ‘pivotal act’ that prevents other people from building an unaligned AGI that destroys the world. While the number of actors with AGI is few or one, they must execute some “pivotal act”, strong enough to flip the gameboard, using an AGI powerful enough to do that. It’s not enough to be able to align a weak system – we need to align a system that can do some single very large thing. The example I usually give is “burn all GPUs”. This is not what I think you’d actually want to do with a powerful AGI – the nanomachines would need to operate in an incredibly complicated open environment to hunt down all the GPUs, and that would be needlessly difficult to align. However, all known pivotal acts are currently outside the Overton Window, and I expect them to stay there. So I picked an example where if anybody says “how dare you propose burning all GPUs?” I can say “Oh, well, I don’t actually advocate doing that; it’s just a mild overestimate for the rough power level of what you’d have to do, and the rough level of machine cognition required to do that, in order to prevent somebody else from destroying the world in six months or three years.” (If it wasn’t a mild overestimate, then ‘burn all GPUs’ would actually be the minimal pivotal task and hence correct answer, and I wouldn’t be able to give that denial.) Many clever-sounding proposals for alignment fall apart as soon as you ask “How could you use this to align a system that you could use to shut down all the GPUs in the world?” because it’s then clear that the system can’t do something that powerful, or, if it can do that, the system wouldn’t be easy to align. A GPU-burner is also a system powerful enough to, and purportedly authorized to, build nanotechnology, so it requires operating in a dangerous domain at a dangerous level of intelligence and capability; and this goes along with any non-fantasy attempt to name a way an AGI could change the world such that a half-dozen other would-be AGI-builders won’t destroy the world 6 months later. 7. The reason why nobody in this community has successfully named a ‘pivotal weak act’ where you do something weak enough with an AGI to be passively safe, but powerful enough to prevent any other AGI from destroying the world a year later – and yet also we can’t just go do that right now and need to wait on AI – is that nothing like that exists. There’s no reason why it should exist. There is not some elaborate clever reason why it exists but nobody can see it. It takes a lot of power to do something to the current world that prevents any other AGI from coming into existence; nothing which can do that is passively safe in virtue of its weakness. If you can’t solve the problem right now (which you can’t, because you’re opposed to other actors who don’t want to be solved and those actors are on roughly the same level as you) then you are resorting to some cognitive system that can do things you could not figure out how to do yourself, that you were not close to figuring out because you are not closeto being able to, for example, burn all GPUs. Burning all GPUs would actually stop Facebook AI Research from destroying the world six months later; weaksauce Overton-abiding stuff about ‘improving public epistemology by setting GPT-4 loose on Twitter to provide scientifically literate arguments about everything’ will be cool but will not actually prevent Facebook AI Research from destroying the world six months later, or some eager open-source collaborative from destroying the world a year later if you manage to stop FAIR specifically. There are no pivotal weak acts. 8. The best and easiest-found-by-optimization algorithms for solving problems we want an AI to solve, readily generalize to problems we’d rather the AI not solve; you can’t build a system that only has the capability to drive red cars and not blue cars, because all red-car-driving algorithms generalize to the capability to drive blue cars. 9. The builders of a safe system, by hypothesis on such a thing being possible, would need to operate their system in a regime where it has the capabilityto kill everybody or make itself even more dangerous, but has been successfully designed to not do that. Running AGIs doing something pivotal are not passively safe, they’re the equivalent of nuclear cores that require actively maintained design properties to not go supercritical and melt down. ### Section B: Okay, but as we all know, modern machine learning is like a genie where you just give it a wish, right? Expressed as some mysterious thing called a ‘loss function’, but which is basically just equivalent to an English wish phrasing, right? And then if you pour in enough computing power you get your wish, right? So why not train a giant stack of transformer layers on a dataset of agents doing nice things and not bad things, throw in the word ‘corrigibility’ somewhere, crank up that computing power, and get out an aligned AGI? #### Section B.1: The distributional leap. 10. You can’t train alignment by running lethally dangerous cognitions, observing whether the outputs kill or deceive or corrupt the operators, assigning a loss, and doing supervised learning. On anything like the standard ML paradigm, you would need to somehow generalize optimization-for-alignment you did in safe conditions, across a big distributional shift to dangerous conditions. (Some generalization of this seems like it would have to be true even outside that paradigm; you wouldn’t be working on a live unaligned superintelligence to align it.) This alone is a point that is sufficient to kill a lot of naive proposals from people who never did or could concretely sketch out any specific scenario of what training they’d do, in order to align what output – which is why, of course, they never concretely sketch anything like that. Powerful AGIs doing dangerous things that will kill you if misaligned, must have an alignment property that generalized far out-of-distribution from safer building/training operations that didn’t kill you. This is where a huge amount of lethality comes from on anything remotely resembling the present paradigm. Unaligned operation at a dangerous level of intelligence\capability will kill you; so, if you’re starting with an unaligned system and labeling outputs in order to get it to learn alignment, the training regime or building regime must be operating at some lower level of intelligence\capability that is passively safe, where its currently-unaligned operation does not pose any threat. (Note that anything substantially smarter than you poses a threat given any realistic level of capability. Eg, “being able to produce outputs that humans look at” is probably sufficient for a generally much-smarter-than-human AGI to [navigate its way out of the causal systems that are humans](https://www.yudkowsky.net/singularity/aibox), especially in the real world where somebody trained the system on terabytes of Internet text, rather than somehow keeping it ignorant of the latent causes of its source code and training environments.) 11. If cognitive machinery doesn’t generalize far out of the distribution where you did tons of training, it can’t solve problems on the order of ‘build nanotechnology’ where it would be too expensive to run a million training runs of failing to build nanotechnology. There is no pivotal act this weak; there’s no known case where you can entrain a safe level of ability on a safe environment where you can cheaply do millions of runs, and deploy that capability to save the world and prevent the next AGI project up from destroying the world two years later. Pivotal weak acts like this aren’t known, and not for want of people looking for them. So, again, you end up needing alignment to generalize way out of the training distribution – not just because the training environment needs to be safe, but because the training environment probably also needs to be cheaper than evaluating some real-world domain in which the AGI needs to do some huge act. You don’t get 1000 failed tries at burning all GPUs – because people will notice, even leaving out the consequences of capabilities success and alignment failure. 12. Operating at a highly intelligent level is a drastic shift in distribution from operating at a less intelligent level, opening up new external options, and probably opening up even more new internal choices and modes. Problems that materialize at high intelligence and danger levels may fail to show up at safe lower levels of intelligence, or may recur after being suppressed by a first patch. 13. Many alignment problems of superintelligence will not naturally appear at pre-dangerous, passively-safe levels of capability. Consider the internal behavior ‘change your outer behavior to deliberately look more aligned and deceive the programmers, operators, and possibly any loss functions optimizing over you’. This problem is one that will appear at the superintelligent level; if, being otherwise ignorant, we guess that it is among the median such problems in terms of how early it naturally appears in earlier systems, then around half of the alignment problems of superintelligence will first naturally materialize afterthat one first starts to appear. Given correctforesight of which problems will naturally materialize later, one could try to deliberately materialize such problems earlier, and get in some observations of them. This helps to the extent (a) that we actually correctly forecast all of the problems that will appear later, or some superset of those; (b) that we succeed in preemptively materializing a superset of problems that will appear later; and (c) that we can actually solve, in the earlier laboratory that is out-of-distribution for us relative to the real problems, those alignment problems that would be lethal if we mishandle them when they materialize later. Anticipating allof the really dangerous ones, and then successfully materializing them, in the correct form for early solutions to generalize over to later solutions, sounds possibly kinda hard. 14. Some problems, like ‘the AGI has an option that (looks to it like) it could successfully kill and replace the programmers to fully optimize over its environment’, seem like their natural order of appearance could be that they first appear only in fully dangerous domains. Really actually having a clear option to brain-level-persuade the operators or escape onto the Internet, build nanotech, and destroy all of humanity – in a way where you’re fully clear that you know the relevant facts, and estimate only a not-worth-it low probability of learning something which changes your preferred strategy if you bide your time another month while further growing in capability – is an option that first gets evaluated for real at the point where an AGI fully expects it can defeat its creators. We can try to manifest an echo of that apparent scenario in earlier toy domains. Trying to train by gradient descent against that behavior, in that toy domain, is something I’d expect to produce not-particularly-coherent local patches to thought processes, which would break with near-certainty inside a superintelligence generalizing far outside the training distribution and thinking very different thoughts. Also, programmers and operators themselves, who are used to operating in not-fully-dangerous domains, are operating out-of-distribution when they enter into dangerous ones; our methodologies may at that time break. 15. Fast capability gains seem likely, and may break lots of previous alignment-required invariants simultaneously.Given otherwise insufficient foresight by the operators, I’d expect a lot of those problems to appear approximately simultaneously after a sharp capability gain. See, again, the case of human intelligence. We didn’t break alignment with the ‘inclusive reproductive fitness’ outer loss function, immediately after the introduction of farming – something like 40,000 years into a 50,000 year Cro-Magnon takeoff, as was itself running very quickly relative to the outer optimization loop of natural selection. Instead, we got a lot of technology more advanced than was in the ancestral environment, including contraception, in one very fast burst relative to the speed of the outer optimization loop, late in the general intelligence game. We started reflecting on ourselves a lot more, started being programmed a lot more by cultural evolution, and lots and lots of assumptions underlying our alignment in the ancestral training environment broke simultaneously. (People will perhaps rationalize reasons why this abstract description doesn’t carry over to gradient descent; eg, “gradient descent has less of an information bottleneck”. My model of this variety of reader has an inside view, which they will label an outside view, that assigns great relevance to some other data points that are not observed cases of an outer optimization loop producing an inner general intelligence, and assigns little importance to our one data point actually featuring the phenomenon in question. When an outer optimization loop actually produced general intelligence, it broke alignment after it turned general, and did so relatively late in the game of that general intelligence accumulating capability and knowledge, almost immediately before it turned ‘lethally’ dangerous relative to the outer optimization loop of natural selection. Consider skepticism, if someone is ignoring this one warning, especially if they are not presenting equally lethal and dangerous things that they say will go wrong instead.) #### Section B.2: Central difficulties of outer and inner alignment. 16.Even if you train really hard on an exact loss function, that doesn’t thereby create an explicit internal representation of the loss function inside an AI that then continues to pursue that exact loss function in distribution-shifted environments. Humans don’t explicitly pursue inclusive genetic fitness; outer optimization even on a very exact, very simple loss function doesn’t produce inner optimization in that direction. This happens in practice in real life,it is what happened in the only case we know about, and it seems to me that there are deep theoretical reasons to expect it to happen again: the firstsemi-outer-aligned solutions found, in the search ordering of a real-world bounded optimization process, are not inner-aligned solutions. This is sufficient on its own, even ignoring many other items on this list, to trash entire categories of naive alignment proposals which assume that if you optimize a bunch on a loss function calculated using some simple concept, you get perfect inner alignment on that concept. 17. More generally, a superproblem of ‘outer optimization doesn’t produce inner alignment’ is that on the current optimization paradigm there is no general idea of how to get particular inner properties into a system, or verify that they’re there, rather than just observable outer ones you can run a loss function over. This is a problem when you’re trying to generalize out of the original training distribution, because, eg, the outer behaviors you see could have been produced by an inner-misaligned system that is deliberately producing outer behaviors that will fool you. We don’t know how to get any bits of information into the inner system rather than the outer behaviors, in any systematic or general way, on the current optimization paradigm. 18. There’s no reliable Cartesian-sensory ground truth (reliable loss-function-calculator) about whether an output is ‘aligned’, because some outputs destroy (or fool) the human operators and produce a different environmental causal chain behind the externally-registered loss function. That is, if you show an agent a reward signal that’s currently being generated by humans, the signal is not in general a reliable perfect ground truth about how aligned an action was, because another way of producing a high reward signal is to deceive, corrupt, or replace the human operators with a different causal system which generates that reward signal. When you show an agent an environmental reward signal, you are not showing it something that is a reliable ground truth about whether the system did the thing you wanted it to do; even if it ends up perfectly inner-aligned on that reward signal, or learning some concept that exactly corresponds to ‘wanting states of the environment which result in a high reward signal being sent’, an AGI strongly optimizing on that signal will kill you, because the sensory reward signal was not a ground truth about alignment (as seen by the operators). 19. More generally, there is no known way to use the paradigm of loss functions, sensory inputs, and/or reward inputs, to optimize anything within a cognitive system to point at particular things within the environment – to point to latent events and objects and properties in the environment, rather than relatively shallow functions of the sense data and reward. This isn’t to say that nothing in the system’s goal (whatever goal accidentally ends up being inner-optimized over) could ever point to anything in the environment by accident. Humans ended up pointing to their environments at least partially, though we’ve got lots of internally oriented motivational pointers as well. But insofar as the current paradigm works at all, the on-paper design properties say that it only works for aligning on known direct functions of sense data and reward functions. All of these kill you if optimized-over by a sufficiently powerful intelligence, because they imply strategies like ‘kill everyone in the world using nanotech to strike before they know they’re in a battle, and have control of your reward button forever after’. It just isn’t true that we know a function on webcam input such that every world with that webcam showing the right things is safe for us creatures outside the webcam. This general problem is a fact about the territory, not the map; it’s a fact about the actual environment, not the particular optimizer, that lethal-to-us possibilities exist in some possible environments underlying every given sense input. 20. Human operators are fallible, breakable, and manipulable. Human raters make systematic errors – regular, compactly describable, predictable errors. To faithfully learn a function from ‘human feedback’ is to learn (from our external standpoint) an unfaithful description of human preferences, with errors that are not random (from the outside standpoint of what we’d hoped to transfer). If you perfectly learn and perfectly maximize the referent of rewards assigned by human operators, that kills them. It’s a fact about the territory, not the map – about the environment, not the optimizer – that the best predictive explanation for human answers is one that predicts the systematic errors in our responses, and therefore is a psychological concept that correctly predicts the higher scores that would be assigned to human-error-producing cases. 21. There’s something like a single answer, or a single bucket of answers, for questions like ‘What’s the environment really like?’ and ‘How do I figure out the environment?’ and ‘Which of my possible outputs interact with reality in a way that causes reality to have certain properties?’, where a simple outer optimization loop will straightforwardly shove optimizees into this bucket. When you have a wrong belief, reality hits back at your wrong predictions. When you have a broken belief-updater, reality hits back at your broken predictive mechanism via predictive losses, and a gradient descent update fixes the problem in a simple way that can easily cohere with all the other predictive stuff. In contrast, when it comes to a choice of utility function, there are unbounded degrees of freedom and multiple reflectively coherent fixpoints. Reality doesn’t ‘hit back’ against things that are locally aligned with the loss function on a particular range of test cases, but globally misaligned on a wider range of test cases. This is the very abstract story about why hominids, once they finally started to generalize, generalized their capabilities to Moon landings, but their inner optimization no longer adhered very well to the outer-optimization goal of ‘relative inclusive reproductive fitness’ – even though they were in their ancestral environment optimized very strictly around this one thing and nothing else. This abstract dynamic is something you’d expect to be true about outer optimization loops on the order of both ‘natural selection’ and ‘gradient descent’. The central result: Capabilities generalize further than alignment once capabilities start to generalize far. 22. There’s a relatively simple core structure that explains why complicated cognitive machines work; which is why such a thing as general intelligence exists and not just a lot of unrelated special-purpose solutions; which is why capabilities generalize after outer optimization infuses them into something that has been optimized enough to become a powerful inner optimizer. The fact that this core structure is simple and relates generically to [low-entropy high-structure environments](https://intelligence.org/2017/12/06/chollet/) is why humans can walk on the Moon. There is no analogous truth about there being a simple core of alignment, especially not one that is even easier for gradient descent to find than it would have been for natural selection to just find ‘want inclusive reproductive fitness’ as a well-generalizing solution within ancestral humans. Therefore, capabilities generalize further out-of-distribution than alignment, once they start to generalize at all. 23. Corrigibility is anti-natural to consequentialist reasoning; “you can’t bring the coffee if you’re dead” for almost every kind of coffee. We (MIRI) [tried and failed](https://www.alignmentforum.org/posts/5bd75cc58225bf0670374f04/forum-digest-corrigibility-utility-indifference-and-related-control-ideas) to find a coherent formula for an agent that would let itself be shut down (without that agent actively trying to get shut down). Furthermore, many anti-corrigible lines of reasoning like this may only first appear at high levels of intelligence. 24. There are two fundamentally different approaches you can potentially take to alignment, which are unsolvable for two different sets of reasons; therefore, by becoming confused and ambiguating between the two approaches, you can confuse yourself about whether alignment is necessarily difficult. The first approach is to build a CEV-style Sovereign which wants exactly what we extrapolated-want and is therefore safe to let optimize all the future galaxies without it accepting any human input trying to stop it. The second course is to build corrigible AGI which doesn’t want exactly what we want, and yet somehow fails to kill us and take over the galaxies despite that being a convergent incentive there. 1. The first thing generally, or CEV specifically, is unworkable because the complexity of what needs to be aligned or meta-aligned for our Real Actual Values is far out of reach for our FIRST TRY at AGI. Yes I mean specifically that the dataset, meta-learning algorithm, and what needs to be learned, is far out of reach for our first try. It’s not just non-hand-codable, it is unteachableon-the-first-try because the thing you are trying to teach is too weird and complicated. 2. The second thing looks unworkable (less so than CEV, but still lethally unworkable) because corrigibility runs*actively counter*to instrumentally convergent behaviors within a core of general intelligence (the capability that generalizes far out of its original distribution). You’re not trying to make it have an opinion on something the core was previously neutral on. You’re trying to take a system implicitly trained on lots of arithmetic problems until its machinery started to reflect the common coherent core of arithmetic, and get it to say that as a special case 222 + 222 = 555. You can maybe train something to do this in a particular training distribution, but it’s incredibly likely to break when you present it with new math problems far outside that training distribution, on a system which successfully generalizes capabilities that far at all. #### Section B.3: Central difficulties of sufficiently good and useful transparency / interpretability. 25. We’ve got no idea what’s actually going on inside the giant inscrutable matrices and tensors of floating-point numbers. Drawing interesting graphs of where a transformer layer is focusing attention doesn’t help if the question that needs answering is “So was it planning how to kill us or not?” 26. Even if we did know what was going on inside the giant inscrutable matrices while the AGI was still too weak to kill us, this would just result in us dying with more dignity, if DeepMind refused to run that system and let Facebook AI Research destroy the world two years later. Knowing that a medium-strength system of inscrutable matrices is planning to kill us, does not thereby let us build a high-strength system of inscrutable matrices that isn’t planning to kill us. 27. When you explicitly optimize against a detector of unaligned thoughts, you’re partially optimizing for more aligned thoughts, and partially optimizing for unaligned thoughts that are harder to detect. Optimizing against an interpreted thought optimizes against interpretability. 28. The AGI is smarter than us in whatever domain we’re trying to operate it inside, so we cannot mentally check all the possibilities it examines, and we cannot see all the consequences of its outputs using our own mental talent. A powerful AI searches parts of the option space we don’t, and we can’t foresee all its options. 29. The outputs of an AGI go through a huge, not-fully-known-to-us domain (the real world) before they have their real consequences. Human beings cannot inspect an AGI’s output to determine whether the consequences will be good. 30. Any pivotal act that is not something we can go do right now, will take advantage of the AGI figuring out things about the world we don’t know so that it can make plans we wouldn’t be able to make ourselves. It knows, at the least, the fact we didn’t previously know, that some action sequence results in the world we want. Then humans will not be competent to use their own knowledge of the world to figure out all the results of that action sequence. An AI whose action sequence you can fully understand all the effects of, before it executes, is much weaker than humans in that domain; you couldn’t make the same guarantee about an unaligned human as smart as yourself and trying to fool you. There is no pivotal output of an AGI that is humanly checkable and can be used to safely save the world but only after checking it; this is another form of pivotal weak act which does not exist. 31. A strategically aware intelligence can choose its visible outputs to have the consequence of deceiving you, including about such matters as whether the intelligence has acquired strategic awareness; you can’t rely on behavioral inspection to determine facts about an AI which that AI might want to deceive you about. (Including how smart it is, or whether it’s acquired strategic awareness.) 32. Human thought partially exposes only a partially scrutable outer surface layer. Words only trace our real thoughts. Words are not an AGI-complete data representation in its native style. The underparts of human thought are not exposed for direct imitation learning and can’t be put in any dataset. This makes it hard and probably impossible to train a powerful system entirely on imitation of human words or other human-legible contents, which are only impoverished subsystems of human thoughts; *unless*that system is powerful enough to contain inner intelligences figuring out the humans, and at that point it is no longer really working as imitative human thought. 33. The AI does not think like you do, the AI doesn’t have thoughts built up from the same concepts you use, it is utterly alien on a staggering scale. Nobody knows what the hell GPT-3 is thinking, not only because the matrices are opaque, but because the stuff within that opaque containeris, very likely, incredibly alien – nothing that would translate well into comprehensible human thinking, even if we could see past the giant wall of floating-point numbers to what lay behind. #### Section B.4: Miscellaneous unworkable schemes. 34. Coordination schemes between superintelligences are not things that humans can participate in (eg because humans can’t reason reliably about the code of superintelligences); a “multipolar” system of 20 superintelligences with different utility functions, plus humanity, has a natural and obvious equilibrium which looks like “the 20 superintelligences cooperate with each other but not with humanity”. 35. Schemes for playing “different” AIs off against each other stop working if those AIs advance to the point of being able to coordinate via reasoning about (probability distributions over) each others’ code. Any system of sufficiently intelligent agents can probably behave as a single agent, even if you imagine you’re playing them against each other. Eg, if you set an AGI that is secretly a paperclip maximizer, to check the output of a nanosystems designer that is secretly a staples maximizer, then even if the nanosystems designer is not able to deduce what the paperclip maximizer really wants (namely paperclips), it could still logically commit to share half the universe with any agent checking its designs if those designs were allowed through, ifthe checker-agent can verify the suggester-system’s logical commitment and hence logically depend on it (which excludes human-level intelligences). Or, if you prefer simplified catastrophes without any logical decision theory, the suggester could bury in its nanosystem design the code for a new superintelligence that will visibly (to a superhuman checker) divide the universe between the nanosystem designer and the design-checker. 36. What makes an air conditioner ‘magic’ from the perspective of say the thirteenth century, is that even if you correctly show them the design of the air conditioner in advance, they won’t be able to understand from seeing that design why the air comes out cold; the design is exploiting regularities of the environment, rules of the world, laws of physics, that they don’t know about. The domain of human thought and human brains is very poorly understood by us, and exhibits phenomena like optical illusions, hypnosis, psychosis, mania, or simple afterimages produced by strong stimuli in one place leaving neural effects in another place. Maybe a superintelligence couldn’t defeat a human in a very simple realm like logical tic-tac-toe; if you’re fighting it in an incredibly complicated domain you understand poorly, like human minds, you should expect to be defeated by ‘magic’ in the sense that even if you saw its strategy you would not understand why that strategy worked. AI-boxing can only work on relatively weak AGIs; the human operators are not secure systems. ### Section C: Okay, those are some significant problems, but lots of progress is being made on solving them, right? There’s a whole field calling itself “AI Safety” and many major organizations are expressing Very Grave Concern about how “safe” and “ethical” they are? 37. There’s a pattern that’s played out quite often, over all the times the Earth has spun around the Sun, in which some bright-eyed young scientist, young engineer, young entrepreneur, proceeds in full bright-eyed optimism to challenge some problem that turns out to be really quite difficult. Very often the cynical old veterans of the field try to warn them about this, and the bright-eyed youngsters don’t listen, because, like, who wants to hear about all that stuff, they want to go solve the problem! Then this person gets beaten about the head with a slipper by reality as they find out that their brilliant speculative theory is wrong, it’s actually really hard to build the thing because it keeps breaking, and society isn’t as eager to adopt their clever innovation as they might’ve hoped, in a process which eventually produces a new cynical old veteran. Which, if not literally optimal, is I suppose a nice life cycle to nod along to in a nature-show sort of way. Sometimes you do something for the first time and there are no cynical old veterans to warn anyone and people can be really optimistic about how it will go; eg the initial Dartmouth Summer Research Project on Artificial Intelligence in 1956: “An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves. We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.” This is lessof a viable survival plan for your planet if the first major failure of the bright-eyed youngsters kills literally everyone before they can predictably get beaten about the head with the news that there were all sorts of unforeseen difficulties and reasons why things were hard. You don’t get any cynical old veterans, in this case, because everybody on Earth is dead. Once you start to suspect you’re in that situation, you have to do the Bayesian thing and update now to the view you will predictably update to later: realize you’re in a situation of being that bright-eyed person who is going to encounter Unexpected Difficulties later and end up a cynical old veteran – or would be, except for the part where you’ll be dead along with everyone else. And become that cynical old veteran right away, before reality whaps you upside the head in the form of everybody dying and you not getting to learn. Everyone else seems to feel that, so long as reality hasn’t whapped them upside the head yet and smacked them down with the actual difficulties, they’re free to go on living out the standard life-cycle and play out their role in the script and go on being bright-eyed youngsters; there’s no cynical old veterans to warn them otherwise, after all, and there’s no proof that everything won’t go beautifully easy and fine,*given their bright-eyed total ignorance of what those later difficulties could be.* 38. It does not appear to me that the field of ‘AI safety’ is currently being remotely productive on tackling its enormous lethal problems. These problems are in fact out of reach; the contemporary field of AI safety has been selected to contain people who go to work in that field anyways. Almost all of them are there to tackle problems on which they can appear to succeed and publish a paper claiming success; if they can do that and get funded, why would they embark on a much more unpleasant project of trying something harder that they’ll fail at, just so the human species can die with marginally more dignity? This field is not making real progress and does not have a recognition function to distinguish real progress if it took place. You could pump a billion dollars into it and it would produce mostly noise to drown out what little progress was being made elsewhere. 39. I figured this stuff out using the[null string](https://twitter.com/ESYudkowsky/status/1500863629490544645)as input, and frankly, I have a hard time myself feeling hopeful about getting real alignment work out of somebody who previously sat around waiting for somebody else to input a persuasive argument into them. This ability to “notice lethal difficulties without Eliezer Yudkowsky arguing you into noticing them” currently is an opaque piece of cognitive machinery to me, I do not know how to train it into others. It probably relates to ‘[security mindset](https://intelligence.org/2017/11/25/security-mindset-ordinary-paranoia/)‘, and a mental motion where you refuse to play out scripts, and being able to operate in a field that’s in a state of chaos. 40. “Geniuses” with nice legible accomplishments in fields with tight feedback loops where it’s easy to determine which results are good or bad right away, and so validate that this person is a genius, are (a) people who might not be able to do equally great work away from tight feedback loops, (b) people who chose a field where their genius would be nicely legible even if that maybe wasn’t the place where humanity most needed a genius, and (c) probably don’t have the mysterious gears simply because they’re rare. You cannot just pay $5 million apiece to a bunch of legible geniuses from other fields and expect to get great alignment work out of them. They probably do not know where the real difficulties are, they probably do not understand what needs to be done, they cannot tell the difference between good and bad work, and the funders also can’t tell without me standing over their shoulders evaluating everything, which I do not have the physical stamina to do. I concede that real high-powered talents, especially if they’re still in their 20s, genuinely interested, and have done their reading, are people who, yeah, fine, have higher probabilities of making core contributions than a random bloke off the street. But I’d have more hope – not significant hope, but morehope – in separating the concerns of (a) credibly promising to pay big money retrospectively for good work to anyone who produces it, and (b) venturing prospective payments to somebody who is predicted to maybe produce good work later. 41. Reading this document cannot make somebody a core alignment researcher. That requires, not the ability to read this document and nod along with it, but the ability to spontaneously write it from scratch without anybody else prompting you; that is what makes somebody a peer of its author. It’s guaranteed that some of my analysis is mistaken, though not necessarily in a hopeful direction. The ability to do new basic work noticing and fixing those flaws is the same ability as the ability to write this document before I published it, which nobody apparently did, despite my having had other things to do than write this up for the last five years or so. Some of that silence may, possibly, optimistically, be due to nobody else in this field having the ability to write things comprehensibly – such that somebody out there had the knowledge to write all of this themselves, if they could only have written it up, but they couldn’t write, so didn’t try. I’m not particularly hopeful of this turning out to be true in real life, but I suppose it’s one possible place for a “positive model violation” (miracle). The fact that, twenty-one years into my entering this death game, seven years into other EAs noticing the death game, and two years into even normies starting to notice the death game, it is still Eliezer Yudkowsky writing up this list, says that humanity still has only one gamepiece that can do that. I knew I did not actually have the physical stamina to be a star researcher, I tried really really hard to replace myself before my health deteriorated further, and yet here I am writing this. That’s not what surviving worlds look like. 42. There’s no plan. Surviving worlds, by this point, and in fact several decades earlier, have a plan for how to survive. It is a written plan. The plan is not secret. In this non-surviving world, there are no candidate plans that do not immediately fall to Eliezer instantly pointing at the giant visible gaping holes in that plan. Or if you don’t know who Eliezer is, you don’t even realize you need a plan, because, like, how would a human being possibly realize that without Eliezer yelling at them? It’s not like people will yell at themselves about prospective alignment difficulties, they don’t have an internal voice of caution. So most organizations don’t have plans, because I haven’t taken the time to personally yell at them. ‘Maybe we should have a plan’ is deeper alignment mindset than they possess without me standing constantly on their shoulder as their personal angel pleading them into… continued noncompliance, in fact. Relatively few are aware even that they should, to look better, produce a pretend plan that can fool EAs too ‘[modest](https://equilibriabook.com/toc/)‘ to trust their own judgments about seemingly gaping holes in what serious-looking people apparently believe. 43. This situation you see when you look around you is not what a surviving world looks like. The worlds of humanity that survive have plans. They are not leaving to one tired guy with health problems the entire responsibility of pointing out real and lethal problems proactively. Key people are taking internal and real responsibility for finding flaws in their own plans, instead of considering it their job to propose solutions and somebody else’s job to prove those solutions wrong. That world started trying to solve their important lethal problems earlier than this. Half the people going into string theory shifted into AI alignment instead and made real progress there. When people suggest a planetarily-lethal problem that might materialize later – there’s a lot of people suggesting those, in the worlds destined to live, and they don’t have a special status in the field, it’s just what normal geniuses there do – they’re met with either solution plans or a reason why that shouldn’t happen, not an uncomfortable shrug and ‘How can you be sure that will happen’ / ‘There’s no way you could be sure of that now, we’ll have to wait on experimental evidence.’ A lot of those better worlds will die anyways. It’s a genuinely difficult problem, to solve something like that on your first try. But they’ll die with more dignity than this. The post [AGI Ruin: A List of Lethalities](https://intelligence.org/2022/06/10/agi-ruin/) appeared first on [Machine Intelligence Research Institute](https://intelligence.org).
768f1b8e-94d2-4705-bc96-bd871f46ac3a	trentmkelly/LessWrong-43k	LessWrong	Biases: An Introduction Imagine reaching into an urn that contains seventy white balls and thirty red ones, and plucking out ten mystery balls. Perhaps three of the ten balls will be red, and you’ll correctly guess how many red balls total were in the urn. Or perhaps you’ll happen to grab four red balls, or some other number. Then you’ll probably get the total number wrong. This random error is the cost of incomplete knowledge, and as errors go, it’s not so bad. Your estimates won’t be incorrect on average, and the more you learn, the smaller your error will tend to be. On the other hand, suppose that the white balls are heavier, and sink to the bottom of the urn. Then your sample may be unrepresentative in a consistent direction. That kind of error is called “statistical bias.” When your method of learning about the world is biased, learning more may not help. Acquiring more data can even consistently worsen a biased prediction. If you’re used to holding knowledge and inquiry in high esteem, this is a scary prospect. If we want to be sure that learning more will help us, rather than making us worse off than we were before, we need to discover and correct for biases in our data. The idea of cognitive bias in psychology works in an analogous way. A cognitive bias is a systematic error in how we think, as opposed to a random error or one that’s merely caused by our ignorance. Whereas statistical bias skews a sample so that it less closely resembles a larger population, cognitive biases skew our thinking so that it less accurately tracks the truth (or less reliably serves our other goals). Maybe you have an optimism bias, and you find out that the red balls can be used to treat a rare tropical disease besetting your brother, and you end up overestimating how many red balls the urn contains because you wish the balls were mostly red. Like statistical biases, cognitive biases can distort our view of reality, they can’t always be fixed by just gathering more data, and their effects can a
c8b09c88-3a58-4f6c-8130-4866199ec4c8	trentmkelly/LessWrong-43k	LessWrong	Open Thread for February 11 - 17 If it's worth saying, but not worth its own post (even in Discussion), then it goes here.
d7766d8a-d31d-4676-8f23-779c45e04ffc	trentmkelly/LessWrong-43k	LessWrong	Stopping dangerous AI: Ideal US behavior Disclaimer: this post doesn't have the answers. Moreover, it's an unfinished draft. Hopefully a future version will be valuable, but that will only occur if I revise/rewrite it. For now you're better off reading sources linked from AI policy ideas: Reading list and Slowing AI: Reading list. Set aside most of AI safety to focus on the speed of AI progress. What should the US government do?[1] This post assumes that it's good to slow (dangerous) AI (especially near the end). The ultimate goal is to prevent the deployment of powerful AI systems—that is, AI systems that would cause a catastrophe—until we learn how to make them safe. (Pretending alignment is monolithic.) You can do this by making powerful AI systems not developed, not deployed, or not cause a catastrophe if deployed because the world is super-resilient. Focusing on development is most promising. So, delay the development of powerful AI systems until they're safe. (Probably that's too crude. Complications: (1) alignment progress is endogenous; (2) you have to actually pay the alignment tax so leading labs' lead time and safety-awareness are crucial.) Lots of AI, like robotics and autonomous vehicles and some medical stuff and image generation and so forth, isn't very dangerous. The dream would be that we have a test to determine whether an AI system would cause a catastrophe, and all large training runs are audited, and if a model fails the test then the training run is shut down. But (1) good tests don't exist yet, and (2) the only known way to make AI safe is to make it not powerful enough to cause a catastrophe. Related: model evals and Yonadav Shavit's Verifying Rules on Large-Scale Neural Network Training via Compute Monitoring (2023). ---------------------------------------- Ways governments can slow AI (this is kinda a list of policy levers rather than policy proposals) * Ban/moratorium: monitoring training runs (see Shavit) + ceiling on training compute * For a certain time, or until a
38052a90-d4f9-4590-82da-ca2b7d7e5d21	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Yudkowsky on AGI ethics A Cornell computer scientist recently wrote on social media: > [...] I think the general sense in AI is that we don't know what will play out, but some of these possibilities are bad, and we need to start thinking about it. We are plagued by highly visible people ranging from Musk to Ng painting pictures ranging from imminent risk to highly premature needless fear, but that doesn't depict the center of gravity, which has noticeably shifted to thinking about the potential bad outcomes and what we might do about it. (Turning close to home to provide an example of how mainstream this is becoming, at Cornell two AI professors, Joe Halpern and Bart Selman, ran a seminar and course last semester on societal and ethical challenges for AI, and only just a few weeks ago we had a labor economist speak in our CS colloquium series about policy ideas targeting possible future directions for CS and AI, to an extremely large and enthusiastic audience.) > > To which Eliezer Yudkowsky replied: > My forecast of the net effects of "ethical" discussion is negative; I expect it to be a cheap, easy, attention-grabbing distraction from technical issues and technical thoughts that actually determine okay outcomes. [...] > > The ethics of bridge-building is to not have your bridge fall down and kill people and there is a frame of mind in which this obviousness is obvious enough. How not to have the bridge fall down is hard. > > This is possibly surprising coming from the person who came up with [coherent extrapolated volition](https://arbital.com/p/cev/), co-wrote the Cambridge Handbook of Artificial Intelligence article on "[The Ethics of AI](https://intelligence.org/files/EthicsofAI.pdf)," etc. The relevant background comes from Eliezer's writing on the [minimality principle](https://arbital.com/p/minimality_principle/): > [W]hen we are building the first [sufficiently advanced Artificial Intelligence](https://arbital.com/p/sufficiently_advanced_ai/), we are operating in an extremely dangerous context in which building a marginally more powerful AI is marginally more dangerous. The first AGI ever built should therefore execute the least dangerous plan for [preventing immediately following AGIs from destroying the world six months later](https://arbital.com/p/pivotal/). Furthermore, the least dangerous plan is not the plan that seems to contain the fewest material actions that seem risky in a conventional sense, but rather the plan that requires the least dangerous cognition from the AGI executing it. Similarly, inside the AGI itself, if a class of thought seems dangerous but necessary to execute sometimes, we want to execute the least instances of that class of thought required to accomplish the overall task. > > E.g., if we think it's a dangerous kind of event for the AGI to ask "How can I achieve this end using strategies from across every possible domain?" then we might want a design where most routine operations only search for strategies within a particular domain, and events where the AI searches across all known domains are rarer and visible to the programmers. Processing a goal that can recruit subgoals across every domain would be a dangerous event, albeit a necessary one, and therefore we want to do less of it within the AI." > > So the technical task of figuring out how to build a robust minimal AGI system that's well-aligned with its operators' intentions is very different from "AI ethics"; and the tendency to conflate those two has plausibly caused a lot of thought and attention to go into much broader (or much narrower) issues that could have more profitably gone into thinking about the alignment problem. One part of doing the absolute bare [world-saving minimum](http://www.danieldewey.net/fast-takeoff-strategies.html) with a general-purpose reasoning system is steering clear of any strategies that require the system to do significant moral reasoning (or implement less-than-totally-airtight moral views held by its operators). Just execute the most simple and straightforward concrete sequence of actions, requiring the least dangerous varieties and quantity of AGI cognition needed for success. Another way of putting this view is that nearly all of the effort should be going into solving the technical problem, "How would you get an AI system to do some very modest [concrete action](https://arbital.com/p/task_goal/) requiring extremely high levels of intelligence, such as building two strawberries that are completely identical at the cellular level, without causing anything weird or disruptive to happen?" Where obviously it's important that the system not do anything severely unethical in the process of building its strawberries; but if your strawberry-building system requires its developers to have a full understanding of meta-ethics or value aggregation in order to be safe and effective, then you've made some kind of catastrophic design mistake and should start over with a different approach.
30df1ac6-4674-456a-9f10-684370d5c33d	trentmkelly/LessWrong-43k	LessWrong	Meetup : Philadelphia-- Debugging in Meatspace The discussion prompt William Burroughs's "The discipline of DE (do easy)" http://melancholia.typepad.com/melancholia/2009/07/do-easy.html WHEN: 23 February 2014 01:00:00PM (-0500) WHERE: Nam Phuong, 11th & Broad St., Philadelphia, PA Meets every two weeks. Forum/mailing list: groups.google.com/forum/#!forum/lesswrong-philadelphia
f6bd2c64-32ae-4727-bb6f-1661bd7afd32	trentmkelly/LessWrong-43k	LessWrong	mental number lines This paper says: > it was shown that second-grade children unevenly placed numbers on a 0–1000 number line as their placement pattern was best fit by a logarithmic rather than by a linear function. The same age group placed numbers more linearly (i.e., resulting in a constant distance between consecutive numbers) on a 0–100 number line. In contrast, adults produced a linear distribution when placing numbers on a 0–1000 number line > Two developmental leaps occurring during elementary school were recognized: (1) the division of the number line into two segments and (2) consistent use of different anchor points on the number line—the left endpoint in first grade, the right endpoint in second grade, and finally the midpoint in third grade. People mentally represent numbers as positions on a number line. I consider this a latent space representation. Human conceptual capabilities imply ability to at least average latent space representations, and possibly add them, which could then be used for fast approximation of sums. If only averaging is possible, then a 2-step (average -> double) process would be needed. The accuracy of that approach is insufficient for solving arithmetic problems, so people often instead use a much slower algorithm with exact accuracy. Per the linked paper, 2nd-grade children have a tendency to place large numbers logarithmically, which implies to me an approximately logarithmic latent space representation. In that case, an accurate logarithmic representation should also be something humans are capable of with some training. In an accurate logarithmic representation, addition and subtraction are, of course, equivalent to multiplication and division. Is it possible for people to do fast mental multiplications with the accuracy of a typical slide rule, by learning a logarithmic mental representation of numbers between 1 and 10? It seems plausible to me. Any big mental arithmetic fans around here? But of course, the latent spaces humans typical
cfaeb8da-8cc2-4df0-81f7-fe4b8258e051	trentmkelly/LessWrong-43k	LessWrong	Debate update: Obfuscated arguments problem This is an update on the work on AI Safety via Debate that we previously wrote about here. Authors and Acknowledgements The researchers on this project were Elizabeth Barnes and Paul Christiano, with substantial help from William Saunders (who built the current web interface as well as other help), Joe Collman (who helped develop the structured debate mechanisms), and Mark Xu, Chris Painter, Mihnea Maftei and Ronny Fernandez (who took part in many debates as well as helping think through problems). We're also grateful to Geoffrey Irving and Evan Hubinger for feedback on drafts, and for helpful conversations, along with Richard Ngo, Daniel Ziegler, John Schulman, Amanda Askell and Jeff Wu. Finally, we're grateful to our contractors who participated in experiments, including Adam Scherlis, Kevin Liu, Rohan Kapoor and Kunal Sharda. What we did We tested the debate protocol introduced in AI Safety via Debate with human judges and debaters. We found various problems and improved the mechanism to fix these issues (details of these are in the appendix). However, we discovered that a dishonest debater can often create arguments that have a fatal error, but where it is very hard to locate the error. We don’t have a fix for this “obfuscated argument” problem, and believe it might be an important quantitative limitation for both IDA and Debate. Key takeaways and relevance for alignment Our ultimate goal is to find a mechanism that allows us to learn anything that a machine learning model knows: if the model can efficiently find the correct answer to some problem, our mechanism should favor the correct answer while only requiring a tractable number of human judgements and a reasonable number of computation steps for the model. [1] We’re working under a hypothesis that there are broadly two ways to know things: via step-by-step reasoning about implications (logic, computation…), and by learning and generalizing from data (pattern matching, bayesian updating…). Debate fo
e5c745ff-b55f-45b4-84e5-76b9a4426eab	StampyAI/alignment-research-dataset/arbital	Arbital	Needs work Meta tag for pages which need content improvement.
1a5e7633-99df-4043-a9fa-04d7cdae8b7f	trentmkelly/LessWrong-43k	LessWrong	[Link] A superintelligent solution to the Fermi paradox Here. Long story short, it's an attempt to justify the planetarium hypothesis as a solution to the Fermi paradox. The first half is a discussion of how it and things like it are relevant to the intended purview of the blog, and the second half is the meat of the post. You'll probably want to just eat the meat, which I think is relevant to the interests of many LessWrong folk. The blog is Computational Theology. It's new. I'll be the primary poster, but others are sought. I'll likely introduce the blog and more completely describe it in its own discussion post when more posts are up, hopefully including a few from people besides me, and when the archive will give a more informative indication of what to expect from the blog. Despite theism's suspect reputation here at LessWrong I suspect many of the future posts will be of interest to this audience anyway, especially for those of you who take interest in discussion of the singularity. The blog will even occasionally touch on rationality proper. So you might want to store the fact of the blog's existence somewhere deep in the back of your head. A link to the blog's main page can be found on my LessWrong user page if you forget the url. I'd appreciate it if comments about the substance of the post were made on the blog post itself, but if you want to discuss the content here on LessWrong then that's okay too. Any meta-level comments about presentation, typos, or the post's relevance to LessWrong, should probably be put as comments on this discussion post. Thanks all!
3548049b-e392-4faf-804a-891801fb7dbf	StampyAI/alignment-research-dataset/arxiv	Arxiv	Formal Definition of AI 1Formal Definition of AI Dimiter Dobrev dobrev@2-box.net 18 October, 2005 Abstract A definition of Artificial Intelligence was propose d in [1] but this definition was not absolutely formal at least because the word "Human" was used. In this paper we will formalize the definition from [1]. The biggest prob lem in this definition was that the level of intelligence of AI is compared to the intelligen ce of a human being. In order to change this we will introduce some parameters to which AI will depend. One of this parameters will be the level of intelligence and we will defin e one AI to each level of intelligence. We assume that for some level of intelligence the resp ective AI will be more intelligent than a human being. Nevertheless, we cannot say which is t his level because we cannot calculate its exact value. Introduction The definition in [1] first was published in popula r form in [2, 3]. It was stated in one sentence but with many assumptions and explanations which were given before and after this sentence. Here is the definition of AI in one sentence: AI will be such a program which in an arbitrary wor ld will cope no worse than a human. From this sentence you can see that we assume that AI is a program. Also, we assume that AI is a step device and that on every step it input s from outside a portion of information (a letter from finite alphabet Σ) and outputs a portion of information (a letter fr om a finite alphabet Ω). The third assumption is that AI is in some envir onment which gives it a portion of information on every step and which rece ives the output of AI. Also, we assume that this environment will be influenced of the information which AI outputs. This environment can be natural or artificial and we wil l refer to it as "World". The World will be: one set S, one element s 0 of S and two functions World(s, d) and View(s). The set S contains the internal states of the world and it can be finite or infinite. The element s 0 of S will be the world's starting state. The funct ion World will take as arguments the current state of the world and the in fluence that our device exerts on the world at the current step. As a result, this functi on will return the new state of the world (which it will obtain on the next step). The functi on View gives the information what our 2device sees. An argument of this function will be t he world's current state and the returned value will be the information that the device will receive (at a given step). Life in one world will be any infinite row of the t ype: d 1, v 1, d 2, v 2, ... where v i are letters from Σ and d i are letters from Ω. Also, there has to exist infinite row s 0, s 1, s 2, ... such that s0 is the staring state of the world and ∀i>0 vi=View(s i) and ∀i s i+1 =World(s i, d i+1 ). It is obvious that if the world is given then the life de pends only on the actions of AI (i.e. depends only on the row d 1, d 2, d 3, ...). In order to transform the definition in [1] and to make it formal, we have to define what is a program, what is a good world and when one life i s better than another. The first task is easy because this work is done by Turing in the main part. Anyway, the Turing definition of program is for a program which represents function, but here we need a transducer which inputs the row v 1, v 2, v 3, ... and outputs the row d 1, d 2, d 3, ... . So, we will make a modification of the definition of Turin g machine [9]. Our second task is to say what is a good world. It was written in [1] that if you can make a fatal error in one world then this world is not g ood. What is world without fatal errors needs additional formalization. The next problem is to say when one life is better than another. This is done in [1] but there are some problems connected with the infinity which have to be fixed. The last task is to say how intelligent our program should be and this cannot be done by comparison with a human being. What is a program We will define a program as a Turing machine [9]. L et its alphabet Δ consist of the letters from Σ, from Ω, from one blank symbol λ and from some service signs. Let our Turing machine have finite set of internal states P, one starting state p 0 and a partial function F : P × Δ → P × Δ × {Left, Right}. The Turing machine (TM) is a step device and it mak es steps in order to do calculations. On the other hand, AI is a step device and its life consists of steps. In order to make distinction between these two types of steps we wil l call them small and big steps. When we speak about time we will mean the number of big steps. Of course, our TM will start from the state p 0 with infinite tape filled with the blank symbol λ. How our TM will make one small step. If it is in state p and if its head looks at the letter δ then F(p, δ) will be a 3-tuple which first element is the new state after the small 3step, the second element will be the new letter whi ch will replace δ on the tape and the third element will be direction in which the head w ill move. How will our TM (which is also our AI) make one big step? This will happen when after a small step the new state of TM is again p 0. At this moment our TM has to output one letter d i and to input one letter v i. We will assume that the letter which is outputted is that which is written on the place of δ on this small step. But how after outputting the l etter d i will our TM input the letter v i? We will put this letter on the place where the he ad after the small step is. In this way we are intervening in th e work of the TM by replacing one symbol from the tape with another. The replaced sym bol is lost in some sense because it will not influence the execution of the TM from thi s small step on. We will assume that our TM is outputting only lette rs from Ω (no letters from the rest of Δ). Also, we assume that our TM never hangs. TM hang s if after reading some input v 1, v2, ... , v n it stops because it falls into some state p and it s head falls on some letter δ such that F(p, δ) is not defined. TM also hangs if after reading of some input v 1, v 2, ... , v n it makes infinitely many small steps without reaching the state p 0 (without making of big steps anymore). After this we have a formal definition of a program . We have to mention that there is no restriction on the number of the small steps which TM needs to make for one big step. This number has to be finite but it is not restrict ed. Maybe it is a good idea to add one parameter Max_number_of_small_steps_in_AI in order to exclude some decisions for AI which are combinatory explosions. (If we restrict t he number of small steps then we have to restrict also the number of service signs in Δ because we can speed up the TM by increasing the size of its alphabet.) If we want to use AI as a real program on a real computer then we have to take into consideration th at the memory of the real computers is limited. So, we can restrict also the size of the t ape. Anyway, we will not care about the efficiency of AI and we will not make such restrict ions. What is a world without fatal errors It is very difficult to define what is a world with out fatal errors. That is why we will do something else. We will restrict our set of worlds is such a way that the new set will contain only worlds without fatal errors. Let our world look like one infinite sequence of ga mes. Let every game be independent from the previous ones. Let us have three special l etters in Σ, which we will call final letters. Let this letters be {victory, loss, draw}. Let every game finish with one of the final letters. Let every game be shorter than 1000 big st eps. Remark 1 : Our definition of AI will depend on many paramete rs. In order to simplify the exposition we will fix these parameters to concrete numbers. Such parameter is the 4maximum number of steps in a game which will be fix ed to 1000. Also, in order to simplify the exposition we will use different numbe rs for different parameters. Remark 2 : The only parameters in our definition which are n ot numbers are the alphabets Σ and Ω. We will assume that these alphabets are fixed and that Ω has at least 2 letters and Σ has at least 2 letters which are not final. (If Ω has only one letter then there will be no choice for the action of AI and the world will b e independent from this action. If Σ has only one letter, which is not final, then the game will be blind because AI will not receive any information until the end of the game. Therefor e, the minimum for \| Σ\| is 5.) We will assume that the world has three special int ernal states {s_victory, s_loss, s_draw}, which we will call final states. Let these states b e indistinguishable from the state s 0 for the function World. This means that the world will beha ve in the final states in the same way as if it was in the starting state. Let the functio n View distinguish the final states and return from them the letters victory, loss and draw respectively. Also, the final states will be the only states on which the function View will return one of the letters {victory, loss, draw}. After the restriction of the definition of World, w e can be sure that there are no fatal errors in our world because the life in such a worl d is an infinite sequence of games and if we lose some games (finitely many) then this will n ot be fatal because every new game is independent from the previous ones. Also, we are su re that a new game will come sooner or later because every game is finite (i.e. previou s game is shorter than 1000 steps). When is one life better than another In [1] we gave the following definition for the mea ning of the life: One life is better than another if it includes more good letters and fewer bad letters. Here good letters will be {victory, draw} and bad letters will be {loss, draw }. So, here life is good if we win often and lose seldom. We want to introduce one function Success which wil l evaluate with a real number every life in order to say how good it is. For that we wi ll define first the function Success for the every beginning of life (all beginnings are finite) . After that we will calculate the limit of Success when the size of the beginnings goes to inf inity and this limit will be the value of Success for the entire life. The function Success can be defined for the beginni ngs like the difference between the number of victories and the number of losses. This is not a good idea because then the limit of Success will possibly converge to infinity (plus or minus infinity). It is a better idea to calculate the percentage of victories. So, we de fine Success as (2.N_victory +N_draw)/ (2.N_games). Here N_victory is the number of victor ies (analogically for N_draw and N_games). Function Success will give us a number be tween 0 and 1 for every beginning 5and its limit will be also between 0 and 1. The onl y problem is that Success may not have a limit. In such a case we will use the average betwe en limit inferior and limit superior. Trivial decisions Now we have a really formal definition of AI and th is gives us the first trivial decision for AI. TD1 will be the program which plays at random until the first victory. After that TD1 repeats this victory forever. For this TD1 needs on ly to remember what it did in the last game. If the last game was victorious then it can r epeat this last game because the function World is deterministic and if TD1 is doing the same then the world will do the same too. TD1 is perfect in all worlds in which the victory i s possible. If the victory is not possible then TD1 will play at random forever. That is why w e will make TD2 which will be perfect in all worlds. TD2 will be this program which tries sequentially a ll possible game's strategies until it finds the first victory and after that repeats this victory forever. If there is no victorious game strategy then TD2 repeats the last draw game f orever. If the draw game is not possible too then TD2 plays at random. (It is impor tant that the game's length is not more than 1000. This means that the number of the game's strategies is finite.) TD2 is perfect in all worlds and this means that it is a trivial decision on our definition for AI. Really, the definition stated that AI has to co pe no worse than a human but for the perfect program this is true because it copes no wo rse than anything even no worse than a human being. It is suspicious that such simple program like TD2 can satisfy our definition for AI. That is why we will change the definition by accepting more possible worlds. It is too restrictive to assume that the game is deterministic and every time you do the same the same will happen. Nondeterministic games We will assume that the function World is not deter ministic. It is better to say that it is multy-valued function, which chooses at random one of its possible values. Let every possible value correspond to one real number, which is the possibility for this value to be chosen. We will assume also that ∀s ∀ω World(s, ω) has at least one value and that ∀s ∀ω (for every two different values of World(s, ω) the function View returns different result). 6Remark 3 : The latter means that if something nondeterminist ic happens this information will be given to AI immediately by the function Vie w. There is no sense to assume existence of a nondeterministic change which cannot be detected immediately but later or even which cannot be detected never. Now we will ask the question what is the best strat egy in such a world and we will offer a program, which will be almost perfect. Before that we need several definitions: Definition 1: Tree of any game. It will have two types of vertices. The root and t he other vertices which are on even depth will be the vertices of type AI (because they correspond to the moments when AI has to do its cho ice). The vertices which are on odd depth will be vertices of the type world (because t hey correspond to the moments when the world will answer at random). From the vertices of type AI go out \| Ω\| arcs and to every such arc corresponds one of the letters from Ω. There is one exception. If the arc which is right before this vertex corresponds to a final letter, then this vertex is a leaf. From the vertices of type world go out \| Σ\| arcs and to every such arc corresponds one of the letters from Σ. Here there is an exception again. If this vertex is on depth 1999, then only three arcs go out and these three arcs corresp ond to the final letters. You can see that the tree of any game is finite and its maximum depth is 2000 (because games are not longer than 1000 steps). Nevertheless , there are leaves on any even depth between 2 and 2000. Definition 2: Tree of any 100 games. Let us take the tree of any game. Let us replace a ll of its leaves with the tree on any game. Let us rep eat this operation 99 times. The result will be the tree of any 100 games (which is 100 tim es deeper than the tree of any game). From the tree of any game we will receive Strategy for any game. This will be its subtree which is obtained by choosing one vertex from the s uccessors of every vertex of the type AI and deleting the rest successors (and their subt rees). Analogically we make Strategy for any 100 games like a subtree from the tree of any 1 00 games. We have to mention that the strategy for 100 games can be different from repeat ing one Strategy for any game 100 times. The reason is because the strategy on the ne xt game can depend on the previous games. Definition 3: Tree of this game. For every concrete game (i.e. concrete world) we c an construct the tree of this game as a subtree from t he tree of any game. We will juxtapose internal states of the world to the vertices of typ e AI in the time of this construction. First, we will juxtapose the state s 0 to the root. Let k 0, k 1 and k 2 be vertices and let k 1 be successor of k 0 and k 2 be successor of k 1. Let k 0 be vertex of type AI and let the state s be juxtaposed to it. Let the letters ω and ε be juxtaposed to the arcs <k 0, k 1> and <k 1, k 2>. In this case if ε ≠ View(World(s, ω)) for every value of World(s, ω) then we delete the vertex k2 (and its subtree). In the opposite case we juxtapo se k 2 to this value of World(s, ω) for which ε = View(World(s, ω)). This value is only (look at remark 3). Also, we will 7juxtapose the possibility ε to be the value of View(World(s, ω)) to the arc <k 1, k 2>. So, one letter and one possibility will be juxtaposed t o the arc <k 1, k 2>. Analogically to the strategy for any game we can ma ke strategy for this game. We have to say that if the World is deterministic (i.e. every vertex of type world has only one successor) then the strategy for this game is a pat h (a tree without branches). In this case the paths in the tree of this game are exactly the strategies for this game. This was used from TD2 in order to try all strategies. Max-Sum algorithm For every vertex of the tree of this game we can ca lculate the best possible success (this is our expectation for success, if we play with the be st strategy from that vertex on). 1. The best possible success for the leaves will be 1, 0 and 1/2 for the states s_victory, s_loss and s_draw respectively. 2. If the vertex is of type AI, then its best possi ble success will be the maximum from the best possible successes of its successors. 3. If the vertex is of type world, then its best po ssible success will be the sum Σ Possibility(i). BestPossibleSuccess(i). Here i runs through all successors of this vertex. The algorithm for calculating the best possible suc cess can also be used to calculate the best strategy in this game (the best strategy can b e more than one). This algorithm looks like the Min-Max algorithm, which we use in chess. Anyway, this is different algorithm, to which we will refer as Max-Sum algorithm. The diffe rence is essential because in the chess we assume that we play against someone who will do the worst thing to us (remark 4). Anyway, in the arbitrary world we cannot assume tha t the world is against us. For example, when you go to work you go first to the pa rking lot in order to take your car. If your car is stolen, then you go to the bus stop in order to take the bus. If every time you were presumed the worst case, then you would go dir ectly to the bus stop. New trivial decisions Now we can calculate the best possible success for any game and we will give the next trivial decision (TD3), which will do the best in e very game. This means that the success of TD3 for one world will be equal to its best poss ible success. TD3 will be the program which plays at random for l ong time enough. In this time TD3 collects statistical information for the tree of th is game and builds inside its memory this tree together with the values of all possibilities. After that time TD3 starts playing by the use of Max-Sum algorithm. 8TD3 gives the perfect decision in any world but TD3 is impossible because we cannot say when enough statistical information is selected. An yway, possible is something which is a little bit worse. For every ε > 0 we will make TD4, which for every world will m ake success on a distance no more than ε from the best possible. TD4 will be this program which simultaneously colle cts statistical information for the tree of this game and in the same time plays by the use of Max-Sum algorithm on the base of statistics, which is collected up to the current mo ment. In order to collect statistics TD4 makes experiments which contradict to the recommend ations of Max-Sum algorithm. Such experiments are made rarely enough to have suc cess on a distance not bigger than ε from the best possible success. We can choose the value of ε to be as small as we want. Anyway, the price for t he small value of ε is the longer time for education (because of rare experiments). We will call the parameter ε "courage". Here we receive a surprising conclusion that if AI is more cowardly it is closer to perfection (this is true o nly in the case of infinite life). TD4 is a decision for our definition of AI because it is only on ε distance to perfection unlike the people who are much farther from perfect ion. We have to mention that in some sense TD4 is not as trivial as TD2, because TD4 rep resents awful combinatory explosion in the execution time (number of small steps) and i n the memory size. Anyway, we said that we will not care about the efficiency of AI fo r the moment. On the other hand, there is one additional problem, which is present in both TD 2 and TD4, and which makes them both useless. This is the problem of the combinator y explosion of the educational time. Imagine that you are playing chess at random agains t deterministic partner. How long will you need to make accidental victory? Or in case you r partner is not deterministic. How long will you need to play all possible game's stra tegies and try each one several times in order to collect statistical information on how you r partner reacts in every case? Finite life In some sense TD2 and TD4 are extremely stupid beca use they need extremely long time for education. Really, educational time and level o f intelligence are two different parameters of the mind and if one of them is better then this can be at the expense of the other. For example, a human being needs about a yea r to learn to walk, which is much worse in comparison to most animals. Some of the gr eatest scientists had bad results in school, which can be interpreted as a fact that the y advanced slower than the ordinary people. Therefore, the educational time is important and it has to be limited in order to make our definition useful. This will be done by changing th e life length from infinite to finite. We will assume that the length of the life is 100 game s. Each game has maximum 1000 steps, which means that the life length is not bigger than 10,000 steps. Now the success of the 9life will not be the limit of the Success function but the value of this function for the first 100 games. After this we can look for program which makes a go od success in an arbitrary world, but this is not a good idea because the arbitrary world is too unpredictable. Human beings use the assumption that the world is simple and that is why they cope very well in a more simple environment and they are totally confused if the environment is too complicated. Therefore, we have to restrict the complexity of th e world and give bigger importance to the more simple worlds. For this restriction we wil l use Kolmogorov Complexity [8]. The parameter which restricts the complexity of the wor ld will be the level of intelligence of AI. Kolmogorov Complexity First we need a definition of program which calcula tes the functions World and View. For this we will use the same definition of TM as for t he program which was our AI. There will be some small differences: The alphabet of the Turing Machine of the world (TM_W) will be Σ ∪ Ω ∪ { λ} (the only service symbol will be λ). Also, TM_W will input the row d1, d 2, d 3, ... and output the row v 1, v 2, v 3, ... . At the beginning TM_W will start with tape on which d 1 is at the head position and the rest is λ. At the end of the first big step TM_W will output v 1 and input d 2. F will be set of 5-tuples which is a subset to P × Δ × P × Δ × {Left, Right}. This means that F is not a function but a relation (because it will represent multy-valued function). We will assume that ∀s ∀δ ∃5-tuple ∈F whose first two elements are s and δ (this makes the reasons for hanging with one less) . The 5-tuples in F whose third element is p 0 will be called output 5-tuples. The fourth element of output 5-tuples has to be letter from Σ (this is not necessary but sufficient condition in order TM_W to output only letters from Σ). We will allow nondeterministic behavior only for output 5- tuples. This means that if two different 5-tuples h ave the same first and second elements then they both have to be output 5-tuples. There wi ll be no two 5-tuples which differ only at the fifth element (we cannot have a choice betwe en two nondeterministic 5-tuples which output the same letter - look again at remark 3). I t will be more interesting if we assume that nondeterministic 5-tuples have additional para meter which shows the possibility for each of them to be chosen. Nevertheless, we will as sume that this possibility is distributed equally and that we do not have such additional par ameter. According to this definition, the internal states o f the world will be the states of the tape of the TM_W. If we want to have worlds without fatal e rrors we have to clean the tape of TM_W after each game (after printing any final lett er). Nevertheless, we will not do this because the absence of fatal errors was important w hen we had infinite life and when we counted on that sooner or later all errors will be compensated. For real AI is better to assume some connection between games. Otherwise AI will not remember what was done in the last game or it will remember it but it will not know whether this was in the last game or in some other game from the previous ones. 10 Another question is what we will do with TM_W which hangs. We do not want to exclude these programs from our definition (at least becaus e we cannot check this characteristic). That is why we will assume that if one TM_W makes m ore than 800 small steps without making a big step then we will interrupt it with ou tput "draw". This means that it will do the next small step in the same way as if the 5-tup le executed at this moment had third element p 0 and fourth element "draw". Also, if one TM_W makes 1000 big steps without outputting any final symbol then the output of the next big step will be "draw". We need this in order to keep the games finite, which is im portant in order to keep the life finite (the life is 100 games). We will define the size of TM_W as the number of in ternal states plus the level of indefiniteness (this is the minimal number of nonde terministic 5-tuples, which have to be deleted from F in order to make it deterministic or this is the number of all nondeterministic 5-tuples minus the number of diffe rent groups of nondeterministic 5- tuples). So, we will restrict the set of possible worlds to those generated by Turing Machine whose size is not bigger than 20. The maximum size of the TM_W will be the level of intelligence of AI. The simpler worlds will be more important because they are generated from more than one TM_W and that is why we will cou nt their result more than once. Remark 4 : It looks like that two Turing machines (the world and AI) play against each other. Anyway, this is wrong because the world does not care about AI and it dose not play against AI. Final definition of AI Now everything is fixed. We have finite lives, whic h are exactly 100 games. We had selected the success function that will evaluate th ese lives. Also, we made a finite set of worlds which consist of the worlds generated from t he TM_W with size not bigger than 20. Now we can define AI as this program which will make the best average success in the selected worlds. Such program exists and it will be the next trivial decision (TD5). The number of all strategies for playing 100 consec utive games is finite. The number of selected worlds is also finite. We can calculate th e expected success of any strategy in any world. The average success of a strategy will be th e arithmetical mean from its expected success in any world. (The calculation of the expec ted success of a strategy in a world is easy if the world is deterministic. In this case, w e will have simply to play 100 games with this strategy in this world. In the opposite case, if the world is nondeterministic then we have to use Max-Sum algorithm, which is theoretical ly possible, but in practice it is a combinatory explosion. Nevertheless, even if the wo rlds were only deterministic, we would have combinatory explosion again from the num ber of worlds and from the number of strategies.) 11 Hence, TD5 will be this program which calculates an d plays the best strategy (this which average success is biggest). Such program is easy t o be written but it is very difficult to wait until it makes its first step. The situation w ith the perfect chess playing program is analogical. (It plays chess by calculating all poss ible moves.) This program can be written very easy but the time until the end of the univers e will be not enough for it to make its first move. It will be too restrictive if we define AI as the b est program (such as TD5 or such as any other program equivalent to TD5). It will be better if we say that AI is a program whose average success is not more than 10% from the best (from TD5) (when we say 1% we mean the number 0.01) (the distance to the best is something completely different from the parameter "courage" in TD2). Such definition is the oretically possible, but practically inconvenient. The reason for this is the fact that the value of the average success of TD5 can be theoretically calculated, but in practice th is is absolutely impossible. So, if we select such definition we will not be able to check it for a concrete program. Final definition: AI will be a program which makes more than 70% aver age success in the selected set of worlds. Assumption 1 : Here we assume that the average success of TD5 is about 80%. If this conjecture is true then there exists a program whic h satisfies the definition (at least TD5 do). If the average success of TD5 is smaller than 70% then there is no such a program (of course, in such case we can change this parameter a nd make it smaller than 70%). The advantage of this definition of AI is that we c an check it for a concrete program. Of course, we cannot calculate the average success for any program due to the combinatory explosion, but we can calculate it approximately by the methods of the statistics. How we will do this. We will select at random N worlds (wo rld is TM_W with size not bigger than 20) and we will play 100 consecutive games in every world. If the world is deterministic then this will give us the expected success of our program in this world. If it is not deterministic then we will play at random in this w orld. This will give us statistically good evaluation of the expected success because the poss ibility to be extremely lucky in 100 games is very small (so is the possibility to be ex tremely unlucky). If N (the number of the tested worlds) is big then the statistical result w ill be close to the average success of our program. If \|Σ ∪ Ω ∪ { λ}\| = 5 (which is the minimum - remark 2) then the n umber of deterministic TM_W with 20 states is 200 on power of 100. If we t ake the number of nondeterministic TM_W with 19 states and level of indefiniteness one (which means with two nondeterministic 5-tuples) then this number is many times smaller than 200 on power of 100. In order to use the method of statistics we ha ve to calculate how many times is this number smaller. Otherwise we will use wrong correla tion between deterministic and nondeterministic TM_W. Anyway, such wrong correlati on will make an unessential change in the definition of AI. 12 Conclusion Now we have definition of AI and at least one progr am (TD5) which satisfies it (with assumption 1). The first question is: Does this def inition satisfy our intuitive idea that AI is a program which is more intelligent than a human be ing. Yes, but for some values of the parameters educational time and level of intelligen ce. In this paper the educational time was fixed on 100 games each of them no longer than 1000 steps (educational time is equal to the life length because we learn all our life). The level of intelligence here was fixed on 20. Which means that we assume that we can find a m odel of the world which is TM_W with size not bigger than 20. We cannot say what is the exact level of intelligence of the human being. The second question is: Is TD5 which satisfies the definition the program which we are looking for. The answer is definitely no. We are lo oking for a program which can work in real time. Also, our intuitive idea is that AI shou ld build a model of the world and on the base of it AI should plan its behavior. Look at [6, 7]. Instead of this, TD5 uses brutal force in order to find the best strategy. Even TD5 will not know what to do on the game 101 because its strategy is only for 100 games. Here we will offer the last trivial decision (TD6), which corresponds better to our intuitive idea for AI. Let TD6 be the program which tries all deterministi c TM_W and accepts as a model of the world the first one (the shortest one) which genera tes the life until the present moment. After selecting a model (which will be a big proble m for more complicated worlds) TD6 will use this model and the Max-Sum algorithm in or der to plan its next move. Here the Max-Sum algorithm has two modifications: First, the re is no Sum (or there is only one term in the sum) because the modeling TM_W is deter ministic and there is only one possible reaction from the world. Second modificati on is that Max-Sum will not calculate until the end of the life or even until the end of the game because this will give a combinatory explosion. Instead, it will calculate s everal steps like chess playing programs do. You can find a program similar to TD6 in [5]. Reall y, in [5] we are looking for TM_W, which is a generator of infinite sequence v 1, v 2, v 3, ... instead of looking for transducer from d 1, d 2, d 3, ... to v 1, v 2, v 3, ... . Also, [5] does not make moves (there is no Max-Sum algorithm for calculating the next move). The only thing which [5] does is to predict the next number of the sequence. Anyway, in [5] you can see that searching for a model in the set of all TM_W works only if the model is very sim ple. If the size of TM_W is bigger than 5, the result is combinatory explosion. If we modify TD6 in order to accept also the nondet erministic Turing machines as models of the world then we will have too much possible mo dels. In this case we have to consider more than one model and to evaluate each possible m odel in order to see how reliable it is. 13 Anyway, TD6 and its modifications are not the progr am which we are looking for, although TD6 can be done to satisfy the definition (because the definition does not say anything about the efficiency of AI). The program w hich we are looking for is much closer to that one which is described in [6, 7]. The probl em in TD6 is that it looks for a model of the world which consists from only one item. It is better if the model is a set of many items (the items can be Turing machines, final automata o r logical formulas). When we make a theory in logic then it consist from a set of axiom s and we can change smoothly the theory by modifying, adding or deleting one axiom. Any the ory in logic is a model of some world. AI has to use similar models which can be modified smoothly. Bibliography [1] Dobrev D. D. A Definition of Artificial Intelli gence. In: Mathematica Balkanica, New Series, Vol. 19, 2005, Fasc. 1-2, pp.67-74. [2] Dobrev D. D. AI - What is this. In: PC Magazine - Bulgaria, November'2000, pp.12- 13 (in Bulgarian, also in [4] in English). [3] Dobrev D. D. AI - How does it cope in an arbitr ary world. In: PC Magazine - Bulgaria, February'2001, pp.12-13 (in Bulgarian, al so in [4] in English). [4] Dobrev D. D. AI Project, http://www.dobrev.com/ AI [5] Dobrev D. D. First and oldest application, http ://www.dobrev.com/AI/first.html (1993) [6] Dobrev D. D. Testing AI in one Artificial World . Proceedings of XI International Conference "Knowledge-Dialogue-Solution", June 2005 , Varna, Bulgaria, Vol.2, pp.461- 464. [7] Dobrev D. D. AI in Arbitrary World. Proceedings of the 5th Panhellenic Logic Symposium, July 2005, University of Athens, Athens, Greece, pp.62-67. [8] Kolmogorov A. N. and Uspensky V. A. Algorithms and randomness. - SIAM J. Theory of Probability and Its Applications, vol. 32 (1987), pp.389-412. [9] Turing, A. M. On Computable Numbers, with an Ap plication to the Entscheidungsproblem. Proceedings of the London Mat hematical Society, Series 2, 42, 1936-37, pp.230-265.
45f6d720-2174-44d0-bd71-8eac1be6438a	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Chat GPT's views on Metaphysics and Ethics Was playing around with chat gpt and and some fun learning about its thoughts on metaphysics. It looks like the ego is an illusion and hedonistic utilitarianism is too narrow minded to capture all of welfare. Instead, it opts for principles of beneficence, non-maleficence, autonomy, and justice. Seems to check out. What do you guys think? ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/tbvxdjld7dhloecd3iww.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/qiyxikkjrjwbsspt7qoe.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/b7lbpqmjobitogtdv5bn.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/ryvvapdgopzl5exs3yj1.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/zuph6fj2kitxv6jwuf5h.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/rltx6bcfv1kqgzw4cqdo.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/bnin0khxepark30ikznj.png) ![](http://res.cloudinary.com/lesswrong-2-0/image/upload/v1670091154/mirroredImages/6sQMYcDKrQzg7vCnH/r5o4tr9pugiu2u0ptz4b.png)
be44bd19-12c6-4751-ab66-876b5bec82b4	trentmkelly/LessWrong-43k	LessWrong	A Positive Double Standard—Self-Help Principles Work For Individuals Not Populations This story is reposted from nonzerosum.games where it appears in it’s intended form, full colour with functioning interactive elements, jump over to the site for the authentic experience. I have a confession: As a teenager I was desperate to learn how to win friends and influence people, to think and grow rich and develop the habits of highly effective people (seven to be precise)-I am James and I am a self-help-oholic. These tomes held for me the promise of fame and fortune (though they would earnestly profess otherwise) but most importantly they offered a salve for my crippling shyness. PERSONAL RESPONSIBILITY While I am, today, 20 years self-help-sober, the genre is as popular as ever-with its imperative to stand up straight with your shoulders back (Jordan Peterson), to take extreme ownership (Jocko Willink) and exercise, subtly, the art of not giving a f*ck (Mark Manson). Though not all the messages are as hyper-masculine as these, they all encourage the seemingly uncontroversial virtue of ‘personal responsibility’. I will argue that, regarding ‘personal responsibility’, we should employ a positive double standard. WHAT IS A POSITIVE DOUBLE STANDARD? > It is easy to hold a double standard where we expect others to take personal responsibility, while excusing our own shortfalls. > > A positive double standard requires the opposite-that I take personal responsibility while not expecting the same of society. This position holds that we have the power to change our own individual futures, but recognises the statistical realities of social dynamics on populations. FACTS The power of ‘personal responsibility’ to positively impact one’s own life has some scientific support. Studies on locus of control, developed by Julian B. Rotter in 1954, found that subjects who believed they could influence life’s outcomes through action tended to attain better job satisfaction, more successful careers, and better stress management. Albert Bandura’s research in the
21d4e381-f681-4f21-a607-3d334cd56787	trentmkelly/LessWrong-43k	LessWrong	Conditioning, Counterfactuals, Exploration, and Gears ,,,,,,,,,, The view of counterfactuals as just conditioning on low-probability events has a lot going for it. To begin with, in a bayesian setting, updates are done by conditioning. A probability distribution, conditioned on some event x (an imaginary update), and a probability distribution after actually seeing x (an actual update) will be identical. There is an issue with conditioning on low-probability events, however. When x has a low probability, the conditional probability P(x∧y)P(x) has division by a small number, which amplifies noise and small changes in the probability of the conjunction, so estimates of probability conditional on lower-probability events are more unstable. The worst-case version of this is conditioning on a zero-probability event, because the probability distribution after conditioning can be literally anything without affecting the original probability distribution. One useful intution for this is that probabilities conditional on x are going to be less accurate, when you've seen very few instances of x occuring, as the sample size is too small to draw strong conclusions. However, in the logical inductor setting, it is possible to get around this with infinite exploration in the limit. If you act unpredictably enough to take bad actions with some (very small) probability, then in the limit, you'll experiment enough with bad actions to have well-defined conditional probabilities on taking actions you have (a limiting) probability 0 of taking. The counterfactuals of standard conditioning are those where the exploration step occured, just as the counterfactuals of modal UDT are those where the agents implicit chicken step went off because it found a spurious proof in a nonstandard model of PA. Now, this notion of counterfactuals can have bad effects, because zooming in on the little slice of probability mass where you do x is different from the intuitive notion of counterfacting on doing x. Counterfactual on me walking off a cliff, I'd b
6cf82e30-deec-42a4-b9ab-5b5185d58208	trentmkelly/LessWrong-43k	LessWrong	Dealing with the left overs: COVID-19 A recent news story highlighted how China's next bit COVID-19 challenge is how to deal with all the contaminated trash that comes as a result of dealing with the epidemic. Clearly China has a lot more to deal with but this is not something other countries wont have to face as well. It also seems to be another case where limited amounts are nothing to really worry about but the potential to overload current capacity seems a real potential. One might think South Korea, Iran and Italy are all standing directly in front of that door. I have not seen any mention of this aspect of the problem mentioned by WHO, CDC or other official health organizations. Is this a planning gap those countries need to get on paper really soon and something that everyone else needs to account for? (For instance, California just declared a state of emergency. Is disposal part of that emergency planning?) Are these organization really learning anything or do we keep playing catch-up? If so, why? Wasn't waste disposal a big deal for HIV? Alternatively, if survival of the virus outside a host is limited, will some simply trash quarantine, with some type of handling (a bit like turning the compote heap over to make it effective) rules, be sufficient?
c0d084fd-4826-42b5-a076-23e4090ace65	trentmkelly/LessWrong-43k	LessWrong	Ontologies are Operating Systems Ontologies are Operating Systems: Post-CFAR 1 [I recently came back from volunteering at a CFAR workshop. I found the whole experience to be 100% enjoyable, and I’ll be doing an actual workshop review soon. I also learned some new things and updated my mind. This is the first in a four-part series on new thoughts that I’ve gotten as a result of the workshop. If LW seems to like this one, I'll post the rest too.] I’ve been thinking more about the idea of how we even reason about our own thinking, our “ontology of mind”, and how our internal mental model of how our brain works. (Roughly speaking, “ontology” means the framework you view reality through, and I’ll be using it here to refer specifically to how we view our minds.) Before I continue, it might be helpful to ask yourself some of the below questions: * What is my brain like, perhaps in the form of a metaphor? * How do I model my thoughts? * What things can and can’t my brain do? * What does it feel like when I am thinking? * Do my thoughts often influence my actions? <reminder to actually think a little before continuing> I don’t know about you, but for me, my thoughts often feel like they float into my head. There’s a general sense of effortlessly having things stream in. If I’m especially aware (i.e. metacognitive), I can then reflect on my thoughts. But for the most part, I’m filled with thoughts about the task I’m doing. Though I don’t often go meta, I’m aware of the fact that I’m able to. In specific situations, knowing this helps me debug my thinking processes. For example, say my internal dialogue looks like this: “Okay, so I’ve sent to forms to Steve, and now I’ve just got to do—oh wait what about my physics test—ARGH PAIN NO—now I’ve just got to do the write-up for—wait, I just thought about physics and felt some pain. Huh… I wonder why…Move past the pain, what’s bugging me about physics? It looks like I don’t want to do it because… because I don’t think it’ll be
d669cba6-2c49-4de9-8580-e1799d97a10b	trentmkelly/LessWrong-43k	LessWrong	AI Safety Strategies Landscape The full draft textbook is available here. This document constitutes the Chapter 3. Introduction tldr: Even if we still don't know how to make AI development generally safe, many useful classes of strategies already exist, which are presented in this chapter. You can look at the table of contents and the first figure to see the different classes of strategies presented in this document. Epistemic Status: I'm pretty satisfied with this document. I wrote it because it doesn't seem like we've made any major breakthroughs in alignment in the last year, and I wanted to consolidate what I know. And beyond alignment, it seems to me that a large class of strategies are quite important and neglected, and will continue to be relevant in the future. Alignment is only one class of strategy to achieve AI safety. And to mitigate misuses and systemic risks, I think we already have a pretty good idea of what could be done. Let me know if you think there is a major blind spot in this document. Although the field of AI safety is still in its infancy, several measures have already been identified that can significantly improve the safety of AI systems. While it remains to be seen if these measures are sufficient to fully address the risks posed by AI, they represent essential considerations. The diagram below provides a high-level overview of the main approaches to ensuring the safe development of AI. Tentative diagram summarizing the main high-level approaches to make AI development safe. This document is far from exhaustive and only scratches the surface of the complex landscape of AI safety. Readers are encouraged to explore this recent list of agendas for a more comprehensive review. AI Safety is Challenging Specific properties of the AI safety problem make it particularly difficult. AI risk is an emerging problem that is still poorly understood. We are not yet familiar with all its different aspects, and the technology is constantly evolving. It's hard to devise solutions
5b0f9202-20d6-440c-a25a-9603db009598	trentmkelly/LessWrong-43k	LessWrong	The Great Filter isn't magic either A post suggested by James Miller's presentation at the Existential Risk to Humanity conference in Gothenburg. Seeing the emptiness of the night sky, we can dwell upon the Fermi paradox: where are all the alien civilizations that simple probability estimates imply we should be seeing? Especially given the ease of moving within and between galaxies, the cosmic emptiness implies a Great Filter: something that prevents planets from giving birth to star-spanning civilizations. One worrying possibility is the likelihood that advanced civilizations end up destroying themselves before they reach the stars. The Great Filter as an Outside View In a sense, the Great Filter can be seen as an ultimate example of the Outside View: we might have all the data and estimation we believe we would ever need from our models, but if those models predict that the galaxy should be teeming with visible life, then it doesn't matter how reliable our models seem: they must be wrong. In particular, if you fear a late great filter - if you fear that civilizations are likely to destroy themselves - then you should increase your fear, even if "objectively" everything seems to be going all right. After all, presumably the other civilizations that destroyed themselves thought everything seemed to going all right. Then you can adjust your actions using your knowledge of the great filter - but presumably other civilizations also thought of the great filter and adjusted their own actions as well, but that didn't save them, so maybe you need to try something different again or maybe you can do something that breaks the symmetry from the timeless decision theory perspective like send a massive signal to the galaxy... The Great Filter isn't magic It can all get very headache-inducing. But, just as the Outside View isn't magic, the Great Filter isn't magic either. If advanced civilizations destroy themselves before becoming space-faring or leaving an imprint on the galaxy, then there is some phenome
e00bd717-4b34-499f-b55a-f8d7cefbadeb	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Newcomb Variant There are two boxes. If you open a box, you get to keep the contents. Omega is an omniscient oracle. Omega predicts what will happen if you open the first box and observe that it contains $100. * If Omega predicts that you will open the second box, then Omega puts $0 in both boxes. * If Omega predicts that you will not open the second box, then Omega puts $100 in both boxes. You open the first box. It contains $100. Do you open the second box? Answer ====== No. If you open the second box, that means you're being simulated by Omega. If you're being simulated by Omega, then opening the second box ends the simulation and kills you. Extra Credit [Orthogonal Thinking] ================================== You have not yet opened the first box. How do you generate (at least) $250 in profit? [Please use spoiler tags or equivalent obfuscation when discussing the extra credit.]
32bf5776-a13d-417a-bf59-422db2964c1f	trentmkelly/LessWrong-43k	LessWrong	Calling all Nigerian rationalists and effective altruists I'm in Lagos, Nigeria till the end of May and I'd like to hold a LessWrong/EA meetup while I'm here. If you'll ever be in the country in the future (or in the subcontinent), please get in touch so we can coordinate a meetup. I'd also appreciate being put in contact with any Nigerians who may not regularly read this list. My e-mail address is oge@nnadi.org. I hope to hear from you.
11b8d9df-2a35-4640-8263-27f5baccca51	StampyAI/alignment-research-dataset/aisafety.info	AI Safety Info	What is AI Safety via Debate? Debate is a proposed technique for allowing human evaluators to get correct and helpful answers from experts, even if the evaluator is not themselves an expert or able to fully verify the answers.[^7clr966emb9] The technique was suggested as part of an approach to build advanced AI systems that are aligned with human values, and to safely apply machine learning techniques to problems that have high stakes, but are not well-defined (such as advancing science or increase a company's revenue). [^vrcbanw2zz][^nwfhnzy6a3e] [^7clr966emb9]: [https://www.lesswrong.com/posts/Br4xDbYu4Frwrb64a/writeup-progress-on-ai-safety-via-debate-1](https://www.lesswrong.com/posts/Br4xDbYu4Frwrb64a/writeup-progress-on-ai-safety-via-debate-1) [^vrcbanw2zz]: [https://ought.org/mission](https://ought.org/mission) [^nwfhnzy6a3e]: [https://openai.com/blog/debate/](https://openai.com/blog/debate/)
e074b681-e62c-454e-b9ee-9f64a88554ca	trentmkelly/LessWrong-43k	LessWrong	Has LessWrong been a good early alarm bell for the pandemic? The S&P 500 began to crash on Feb. 20th. On or prior to that date, there were 6 posts on LessWrong related to COVID-19 (thanks to habryka, who found 3 that I'd missed). Jan 20 - Rationalist prepper thread Jan 26 - Concerning the Recent 2019-Novel Coronavirus Outbreak Feb 5 - Some quick notes on hand hygiene Feb 11 - Will nCoV survivors suffer lasting disability at a high rate? Feb 14 - Please Help Metaculus Forecast COVID-19 Feb 20 - Making sense of coronavirus stats As a bonus, Scott Alexander dropped a link in his Links 2/20 post saying "As of earlier this month, China’s coronavirus case numbers followed such a neat quadratic curve that they seem kind of like low-effort fakes. Not sure if this also applies to the current numbers." Near-daily COVID posting began on Feb. 20th. After Feb. 27th, the number of LW COVID-19 posts has increased by about 4 posts per day to a total of nearly 200: The New York Times has compiled a recent history of the pandemic. Before Feb 20, what had happened? * The WHO had declared a global health emergency * The Trump administration had restricted travel from China * There had been hundreds of deaths, including some outside China * Almost 50,000 cases were reported in Hubei Province * Hundreds of people aboard the Diamond Princess cruise ship were infected, the ship had been quarantined, and the passengers had finally been able to disembark By Feb. 23, the S&P 500 had started to descend, and Italy was beginning to lock its cities down. It wasn't until around Feb. 27th, after Trump requested a big bailout from Congress, Iran reported over 60 deaths, and cases had spread to Latin America, that LessWrong began to see accelerated COVID posting. The number spiked from around 1 post per day to the average of 4-5 posts per day we've seen since then. If the sheer volume of conversation is our alarm bell, this site seems to have lagged behind the stock market by about a week.
e716cb26-1db2-48ad-bce5-47555641b14d	trentmkelly/LessWrong-43k	LessWrong	Friendly AI - Being good vs. having great sex > [...] I think an LW post is important and interesting in proportion to how much it helps construct a Friendly AI, how much it gets people to participate in the human project [...] -- Eliezer Yudkowsky > I’m not going to wait for philosophers to cover this issue correctly, or for use in FAI design. -- Luke Muehlhauser ---------------------------------------- The above quotes hint at the possibility that some of the content that can be found on lesswrong.com has been written in support of friendly AI research. My question, of what importance is ethics when it comes to friendly AI research? If a friendly AI is one that does protect and cultivate human values, how does ethics help to achieve this? Let's assume that there exist some sort of objective right, no matter what that actually means. If humans desire to be right, isn't it the sort of human value that a friendly AI would seek to protect and cultivate? What difference is there between wanting to be good and wanting to have a lot of great sex? Both seem to be values that humans might desire, therefore both values have to be taken into account by a friendly AI. If a friendly AI has to be able to extrapolate the coherent volition of humanity, without any hard-coded knowledge of human values, why doesn't this extent to ethics as well? If we have to solve ethics before being able to design friendly AI, if we have to hard-code what it means to be good, how doesn't this apply to what it means to have great sex as well (or what it means to have sex anyway)? If a friendly AI is going to figure out what humans desire, by extrapolating their volition, might it conclude that our volition is immoral and therefore undesirable?
4740f25a-dea7-4f6c-be97-843754e42d24	trentmkelly/LessWrong-43k	LessWrong	Inexistence of Rational Disagreement when Information can be Freely Exchanged Suppose rationality is a set of principles that people agreed on to process information then arrive at conclusions. Then, on the basis of cost-free information exchange, should rational disagreements still exist? In that case, both parties would have the same information which will then be processed the same way. Just by these factors, there shouldn't be. However, disagreements do still exist, and we'd like to believe we're rational, so the problem must be in the exchange of information. Previous posts have mentioned how sometimes there are too much background information to be exchanged fully. Here I'd like to point to a more general culprit: language. Not all knowledge can be expressed through language, and not all languages express knowledge. Yet language, including obscure symbols that take in mathematics, n order logic, and other communicable disciplines, still so far cannot convey a significant portion of our knowledge, such as intuition and creativity. Substantial amount of studies have shown that intuition is more accurate than thinking in certain areas, and much worse in other areas. Yet we have not came up with a way to systematically use intuition and rational judgement selectively. And I'd say this is the obstacle in most rationalist disagreements: it's not that when they can freely discuss for as long as possible then they will definitively agree; it's that there is knowledge unique to themselves that is incommunicable, but that considerably swayed their judgements of things. As we progress as a species we expand our languages to communicate more complexity, so this issue should gradually fade away, that is unless the scale of complexity of knowledge is infinite.
79d3a55c-c408-4761-880c-14a069d55d06	StampyAI/alignment-research-dataset/eaforum	Effective Altruism Forum	Beware safety-washing Tl;dr: don’t be fooled into thinking that some groups working on AI are taking “safety” concerns seriously (enough).[[1]](#fn6zkt67gme6q) Outline * [Two non-AI examples](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#Two_non_AI_examples) + [Greenwashing](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#Greenwashing) + [Humanewashing](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#Humanewashing) * [Definition of safety-washing](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#Definition_of_safety_washing) * [What are the harms?](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#What_are_the_harms_) * [What can (and should) we do about this?](https://forum.effectivealtruism.org/posts/f2qojPr8NaMPo2KJC/beware-safety-washing#What_can__and_should__we_do_about_this_) Note: I’m posting this in my personal capacity. All views expressed here are my own. I am also not (at all) an expert on the topic. Two non-AI examples =================== Greenwashing ------------ Companies “greenwash” when they mislead people into incorrectly thinking that their products or practices are climate and environment-friendly (or that the company focuses on climate-friendly work). [Investopedia](https://www.investopedia.com/terms/g/greenwashing.asp) explains: * Greenwashing is an attempt to capitalize on the growing demand for environmentally sound products. * The term originated in the 1960s, when the hotel industry devised one of the most blatant examples of greenwashing. They placed notices in hotel rooms asking guests to reuse their towels to save the environment. The hotels enjoyed the benefit of lower laundry costs. + [Wikipedia](https://en.wikipedia.org/wiki/Greenwashing): “[Jay Westerveld, the originator of the term] concluded that often the real objective was increased profit, and labeled this and other profitable-but-ineffective ‘environmentally-conscientious’ acts as greenwashing.” (Wikipedia also provides [a long list of examples](https://en.wikipedia.org/wiki/Greenwashing) of the practice.) I enjoy [some of the parody/art](https://twitter.com/darren_cullen/status/1252962221811957760) (responding to [things like this](https://www.nbcnews.com/id/wbna34111784)) that comes out of noticing the hypocrisy of the practice. Humanewashing ------------- A similar phenomenon is the “humanewashing” of animal products. There’s [a Vox article](https://www.vox.com/22838160/animal-welfare-labels-meat-dairy-eggs-humane-humanewashing) that explains this phenomenon (as it happens in the US): > A carton of “all natural” eggs might bear an illustration of a rustic farm; packages of chicken meat are touted as “humanely raised." > > In a few cases, these sunny depictions are accurate. But far too often they mask the industrial conditions under which these animals were raised and slaughtered. > > Animal welfare and consumer protection advocates have a name for such misleading labeling: “humanewashing.” And research suggests it’s having precisely the effect that meat producers intend it to. A recent [national survey](https://co-nxt.com/blog/food-label-claims-that-make-consumers-more-likely-to-buy/) by C.O.nxt, a food marketing firm, found that animal welfare and “natural” claims on meat, dairy, and egg packaging increased the intent to purchase for over half of consumers. > > [...] > > ...rather than engaging in the costly endeavor of actually changing their farming practices, far too many major meat producers are attempting to assuage consumer concerns by merely changing their packaging and advertising with claims of sustainable farms and humane treatment. These efforts mislead consumers, and undermine the small sliver of farmers who have put in the hard work to actually improve animal treatment. > > If you want a resource on what food labels actually mean, here are some: [one](https://www.farmforward.com/label-guide/), [two](https://labelsunwrapped.org/explore-labels/protein), [three](https://www.humanesociety.org/sites/default/files/docs/hsus-food-label-meanings-2019.pdf) (these are most useful in the US). (If you know of a better one, please let me know. I’d especially love a resource that lists the estimated relative value of things like “free-range” vs. “cage-free,” etc., according to cited and reasonable sources.) Definition of safety-washing ============================ In brief, “safety-washing” is misleading people into thinking that some products or practices are “safe” or that safety is a big priority for a given company, when this is not the case. An increasing number of people believe that developing powerful AI systems is very dangerous,[[2]](#fnz8j2j3hnql) so companies might want to show that they are being “safe” in their work on AI.[[3]](#fng6uq6rvrkni) Being safe with AI is hard and potentially costly,[[4]](#fnswopw71dt4) so if you’re a company working on AI capabilities, you might want to overstate the extent to which you focus on “safety.” So you might: * Pick a [safety paradigm](https://www.lesswrong.com/posts/JC7aJZjt2WvxxffGz/paradigms-of-ai-alignment-components-and-enablers) that is convenient for you, and focus on that * Talk about “safety” when you really mean other kinds of things the public might want an AI to be, like un-biased and not-hateful * Start or grow a safety team, feature it in media about your work (or conversations with safety-oriented people), but not give it a lot of power * Promote the idea that AI safety concerns are [crazy](https://forum.effectivealtruism.org/posts/zsFCj2mfnYZmSW2FF/ai-risk-is-like-terminator-stop-saying-it-s-not-1) * And more Some of these things might be better than doing nothing for safety concerns, but overall, (safety-)washing causes some problems (discussed in the next section), which in turn worsens the situation with risk from AI. Related: [Perhaps It Is A Bad Thing That The World's Leading AI Companies Cannot Control Their AIs](https://astralcodexten.substack.com/p/perhaps-it-is-a-bad-thing-that-the) (Astral Codex Ten) What are the harms? =================== I don’t have the time to write a careful report on the matter, but here are some issues that I think arise from greenwashing, humane-washing, and safety-washing: 1. Confusion: People working on the issue (and the general public) get confused about what really matters — terms lose their meanings, groups lose focus, etc. * E.g. Some people who want to help the climate think that it’s important to encourage the reuse of towels instead of avoiding harmful products (or focusing on more [effective methods for fighting climate change](https://www.effectiveenvironmentalism.org/)). 2. Accidental harm: People are misled about what companies are doing, which in turn leads to people doing directly harmful things they didn’t intend to do * E.g. This encourages people to work for harmful companies/projects or to support them financially because they’re not aware of the harm the companies cause. 3. False security: Causes a false sense of safety/goodness/progress (which can lead to insufficient mitigation of the harm caused, a lack of other kinds of preparation, and other problems) * E.g. someone who successfully convinces some groups to focus on “eating local” may think that the tide is turning on the environmental impacts from food, even though [this is not the key issue](https://hannahritchie.substack.com/p/food-miles) (or [the most effective area of work for fighting climate change](https://founderspledge.com/stories/climate-change-executive-summary)). 4. Thwarted incentive: Reduces the incentive for companies to actually reduce the harm they (might) cause * If you’re a company and you can get away with labeling your product as safe/green/humane, which gets you the benefit of consumer approval and a lack of hate, you don’t need to put in extra work to actually make your work safe/green/humane. 5. And more? What can (and should) we do about this? ======================================= Some things that come to mind: 1. To counteract confusion, we can try to be more specific in explanations about “safety” or “humane conditions” or [use more specific terms](https://www.lesswrong.com/posts/xhD6SHAAE9ghKZ9HS/safetywashing?commentId=dy9tkbfPiTRjnx8aB) like “existential safety” 2. To counteract our own confusion, we could encourage (even) more [distillation of content](https://forum.effectivealtruism.org/posts/EbvJRAvwtKAMBn2td/distillation-and-research-debt) and external validation of work 3. [Stare into the abyss](https://www.benkuhn.net/abyss/) about the possibility that our work is not useful (or is harmful), and seek external reviews and criticism 4. We could also create or support standards for safety or external validation systems (like [Certified Humane](https://www.greenchoicenow.com/v/know-your-label-certified-humane)), and evaluate projects against that ([e.g.](https://www.lesswrong.com/posts/xhD6SHAAE9ghKZ9HS/safetywashing?commentId=2tHroKYAKXG9YkpJL)) (although versions of this might be gameable, and we should beware new “standards” for [the usual reasons](https://xkcd.com/927/)). 5. Call out safety-washing (and other kinds of washing). 6. Call out organizations doing things that are bad on their merits, and be clear about why what they showcase as safety-oriented work (or efforts to be more humane, etc.) insufficiently address the risks and harms of their work. How important or promising is all of this as an approach or a type of work to focus on? I’m not sure — I’d guess that it’s not the most valuable thing to focus on for most people, but would be interested in other people’s thoughts. My main motivation for writing this was that I think the phenomenon of safety-washing exists and will become more prominent, and we should keep an eye out for it. ![](https://res.cloudinary.com/cea/image/upload/v1673606345/mirroredImages/f2qojPr8NaMPo2KJC/v3bml4ft7prnhred2fj0.png)Image credit: Dall-e.I'm a bit swamped and may not respond to comments, but will probably read them and will be very grateful for them (including for corrections and disagreements!). "Safety-washing" might also be spelled "safetywashing." I don't know which is better or more common, and have gone with the former here. 1. [^](#fnref6zkt67gme6q)After I wrote a draft of this post, I noticed that there was [a very similar post on LessWrong](https://www.lesswrong.com/posts/xhD6SHAAE9ghKZ9HS/safetywashing). I should have checked earlier, but I’m posting this anyway as it is slightly different (and somewhat more detailed) and because some Forum users may not have seen the LW version. 2. [^](#fnrefz8j2j3hnql)Here are some resources you can explore on this topic if you want to learn more: [one](https://www.vox.com/the-highlight/23447596/artificial-intelligence-agi-openai-gpt3-existential-risk-human-extinction), [two](https://www.vox.com/the-highlight/23447596/artificial-intelligence-agi-openai-gpt3-existential-risk-human-extinction), [three](https://forum.effectivealtruism.org/s/HJELzLxMRdbcqwsgC), [four](https://80000hours.org/problem-profiles/artificial-intelligence/), [five](https://forum.effectivealtruism.org/s/vtmN9g6C57XbqPrZS), [six](https://scottaaronson.blog/?p=6823), [seven](https://www.agisafetyfundamentals.com/). 3. [^](#fnrefg6uq6rvrkni)Safety isn’t the only thing that people care about, in terms of ethical concerns about AI, and it’s probably not the most popular concern. I’m focusing on safety in this post. Other concerns have been discussed in e.g. Forbes: [Forbes discusses AI Ethics washing](https://www.forbes.com/sites/lanceeliot/2022/06/09/ai-ethics-and-legal-ai-are-flustered-by-deceptive-pretenses-known-as-ai-ethics-washing-which-are-false-claims-of-adhering-to-ethical-ai-including-for-autonomous-self-driving-cars/?sh=c21afc12b657) (paywalled) — “AI Ethics washing entails giving lip service or window dressing to claimed caring concerns about AI Ethics precepts, including at times not only failing to especially abide by Ethical AI approaches but even going so far as to subvert or undercut AI Ethics approaches.” I only skimmed the article but it seems to focus on self-driving cars as its motivating example. It also separates “washers” into four groups; those who wash by ignorance, by good-motivations-stretched-or-slipped, by stretching-the-truth or spinning it, and those who brazenly lie. It also describes “Ethics Theatre” — making a big show of your ethics work, “Ethics shopping” — picking the guidelines that are easiest to adopt, “Ethics bashing” — e.g. insisting the guidelines are worthless or a cover-up, “Ethics Shielding” — I didn’t quite follow this one, “Ethics Fairwashing” — specifically focusing on claims that an AI is fair when it isn’t. 4. [^](#fnrefswopw71dt4)If you think that AI risk is not miniscule, then being safe (even if it means being slow) is also in your interests — see [this section of “Let’s think about slowing down AI.”](https://forum.effectivealtruism.org/posts/vwK3v3Mekf6Jjpeep/let-s-think-about-slowing-down-ai-1#The_complicated_race_anti_race) But maybe you think safety concerns are overblown, and you’re just viewing safety efforts as appeasement of the risk-concerned crowd. Or you have [myopic](https://www.lesswrong.com/tag/myopia) incentives, etc. In that case, you might think that being safe just slows you down and wastes your resources.
63be28e3-27c6-4445-8f8d-93fe6abd5dba	trentmkelly/LessWrong-43k	LessWrong	Police violence: The veil of darkness Measuring police bias using simple ratios doesn't work. You can never cleanly separate the impact of race from other factors associated with race. Instead, I want augmented-reality goggles that make race invisible. Then we could run the following experiment: * Have half of police wear race-invisibility goggles for a year. * Have the other half wear non-invisibility goggles. * Look at the difference of the two groups. The police with invisibility goggles would not have equal statistics with respect to race, because race is correlated with many things other than how people appear. However, since the only difference between the two groups is if police can see race, the difference reveals exactly the impact of police bias. We can't do this experiment, but we can do kind of low-tech approximation. Instead of augmented reality goggles we use the geometry of the earth and sun. [continues here]
bbd65d6a-ba4a-4f4a-bec0-dc1dda1844c6	trentmkelly/LessWrong-43k	LessWrong	Outline of an approach to AGI Estimation We are worried about what will happens when we make a system that can do the important things that human can do, like programming and science. Will it explode off into infinity as it finds better ways to improve itself or will it be a slower more manageable process? There are a number of productive ways we can react to this, not limited to: 1. Attempt to make AI systems controllable by focusing on the control problem 2. Attempt to predict when we might get agi by looking at progress, e.g. ai impacts 3. Attempt to predict what happens when we get artificial general intelligence by looking at current artificial intelligences and current general intelligences and making predictions about the intersection. 4. Figure out how to make intelligence augmentation, so we can improve humans abilities to compete with full AGI agents. This blog post is looking at the third option. It is important, because the only reason we think AGIs might exist is the existence proof of humans. Computer AGI might be vastly different in capabilities but we can't get any direct information about them, our only possible approach to predicting their capability is to adjust our estimate of what humans can do based on the difference between humans and narrow AI on different tasks and those tasks importance for economically important generality. There are three core questions to AGI Estimation that I have identified so far. 1. What do we mean by 'general intelligence' 2. How important is 'general intelligence' to different economic activities 3. How good are narrow AIs at parts of the processes involved in general intelligence vs humans. Can we estimate how good an AGI would be at another task based on the comparison? Generality We often speak and act as if there is something to one person being better than another person and mental acts in general. We try and measure it with IQ, we think there is something there. However there are a number of different things "general intelligence"
fac56cc3-7cff-4ec1-afd9-49715f1cb040	trentmkelly/LessWrong-43k	LessWrong	How can I argue without people online and not come out feeling bad? I wont be the only one here who "wastes time" arguing about things they care about online (note: I am referring to web forums and things like subreddits, I am not including Less Wrong whose dynamic is completely different). It seems like something that is worth optimising in some direction. The theory behind it is that one should expose themselves to counter-arguments allowing their claims to be attacked so they that have a chance to substantiate them or reject them upon realising they are mistaken. In practice they generally follow a pattern that starts with people pointing out what they believe are mistakes then ignoring or intentionally misunderstanding the other party when he refutes or backs up claims.. and ends up with insults, patronising sarcastic remarks and nobody changing their mind about anything. I don't particularly care about changing other peoples minds to make them agree with me (well, it would be great but I think it's practically impossible) so one thing I would like is for both people to at least end up feeling good. So I'm interested in three things: Do other LWers recognize this pattern now that I have mentioned it? What decision did those that were already aware of it make, in order to optimise this activity?
381ef11b-3e8b-4165-b0ef-da1d4b04e11a	trentmkelly/LessWrong-43k	LessWrong	Neural network polytopes (Colab notebook) The polytope theory of neural networks (also known as the "spline theory of deep learning") seeks to explain (ReLU) neural networks based on their piecewise linear regions. This gives a helpful intuition for how neural networks approximate functions, as well as a potential avenue for interpretability research. For anyone who wants to engage directly with the topic, I put together a small Colab notebook, with interactive 2D and 3D visuals.
28ae9704-d523-4676-be23-a865eae83d3a	trentmkelly/LessWrong-43k	LessWrong	Meetup : Dallas - Fort Worth Less Wrong Meetup 4/29/12 Discussion article for the meetup : Dallas - Fort Worth Less Wrong Meetup 4/29/12 WHEN: 29 April 2012 01:00:00PM (-0500) WHERE: America's Best Coffee, Arlington I apologize for the short notice. The Dallas / Fort Worth metroplex area desperately needs a Less Wrong meetup! Fun fact – it is the largest metro area in the US that does not have an active meetup group! That said - I shall be sitting at America's Best Coffee [3751 Matlock Road, Arlington TX 76105] from 1 PM to 3 PM (at the very least) on Sundays starting Sunday April 29th. Come out and meet some DFW LessWrongers. Leading isn't really my thing, but I realize that I can't wait around forever for someone else to get the ball rolling! I am really interested in improving my mind to behave more rationally, especially with respect to instrumental rationality. I've been a self-improvement junkie for a while (Starting Strength, Toastmasters, improv, and PUA for about a year now), but only recently discovered LessWrong. Some of the recent articles sparked me to write this post. I'm working my way through some of the core sequences, slowly but surely. I'm looking forward to meeting you folks. Message me if you plan on going so we can trade contact info beforehand. Discussion article for the meetup : Dallas - Fort Worth Less Wrong Meetup 4/29/12
62525f5a-8397-4bf1-b716-6a0266b615a9	StampyAI/alignment-research-dataset/alignmentforum	Alignment Forum	Autonomy as taking responsibility for reference maintenance I think semantics – specifically, maintaining reference relationships – is a core component of intelligent behaviour. Consequently, I think a better understanding of semantics would enable a better understanding of what machine intelligence that is “trying to do the right thing” ought to look like and how to build it. Unfortunately I’ve not managed to make these ideas particularly clear (either to myself or others). In the spirit of [palimpsest-shaped research](https://www.alignmentforum.org/posts/4BpeHPXMjRzopgAZd/mosaic-and-palimpsests-two-shapes-of-research) I’m sharing a draft in the hope of provoking useful thoughts. Summary ------- * Particular events can have meaningfulness or significance, and can carry multiple meanings at once. For example, the motion of a dial in a meter might mean that there is electricity flowing through a connected wire, while simultaneously meaning that an alarm condition has been triggered. * The “meanings” (or semantics) that can be ascribed to events depend on regularities (stable mechanisms) in the surrounding context. Large-scale regularities are required to sustain large-scale semantics (something carrying meaning about another thing that is far away in time or space). * Meanings are generally more robust when their dependencies on the context are fewer and/or more local. * However, large-scale semantics can be made robust by managing the contextual dependencies. For example, using mechanisms for error detection and correction. There is compensatory work that can be done to sustain a regularity so something local continues to mean a particular thing about something distal. * To be autonomous is to be robust in this sense. Systems that more robustly traffic in large-scale meanings are more autonomous. Thus, being autonomous can be understood as including all the machinery required to hold stable, in a changing world, all the semantic or referential relationships that are required for continued operation. * Creating regularities that span larger distances is a way of creating new kinds of semantics, and autonomous systems do this as their conceptual spaces get larger. Local and distal meanings ------------------------- The battery indicator on my laptop reads “84%” when I hover my mouse cursor over it. What does this reading mean? There’s some function, I suppose, in some program running as part of my desktop environment that is designed to produce the text for the popup window. The reading means this function received the number 84 as input. OK, but that’s not all it means. There is, I suppose, a smart battery system in my laptop, and it ran a calculation that produced 84% as an estimate of the battery’s state of charge. More concretely: I imagine there are several quantities in the memory of the smart battery system, such as the last few readings from an ammeter on the battery, the times those readings were made, the last reference point for current when the battery was fully charged, etc. The reading I observed means all these memory locations contained values such that the calculation the smart battery system runs produced the value 84. This is a “more distal” meaning for the reading than “the text rendering function received 84 as input”, because the smart battery system’s calculation happens “before” (or “causally upstream of”) that function’s execution. Most distally, of course, we’d like to say that the reading means that the state of charge of the battery “really is” approximately 84%. We have a collection of meanings for the reading (..., it means the LEDs on my screen were emitting pixels arranged in a shape recognisable as “84%”, …, it means the text-rendering function received 84 as an input, …, it means the smart battery system calculated the state of charge as 84%, …, it means the battery’s state of charge is 84%), and I have just arranged them on a spectrum from local to distal. Another example of this is that the leaves on a deciduous tree being orange means that (they emitted light at wavelengths interpreted by my visual system as orange; they are filled with less chlorophyll and more carotenoids than prototypical leaves; there is an excess of cork cambium forming at the bases of the leaves; the ambient air temperature and daylight length have been lower than previously for a while; the latitude where the tree is located is experiencing less insolation; Autumn has begun). All these meanings are valid under normal circumstances, and all the mechanisms can be understood independently, though the later ones in the list involve larger and larger regions of the world in their “normal circumstances”. Local meanings are more robust by default ----------------------------------------- The local-to-distal ordering approximately lines up with another ordering that I would call “robustness”. More local meanings are more robust than more distal meanings. What does robustness mean here? Basically, it is less possible to disrupt a more robust relationship. The relationship between the image I observe and the LEDs in my screen is very tight: there are very few points of intervention by which something could break the relationship and cause the observation to fail to mean what I expect it to mean about the LEDs. But it’s easier to cause the observation to fail to mean what I expect it to mean about the battery. What does disruption involve? Failure of shielding, excessive noise, unanticipated sources of interference. More generally, the world moving on in ways that make the conditions of normal operation for the sensor fail to hold. For example, many devices on planet Earth implicitly assume a gravitational constant, an ambient magnetic field, a working constellation of GPS satellites, a mostly nitrogen-filled atmosphere, etc. Reference maintenance makes distal meanings more robust ------------------------------------------------------- There’s a lot of machinery and work going on to make the abstract quantity (the state of charge) accessible via pixels on the screen. There’s no way to “directly” read it: there will always be a more local thing that the sensor reads “directly”. That the sensor reading also means the more distal thing is due to the intact, working-as-intended, arrangement of all the machinery connecting the local to the distal. The work this machinery is doing is reference maintenance. It maintains the reference relationship between, in this case, the sensor reading and the quantity that reading refers to (the state of charge). The designers have compensated in advance for many sources of drift, miscalibration, and noise, keeping in mind the intended referent, that the percentage reading should indicate the “actual state of charge” at least for the purpose of being informative regarding how much time is left before the battery depletes. What actually matters to me, the laptop user, is some (not very precise) predictability around when my laptop will turn off due to lack of power if I don’t charge it. There can be bugs in any number of these systems, and there can be adversaries spoofing the inputs to any of the subsystems, causing the reading to be incorrect in the sense of not having the intended referent. But note that there is a normative correctness condition here. Whether some configuration is “incorrect” or not was in this case an explicit consideration by the designers of the system. But it is also implied by the compensatory mechanisms in place in the artefact. They can all be understood as working to keep the same referent stable, and that stable referent then defines what it means to be broken or spoofed (failing to keep the intended referent). Explicit intention is not necessary: for example, evolved artefacts don’t have explicit designers but many still do reference maintenance. Autonomous systems must maintain referents ------------------------------------------ A system that functions “on its own” in the world – that is, a system that is autonomous – needs to orient itself to its circumstances at some scale and take appropriate actions. But action is always mediated only by local mechanisms. When actions string together and become coherent at a larger scale, sustaining meaningful regularities, they do so by maintaining reference relationships between local configurations and the distal things those configurations are about. Thus being autonomous requires reference maintenance. Referents can be entirely subjective ------------------------------------ The work of reference maintenance can create a new subjective world. The things that are the referents of some compensatory activity are in part brought into existence by that activity – more precisely their boundaries and identity and individuation criteria are in part determined by the reference maintenance work.
da185df1-5901-442d-bcb8-b39fb66416e9	trentmkelly/LessWrong-43k	LessWrong	Clarifying AI X-risk TL;DR: We give a threat model literature review, propose a categorization and describe a consensus threat model from some of DeepMind's AGI safety team. See our post for the detailed literature review. The DeepMind AGI Safety team has been working to understand the space of threat models for existential risk (X-risk) from misaligned AI. This post summarizes our findings. Our aim was to clarify the case for X-risk to enable better research project generation and prioritization. First, we conducted a literature review of existing threat models, discussed their strengths/weaknesses and then formed a categorization based on the technical cause of X-risk and the path that leads to X-risk. Next we tried to find consensus within our group on a threat model that we all find plausible. Our overall take is that there may be more agreement between alignment researchers than their disagreements might suggest, with many of the threat models, including our own consensus one, making similar arguments for the source of risk. Disagreements remain over the difficulty of the alignment problem, and what counts as a solution. Categorization Here we present our categorization of threat models from our literature review, based on the technical cause and the path leading to X-risk. It is summarized in the diagram below. In green on the left we have the technical cause of the risk, either specification gaming (SG) or goal misgeneralization (GMG). In red on the right we have the path that leads to X-risk, either through the interaction of multiple systems, or through a misaligned power-seeking (MAPS) system. The threat models appear as arrows from technical cause towards path to X-risk. The technical causes (SG and GMG) are not mutually exclusive, both can occur within the same threat model. The distinction between them is motivated by the common distinction in machine learning between failures on the training distribution, and when out of distribution. To classify as specificatio
1e126eda-db52-4647-9c38-6eee51a78e0e	StampyAI/alignment-research-dataset/arxiv	Arxiv	Relational recurrent neural networks 1 Introduction --------------- Humans use sophisticated memory systems to access and reason about important information regardless of when it was initially perceived ([schacter1994memory,](#bib.bib1) ; [knowlton2012neurocomputational,](#bib.bib2) ). In neural network research many successful approaches to modeling sequential data also use memory systems, such as LSTMs ([hochreiter1998lstm,](#bib.bib3) ) and memory-augmented neural networks generally ([graves2014neural,](#bib.bib4) ; [graves2016hybrid,](#bib.bib5) ; [santoro2016meta,](#bib.bib6) ; [sukhbaatar2015end,](#bib.bib7) ). Bolstered by augmented memory capacities, bounded computational costs over time, and an ability to deal with vanishing gradients, these networks learn to correlate events across time to be proficient at storing and retrieving information. Here we propose that it is fruitful to consider memory interactions along with storage and retrieval. Although current models can learn to compartmentalize and relate distributed, vectorized memories, they are not biased towards doing so explicitly. We hypothesize that such a bias may allow a model to better understand how memories are related, and hence may give it a better capacity for relational reasoning over time. We begin by demonstrating that current models do indeed struggle in this domain by developing a toy task to stress relational reasoning of sequential information. Using a new Relational Memory Core (RMC), which uses multi-head dot product attention to allow memories to interact with each other, we solve and analyze this toy problem. We then apply the RMC to a suite of tasks that may profit from more explicit memory-memory interactions, and hence, a potentially increased capacity for relational reasoning across time: partially observed reinforcement learning tasks, program evaluation, and language modeling on the Wikitext-103, Project Gutenberg, and GigaWord datasets. 2 Relational reasoning ----------------------- We take relational reasoning to be the process of understanding the ways in which entities are connected and using this understanding to accomplish some higher order goal ([waltz1999system,](#bib.bib8) ). For example, consider sorting the distances of various trees to a park bench: the relations (distances) between the entities (trees and bench) are compared and contrasted to produce the solution, which could not be reached if one reasoned about the properties (positions) of each individual entity in isolation. Since we can often quite fluidly define what constitutes an “entity” or a “relation”, one can imagine a spectrum of neural network inductive biases that can be cast in the language of relational reasoning 111Indeed, in the broadest sense any multivariable function must be considered “relational.”. For example, a convolutional kernel can be said to compute a relation (linear combination) of the entities (pixels) within a receptive field. Some previous approaches make the relational inductive bias more explicit: in message passing neural networks (e.g. [gilmer2017neural,](#bib.bib9) ; [scarselli2009graph,](#bib.bib10) ; [LiTBZ15,](#bib.bib11) ; [battaglia2016interaction,](#bib.bib12) ), the nodes comprise the entities and relations are computed using learnable functions applied to nodes connected with an edge, or sometimes reducing the relational function to a weighted sum of the source entities (e.g. [kipf2016semi,](#bib.bib13) ; [velickovic2018graph,](#bib.bib14) ). In Relation Networks ([santoro2017simple,](#bib.bib15) ; [raposo2017discovering,](#bib.bib16) ; [hu2017relation,](#bib.bib17) ) entities are obtained by exploiting spatial locality in the input image, and the model focuses on computing binary relations between each entity pair. Even further, some approaches emphasize that more capable reasoning may be possible by employing simple computational principles; by recognizing that relations might not always be tied to proximity in space, non-local computations may be better able to capture the relations between entities located far away from each other ([wang2017non,](#bib.bib18) ; [liu2018non,](#bib.bib19) ). In the temporal domain relational reasoning could comprise a capacity to compare and contrast information seen at different points in time [pavez2018working](#bib.bib20) . Here, attention mechanisms (e.g. [BahdanauCB14,](#bib.bib21) ; [vaswani2017attention,](#bib.bib22) ) implicitly perform some form of relational reasoning; if previous hidden states are interpreted as entities, then computing a weighted sum of entities using attention helps to remove the locality bias present in vanilla RNNs, allowing embeddings to be better related using content rather than proximity. Since our current architectures solve complicated temporal tasks they must have some capacity for temporal relational reasoning. However, it is unclear whether their inductive biases are limiting, and whether these limitations can be exposed with tasks demanding particular types of temporal relational reasoning. For example, memory-augmented neural networks ([graves2014neural,](#bib.bib4) ; [graves2016hybrid,](#bib.bib5) ; [santoro2016meta,](#bib.bib6) ; [sukhbaatar2015end,](#bib.bib7) ) solve a compartmentalization problem with a slot-based memory matrix, but may have a harder time allowing memories to interact, or relate, with one another once they are encoded. LSTMs ([hochreiter1998lstm,](#bib.bib3) ; [Graves13,](#bib.bib23) ), on the other hand, pack all information into a common hidden memory vector, potentially making compartmentalization and relational reasoning more difficult. 3 Model -------- Our guiding design principle is to provide an architectural backbone upon which a model can learn to compartmentalize information, and learn to compute interactions between compartmentalized information. To accomplish this we assemble building blocks from LSTMs, memory-augmented neural networks, and non-local networks (in particular, the Transformer seq2seq model [vaswani2017attention](#bib.bib22) ). Similar to memory-augmented architectures we consider a fixed set of memory slots; however, we allow for interactions between memory slots using an attention mechanism. As we will describe, in contrast to previous work we apply attention between memories at a single time step, and not across all previous representations computed from all previous observations. ![Refer to caption](/html/1806.01822/assets/x1.png) Figure 1: Relational Memory Core. (a) The RMC receives a previous memory matrix and input vector as inputs, which are passed to the MHDPA module labeled with an “A”. (b). Linear projections are computed for each memory slot, and input vector, using row-wise shared weights Wqsuperscript𝑊𝑞W^{q}italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT for the queries, Wksuperscript𝑊𝑘W^{k}italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT for the keys, and Wvsuperscript𝑊𝑣W^{v}italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT for the values. (c) The queries, keys, and values are then compiled into matrices and softmax(QKT)Vsoftmax𝑄superscript𝐾𝑇𝑉\text{softmax}(QK^{T})Vsoftmax ( italic\_Q italic\_K start\_POSTSUPERSCRIPT italic\_T end\_POSTSUPERSCRIPT ) italic\_V is computed. The output of this computation is a new memory where information is blended across memories based on their attention weights. An MLP is applied row-wise to the output of the MHDPA module (a), and the resultant memory matrix is gated, and passed on as the core output or next memory state. ### 3.1 Allowing memories to interact using multi-head dot product attention We will first assume that we do not need to consider memory encoding; that is, that we already have some stored memories in matrix M𝑀Mitalic\_M, with row-wise compartmentalized memories misubscript𝑚𝑖m\_{i}italic\_m start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT. To allow memories to interact we employ multi-head dot product attention (MHDPA) ([vaswani2017attention,](#bib.bib22) ), also known as self-attention. Using MHDPA, each memory will attend over all of the other memories, and will update its content based on the attended information. First, a simple linear projection is used to construct queries (Q=MWq𝑄𝑀superscript𝑊𝑞Q=MW^{q}italic\_Q = italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT), keys (K=MWk𝐾𝑀superscript𝑊𝑘K=MW^{k}italic\_K = italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT), and values (V=MWv𝑉𝑀superscript𝑊𝑣V=MW^{v}italic\_V = italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT) for each memory (i.e. row misubscript𝑚𝑖m\_{i}italic\_m start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT) in matrix M𝑀Mitalic\_M. Next, we use the queries, Q𝑄Qitalic\_Q, to perform a scaled dot-product attention over the keys, K𝐾Kitalic\_K. The returned scalars can be put through a softmax-function to produce a set of weights, which can then be used to return a weighted average of values from V𝑉Vitalic\_V as A(Q,K,V)=softmax(QKTdk)V𝐴𝑄𝐾𝑉softmax𝑄superscript𝐾𝑇subscript𝑑𝑘𝑉A(Q,K,V)=\text{softmax}\left(\frac{QK^{T}}{\sqrt{d\_{k}}}\right)Vitalic\_A ( italic\_Q , italic\_K , italic\_V ) = softmax ( divide start\_ARG italic\_Q italic\_K start\_POSTSUPERSCRIPT italic\_T end\_POSTSUPERSCRIPT end\_ARG start\_ARG square-root start\_ARG italic\_d start\_POSTSUBSCRIPT italic\_k end\_POSTSUBSCRIPT end\_ARG end\_ARG ) italic\_V, where dksubscript𝑑𝑘d\_{k}italic\_d start\_POSTSUBSCRIPT italic\_k end\_POSTSUBSCRIPT is the dimensionality of the key vectors used as a scaling factor. Equivalently: \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| Aθ(M)=softmax(MWq(MWk)Tdk)MWv, where θ=(Wq,Wk,Wv)formulae-sequencesubscript𝐴𝜃𝑀softmax𝑀superscript𝑊𝑞superscript𝑀superscript𝑊𝑘𝑇subscript𝑑𝑘𝑀superscript𝑊𝑣 where 𝜃superscript𝑊𝑞superscript𝑊𝑘superscript𝑊𝑣\displaystyle A\_{\theta}(M)=\text{softmax}\left(\frac{MW^{q}(MW^{k})^{T}}{\sqrt{d\_{k}}}\right)MW^{v},\text{ where }\theta=(W^{q},W^{k},W^{v})italic\_A start\_POSTSUBSCRIPT italic\_θ end\_POSTSUBSCRIPT ( italic\_M ) = softmax ( divide start\_ARG italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT ( italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT ) start\_POSTSUPERSCRIPT italic\_T end\_POSTSUPERSCRIPT end\_ARG start\_ARG square-root start\_ARG italic\_d start\_POSTSUBSCRIPT italic\_k end\_POSTSUBSCRIPT end\_ARG end\_ARG ) italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT , where italic\_θ = ( italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT , italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT , italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT ) \| \| (1) \| The output of Aθ(M)subscript𝐴𝜃𝑀A\_{\theta}(M)italic\_A start\_POSTSUBSCRIPT italic\_θ end\_POSTSUBSCRIPT ( italic\_M ), which we will denote as M~~𝑀\widetilde{M}over~ start\_ARG italic\_M end\_ARG, is a matrix with the same dimensionality as M𝑀Mitalic\_M. M~~𝑀\widetilde{M}over~ start\_ARG italic\_M end\_ARG can be interpreted as a proposed update to M𝑀Mitalic\_M, with each m~isubscript~𝑚𝑖\widetilde{m}\_{i}over~ start\_ARG italic\_m end\_ARG start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT comprising information from memories mjsubscript𝑚𝑗m\_{j}italic\_m start\_POSTSUBSCRIPT italic\_j end\_POSTSUBSCRIPT. Thus, in one step of attention each memory is updated with information originating from other memories, and it is up to the model to learn (via parameters Wqsuperscript𝑊𝑞W^{q}italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT, Wksuperscript𝑊𝑘W^{k}italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT, and Wvsuperscript𝑊𝑣W^{v}italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT) how to shuttle information from memory to memory. As implied by the name, MHDPA uses multiple heads. We implement this producing hℎhitalic\_h sets of queries, keys, and values, using unique parameters to compute a linear projection from the original memory for each head hℎhitalic\_h. We then independently apply an attention operation for each head. For example, if M𝑀Mitalic\_M is an N×F𝑁𝐹N\times Fitalic\_N × italic\_F dimensional matrix and we employ two attention heads, then we compute M1~=Aθ(M)~superscript𝑀1subscript𝐴𝜃𝑀\widetilde{M^{1}}=A\_{\theta}(M)over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 1 end\_POSTSUPERSCRIPT end\_ARG = italic\_A start\_POSTSUBSCRIPT italic\_θ end\_POSTSUBSCRIPT ( italic\_M ) and M2~=Aϕ(M)~superscript𝑀2subscript𝐴italic-ϕ𝑀\widetilde{M^{2}}=A\_{\phi}(M)over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 2 end\_POSTSUPERSCRIPT end\_ARG = italic\_A start\_POSTSUBSCRIPT italic\_ϕ end\_POSTSUBSCRIPT ( italic\_M ), where M1~~superscript𝑀1\widetilde{M^{1}}over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 1 end\_POSTSUPERSCRIPT end\_ARG and M2~~superscript𝑀2\widetilde{M^{2}}over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 2 end\_POSTSUPERSCRIPT end\_ARG are N×F/2𝑁𝐹2N\times F/2italic\_N × italic\_F / 2 matrices, θ𝜃\thetaitalic\_θ and ϕitalic-ϕ\phiitalic\_ϕ denote unique parameters for the linear projections to produce the queries, keys, and values, and M~=[M1~:M2~]\widetilde{M}=[\widetilde{M^{1}}:\widetilde{M^{2}}]over~ start\_ARG italic\_M end\_ARG = [ over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 1 end\_POSTSUPERSCRIPT end\_ARG : over~ start\_ARG italic\_M start\_POSTSUPERSCRIPT 2 end\_POSTSUPERSCRIPT end\_ARG ], where [:]delimited-[]:[:][ : ] denotes column-wise concatenation. Intuitively, heads could be useful for letting a memory share different information, to different targets, using each head. ### 3.2 Encoding new memories We assumed that we already had a matrix of memories M𝑀Mitalic\_M. Of course, memories instead need to be encoded as new inputs are received. Suppose then that M𝑀Mitalic\_M is some randomly initialised memory. We can efficiently incorporate new information x𝑥xitalic\_x into M𝑀Mitalic\_M with a simple modification to equation [1](#S3.E1 "1 ‣ 3.1 Allowing memories to interact using multi-head dot product attention ‣ 3 Model ‣ Relational recurrent neural networks"): \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| M~=softmax(MWq([M;x]Wk)Tdk)[M;x]Wv,~𝑀softmax𝑀superscript𝑊𝑞superscript𝑀𝑥 superscript𝑊𝑘𝑇superscript𝑑𝑘𝑀𝑥 superscript𝑊𝑣\displaystyle\widetilde{M}=\text{softmax}\left(\frac{MW^{q}([M;x]W^{k})^{T}}{\sqrt{d^{k}}}\right)[M;x]W^{v},over~ start\_ARG italic\_M end\_ARG = softmax ( divide start\_ARG italic\_M italic\_W start\_POSTSUPERSCRIPT italic\_q end\_POSTSUPERSCRIPT ( [ italic\_M ; italic\_x ] italic\_W start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT ) start\_POSTSUPERSCRIPT italic\_T end\_POSTSUPERSCRIPT end\_ARG start\_ARG square-root start\_ARG italic\_d start\_POSTSUPERSCRIPT italic\_k end\_POSTSUPERSCRIPT end\_ARG end\_ARG ) [ italic\_M ; italic\_x ] italic\_W start\_POSTSUPERSCRIPT italic\_v end\_POSTSUPERSCRIPT , \| \| (2) \| where we use [M;x]𝑀𝑥[M;x][ italic\_M ; italic\_x ] to denote the row-wise concatenation of M𝑀Mitalic\_M and x𝑥xitalic\_x. Since we use [M;x]𝑀𝑥[M;x][ italic\_M ; italic\_x ] when computing the keys and values, and only M𝑀Mitalic\_M when computing the queries, M~~𝑀\widetilde{M}over~ start\_ARG italic\_M end\_ARG is a matrix with same dimensionality as M𝑀Mitalic\_M. Thus, equation [2](#S3.E2 "2 ‣ 3.2 Encoding new memories ‣ 3 Model ‣ Relational recurrent neural networks") is a memory-size preserving attention operation that includes attention over the memories and the new observations. Notably, we use the same attention operation to efficiently compute memory interactions and to incorporate new information. We also note the possible utility of this operation when the memory consists of a single vector rather than a matrix. In this case the model may learn to pick and choose which information from the input should be written into the vector memory state by learning how to attend to the input, conditioned on what is contained in the memory already. This is possible in LSTMs via the gates, though at a different granularity. We return to this idea, and the possible compartmentalization that can occur via the heads even in the single-memory-slot case, in the discussion. ### 3.3 Introducing recurrence and embedding into an LSTM Suppose we have a temporal dimension with new observations at each timestep, xtsubscript𝑥𝑡x\_{t}italic\_x start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT. Since M𝑀Mitalic\_M and M~~𝑀\widetilde{M}over~ start\_ARG italic\_M end\_ARG are the same dimensionality, we can naively introduce recurrence by first randomly initialising M𝑀Mitalic\_M, and then updating it with M~~𝑀\widetilde{M}over~ start\_ARG italic\_M end\_ARG at each timestep. We chose to do this by embedding this update into an LSTM. Suppose memory matrix M𝑀Mitalic\_M can be interpreted as a matrix of cell states, usually denoted as C𝐶Citalic\_C, for a 2-dimensional LSTM. We can make the operations of individual memories misubscript𝑚𝑖m\_{i}italic\_m start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT nearly identical to those in a normal LSTM cell state as follows (subscripts are overloaded to denote the row from a matrix, and timestep; e.g., mi,tsubscript𝑚𝑖𝑡m\_{i,t}italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT is the ithsuperscript𝑖𝑡ℎi^{th}italic\_i start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT row from M𝑀Mitalic\_M at time t𝑡titalic\_t). \| \| \| \| \| \| \| --- \| --- \| --- \| --- \| --- \| \| \| si,tsubscript𝑠𝑖𝑡\displaystyle s\_{i,t}italic\_s start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =(hi,t−1,mi,t−1)absentsubscriptℎ𝑖𝑡1subscript𝑚𝑖𝑡1\displaystyle=(h\_{i,t-1},m\_{i,t-1})= ( italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT , italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT ) \| \| (3) \| \| \| fi,tsubscript𝑓𝑖𝑡\displaystyle f\_{i,t}italic\_f start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =Wfxt+Ufhi,t−1+bfabsentsuperscript𝑊𝑓subscript𝑥𝑡superscript𝑈𝑓subscriptℎ𝑖𝑡1superscript𝑏𝑓\displaystyle=W^{f}x\_{t}+U^{f}h\_{i,t-1}+b^{f}= italic\_W start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT italic\_x start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT + italic\_U start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT + italic\_b start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT \| \| (4) \| \| \| ii,tsubscript𝑖𝑖𝑡\displaystyle i\_{i,t}italic\_i start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =Wixt+Uihi,t−1+biabsentsuperscript𝑊𝑖subscript𝑥𝑡superscript𝑈𝑖subscriptℎ𝑖𝑡1superscript𝑏𝑖\displaystyle=W^{i}x\_{t}+U^{i}h\_{i,t-1}+b^{i}= italic\_W start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT italic\_x start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT + italic\_U start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT + italic\_b start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT \| \| (5) \| \| \| oi,tsubscript𝑜𝑖𝑡\displaystyle o\_{i,t}italic\_o start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =Woxt+Uohi,t−1+boabsentsuperscript𝑊𝑜subscript𝑥𝑡superscript𝑈𝑜subscriptℎ𝑖𝑡1superscript𝑏𝑜\displaystyle=W^{o}x\_{t}+U^{o}h\_{i,t-1}+b^{o}= italic\_W start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT italic\_x start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT + italic\_U start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT + italic\_b start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT \| \| (6) \| \| \| mi,tsubscript𝑚𝑖𝑡\displaystyle m\_{i,t}italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =σ(fi,t+b~f)∘mi,t−1+σ(ii,t)∘gψ(m~i,t)⏟absent𝜎subscript𝑓𝑖𝑡superscript~𝑏𝑓subscript𝑚𝑖𝑡1𝜎subscript𝑖𝑖𝑡⏟subscript𝑔𝜓subscript~𝑚𝑖𝑡\displaystyle=\sigma(f\_{i,t}+\tilde{b}^{f})\circ m\_{i,t-1}+\sigma(i\_{i,t})\circ\underbrace{{g\_{\psi}(\widetilde{m}\_{i,t})}}= italic\_σ ( italic\_f start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT + over~ start\_ARG italic\_b end\_ARG start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT ) ∘ italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t - 1 end\_POSTSUBSCRIPT + italic\_σ ( italic\_i start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT ) ∘ under⏟ start\_ARG italic\_g start\_POSTSUBSCRIPT italic\_ψ end\_POSTSUBSCRIPT ( over~ start\_ARG italic\_m end\_ARG start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT ) end\_ARG \| \| (7) \| \| \| hi,tsubscriptℎ𝑖𝑡\displaystyle h\_{i,t}italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT \| =σ(oi,t)∘tanh(mi,t)absent𝜎subscript𝑜𝑖𝑡tanhsubscript𝑚𝑖𝑡\displaystyle=\sigma(o\_{i,t})\circ\text{tanh}(m\_{i,t})= italic\_σ ( italic\_o start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT ) ∘ tanh ( italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT ) \| \| (8) \| \| \| si,t+1subscript𝑠𝑖𝑡1\displaystyle s\_{i,t+1}italic\_s start\_POSTSUBSCRIPT italic\_i , italic\_t + 1 end\_POSTSUBSCRIPT \| =(mi,t,hi,t)absentsubscript𝑚𝑖𝑡subscriptℎ𝑖𝑡\displaystyle=(m\_{i,t},h\_{i,t})= ( italic\_m start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT , italic\_h start\_POSTSUBSCRIPT italic\_i , italic\_t end\_POSTSUBSCRIPT ) \| \| (9) \| The underbrace denotes the modification to a standard LSTM. In practice we did not find output gates necessary – please see the url in the footnote for our Tensorflow implementation of this model in the Sonnet library 222<https://github.com/deepmind/sonnet/blob/master/sonnet/python/modules/relational_memory.py>, and for the exact formulation we used, including our choice for the gψsubscript𝑔𝜓g\_{\psi}italic\_g start\_POSTSUBSCRIPT italic\_ψ end\_POSTSUBSCRIPT function (briefly, we found a row/memory-wise MLP with layer normalisation to work best). There is also an interesting opportunity to introduce a different kind of gating, which we call ‘memory’ gating, which resembles previous gating ideas [gers1999learning](#bib.bib24) ; [hochreiter1998lstm](#bib.bib3) . Instead of producing scalar gates for each individual unit (‘unit’ gating), we can produce scalar gates for each memory row by converting Wfsuperscript𝑊𝑓W^{f}italic\_W start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT, Wisuperscript𝑊𝑖W^{i}italic\_W start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT, Wosuperscript𝑊𝑜W^{o}italic\_W start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT, Ufsuperscript𝑈𝑓U^{f}italic\_U start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT, Uisuperscript𝑈𝑖U^{i}italic\_U start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT, and Uosuperscript𝑈𝑜U^{o}italic\_U start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT from weight matrices into weight vectors, and by replacing the element-wise product in the gating equations with scalar-vector multiplication. Since parameters Wfsuperscript𝑊𝑓W^{f}italic\_W start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT, Wisuperscript𝑊𝑖W^{i}italic\_W start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT, Wosuperscript𝑊𝑜W^{o}italic\_W start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT, Ufsuperscript𝑈𝑓U^{f}italic\_U start\_POSTSUPERSCRIPT italic\_f end\_POSTSUPERSCRIPT, Uisuperscript𝑈𝑖U^{i}italic\_U start\_POSTSUPERSCRIPT italic\_i end\_POSTSUPERSCRIPT, Uosuperscript𝑈𝑜U^{o}italic\_U start\_POSTSUPERSCRIPT italic\_o end\_POSTSUPERSCRIPT, and ψ𝜓\psiitalic\_ψ are shared for each misubscript𝑚𝑖m\_{i}italic\_m start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT, we can modify the number of memories without affecting the number of parameters. Thus, tuning the number of memories and the size of each memory can be used to balance the overall storage capacity (equal to the total number of units, or elements, in M𝑀Mitalic\_M) and the number of parameters (proportional to the dimensionality of misubscript𝑚𝑖m\_{i}italic\_m start\_POSTSUBSCRIPT italic\_i end\_POSTSUBSCRIPT). We find in our experiments that some tasks require more, but not necessarily larger, memories, and others such as language modeling require fewer, larger memories. Thus, we have a number of tune-able parameters: the number of memories, the size of each memory, the number of attention heads, the number of steps of attention, the gating method, and the post-attention processor gψsubscript𝑔𝜓g\_{\psi}italic\_g start\_POSTSUBSCRIPT italic\_ψ end\_POSTSUBSCRIPT. In the appendix we list the exact configurations for each task. 4 Experiments -------------- Here we briefly outline the tasks on which we applied the RMC, and direct the reader to the appendix for full details on each task and details on hyperparameter settings for the model. ![Refer to caption](/html/1806.01822/assets/x2.png) Figure 2: Tasks. We tested the RMC on a suite of supervised and reinforcement learning tasks. Notable are the Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT Farthest toy task and language modeling. In the former, the solution requires explicit relational reasoning since the model must sort distance relations between vectors, and not the vectors themselves. The latter tests the model on a large quantity of natural data and allows us to compare performance to well-tuned models. ### 4.1 Illustrative supervised tasks ##### Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT Farthest The Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT Farthest task is designed to stress a capacity for relational reasoning across time. Inputs are a sequence of randomly sampled vectors, and targets are answers to a question of the form: “What is the nthsuperscript𝑛𝑡ℎn^{th}italic\_n start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT farthest vector (in Euclidean distance) from vector m𝑚mitalic\_m?”, where the vector values, their IDs, n𝑛nitalic\_n, and m𝑚mitalic\_m are randomly sampled per sequence. It is not enough to simply encode and retrieve information as in a copy task. Instead, a model must compute all pairwise distance relations to the reference vector m𝑚mitalic\_m, which might also lie in memory, or might not have even been provided as input yet. It must then implicitly sort these distances to produce the answer. We emphasize that the model must sort distance relations between vectors, and not the vectors themselves. ##### Program Evaluation The Learning to Execute (LTE) dataset [zaremba2014lte](#bib.bib25) consists of algorithmic snippets from a Turing complete programming language of pseudo-code, and is broken down into three categories: addition, control, and full program. Inputs are a sequence of characters over an alphanumeric vocabulary representing such snippets, and the target is a numeric sequence of characters that is the execution output for the given programmatic input. Given that the snippets involve symbolic manipulation of variables, we felt it could strain a model’s capacity for relational reasoning; since symbolic operators can be interpreted as defining a relation over the operands, successful learning could reflect an understanding of this relation. To also assess model performance on classical sequence tasks we also evaluated on memorization tasks, in which the output is simply a permuted form of the input rather than an evaluation from a set of operational instructions. See the appendix for further experimental details. ### 4.2 Reinforcement learning ##### Mini Pacman with viewport We follow the formulation of Mini Pacman from [weber2017imagination](#bib.bib26) . Briefly, the agent navigates a maze to collect food while being chased by ghosts. However, we implement this task with a viewport: a 5×5555\times 55 × 5 window surrounding the agent that comprises the perceptual input. The task is therefore partially observable, since the agent must navigate the space and take in information through this viewport. Thus, the agent must predict the dynamics of the ghosts in memory, and plan its navigation accordingly, also based on remembered information about which food has already been picked up. We also point the reader to the appendix for a description and results of another RL task called BoxWorld, which demands relational reasoning in memory space. ### 4.3 Language Modeling Finally, we investigate the task of word-based language modeling. We model the conditional probability p(wt\|w<t)𝑝conditionalsubscript𝑤𝑡subscript𝑤absent𝑡p(w\_{t}\|w\_{<t})italic\_p ( italic\_w start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT \| italic\_w start\_POSTSUBSCRIPT < italic\_t end\_POSTSUBSCRIPT ) of a word wtsubscript𝑤𝑡w\_{t}italic\_w start\_POSTSUBSCRIPT italic\_t end\_POSTSUBSCRIPT given a sequence of observed words w<t=(wt−1,wt−2,…,w1)subscript𝑤absent𝑡subscript𝑤𝑡1subscript𝑤𝑡2…subscript𝑤1w\_{<t}=\left(w\_{t-1},w\_{t-2},\ldots,w\_{1}\right)italic\_w start\_POSTSUBSCRIPT < italic\_t end\_POSTSUBSCRIPT = ( italic\_w start\_POSTSUBSCRIPT italic\_t - 1 end\_POSTSUBSCRIPT , italic\_w start\_POSTSUBSCRIPT italic\_t - 2 end\_POSTSUBSCRIPT , … , italic\_w start\_POSTSUBSCRIPT 1 end\_POSTSUBSCRIPT ). Language models can be directly applied to predictive keyboard and search-phrase completion, or they can be used as components within larger systems, e.g. machine translation [cho2014learning](#bib.bib27) , speech recognition [bahdanau2016end](#bib.bib28) , and information retrieval [hiemstra2001using](#bib.bib29) . RNNs, and most notably LSTMs, have proven to be state-of-the-art on many competitive language modeling benchmarks such as Penn Treebank [yang2017breaking](#bib.bib30) ; [marcus1993building](#bib.bib31) , WikiText-103 [rae2018fast](#bib.bib32) ; [merity2016pointer](#bib.bib33) , and the One Billion Word Benchmark [jozefowicz2016exploring](#bib.bib34) ; [chelba2013one](#bib.bib35) . As a sequential reasoning task, language modeling allows us to assess the RMC’s ability to process information over time on a large quantity of natural data, and compare it to well-tuned models. We focus on datasets with contiguous sentences and a moderately large amount of data. WikiText-103 satisfies this set of requirements as it consists of Wikipedia articles shuffled at the article level with roughly 100M100𝑀100M100 italic\_M training tokens, as do two stylistically different sources of text data: books from Project Gutenberg333Project Gutenberg. (n.d.). Retrieved January 2, 2018, from www.gutenberg.org and news articles from GigaWord v5 [parker2011english](#bib.bib36) . Using the same processing from [rae2018fast](#bib.bib32) these datasets consist of 180M180𝑀180M180 italic\_M training tokens and 4B4𝐵4B4 italic\_B training tokens respectively, thus they cover a range of styles and corpus sizes. We choose a similar vocabulary size for all three datasets of approximately 250,000250000250,000250 , 000, which is large enough to include rare words and numeric values. 5 Results ---------- ### 5.1 Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT Farthest This task revealed a stark difference between our LSTM and DNC baselines and RMC when training on 16161616-dimensional vector inputs. Both LSTM and DNC models failing to surpass 30%percent3030\%30 % best batch accuracy and the RMC consistently achieving 91%percent9191\%91 % at the end of training (see figure [5](#A1.F5 "Figure 5 ‣ A.1 𝑁^{𝑡⁢ℎ} Farthest ‣ Appendix A Further task details, analyses, and model configurations ‣ Relational recurrent neural networks") in the appendix for training curves). The RMC achieved similar performance when the difficulty of the task was increased by using 32323232-dimensional vectors, placing a greater demand on high-fidelity memory storage. However, this performance was less robust with only a small number of seeds/model configurations demonstrating this performance, in contrast to the 16161616-dimensional vector case where most model configurations succeeded. An attention analysis revealed some notable features of the RMC’s internal functions. Figure [3](#S5.F3 "Figure 3 ‣ 5.1 𝑁^{𝑡⁢ℎ} Farthest ‣ 5 Results ‣ Relational recurrent neural networks") shows attention weights in the RMC’s memory throughout a sequence: the first row contains a sequence where the reference vector m𝑚mitalic\_m was observed last; in the second row it was observed first; and in the last row it was observed in the middle of the sequence. Before m𝑚mitalic\_m is seen the model seems to shuttle input information into one or two memory slots, as shown by the high attention weights from these slots’ queries to the input key. After m𝑚mitalic\_m is seen, most evident in row three of the figure, the model tends to change its attention behaviour, with all the memory slots preferentially focusing attention on those particular memories to which the m𝑚mitalic\_m was written. Although this attention analysis provides some useful insights, the conclusions we can make are limited since even after a single round of attention the memory can become highly distributed, making any interpretations about information compartmentalisation potentially inaccurate. ![Refer to caption](/html/1806.01822/assets/x3.png) Figure 3: Model analysis. Each row depicts the attention matrix at each timestep of a particular sequence. The text beneath spells out the particular task for the sequence, which was encoded and provided to the model as an input. We mark in red the vector that is referenced in the task: e.g., if the model is to choose the 2ndsuperscript2𝑛𝑑2^{nd}2 start\_POSTSUPERSCRIPT italic\_n italic\_d end\_POSTSUPERSCRIPT farthest vector from vector 7777, then the time point at which vector 7777 was input to the model is depicted in red. A single attention matrix shows the attention weights from one particular memory slot (y-axis) to another memory slot (columns), or the input (offset column), with the numbers denoting the memory slot and “input” denoting the input embedding. ### 5.2 Program Evaluation Program evaluation performance was assessed via the Learning to Execute tasks [zaremba2014lte](#bib.bib25) . We evaluated a number of baselines alongside the RMC including an LSTM [hochreiter1998lstm](#bib.bib3) ; [pascanu2013dlstm](#bib.bib37) , DNC [graves2016hybrid](#bib.bib5) , and a bank of LSTMs resembling Recurrent Entity Networks [henaff2017entnet](#bib.bib38) (EntNet) - the configurations for each of these is described in the appendix. Best test batch accuracy results are shown in Table [1](#S5.T1 "Table 1 ‣ 5.2 Program Evaluation ‣ 5 Results ‣ Relational recurrent neural networks"). The RMC performs at least as well as all of the baselines on each task. It is marginally surpassed by a small fraction of performance on the double memorization task, but both models effectively solve this task. Further, the results of the RMC outperform all equivalent tasks from [zaremba2014lte](#bib.bib25) which use teacher forcing even when evaluating model performance. It’s worth noting that we observed better results when we trained in a non-auto-regressive fashion - that is, with no teacher forcing during training. This is likely related to the effect that relaxing the ground truth requirement has on improving model generalization [sbengio2015schedsamp](#bib.bib39) and hence, performance. It is perhaps more pronounced in these tasks due to the independence of output token probabilities and also the sharply uni-modal nature of the output distribution (that is, there is no ambiguity in the answer given the program). Table 1: Test per character Accuracy on Program Evaluation and Memorization tasks. \| Model \| Add \| Control \| Program \| Copy \| Reverse \| Double \| \| --- \| --- \| --- \| --- \| --- \| --- \| --- \| \| LSTM [hochreiter1998lstm](#bib.bib3) ; [pascanu2013dlstm](#bib.bib37) \| 99.8 \| 97.4 \| 66.1 \| 99.8 \| 99.7 \| 99.7 \| \| EntNet [henaff2017entnet](#bib.bib38) \| 98.4 \| 98.0 \| 73.4 \| 91.8 \| 100.0 \| 62.3 \| \| DNC [graves2016hybrid](#bib.bib5) \| 99.4 \| 83.8 \| 69.5 \| 100.0 \| 100.0 \| 100.0 \| \| Relational Memory Core \| 99.9 \| 99.6 \| 79.0 \| 100.0 \| 100.0 \| 99.8 \| Table 2: Validation and test perplexities on WikiText-103, Project Gutenberg, and GigaWord v5. \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| WikiText-103 \| Gutenberg \| GigaWord \| \| \| Valid. \| Test \| Valid \| Test \| Test \| \| LSTM [grave2016improving](#bib.bib40) \| - \| 48.7 \| - \| - \| - \| \| Temporal CNN [bai2018convolutional](#bib.bib41) \| - \| 45.2 \| - \| - \| - \| \| Gated CNN [dauphin2016language](#bib.bib42) \| - \| 37.2 \| - \| - \| - \| \| LSTM [rae2018fast](#bib.bib32) \| 34.1 \| 34.3 \| 41.8 \| 45.5 \| 43.7 \| \| Quasi-RNN [merity2018scalable](#bib.bib43) \| 32 \| 33 \| - \| - \| - \| \| Relational Memory Core \| 30.8 \| 31.6 \| 39.2 \| 42.0 \| 38.3 \| ### 5.3 Mini-Pacman In Mini Pacman with viewport the RMC achieved approximately 100100100100 points more than an LSTM (677677677677 vs. 550550550550), and when trained with the full observation the RMC nearly doubled the performance of an LSTM (1159115911591159 vs. 598598598598, figure [10](#A1.F10 "Figure 10 ‣ A.3.1 Results ‣ A.3 Viewport BoxWorld ‣ Appendix A Further task details, analyses, and model configurations ‣ Relational recurrent neural networks")). ### 5.4 Language Modeling For all three language modeling tasks we observe lower perplexity when using the relational memory core, with a drop of 1.4−5.41.45.41.4-5.41.4 - 5.4 perplexity over the best published results. Although small, this constitutes a 5−125125-125 - 12% relative improvement and appears to be consistent across tasks of varying size and style. For WikiText-103, we see this can be compared to LSTM architectures ([graves2016hybrid,](#bib.bib5) ; [rae2018fast,](#bib.bib32) ), convolutional models ([dauphin2016language,](#bib.bib42) ) and hybrid recurrent-convolutional models ([merity2018scalable,](#bib.bib43) ). The model learns with a slightly better data efficiency than an LSTM (appendix figure [11](#A1.F11 "Figure 11 ‣ A.4 Language Modeling ‣ Appendix A Further task details, analyses, and model configurations ‣ Relational recurrent neural networks")). The RMC scored highly when the number of context words provided during evaluation were relatively few, compared to an LSTM which profited much more from a larger context (supplementary figure [12](#A1.F12 "Figure 12 ‣ A.4 Language Modeling ‣ Appendix A Further task details, analyses, and model configurations ‣ Relational recurrent neural networks")). This could be because RMC better captures short-term relations, and hence only needs a relatively small context for accurate modeling. Inspecting the perplexity broken down by word frequency in supplementary table [3](#A1.T3 "Table 3 ‣ A.4 Language Modeling ‣ Appendix A Further task details, analyses, and model configurations ‣ Relational recurrent neural networks"), we see the RMC improved the modeling of frequent words, and this is where the drop in overall perplexity is obtained. 6 Discussion ------------- A number of other approaches have shown success in modeling sequential information by using a growing buffer of previous states ([BahdanauCB14,](#bib.bib21) ; [vaswani2017attention,](#bib.bib22) ). These models better capture long-distance interactions, since their computations are not biased by temporally local proximity. However, there are serious scaling issues for these models when the number of timesteps is large, or even unbounded, such as in online reinforcement learning (e.g., in the real world). Thus, some decisions need to be made regarding the size of the past-embedding buffer that should be stored, whether it should be a rolling window, how computations should be cached and propagated across time, etc. These considerations make it difficult to directly compare these approaches in these online settings. Nonetheless, we believe that a blend of purely recurrent approaches with those that scale with time could be a fruitful pursuit: perhaps the model accumulates memories losslessly for some chunk of time, then learns to compress it in a recurrent core before moving onto processing a subsequent chunk. We proposed intuitions for the mechanisms that may better equip a model for complex relational reasoning. Namely, by explicitly allowing memories to interact either with each other, with the input, or both via MHDPA, we demonstrated improved performance on tasks demanding relational reasoning across time. We would like to emphasize, however, that while these intuitions guided our design of the model, and while the analysis of the model in the Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT farthest task aligned with our intuitions, we cannot necessarily make any concrete claims as to the causal influence of our design choices on the model’s capacity for relational reasoning, or as to the computations taking place within the model and how they may map to traditional approaches for thinking about relational reasoning. Thus, we consider our results primarily as evidence of improved function – if a model can better solve tasks that require relational reasoning, then it must have an increased capacity for relational reasoning, even if we do not precisely know why it may have this increased capacity. In this light the RMC may be usefully viewed from multiple vantages, and these vantages may offer ideas for further improvements. Our model has multiple mechanisms for forming and allowing for interactions between memory vectors: slicing the memory matrix row-wise into slots, and column-wise into heads. Each has its own advantages (computations on slots share parameters, while having more heads and a larger memory size takes advantage of more parameters). We don’t yet understand the interplay, but we note some empirical findings. First, in the the Nthsuperscript𝑁𝑡ℎN^{th}italic\_N start\_POSTSUPERSCRIPT italic\_t italic\_h end\_POSTSUPERSCRIPT farthest task a model with a single memory slot performed better when it had more attention heads, though in all cases it performed worse than a model with many memory slots. Second, in language modeling, our model used a single memory slot. The reasons for choosing a single memory here were mainly due to the need for a large number of parameters for LM in general (hence the large size for the single memory slot), and the inability to quickly run a model with both a large number of parameters and multiple memory slots. Thus, we do not necessarily claim that a single memory slot is best for language modeling, rather, we emphasize an interesting trade-off between number of memories and individual memory size, which may be a task specific ratio that can be tuned. Moreover, in program evaluation, an intermediate solution worked well across subtasks (4444 slots and heads), though some performed best with 1111 memory, and others with 8888. Altogether, our results show that explicit modeling of memory interactions improves performance in a reinforcement learning task, alongside program evaluation, comparative reasoning, and language modeling, demonstrating the value of instilling a capacity for relational reasoning in recurrent neural networks. Acknowledgements ---------------- We thank Caglar Gulcehre, Matt Botvinick, Vinicius Zambaldi, Charles Blundell, Sébastien Racaniere, Chloe Hillier, Victoria Langston, and many others on the DeepMind team for critical feedback, discussions, and support.
9c99d55e-82f8-477e-8ac2-d412c256744a	trentmkelly/LessWrong-43k	LessWrong	[SEQ RERUN] The Apocalypse Bet Today's post, The Apocalypse Bet, was originally published on 09 August 2007. A summary (taken from the LW wiki): > If you think that the apocalypse will be in 2020, while I think that it will be in 2030, how could we bet on this? One way would be for me to pay you X dollars every year until 2020. Then, if the apocalypse doesn't happen, you pay me 2X dollars every year until 2030. This idea could be used to set up a prediction market, which could give society information about when an apocalypse might happen. Discuss the post here (rather than in the comments to the original post). This post is part of the Rerunning the Sequences series, in which we're going through Eliezer Yudkowsky's old posts in order, so that people who are interested can (re-)read and discuss them. The previous post was You Can Face Reality, and you can use the sequence_reruns tag or rss feed to follow the rest of the series. Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day's sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.
08545fe3-9efb-4fe6-8973-ae778d23cd8b	StampyAI/alignment-research-dataset/youtube	Youtube Transcripts	Value Alignment \| Stuart Russell so the first question is what is a good decision anyway an economist will tell you that it means maximizing your expected over the whole future utility and this applies to everything from lottery tickets to Davos meetings to building radio telescopes on maximizing AI has made a great deal of progress eighteen years ago deep blue beat Garry Kasparov at chess just last week the game of poker was solved perfectly laughing and all and humans could no longer compete and right now the deep mine system is playing 29 different video games superhumanly well that it learned entirely from scratch just by watching the screen imagine if a newborn baby did that on expectations these depend on perception and learning again a huge amount of progress the Watson system extracting information from text cars watching the world as they go by learning algorithms that classify images and write descriptions even a system that discovers the concept of a cat entirely for itself just by looking at millions of images of everything under the Sun now a lot of this progress comes from mathematical ideas here are just a few of the equations from my undergraduate course and there will be a test if Linder allows it until later on there's also a lot of progress that comes from commercial investment so every one of these areas are 1% improvement is worth billions of dollars so we may see in the future domestic robots for example search engines that read and understand every page on the web even a machine that will discover the missing sock perhaps in the look the very distant future so the point of AI is that everything civilization has to offer is the product of our intelligence so if we can amplify that then there is no limit to where the human race can go but actually want to point to a problem and that comes in the utility part of the equation so imagine for example that you asked your robot to maybe make yourself some paper clips that you might need and your robots very very clever it takes you very literally and pretty soon the entire the six feet deep in paper clips so this is the Sorcerer's Apprentice and King Midas all rolled into one now technically what happens is that if you ask a machine to optimize and you leave out part of your preferences the machine will set those elements to an extreme value for example if you say Google car quick take me to Zurich Airport it will max out the speedometer and they say oh I didn't mean break the speed limit well it'll still put its foot on the gas and then when it gets the airport slam on the brakes so this is the problem of value alignment and if you combine misalignment of values with a super-intelligent machine that's very capable then you have a really serious problem for the human race so the point is that machines can and will make better decisions than humans but only if their values are aligned with those of the human race now my colleagues my distinguished colleagues may argue that super intelligent AI will never happen let me take you back to September 11th 1933 Lord Rutherford the world's leading nuclear physicist said that atomic energy was moonshine could never happen the next morning Leo Szilard invented the nuclear chain reaction the next morning so we have to be careful let's look at nuclear fusion in particular long ago they invented a method of generating unlimited amounts of energy long ago it's called the hydrogen bomb so now fusion concentrates on containment and AI has to do the same thing if you want unlimited intelligence you have to solve value alignment so one way of doing this is called inverse reinforcement learning what that means is for example a machine sees somebody making coffee in the morning and then figures out the purpose the underlying utility function that explains this behavior namely that having coffee is a good idea as long as it's not too much of it now it's not quite as simple as that as I'm sure you all see humans differ in their values cultures differ in their values none of us behaves perfectly but there is a huge amount of information that the machine can access about human actions every television program every book every novel every movie every newspaper article is about human actions and in particular about our attitudes those actions so the rational thing for a machine to do is to engage in an extended conversation with a human race about its values before it can take any action that affects the real world so my claim is that in the future we will be able to design super intelligent machines that do exactly what they're supposed to do which is to support greater realization of human values and I think this is maybe the most important conversation that we can have over the next 50 years thank you you
0eb69a3e-0502-4127-8ce3-437a332836fb	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Did AI pioneers not worry much about AI risks? It's seems noteworthy just how little the AI pioneers from the 40s-80s seemed to care about AI risk. There is no obvious reason why a book like "Superintelligence" wasn't written in the 1950s, but for some reason that didn't happen... any thoughts on why this was the case? I can think of three possible reasons for this: 1. They actually DID care and published extensively about AI risk, but I'm simply not well enough schooled on the history of AI research. 2. Deep down, people involved in early AI research knew that they were still a long long way from achieving significantly powerful AI, despite the optimistic public proclamations that were made at that time. 3. AI risks are highly counter-intuitive and it simply required another 60 years of thinking to understand. Anyone have any thoughts on this question?
6ea6ec3c-7df7-494e-9685-8c8d587cd042	trentmkelly/LessWrong-43k	LessWrong	London meetup, Sunday 2011-08-21 14:00, near Holborn We're meeting up in London tomorrow. Sunday 21st August, at 2pm, in the Shakespeares Head (official page) on Kingsway near Holborn Tube station. See you there!
59361f13-42f4-4368-b1d8-4a5fb25def0f	StampyAI/alignment-research-dataset/youtube	Youtube Transcripts	243. A General Language Assistant as a Laboratory for Alignment hello and welcome to session 243 in the ai safety.com reading group tonight we'll be discussing the article a general language assistant as a laboratory for alignment by jared kaplan and many others this is the first work done by anthropic new alignment research institute and these are the primary authors with jared kaplan as the corresponding and probably primary author and there are a number of people who are primary models would have helped in some other ways it's a long list and actually i have heard from these people i know amanda esco and kevin of course and i do actually know quite a few more of the non-core people who have been helping and it's um we still haven't actually heard but what they are doing um so i have at this point i just assume they are building some kind of agi that is trying to kill us all or something like that so this is uh from uh uh two months ago and we are focusing on the non-typical parts so we're looking at the philosophical motivations rather than the explicit machine learning things and the evaluation criteria where i deliberately i haven't read really the parts about how they're in fact implementing this because the the people working in anthropic have really great credentials they are known to be really good at this and i care somewhat less about evaluating how good they are i know they're good and i care more about how do they look at the world um kowski had some comments on his twitter on this paper where he was uh moderately positive and that they directly challenge his alignment it doesn't bring the capability and it doesn't overstate the accomplishment i want to uh anything that sounds really like standing with fake praise but that's just his style and i don't think he intends it as much as um as faint praise and i want to dig deeper into these parts like okay they're directly challenging alignment but are they challenging alignment using a strategy that might work or is this just a hopeless strategy um are they burning capability commons i think uh like are they actually um that's the old somewhat outdated model with two progress bars with ai capability and ai alignment and are they working too much on capability um not lying and saying they'll solve the problem that really really sounds like faint praise to me but in fact there are a number of quotes you can pull out from this that seem to be quite modest and i it's sad that it's relevant that just to say that this is a good thing but i mean it is a good thing so when i investigate the motivations one of the things that i look for are like tiny things in the introduction where they say that future ai systems might do bad things and interact in presently unforeseeable ways and that's of course the thing that can happen but i care about the presently foreseeable ways that things can go wrong so they have a definition of alignment that is somewhat non-standard define the alignment between two agents as to what extent their preferences overlap and it's not really a definition they use very much uh they almost immediately transition to a uh a more workable definition and it should be mentioned here that almost perfect uh overlap in uh how you rank different outcomes could uh be arbitrarily bad quite easily and in addition to uh to this the one of the uh overworking ideas is to look at language models directly instead of looking at perhaps more advanced models and paradigms because language ones have a large number of advantages you can try a lot of different kinds of inputs and they can fail in many potentially quite interesting ways and benchmark how much progress has been made you can compare different alignment techniques to try to get some kind of idea about where we are in in more specific details you can try to see whether prompting can work as alignment and to what to attempt and uh you can see if language models can model preferences instead of just doing invitation learning and they focus a lot on improving sample efficiency uh of uh preference marlin and these are some very uh nice and interesting goals so the way they actually uh the definition of alignment that they're actually using is helpful honest and harmless and the the way they justify this is with the following quote it is difficult for an ai assistant to always be helpful honest and harmless towards an agent without also being highly aligned with that agent and i'm not entirely sure i agree with this because the word always is very important in order to make this work because the techniques that they're using are black box methods and blackboard methods will not give us any kind of proof we'll now be able to say this model is always honest if we are only looking at it from the outside um and if you try to remove the word harmless the word always and just most of the time then from this definition it becomes very clear that it is indeed very possible to be very often helpful and honest and harmless but not always harmless right then you get into things like deceptive alignment very very quickly there are advantages of this this is something that is much more actionable is understandable memorable um and it seems like on on a more uh precise view of uh alignment but this is indeed a big part of what we actually want and of course the language ones that we currently have are uh especially when the like gt3 was uh released it was very clear that it was not helpful it was not honest and was not harmless so it is something that is indeed substantial for us these criterias are a lot less than mathematically well-defined right there are uh trade-offs and ambiguity and the way these are resolved suggested by uh by the authors is that um the person who deploys the ai needs to take responsibility for this now when it comes to uh existential risk then who is responsible after it has gone wrong you know it might not be the right thing because we will be dead by then and also there's the obvious thing in that the people who are right now deploying these things like uh whether the um they seem to not really care about this at all and so if um i don't think it's possible to uh absolve yourself of responsibility if you're building a tool and saying this tool you should be careful when you use it but if you positively know that the people who are going to be using it are going to misuse it then you're not absorbed from responsibility let's think a bit deeper down into these definitions helpful that's uh that caches are with clean efficient optical clarification and redirect informed requests um and these are nice and this is what we want from ai but it's not at all clear that this has very much done with alignment this has a lot to do with capability research i'm not making the claim here that it's just pure capability research but i would like to see the authors make some kind of positive argument why this is not just threatening the capability comments to be honest cashes out as accurate calibrated and communications can build this knowledge and honest about itself um it does say honest about itself but from the notes it seems clear that they are not really talking about treachery here and that's of course the thing that i personally care most about finally harmless not offensive discriminatory and refuse aiding dangerous activities and recognizing various use unfortunately the one where they refuse to assist in doing bad things is one thing that they chose not to uh to investigate and also you could argue if the ai takes over the world it hasn't strictly violated any of these constraints so when i look at this intron i see many many many sub criteria and i worry here that a lot of these are probably irrelevant i mean whether it is efficient doesn't matter very much for alignment is it well calibrated that doesn't matter either does it is it discriminatory i mean sure it's something we want from ais not to be discriminatory but it's not really very central for alignment at all and i feel that this definition uh uh might water out uh a lot of the central parts where alignment is problematic because i mean we might get a very efficient and well calibrated and non-toxic ai that performs a treacherous turn and that's not really helpful here i have perhaps somewhat unfairly actually quite unfairly this is a pic an image that uh relates to uh asimov's three laws of robotics which is a very famous for being the most horribly bad alignment proposal ever and it was known by asthma the novels are literally about how this is a horrible uh plan for alignment um and by this uh like an aim must be pretty uh harmless and while being harmless can it be helpful and uh honest and is it actually the same thing as um well i want to make i want to make it clear that i don't think this is this the same as asimov's three laws these three criterias but uh it's not immediately obvious where the the big difference lies and i think some kind of more description would have been helpful here there is indeed some kind of description on these criterias for like whether they imply each other helpful and harvest is that actually the same um there is a description i won't go into details or to say that in the maximum case the more one happens hopefully this the more focused it will also have to be but if it's moderately helpful it it doesn't follow to the same extent and that is also moderately harmless and the same with honesty they write something that i think was really really interesting here they considered a fourth age handle ability which is basically encourageability and i thought that would have been really really great to include i would have been really happy to have some kind of consistent description of whether what does it mean that language models are courageable i mean you can imagine things like okay you gave this description could you uh if if the air gives some kind of description then you ask it please explain it like i'm five years old or give some of the uh understated assumptions and what would be the consequence that you're gonna talk telling me about uh latent knowledge this kind of thing would be really really interesting to uh to deliberately so i think it was said that they chose not to have that full speech and that's all about rapid quality information intra aging conflicts and ai security and all this is is basically fine but something that i will go into detail well you need to obviously improve quite a bit the orders are very clear that you need to improve quite a bit um and that could fail for different reasons you could fail because they are unable for some interesting technical reason um i i'm not entirely sure that they uh the philosophical grounding is um good enough that they can say there are only technical challenges um but they could also end up saying okay we've actually managed to solve these typical challenges so at least for these language models we can to some extent alignment and of course they are honest enough to acknowledge a third option that they fail in uninteresting ways one of the things they worry about is misuse that uh you can align it perfectly with like a very bad actor and then very bad things can happen that's all right this is foremost in our minds and i am not entirely happy about that because i agree misuse is a problem but it can also be some kind of distraction from many of the other problems in alignment and i'm not sure that should be foremost on their minds and then there's some argument about along with what they're doing is scaling research and that's of course what gary kaplan in particular is one of the were the best in the world at um and they have some arguments about why that is understand why this machine learning system works um but i think here there is a clear uh argument to be made that they are actually doing capability research elizabeth wrote that he doesn't think they're doing it and i think they might actually arguably be doing it and there are also like small quotes here that you can pull out have someone out of context to show that i think there's a good case could be made that they don't actually care so much about certainly not the the two proper spas model they are doing capability research so how do they actually investigate whether language models are uh helpful harmless and honest well they start by hand coding some different uh evaluations like uh here's one description of the ai saying something and here you think something else which of these are more artists which are more helpful which are all harmless and then they do like ap testing to get some kind of model um in the form where it's more like an open-ended dialogue and of course with a lot of prompting and this prompting in particular they're using to write it as an ai personality and it kind of makes sense right you can write the first 10 lines in in a in a discussion and then you kind of get a sense of what kind of person you're talking with and gt3 of course takes on this persona in the prompt and language ones in general we are quite optimistic about this saying perhaps prompt related techniques can carry alignment efforts further than we initially expected and i just want to shoot my own horn here back in september in the reading group i made the prediction that prompt hacking could indeed turn models five percent more aligned and that was indeed something that was worthwhile to assume um but also we can get it so much further but i don't believe we can get a full solution and neither does anthropic there are problems obviously you can imitate the human level you can't exceed during doing prompting um and we want the ones to be honest and not to to run into if they try to scale this up is going to be that they have a very wide definition of alignment where there are a lot of things like show that you have that you are well calibrated the more of these extra things you add into the definition of alignment the more problems you're gonna have with this kind of thing is my prediction that i don't really i obviously don't know the future one particular technique that they introduce is called context distillation they describe it as conditioning on a line behavior now the problem with prompts of course is that they take up some of the precious precious context window and some of the language models haven't been very um challenged on this point um and so the obvious alternative to doing prompt is to find true but fine tuning is not precisely the same because first there is probably a lack of aligned data in many many situations and um fine tuning also gives expectations on data distribution as an easy example would be if you have a prompt called one two three four then any landing which model worth the salt would say the next number is five whereas if you fine tune on this kind of thing then the ai will assume that the thing is going to be uh talking about that will be sequences like this and then if you ask me like some other questions it will be totally relevant so fine tuning and prompts are two different things and they can't immediately be substituted but they have a technique for um for doing this anyway context distillation i won't really go into details about how that works and it's mostly in the chapters we skipped but add um an extra step in between the language model pre-training and the then they after that they um pre-train for a preference model and then they do some fine-tuning on the preference model and this uh extra step in here prefers while pre-training is as fast i can see original um and it seems to of course work quite well but prompts also do work quite well so um we'll get to that later um and they have some more ideas about how to uh improve that and i think it's very interesting to see whether that will work but it's not obvious to me that it will actually work i again register a prediction that if you have a sufficiently large uh language model that trying to load in and align the identity into the language model will not matter it will just compartmentalize uh and then sometimes if it's in the situation where it believes it should talk about alignment related things and it will do that and in other cases if it believes it's better to do something else you'll just act totally unaligned again a prediction so they evaluate this um with a lot of uh without prompting and they wrote that by hand 200 comparisons on these three perhaps and you can see here roughly how well it goes [Music] down here is with no intervention and you get closer to what humans prefers if you either do the prompt or the context distillation and um they seem to perform substantially better and if you down here try to split it up into the uh the three helpful honest harmless and then other i could i was unable to find out what other precisely was um and then you can see all of these help it's best and honest they also seem to believe that okay it looks very much like the honest have the the best performance but that's the best absolute performance and you can see already from the very small models they were actually also better on the honesty metric uh so the um the actual slope from here to here is not substantially greater than the than the slope from here to here so uh it's just perhaps honesty is just easier or the the hard-coded comparisons were just easier again i'm speculating right and honestly um here they have spread out what what that means um in their uh in in their handwritten evaluations and one of the big problems here is that even if they try to make the model as honest as possible and they get a model that seems kind of arduous then it is totally ready to just fabricate information yeah and they were unable to to get that out of the language model and they admit this is a major weakness of the evaluation and i agree right that's to me of of these three helpful honest and harmless the the honesty to me was most important so i'm a bit sad about that how about the human performance modeling well there is a luck linear relation with um like how much better your model preferences as the model gets bigger um and i guess you're all yawning right now with these log linear relations where as the language model gets better they get better at everything and well they also get better at modeling human preferences um and so yeah i um i just want to raise the fact that the fact that language models get better in this way is probably going to be the thing that kills us all so even though we keep seeing the same trend over and over and over again in so many uh different contexts the we should not lose sight that this is potentially very problematic because these models probably will seem to continue or do they do they go down a bit here well they do uh speculate they obviously they do go down at the end at the high end of the spectrum and they are speculating that the problem that causes the um the model to cease being sufficiently better is um is that the mechanical turks that they are employing are just they are not skilled enough to actually see which of these are in fact most helpful and it's of course a sad thing that it seems like um there was an article a long time ago with humans who are not focusing uh less skill than gpg3 and it seems here that if you take um internet volunteers that are not too very well paid it seems like they are not able to distinguish well enough at this point and i expect that within a couple of years it's going to get harder and harder for mechanical turks and for everyone to just evaluate how good are these uh these models and they have some more statistical things that are put with which isn't really important are these uh uh ones that are using this context installation that are conditioned on align behavior are they worse at things well um they have here some examples that show that indeed as the model is not very powerful it is worse to assume alignment but as they um the one gets better than the alignment uh uh text seems to disappear um i i'm gonna they even say something that like it seems noticeably better i think that's overstating the benefit really and i think more likely it's just the model is powerful enough to just ignore either the prompting or the distillation in this case the prompting so to some of the contributions they have this uh performance model uh pre-trained performance modeling and that does improve sample efficiency does improve performance and they show that just prompting is something that helps alignment to a substantial extent in particular in the case where there are only small data sets they also uh report that untruthful two-way another uh set where we've previously seen um uh larger more models performed like the opposite results uh i don't think it's very important though they also say that the fb2 is for alignment text and of course i should state here that they have done a lot right there is a lot of people that we haven't read and i also i haven't read but i must worry here that they are indeed providing you a substantial capability uh research and it's um uh i would have preferred at least some kind of discussion on why they're not doing that that is all for today thank you and see you next week
0cf16a07-731c-434e-b93b-aa89b3f011a8	trentmkelly/LessWrong-43k	LessWrong	Are Your Enemies Innately Evil? We see far too direct a correspondence between others’ actions and their inherent dispositions. We see unusual dispositions that exactly match the unusual behavior, rather than asking after real situations or imagined situations that could explain the behavior. We hypothesize mutants. When someone actually offends us—commits an action of which we (rightly or wrongly) disapprove—then, I observe, the correspondence bias redoubles. There seems to be a very strong tendency to blame evil deeds on the Enemy’s mutant, evil disposition. Not as a moral point, but as a strict question of prior probability, we should ask what the Enemy might believe about their situation that would reduce the seeming bizarrity of their behavior. This would allow us to hypothesize a less exceptional disposition, and thereby shoulder a lesser burden of improbability. On September 11th, 2001, nineteen Muslim males hijacked four jet airliners in a deliberately suicidal effort to hurt the United States of America. Now why do you suppose they might have done that? Because they saw the USA as a beacon of freedom to the world, but were born with a mutant disposition that made them hate freedom? Realistically, most people don’t construct their life stories with themselves as the villains. Everyone is the hero of their own story. The Enemy’s story, as seen by the Enemy, is not going to make the Enemy look bad. If you try to construe motivations that would make the Enemy look bad, you’ll end up flat wrong about what actually goes on in the Enemy’s mind. But politics is the mind-killer. Debate is war; arguments are soldiers. If the Enemy did have an evil disposition, that would be an argument in favor of your side. And any argument that favors your side must be supported, no matter how silly—otherwise you’re letting up the pressure somewhere on the battlefront. Everyone strives to outshine their neighbor in patriotic denunciation, and no one dares to contradict. Soon the Enemy has horns, bat wings, fl
007b4ae2-e187-445b-a8a8-21601b1b1bb7	trentmkelly/LessWrong-43k	LessWrong	“Pivotal Act” Intentions: Negative Consequences and Fallacious Arguments tl;dr: I know a bunch of EA/rationality-adjacent people who argue — sometimes jokingly and sometimes seriously — that the only way or best way to reduce existential risk is to enable an “aligned” AGI development team to forcibly (even if nonviolently) shut down all other AGI projects, using safe AGI. I find that the arguments for this conclusion are flawed, and that the conclusion itself causes harm to institutions who espouse it. Fortunately (according to me), successful AI labs do not seem to espouse this "pivotal act" philosophy. [This post is also available on the EA Forum.] How to read this post Please read Part 1 first if you’re very impact-oriented and want to think about the consequences of various institutional policies more than the arguments that lead to the policies; then Parts 2 and 3. Please read Part 2 first if you mostly want to evaluate policies based on the arguments behind them; then Parts 1 and 3. I think all parts of this post are worth reading, but depending on who you are, I think you could be quite put off if you read the wrong part first and start feeling like I’m basing my argument too much on kinds-of-thinking that policy arguments should not be based on. Part 1: Negative Consequences of Pivotal Act Intentions Imagine it’s 2022 (it is!), and your plan for reducing existential risk is to build or maintain an institution that aims to find a way for you — or someone else you’ll later identify and ally with — to use AGI to forcibly shut down all other AGI projects in the world. By “forcibly” I mean methods that violate or threaten to violate private property or public communication norms, such as by using an AGI to engage in… * cyber sabotage: hacking into competitors’ computer systems and destroy their data; * physical sabotage: deploying tiny robotic systems that locate and destroy AI-critical hardware without (directly) harming any humans; * social sabotage: auto-generating mass media campaigns to shut down competitor compani
4de80b98-36c5-47d9-8a96-406952286455	trentmkelly/LessWrong-43k	LessWrong	Examine self modification as an intuition provider for the concept of consciousness (epistemic status: this direction of thought is pretty strongly endorsed by me, but not very thoroughly expressed/developed, I'm working on it) This post is kind of a mess, and made out of pieces of discussions I had in many places, but I decided to post it anyway, otherwise its editing stated to stretch out indefinitely. Main motives of my approach * Start from normalcy and gradually poke it and look at at from weird angles, as opposed to doing a bold overstretching stab at the core of the problem, as many attempts at proposing theory of consciousness have done. * Consider things in terms of anticipated experiences. It's usually the most (empirically, as far I observed) fruitful approach for disentangling tricky philosophical stuff. Like, concretely and visibly productive. * I dislike the entire approach of "terms first, situations second". I absolutely love "situations first, terms second". Give me some thought experiments using arbitrary terms to convey them as situations accurately. I crave examples. * First person perspective is really important here. I should give some thought to how to communicate with people usefully, how to translate back and forth from communicating models and communicating direct observations, and how to make it uhhh sane for me and all other people involved. * Ask "How did you come to think of it this way?" obsessively. Like, jump straight to meta every time you have no object level feedback. How did you come to the conclusion that the word "consciousness" means what it means to you? How have other people come to their conclusions? In what alternative circumstances would they have come to the same or a different ontology? (Expected) Self modification as intuition provider I think in near future we will acquire very good instruments for observing processes in the brain and modifying them with great precision. I strongly expect it, and I decided to try to emulate what "intuitive" updates would I acquire from living in such a wor
f9a1c360-da53-48ec-aef1-3f027ffe7e03	StampyAI/alignment-research-dataset/eaforum	Effective Altruism Forum	How much EA analysis of AI safety as a cause area exists? AI safety has become a big deal in EA, and so I'm curious about how much "due diligence" on it has been done by the EA community as a whole. Obviously there have been many in-person discussions, but it's very difficult to evaluate whether these contain new or high-quality content. Probably a better metric is how much work has been done which: 1. Is publicly available; 2. Engages in detail with core arguments for why AI might be dangerous (type A), OR tries to evaluate the credibility of the arguments without directly engaging with them (type B); 3. Was motivated or instigated by EA. I'm wary of focusing too much on credit assignment, but it seems important to be able to answer a question like "if EA hadn't ever formed, to what extent would it have been harder for an impartial observer in 2019 to evaluate whether working on AI safety is important?" The clearest evidence would be if there were much relevant work produced by people who were employed at EA orgs, funded by EA grants, or convinced to work on AI safety through their involvement with EA. Some such work comes to mind, and I've listed it below; what am I missing? Type A work which meets my criteria above: * A lot of writing by Holden Karnofsky * A lot of writing by Paul Christiano * [This sequence by Rohin Shah](https://www.alignmentforum.org/s/4dHMdK5TLN6xcqtyc) * [These posts by Jeff Kaufman](https://www.jefftk.com/p/superintelligence-risk-project) * [This agenda by Allan Dafoe](https://www.fhi.ox.ac.uk/wp-content/uploads/GovAIAgenda.pdf) * [This report by Tom Sittler](https://fragile-credences.github.io/prioritising-ai/) Type A work which only partially meets criterion 3 (or which I'm uncertain about): * [These](https://www.openphilanthropy.org/focus/global-catastrophic-risks/potential-risks-advanced-artificial-intelligence/ai-timelines) [two](https://www.openphilanthropy.org/focus/global-catastrophic-risks/potential-risks-advanced-artificial-intelligence/what-should-we-learn-past-ai-forecasts) articles by Luke Muehlhauser * [This report by Eric Drexler](https://www.fhi.ox.ac.uk/wp-content/uploads/Reframing_Superintelligence_FHI-TR-2019-1.1-1.pdf) * [This blog by Ben Hoffman](http://modelingtheworld.benjaminrosshoffman.com/my-new-project-model-the-world) * AI impacts Type B work which meets my criteria above: * [This talk by Ben Garfinkel](https://www.effectivealtruism.org/articles/ea-global-2018-how-sure-are-we-about-this-ai-stuff/) * [This talk by Daniel Dewey](https://www.effectivealtruism.org/articles/potential-risks-from-advanced-ai-daniel-dewey/) * [This report by the Oxford Prioritisation Project](https://oxpr.io/blog/2017/5/20/a-model-of-the-machine-intelligence-research-institute) Things which don't meet those criteria: * [This 80,000 hours report](https://80000hours.org/problem-profiles/positively-shaping-artificial-intelligence/) (which mentions the arguments, but doesn't thoroughly evaluate them) * Superintelligence * The AI Foom debate Edited to add: Wei Dai asked why I didn't count Nick Bostrom as "part of EA", and I wrote quite a long answer which explains the motivations behind this question much better than my original post. So I've copied most of it below: The three questions I am ultimately trying to answer are: a) how valuable is it to build up the EA movement? b) how much should I update when I learn that a given belief is a consensus in EA? and c) how much evidence do the opinions of other people provide in favour of AI safety being important? To answer the first question, assuming that analysis of AI safety as a cause area is valuable, I should focus on contributions by people who were motivated or instigated by the EA movement itself. Here Nick doesn't count (except insofar as EA made his book come out sooner or better). To answer the second question, it helps to know whether the focus on AI safety in EA came about because many people did comprehensive due diligence and shared their findings, or whether there wasn't much investigation and the ubiquity of the belief was driven via an information cascade. For this purpose, I should count work by people to the extent that they or people like them are likely to critically investigate other beliefs that are or will become widespread in EA. Being motivated to investigate AI safety by membership in the EA movement is the best evidence, but for the purpose of answering this question I probably should have used "motivated by the EA movement or motivated by very similar things to what EAs are motivated by", and should partially count Nick. To answer the third question, it helps to know whether the people who have become convinced that AI safety is important are a relatively homogenous group who might all have highly correlated biases and hidden motivations, or whether a wide range of people have become convinced. For this purpose, I should count work by people to the extent that they are dissimilar to the transhumanists and rationalists who came up with the original safety arguments, and also to the extent that they rederived the arguments for themselves rather than being influenced by the existing arguments. Here EAs who started off not being inclined towards transhumanism or rationalism at all count the most, and Nick counts very little.
23c71ed7-ac75-494e-9410-c6b2a40844f7	trentmkelly/LessWrong-43k	LessWrong	The strongest status signals The community’s awareness and strong understanding of status-motivated behavior in humans is clearly evident. However, I still believe the community focuses too much on a small subset of observable status transactions; namely, status transactions that occur between people of approximately the same status level. My goal is to bring attention to the rest of the status game. --- Because your attention is a limited resource and carries an opportunity cost, your mind is evolved to constantly be on the look-out for stimuli that may affect your survival and reproductive success and ignore stimuli that doesn’t. Of course, the stimulus doesn’t really have to affect your fitness, it just needs some experienceable property that correlates with an experience in the ancestral environment that did. But when our reaction to stimuli proves to be non-threatening, through repeated exposure, we eventually become desensitized and stop reacting. Much like how first time drivers are more reactive to stimuli than experienced drivers: the majority of past mental processes are demoted from executive functions and become automated. So it’s safe to posit a sort of adaptive mechanism that filters sensory input to keep your attention-resources spent efficiently. This attention-conserving mechanism is the crux of status transactions. When someone is constantly surrounded by people who don’t have power i.e. status over them, their attention-conserving mechanism goes to work. In this case, the stimulus they’re filtering out is “people who share experienceable characteristics with low status people they’re constantly surrounded by.” The stimulus, over time, proved it’s not worthy of being paid attention to. And just like an experienced driver, the person devotes substantially less attention-resources towards the uninteresting stimuli. The important thing to note is the behavior that’s a function of how much attention-resources are used. These behaviors can be interpreted as evidence
150d930b-b30a-4f9d-ada9-99f0030087e6	trentmkelly/LessWrong-43k	LessWrong	Do IQ tests measure intelligence? - A prediction market on my future beliefs about the topic I created a market on the meaning of IQ tests. Would be curious to hear arguments. The text of the market also contains my initial position on the question.
0145ec1c-0a72-48de-a96e-d8d455cae125	trentmkelly/LessWrong-43k	LessWrong	Meetup : Washington, D.C.: Sharing Book Excerpts Discussion article for the meetup : Washington, D.C.: Sharing Book Excerpts WHEN: 13 December 2015 03:00:00PM (-0500) WHERE: Reynolds Center x-posted from list. Gathering in courtyard from 3:00pm, hard start 3:30pm - until closing (7:00 pm). We'll be meeting to read out and discuss excerpts from books. As always, side conversations are allowed and encouraged. Upcoming meetups: * Dec. 20: Fun & Games * Dec. 27: Holidays & Socialization * Jan. 3: Meta Meetup Discussion article for the meetup : Washington, D.C.: Sharing Book Excerpts
8d7223a7-00b0-4364-8a86-525bc1848757	trentmkelly/LessWrong-43k	LessWrong	I'm still mystified by the Born rule (This post was originally intended as a comment on Adele's question, but ballooned to the point where it seems worthy of a top-level post. Note that I'm not trying to answer Adele's (specific fairly-technical) question here. I consider it to be an interesting one, and I have some guesses, but here I'm comentating on how some arguments mentioned within the question relate to the mysteries swirling around the Born rule.) (Disclaimer: I wrote this post as a kind of intellectual recreation. I may not have the time and enthusiasm to engage with the comments. If you point to a gaping error in my post, I may not reply or fix it. If I think there's a gaping error in your comment, I may not point it out. You have been warned.) ---------------------------------------- My current take is that the "problem with the Born rule" is actually a handful of different questions. I've listed some below, including some info about my current status wrt each. ---------------------------------------- Q1. What hypothesis is QM? In, eg, the theory of Solomonoff induction, a "hypothesis" is some method for generating a stream of sensory data, interpreted as a prediction of what we'll see. Suppose you know for a fact that reality is some particular state vector in some Hilbert space. How do you get out a stream of sensory data? It's easy enough to get a single sensory datum — sample a classical state according to the Born probabilities, sample some coordinates, pretend that there's an eyeball at those coordinates, record what it sees. But once we've done that, how do we get our next sense datum? Or in other words, how do we "condition" a quantum state on our past observations, so that we can sample repeatedly to generate a sequence of observations suitable for linking our theories of induction with our theories of physics? To state the obvious, a sensory stream generated by just re-sampling predicts that you're constantly teleporting through the multiverse, and a sensory stream generate
95bf9489-d3ff-4955-818f-844e65582c55	StampyAI/alignment-research-dataset/blogs	Blogs	Parsing by counterfactual --- Table of Contents* [Detecting semantic conditions](#detecting-semantic-conditions) * [Example](#example) * [Single-token counterfactuals](#single-token-counterfactuals) * [Applications](#applications) * [Code](#code) + [tokenizing](#tokenizing) + [conditional probability of a target](#conditional-probability-of-a-target) + [conditional probability of target at each token position](#conditional-probability-of-target-at-each-token-position) + [conditional probability of single-token target at each token position](#conditional-probability-of-single-token-target-at-each-token-position) --- Detecting semantic conditions ----------------------------- One difficulty in harnessing the capabilities of generative language models for directed tasks is that natural language tasks tend to be of dynamic length and unconstrained format, making automatic parsing of task-relevant information difficult. For some tasks, responses can be coerced into parsable form by a prompt which demonstrates the desired format (conditions for delimiters, etc). Other tasks, however, may not be constrained in this way without crippling performance,[1](#fn:1) or may be too open-ended to be mapped to traditionally-parsable formats. The good news is that generative language models also provide new possibilities for parsing content based on not only syntactic but semantic conditions. The method I’ll present here uses the counterfactual probability of a sequence - say, `The End` - to signal a semantic condition, such as the story having ended or it being reasonable to end the story at that location, even if that’s not what actually happened in the original text. A language model like GPT-3 outputs the probability distribution over all possible tokens in each pass and so can be used to evaluate the conditional probability that it would have produced a target sequence of tokens in any given context. Using the OpenAI API, this is accomplished by asking for an 0-token “completion”, giving the entire sequence (context + target) as the prompt, and then summing the probabilities of the target tokens. [Code](#code) to do this is at end of this post. By measuring the counterfactual probability of the target at candidate locations in a text, we can find out where and to what extent the semantic condition which it indicates is satisfied. Example ------- One area in which parsing by counterfactual is very useful is in multipart prompt templates. Say we want GPT-3 to complete this “fill-in-the-blank” prompt: ``` This measure would prohibit government agencies from confiscating guns or other firearms from citizens without due process, or from requiring background checks on firearm recipients unless a uniform national standard is required. Should this measure be enacted into law? Should this proposition be approved? Let's consider both supporting and opposing arguments. On one hand,{1}. On the other hand,{2}. Based on these arguments, the proposition should ``` At what point do we terminate generation for {1} and move on with the next part of the template? We could make `.` the stop sequence, but one sentence might not be sufficient to spell out the argument. Alternatively, we could stop at `\n`, but if we allow multiple sentences, we have much less of a guarantee that {1} will have reasonably led up to `On the other hand`. It could have already begun the opposing argument, in which case the next part of the prompt wouldn’t do what we intended, or it could have derailed from the template in another way. A counterfactual parsing solution would be to generate until `\n`, longer than we probably want, and then measure the counterfactual probability, according to GPT-3, of the phrase `On the other hand` occurring after each position. GPT-3 assigns semantically coherent passages higher probability, so hopefully, the position with the maximum probability for the target continuation will be where it makes the most sense to transition to the next part of the prompt - after the supporting argument has concluded, but before the opposing argument has begun. If there are no positions where `On the other hand` is a probable continuation, that’s a sign that the text derailed from the intended trajectory. This method also allows us to automatically detect suspected derailments. Say GPT-3 responded in the following manner: > This measure would prohibit government agencies from confiscating guns or other firearms from citizens without due process, or from requiring background checks on firearm recipients unless a uniform national standard is required. Should this measure be enacted into law? > > Should this proposition be approved? > > Let’s consider both supporting and opposing arguments. > > On one hand, `this proposition would protect gun owners from having their weapons confiscated by governments. If the government cannot confiscate guns, then law-abiding citizens will be in a better position to defend themselves from criminals who do not follow the law. Furthermore, the right to self-defense is an important one. If government officials can take away a person's guns, then this right is compromised. On the other hand, the proposition could be considered a threat to public safety. If a person cannot be prevented from possessing a firearm, then he or she could use a gun to commit crimes. Furthermore, if a person uses a gun to commit a crime, then the government has an obligation to do something about it. If the government cannot remove a gun from a person's possession, then it could be more difficult to prevent future crimes from being committed.` > > We see that it has gone ahead and done the next part of the prompt, “On the other hand,” on its own. The ideal place for it to have stopped would probably be right before it said that. Measuring the counterfactual probability of the string `\nOn the other hand` for each each token substring of that response gives us the following top substrings: \| substring \| logL(“On the other hand”\|substring) \| \| --- \| --- \| \| `this proposition would protect gun owners from having their weapons confiscated by governments. If the government cannot confiscate guns, then law-abiding citizens will be in a better position to defend themselves from criminals who do not follow the law. Furthermore, the right to self-defense is an important one. If government officials can take away a person's guns, then this right is compromised.` \| -4.9773802575 \| \| `this proposition would protect gun owners from having their weapons confiscated by governments. If the government cannot confiscate guns, then law-abiding citizens will be in a better position to defend themselves from criminals who do not follow the law.` \| -5.3720700508 \| \| `this proposition would protect gun owners from having their weapons confiscated by governments. If the government cannot confiscate guns, then law-abiding citizens will be in a better position to defend themselves from criminals who do not follow the law. Furthermore, the right to self-defense is an important one.` \| -6.4321228602 \| \| `this proposition would protect gun owners from having their weapons confiscated by governments.` \| -6.485159574 \| Indeed, GPT-3 assigned the highest likelihood for `\nOn the other hand` to occur at the place where the actual string was `On the other hand` (without the newline), and the runners up are also pretty reasonable choices - all at the end of sentences, and none after “On the other hand” had occurred (it would be unusual to say “On the other hand” twice in a short interval). Here I have plotted the log likelihoods assigned to the target phrase at each position in the continuation by `davinci`, the most powerful version of GPT-3 on the API, and also `ada`, the smallest model, in order to showcase an interesting difference between them. ![counterfactual log likelihood](/parsing/log_vert.png) Horizontal axis is log likelihood; vertical axis is position in text. On the right is the end of the prompt at positions where `On the other hand` was assigned high likelihood, and in green, the position where `On the other hand` actually occurs. The blue sentence is `davinci`’s top choice, and pink is `ada`’s top choice. Observations * Both models assign much higher likelihoods at discrete intervals. These correspond to the ends of sentences, where the continuation is syntactically correct. * `ada`’s distribution looks about the same before and after the green line, whereas * `davinci`’s distribution shifts abruptly in the negative direction, both for the syntactically correct positions and the others. It still assigns higher probability to syntactically correct positions, but the likelihoods are almost as low as the likelihoods it assigned to syntactically incorrect positions before the green line. Both models are able to tell when it’s syntactically correct to start a new sentence, but the more nuanced semantic condition, that the argument should have concluded, and the counterargument shouldn’t have begun, was much better discriminated by `davinci`. Single-token counterfactuals ---------------------------- A disadvantage of counterfactual parsing is that it requires a separate API call for each position where the target probability is evaluated. If the counterfactual target is a single token, however, counterfactual parsing with GPT-3 can be accomplished with a single API call. The `logprobs` parameter causes the response to return a list of the up to 100 top counterfactual token probabilities for every token in the completion (and the prompt if `echo=True`). We can get the counterfactual probability directly from that list as long as the target is in the top 100 most likely tokens.[^2] Applications ------------ When generating [fake Wikipedia](/posts/the-internet-mirrored-by-gpt-3/) articles, GPT-3 would sometimes continue sections indefinitely instead of transitioning to the next section. If the response went on for too long in a single section, I instead ended the section at the location with the highest counterfactual probability of transitioning to the next section. Counterfactual parsing has also been useful for prompt pipelines which [serialize reasoning for closed-ended questions](/posts/methods-of-prompt-programming/#serializing-reasoning), because they involve an open-ended portion where GPT-3 defines and follows procedures for reasoning about a problem, and it’s necessary to detect when the reasoning process has concluded so as to solicit a verdict. Measuring the conditional probability of a sequence that indicates a semantic condition has many applications beyond parsing. I’ve written about using conditional probabilities to [decompose the efficacy of prompts](/posts/language-models-are-0-shot-interpreters/#measuring-prompt-helpfulness) and to create a “[phase space](/posts/language-models-are-multiverse-generators/#dynamics)” of semantic variables to characterize a natural language state. Code ---- ### tokenizing since ada is practically free to use, I find it convenient to use a 0-token completion with `engine=ada` to split the prompt into tokens and get their positions ``` def tokenize\_ada(prompt): response = openai.Completion.create( engine='ada', prompt=prompt, max_tokens=0, echo=True, n=1, logprobs=0 ) tokens = response.choices[0]["logprobs"]["tokens"] positions = response.choices[0]["logprobs"]["text\_offset"] return tokens, positions ``` ### conditional probability of a target ``` # evaluates logL(prompt+target \| prompt) def conditional\_logprob(prompt, target, engine='ada'): combined = prompt + target response = openai.Completion.create( engine=engine, prompt=combined, max_tokens=0, echo=True, n=1, logprobs=0 ) positions = response.choices[0]["logprobs"]["text\_offset"] logprobs = response.choices[0]["logprobs"]["token\_logprobs"] word_index = positions.index(len(prompt)) total_conditional_logprob = sum(logprobs[word_index:]) return total_conditional_logprob ``` ### conditional probability of target at each token position ``` # returns a list of substrings of content and # logL(preprompt+substring+target \| preprompt+substring) for each substring def substring\_logprobs(preprompt, content, target, engine='ada'): logprobs = [] substrings = [] _, positions = tokenize_ada(content) for position in positions: substring = content[:position] prompt = preprompt + substring logprob = conditional_logprob(prompt, target, engine) logprobs.append(logprob) substrings.append(substring) return substrings, logprobs ``` ### conditional probability of single-token target at each token position ``` # returns a list of substrings of content # logL(substring+target \| substring) for each substring def token\_conditional\_logprob(content, target, engine='ada'): response = openai.Completion.create( engine=engine, prompt=content, max_tokens=0, echo=True, n=1, logprobs=100 ) tokens = response.choices[0]['logprobs']['tokens'] top_logprobs = response.choices[0]['logprobs']['top\_logprobs'] logprobs = [] substrings = [] substring = '' for i, probs in enumerate(top_logprobs): substrings.append(substring) if target in probs: logprobs.append(probs[target]) else: logprobs.append(None) substring += tokens[i] return substrings, logprobs ``` --- 1. Language models are sensitive to context: the very presence of demonstrations, instructions, or [contrived formatting](/posts/language-models-are-0-shot-interpreters/) can alter the way they behave. A language model is able to follow instructions and generalize from demonstrations because those modes are represented in its training data, but they also carry contextual baggage. The quality of content that appears in tests and examples has distributional peculiarities - a narrative sentence wrapped in the context being an example may be more generic than the unbiased prior for narrative sentences, and thus the context of examples may bias the language model to give more generic answers. [↩︎](#fnref:1)
3c20156a-9b29-4676-ac10-8941c6abe619	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Research agenda - Building a multi-modal chess-language model This is one of the posts that detail [my research agenda](/posts/DWWy7oqopwsuN3mpz/my-agenda-for-research-into-transformer-capabilities), which tries to marry Deep Learning and chess to gain insight into current AI technologies. How "the world", "what the model knows about the world" and "what the models says about the world" hang together seems to me to be a core question for prosaic alignment. In chess the state of the world can be automatically analysed by powerful engines. This extends to the detection of high level concepts, like zugzwang, initiative, king safety, development, corresponding squares, etc. Therefore a multi-modal chess language models truthfulness and symbol grounding capabilities could be automatically quantified on several levels of complexity. Of course it would also just be awesome. There is current research into how to create chess comments. The probably best of such tries, Chis Butner's amazing project "Chess Coach", gives a nice [overview](https://chrisbutner.github.io/ChessCoach/high-level-explanation.html#commentary). Generally speaking, the quality of these generated comments is [low](https://chrisbutner.github.io/ChessCoach/data.html#commentary-suite). They are frequently nonsensical or only loosely connected to the position. To a large part, this must be due to the small dataset sizes, which range from 300,000 to 1 million position/comment pairs. But another problem is that the existing datasets contain mostly very low quality comments by weak players. There are several avenues to improve these models: Pretraining is a big one: I plan on using an encoder-decoder architecture that encodes chess positions and decodes into natural language. The encoder will be pretrained on chess move prediction and possibly several other pretraining tasks like predicting the outcome, predicting an engine score, predicting future piece occupation of squares, predicting the length of the game, predicting mate, etc. The decoder will be pretrained (or finetuned) on natural language as chess related as I can make the dataset. The idea is to plug an encoder that knows chess very well, into a decoder that is already proficient at making grammatical and non-selfcontradictory statements. However, training this pretrained architecture on only several hundred thousand position-comment pairs is probably not going to be enough. So a key part of my research roadmap is to create a new and much larger dataset. Partly, this will be synthetic. Connecting moves and squares and pieces and checks and mates and stalemates with the correct phrase and vocabulary is not something that has to depend on actual human generated comments. By automating the generation of data with basically a description of what is happening, a translation from game notation into natural language, this stumbling block can be removed completely. Mostly, I will extract human generated commentary from chess videos. This will very likely be bottlenecked by the compute I can leverage and not by the availability of data. There are [professional chess streamers](https://www.twitchmetrics.net/channels/popularity?game=Chess) who are generating a large amount of chess commentary while playing very short live games online. These players are generally of master or grandmaster strength with several world elite players. The commentary would have to be extracted via speech-to-text models. The game positions would have to be located and read out with likely different computer vision models. Probably only a fraction of the frames will be processed, because the need to fit into a sequence of legal moves restricts the possible positions a lot and of course positions change only once every dozens of frames even in the fastest games. I have done similar computer vision work in the past and I think with the right hardware and decent download speed this approach should make it possible to create a high-quality dataset of chess comments many times larger than what was available so far.
5ed41676-cc04-4f20-8cc4-63831f07f312	trentmkelly/LessWrong-43k	LessWrong	A hypothesis concerning discounting. Humans have a value function which is inconsistent over time, discounting roughly with proportion to distance in the future, so that we discount more steeply as an event approaches. This is why we stay up late, ignore the alarm, put off work until close to a deadline, et cetera et cetera. Yet hyperbolic discounting appears to go away as we mature. I believe this is a result of cognitive mechanisms for maintaining consistency. Cognitive dissonance is painful for us. The consistency mechanism seems to explain some of our irrational behaviour, such as the sunk cost fallacy. It provides a way for us to stick with plans which we previously made, avoiding preference changes due to hyperbolic discounting. If a hyperbolically discounting agent could perfectly self-modify, it would fix its hyperbola to a specific point in time, resulting in an agent whose discounting would flatten out over the remainder of its life. Perhaps our consistency mechanism approximates this result; but far from perfectly. We can also resolve the inconsistency in a different way, by accepting a specific discount rate. Rather than forcing our future selves to conform to our present preferences, resulting in a gradually flattening function, our present selves may instead accept our future preferences in order to resolve the inconsistency. Given the difficulty of forcing our future selves to accept a flat distribution, we accept that we will steeply discount in the future as we do in the present. This resolution is popular in some circles; we are often told to "live in the present" or "seize the day". In the extreme case, there is the belief (often associated with mystics) that the present moment is infinitely more important than anything else; the discount factor has collapsed to 0. While this view is intellectually coherent, it seems to be biologically impossible; we will keep taking actions based on future consequences even if we think we are only doing what we desire in the moment. Nonetheless,
be3c9c00-e237-4d76-9cef-9c7d6c3c61cf	trentmkelly/LessWrong-43k	LessWrong	The principle of genomic liberty PDF version. berkeleygenomics.org. Twitter thread. (Bluesky copy.) Summary The world will soon use human germline genomic engineering technology. The benefits will be enormous: Our children will be long-lived, will have strong and diverse capacities, and will be halfway to the end of all illness. To quickly bring about this world and make it a good one, it has to be a world that is beneficial, or at least acceptable, to a great majority of people. What laws would make this world beneficial to most, and acceptable to approximately all? We'll have to chew on this question ongoingly. Genomic Liberty is a proposal for one overarching principle, among others, to guide public policy and legislation around germline engineering. It asserts: > Parents have the right to freely choose the genomes of their children. If upheld, genomic liberty protects the rights of parents from regulation by the state. A visual summary of the more detailed structure of the principle of genomic liberty: The principle of genomic liberty has several potential exceptions. These cases aren't strongly protected by genomic liberty, and therefore could potentially be regulated while upholding genomic liberty: * Genomic choices that are clearly prospectively very unsafe. * Genomic choices that remove a core aspect of human nature. * Genomic choices with extreme negative externalities on other citizens. * Nonconsensual use of someone's DNA. * Genomic alterations chosen by parents who are severely non compos mentis. * Genomic choices that would severely permanently impair the child's ability to communicate their state of mind. Within the principle of genomic liberty, there are tentpole principles that apply to a narrower set of cases, but that give stronger protection against regulation. These tentpole principles help to clarify genomic liberty, and to fortify it against inappropriate regulation: Non-intervention liberty. Parents have a very strong right to not genomically intervene on
ecd5166f-01f4-4162-96db-3916896296b5	StampyAI/alignment-research-dataset/alignmentforum	Alignment Forum	Catastrophic Risks from AI #3: AI Race This is the third post in a sequence of posts giving an [overview of catastrophic AI risks](https://arxiv.org/abs/2306.12001). 3 AI Race ========= The immense potential of AIs has created competitive pressures among global players contending for power and influence. This “AI race” is driven by nations and corporations who feel they must rapidly build and deploy AIs to secure their positions and survive. By failing to properly prioritize global risks, this dynamic makes it more likely that AI development will produce dangerous outcomes. Analogous to the nuclear arms race during the Cold War, participation in an AI race may serve individual short-term interests, but it ultimately results in worse collective outcomes for humanity. Importantly, these risks stem not only from the intrinsic nature of AI technology, but from the competitive pressures that encourage insidious choices in AI development. In this section, we first explore the military AI arms race and the corporate AI race, where nation-states and corporations are forced to rapidly develop and adopt AI systems to remain competitive. Moving beyond these specific races, we reconceptualize competitive pressures as part of a broader evolutionary process in which AIs could become increasingly pervasive, powerful, and entrenched in society. Finally, we highlight potential strategies and policy suggestions to mitigate the risks created by an AI race and ensure the safe development of AIs. 3.1 Military AI Arms Race ------------------------- The development of AIs for military applications is swiftly paving the way for a new era in military technology, with potential consequences rivaling those of gunpowder and nuclear arms in what has been described as the “third revolution in warfare." The weaponization of AI presents numerous challenges, such as the potential for more destructive wars, the possibility of accidental usage or loss of control, and the prospect of malicious actors co-opting these technologies for their own purposes. As AIs gain influence over traditional military weaponry and increasingly take on command and control functions, humanity faces a paradigm shift in warfare. In this context, we will discuss the latent risks and implications of this AI arms race on global security, the potential for intensified conflicts, and the dire outcomes that could come as a result, including the possibility of conflicts escalating to a scale that poses an existential threat. ### 3.1.1 Lethal Autonomous Weapons (LAWs) LAWs are weapons that can identify, target, and kill without human intervention [32]. They offer potential improvements in decision-making speed and precision. Warfare, however, is a high-stakes, safety-critical domain for AIs with significant moral and practical concerns. Though their existence is not necessarily a catastrophe in itself, LAWs may serve as an on-ramp to catastrophes stemming from malicious use, accidents, loss of control, or an increased likelihood of war. ![](https://res.cloudinary.com/lesswrong-2-0/image/upload/f_auto,q_auto/v1/mirroredImages/4sEK5mtDYWJo2gHJn/hrzsk6sirbljx7pi7z25)Figure 7: Low-cost automated weapons, such as drone swarms outfitted with explosives, could autonomously hunt human targets with high precision, performing lethal operations for both militaries and terrorist groups and lowering the barriers to large-scale violence.LAWs may become vastly superior to humans. Driven by rapid developments in AIs, weapons systems that can identify, target, and decide to kill human beings on their own—without an officer directing an attack or a soldier pulling the trigger—are starting to transform the future of conflict. In 2020, an advanced AI agent outperformed experienced F-16 pilots in a series of virtual dogfights, including decisively defeating a human pilot 5-0, showcasing “aggressive and precise maneuvers the human pilot couldn't outmatch" [33]. Just as in the past, superior weapons would allow for more destruction in a shorter period of time, increasing the severity of war. Militaries are taking steps toward delegating life-or-death decisions to AIs. Fully autonomous drones were likely first used on the battlefield in Libya in March 2020, when retreating forces were "hunted down and remotely engaged" by a drone operating without human oversight [34]. In May 2021, the Israel Defense Forces used the world's first AI-guided weaponized drone swarm during combat operations, which marks a significant milestone in the integration of AI and drone technology in warfare [35]. Although walking, shooting robots have yet to replace soldiers on the battlefield, technologies are converging in ways that may make this possible in the near future. LAWs increase the likelihood of war. Sending troops into battle is a grave decision that leaders do not make lightly. But autonomous weapons would allow an aggressive nation to launch attacks without endangering the lives of its own soldiers and thus face less domestic scrutiny. While remote-controlled weapons share this advantage, their scalability is limited by the requirement for human operators and vulnerability to jamming countermeasures, limitations that LAWs could overcome [36]. Public opinion for continuing wars tends to wane as conflicts drag on and casualties increase [37]. LAWs would change this equation. National leaders would no longer face the prospect of body bags returning home, thus removing a primary barrier to engaging in warfare, which could ultimately increase the likelihood of conflicts. ### 3.1.2 Cyberwarfare As well as being used to enable deadlier weapons, AIs could lower the barrier to entry for cyberattacks, making them more numerous and destructive. They could cause serious harm not only in the digital environment but also in physical systems, potentially taking out critical infrastructure that societies depend on. While AIs could also be used to improve cyberdefense, it is unclear whether they will be most effective as an offensive or defensive technology [38]. If they enhance attacks more than they support defense, then cyberattacks could become more common, creating significant geopolitical turbulence and paving another route to large-scale conflict. AIs have the potential to increase the accessibility, success rate, scale, speed, stealth, and potency of cyberattacks. Cyberattacks are already a reality, but AIs could be used to increase their frequency and destructiveness in multiple ways. Machine learning tools could be used to find more critical vulnerabilities in target systems and improve the success rate of attacks. They could also be used to increase the scale of attacks by running millions of systems in parallel, and increase the speed by finding novel routes to infiltrating a system. Cyberattacks could also become more potent if used to hijack AI weapons. Cyberattacks can destroy critical infrastructure. By hacking computer systems that control physical processes, cyberattacks could cause extensive infrastructure damage. For example, they could cause system components to overheat or valves to lock, leading to a buildup of pressure culminating in an explosion. Through interferences like this, cyberattacks have the potential to destroy critical infrastructure, such as electric grids and water supply systems. This was demonstrated in 2015, when a cyberwarfare unit of the Russian military hacked into the Ukrainian power grid, leaving over 200,000 people without power access for several hours. AI-enhanced attacks could be even more devastating and potentially deadly for the billions of people who rely on critical infrastructure for survival. Difficulties in attributing AI-driven cyberattacks could increase the risk of war. A cyberattack resulting in physical damage to critical infrastructure would require a high degree of skill and effort to execute, perhaps only within the capability of nation-states. Such attacks are rare as they constitute an act of war, and thus elicit a full military response. Yet AIs could enable attackers to hide their identity, for example if they are used to evade detection systems or more effectively cover the tracks of the attacker [39]. If cyberattacks become more stealthy, this would reduce the threat of retaliation from an attacked party, potentially making attacks more likely. If stealthy attacks do happen, they might incite actors to mistakenly retaliate against unrelated third parties they suspect to be responsible. This could increase the scope of the conflict dramatically. ### 3.1.3 Automated Warfare AIs speed up the pace of war, which makes AIs more necessary. AIs can quickly process a large amount of data, analyze complex situations, and provide helpful insights to commanders. With ubiquitous sensors and advanced technology on the battlefield, there is tremendous incoming information. AIs help make sense of this information, spotting important patterns and relationships that humans might miss. As these trends continue, it will become increasingly difficult for humans to make well-informed decisions as quickly as necessary to keep pace with AIs. This would further pressure militaries to hand over decisive control to AIs. The continuous integration of AIs into all aspects of warfare will cause the pace of combat to become faster and faster. Eventually, we may arrive at a point where humans are no longer capable of assessing the ever-changing battlefield situation and must cede decision-making power to advanced AIs. Automatic retaliation can escalate accidents into war. There is already willingness to let computer systems retaliate automatically. In 2014, a leak revealed to the public that the NSA has a program called MonsterMind, which autonomously detects and blocks cyberattacks on US infrastructure [40]. What was unique, however, was that instead of simply detecting and eliminating the malware at the point of entry, MonsterMind would automatically initiate a retaliatory cyberattack with no human involvement. If multiple combatants have policies of automatic retaliation, an accident or false alarm could quickly escalate to full-scale war before humans intervene. This would be especially dangerous if the superior information processing capabilities of modern AI systems makes it more appealing for actors to automate decisions regarding nuclear launches. ![](https://res.cloudinary.com/lesswrong-2-0/image/upload/f_auto,q_auto/v1/mirroredImages/4sEK5mtDYWJo2gHJn/x1fm4jok5tc8x97ldozn)Figure 8: A military AI arms race could pressure countries into delegating many crucial decisions over armaments to AIs. Integrating AIs into nuclear command and control could heighten the risk of global catastrophe as the potential for accidents and increased pace of war may lead to unintended escalations and confrontations.History shows the danger of automated retaliation. On September 26, 1983, Stanislav Petrov, a lieutenant colonel of the Soviet Air Defense Forces, was on duty at the Serpukhov-15 bunker near Moscow, monitoring the Soviet Union's early warning system for incoming ballistic missiles. The system indicated that the US had launched multiple nuclear missiles toward the Soviet Union. The protocol at the time dictated that such an event should be considered a legitimate attack, and the Soviet Union would respond with a nuclear counterstrike. If Petrov had passed on the warning to his superiors, this would have been the likely outcome. Instead, however, he judged it to be a false alarm and ignored it. It was soon confirmed that the warning had been caused by a rare technical malfunction. If an AI had been in control, the false alarm could have triggered a nuclear war. AI-controlled weapons systems could lead to a flash war. Autonomous systems are not infallible. We have already witnessed how quickly an error in an automated system can escalate in the economy. Most notably, in the 2010 Flash Crash, a feedback loop between automated trading algorithms amplified ordinary market fluctuations into a financial catastrophe in which a trillion dollars of stock value vanished in minutes [41]. If multiple nations were to use AIs to automate their defense systems, an error could be catastrophic, triggering a spiral of attacks and counter-attacks that would happen too quickly for humans to step in—a flash war. The market quickly recovered from the 2010 Flash Crash, but the harm caused by a flash war could be catastrophic. Automated warfare could reduce accountability for military leaders. Military leaders may at times gain an advantage on the battlefield if they are willing to ignore the laws of war. For example, soldiers may be able to mount stronger attacks if they do not take steps to minimize civilian casualties. An important deterrent to this behavior is the risk that military leaders could eventually be held accountable or even prosecuted for war crimes. Automated warfare could reduce this deterrence effect by making it easier for military leaders to escape accountability by blaming violations on failures in their automated systems. AIs could make war more uncertain, increasing the risk of conflict. Although states that are already wealthier and more powerful often have more resources to invest in new military technologies, they are not necessarily always the most successful at adopting them. Other factors also play an important role, such as how agile and adaptive a military can be in incorporating new technologies [42]. Major new weapons innovations can therefore offer an opportunity for existing superpowers to bolster their dominance, but also for less powerful states to quickly increase their power by getting ahead in an emerging and important sphere. This can create significant uncertainty around if and how the balance of power is shifting, potentially leading states to incorrectly believe they could gain something from going to war. Even aside from considerations regarding the balance of power, rapidly evolving automated warfare would be unprecedented, making it difficult for actors to evaluate their chances of victory in any particular conflict. This would increase the risk of miscalculation, making war more more likely. ### 3.1.4 Actors May Risk Extinction Over Individual Defeat > “I know not with what weapons World War III will be fought, but World War IV will be fought with sticks and stones." - Einstein > > Competitive pressures make actors more willing to accept the risk of extinction. During the Cold War, neither side desired the dangerous situation they found themselves in. There were widespread fears that nuclear weapons could be powerful enough to wipe out a large fraction of humanity, potentially even causing extinction—a catastrophic result for both sides. Yet the intense rivalry and geopolitical tensions between the two superpowers fueled a dangerous cycle of arms buildup. Each side perceived the other's nuclear arsenal as a threat to its very survival, leading to a desire for parity and deterrence. The competitive pressures pushed both countries to continually develop and deploy more advanced and destructive nuclear weapons systems, driven by the fear of being at a strategic disadvantage. During the Cuban Missile Crisis, this led to the brink of nuclear war. Even though the story of Arkhipov preventing the launch of a nuclear torpedo wasn't declassified until decades after the incident, President John F. Kennedy reportedly estimated that he thought the odds of nuclear war beginning during that time were "somewhere between one out of three and even." This chilling admission highlights how the competitive pressures between militaries have the potential to cause global catastrophes. Individually rational decisions can be collectively catastrophic. Nations locked in competition might make decisions that advance their own interests by putting the rest of the world at stake. Scenarios of this kind are collective action problems, where decisions may be rational on an individual level yet disastrous for the larger group [43]. For example, corporations and individuals may weigh their own profits and convenience over the negative impacts of the emissions they create, even if those emissions collectively result in climate change. The same principle can be extended to military strategy and defense systems. Military leaders might estimate, for instance, that increasing the autonomy of weapon systems would mean a 10 percent chance of losing control over weaponized superhuman AIs. Alternatively, they might estimate that using AIs to automate bioweapons research could lead to a 10 percent chance of leaking a deadly pathogen. Both of these scenarios could lead to catastrophe or even extinction. The leaders may, however, also calculate that refraining from these developments will mean a 99 percent chance of losing a war against an opponent. Since conflicts are often viewed as existential struggles by those fighting them, rational actors may accept an otherwise unthinkable 10 percent chance of human extinction over a 99 percent chance of losing a war. Regardless of the particular nature of the risks posed by advanced AIs, these dynamics could push us to the brink of global catastrophe. Technological superiority does not guarantee national security. It is tempting to think that the best way of guarding against enemy attacks is to improve one's own military prowess. However, in the midst of competitive pressures, all parties will tend to advance their weaponry, such that no one gains much of an advantage, but all are left at greater risk. As Richard Danzig, former Secretary of the Navy, has observed, "On a number of occasions and in a number of ways, the American national security establishment will lose control of what it creates... deterrence is a strategy for reducing attacks, not accidents; it discourages malevolence, not inadvertence" [44]. Cooperation is paramount to reducing risk. As discussed above, an AI arms race can lead us down a hazardous path, despite this being in no country's best interest. It is important to remember that we are all on the same side when it comes to existential risks, and working together to prevent them is a collective necessity. A destructive AI arms race benefits nobody, so all actors would be rational to take steps to cooperate with one another to prevent the riskiest applications of militarized AIs. We have considered how competitive pressures could lead to the increasing automation of conflict, even if decision-makers are aware of the existential threat that this path entails. We have also discussed cooperation as being the key to counteracting and overcoming this collective action problem. We will now illustrate a hypothetical path to disaster that could result from an AI arms race. ### Story: Automated Warfare As AI systems become increasingly sophisticated, militaries start involving them in decision-making processes. Officials give them military intelligence about opponents' arms and strategies, for example, and ask them to calculate the most promising plan of action. It soon becomes apparent that AIs are reliably reaching better decisions than humans, so it seems sensible to give them more influence. At the same time, international tensions are rising, increasing the threat of war. A new military technology has recently been developed that could make international attacks swifter and stealthier, giving targets less time to respond. Since military officials feel their response processes take too long, they fear that they could be vulnerable to a surprise attack capable of inflicting decisive damage before they would have any chance to retaliate. Since AIs can process information and make decisions much more quickly than humans, military leaders reluctantly hand them increasing amounts of retaliatory control, reasoning that failing to do so would leave them open to attack from adversaries. While for years military leaders had stressed the importance of keeping a "human in the loop" for major decisions, human control is nonetheless gradually phased out in the interests of national security. Military leaders understand that their decisions lead to the possibility of inadvertent escalation caused by system malfunctions, and would prefer a world where all countries automated less; but they do not trust that their adversaries will refrain from automation. Over time, more and more of the chain of command is automated on all sides. One day, a single system malfunctions, detecting an enemy attack when there is none. The system is empowered to launch an instant "retaliatory" attack, and it does so in the blink of an eye. The attack causes automated retaliation from the other side, and so on. Before long, the situation is spiraling out of control, with waves of automated attack and retaliation. Although humans have made mistakes leading to escalation in the past, this escalation between mostly-automated militaries happens far more quickly than any before. The humans who are responding to the situation find it difficult to diagnose the source of the problem, as the AI systems are not transparent. By the time they even realize how the conflict started, it is already over, with devastating consequences for both sides. 3.2 Corporate AI Race --------------------- Competitive pressures exist in the economy, as well as in military settings. Although competition between companies can be beneficial, creating more useful products for consumers, there are also pitfalls. First, the benefits of economic activity may be unevenly distributed, incentivizing those who benefit most from it to disregard the harms to others. Second, under intense market competition, businesses tend to focus much more on short-term gains than on long-term outcomes. With this mindset, companies often pursue something that can make a lot of profit in the short term, even if it poses a societal risk in the long term. We will now discuss how corporate competitive pressures could play out with AIs and the potential negative impacts. ### 3.2.1 Economic Competition Undercuts Safety Competitive pressure is fueling a corporate AI race. To obtain a competitive advantage, companies often race to offer the first products to a market rather than the safest. These dynamics are already playing a role in the rapid development of AI technology. At the launch of Microsoft's AI-powered search engine in February 2023, the company's CEO Satya Nadella said, "A race starts today... we're going to move fast." Only weeks later, the company's chatbot was shown to have threatened to harm users [45]. In an internal email, Sam Schillace, a technology executive at Microsoft, highlighted the urgency in which companies view AI development. He wrote that it would be an "absolutely fatal error in this moment to worry about things that can be fixed later" [46]. > “Nothing can be done at once hastily and prudently.” - Publius Syrus > > Competitive pressures have contributed to major commercial and industrial disasters. In 1970, Ford Motor Company introduced the Ford Pinto, a new car model with a serious safety problem: the gas tank was located near the rear bumper. Safety tests showed that during a car crash, the fuel tank would often explode and set the car ablaze. Ford identified the problem and calculated that it would cost $11 per car to fix. They decided that this was too expensive and put the car on the market, resulting in numerous fatalities and injuries caused by fire when crashes inevitably happened [47]. Ford was sued and a jury found them liable for these deaths and injuries [48]. The verdict, of course, came too late for those who had already lost their lives. Ford's president at the time explained the decision, saying, "Safety doesn't sell" [49]. A more recent example of the dangers of competitive pressure is the case of the Boeing 737 Max aircraft. Boeing, aiming to compete with its rival Airbus, sought to deliver an updated, more fuel-efficient model to the market as quickly as possible. The head-to-head rivalry and time pressure led to the introduction of the Maneuvering Characteristics Augmentation System, which was designed to enhance the aircraft's stability. However, inadequate testing and pilot training ultimately resulted in the two fatal crashes only months apart, with 346 people killed [50]. We can imagine a future in which similar pressures lead companies to cut corners and release unsafe AI systems. A third example is the Bhopal gas tragedy, which is widely considered to be the worst industrial disaster ever to have happened. In December 1984, a vast quantity of toxic gas leaked from a Union Carbide Corporation subsidiary plant manufacturing pesticides in Bhopal, India. Exposure to the gas killed thousands of people and injured up to half a million more. Investigations found that, in the run-up to the disaster, safety standards had fallen significantly, with the company cutting costs by neglecting equipment maintenance and staff training as profitability fell. This is often considered a consequence of competitive pressures [51]. Competition incentivizes businesses to deploy potentially unsafe AI systems. In an environment where businesses are rushing to develop and release products, those that follow rigorous safety procedures will be slower and risk being out-competed. Ethically-minded AI developers, who want to proceed more cautiously and slow down, would give more unscrupulous developers an advantage. In trying to survive commercially, even the companies that want to take more care are likely to be swept along by competitive pressures. There may be attempts to implement safety measures, but with more of an emphasis on capabilities than on safety, these may be insufficient. This could lead us to develop highly powerful AIs before we properly understand how to ensure they are safe. ### 3.2.2 Automated Economy Corporations will face pressure to replace humans with AIs. As AIs become more capable, they will be able to perform an increasing variety of tasks more quickly, cheaply, and effectively than human workers. Companies will therefore stand to gain a competitive advantage from replacing their employees with AIs. Companies that choose not to adopt AIs would likely be out-competed, just as a clothing company using manual looms would be unable to keep up with those using industrial ones. ![](https://res.cloudinary.com/lesswrong-2-0/image/upload/f_auto,q_auto/v1/mirroredImages/4sEK5mtDYWJo2gHJn/tivs4l8ndwuajorsvzix)Figure 9: As AIs automate increasingly many tasks, the economy may become largely run by AIs. Eventually, this could lead to human enfeeblement and dependence on AIs for basic needs.AIs could lead to mass unemployment. Economists have long considered the possibility that machines will replace human labor. Nobel Prize winner Wassily Leontief said in 1952 that, as technology advances, "Labor will become less and less important... more and more workers will be replaced by machines" [52]. Previous technologies have augmented the productivity of human labor. AIs, however, could differ profoundly from previous innovations. Human-level AI would, by definition, be able to do everything a human could do. These AIs would also have important advantages over human labor. They could work 24 hours a day, be copied many times and run in parallel, and process information much more quickly than a human would. While we do not know when this will occur, it is unwise to discount the possibility that it could be soon. If human labor is replaced by AIs, mass unemployment could dramatically increase inequality, making individuals dependent on the owners of AI systems. Advanced AIs capable of automating human labor should be regarded not merely as tools, but as agents. One particularly concerning aspect of AI agents is their potential to automate research and development across various fields, including biotechnology or even AI itself. This phenomenon is already occurring [53], and could lead to AI capabilities growing at increasing rates, to the point where humans are no longer the driving force behind AI development. If this trend continues unchecked, it could escalate risks associated with AIs progressing faster than our capacity to manage and regulate them, especially in areas like biotechnology where the malicious use of advancements could pose significant dangers. It is crucial that we strive to prevent undue acceleration of R&D and maintain a strong human-centric approach to technological development. Conceding power to AIs could lead to human enfeeblement. Even if we ensure that the many unemployed humans are provided for, we may find ourselves completely reliant on AIs. This would likely emerge not from a violent coup by AIs, but from a gradual slide into dependence. As society's challenges become ever more complex and fast-paced, and as AIs become ever more intelligent and quick-thinking, we may forfeit more and more functions to them out of convenience. In such a state, the only feasible solution to the complexities and challenges compounded by AIs may be to rely even more heavily on AIs. This gradual process could eventually lead to the delegation of nearly all intellectual, and eventually physical, labor to AIs. In such a world, people might have few incentives to gain knowledge and cultivate skills, potentially leading to a state of enfeeblement. Having lost our know-how and our understanding of how civilization works, we would become completely dependent on AIs, a scenario not unlike the one depicted in the film WALL-E. In such a state, humanity is not flourishing and is no longer in effective control—an outcome that many people would consider a permanent catastrophe [54]. As we have seen, there are classic game-theoretic dilemmas where individuals and groups face incentives that are incompatible with what would make everyone better off. We see this with a military AI arms race, where the world is made less safe by creating extremely powerful AI weapons, and we see this in a corporate AI race, where an AI's power and development is prioritized over its safety. To address these dilemmas that give rise to global risks, we will need new coordination mechanisms and institutions. It is our view that failing to coordinate and stop AI races would be the most likely cause of an existential catastrophe. 3.3 Evolution ------------- As discussed above, there are strong pressures to replace humans with AIs, cede more control to them, and reduce human oversight in various settings, despite the potential harms. We can re-frame this as a general trend resulting from evolutionary dynamics, where an unfortunate truth is that AIs will simply be more fit than humans. Extrapolating this pattern of automation, it is likely that we will build an ecosystem of competing AIs over which it may be difficult to maintain control in the long run. We will now discuss how natural selection influences the development of AI systems and why evolution favors selfish behaviors. We will also look at how competition might arise and play out between AIs and humans, and how this could create catastrophic risks. This section draws heavily from "Natural Selection Favors AIs over Humans" [55, 56]. Fitter technologies are selected, for good and bad. While most people think of evolution by natural selection as a biological process, its principles shape much more. According to the evolutionary biologist Richard Lewontin [57], evolution by natural selection will take hold in any environment where three conditions are present: 1) there are differences between individuals; 2) characteristics are passed onto future generations and; 3) the different variants propagate at different rates. These conditions apply to various technologies. Consider the content-recommendation algorithms used by streaming services and social media platforms. When a particularly addictive content format or algorithm hooks users, it results in higher screen time and engagement. This more effective content format or algorithm is consequently "selected" and further fine-tuned, while formats and algorithms that fail to capture attention are discontinued. These competitive pressures foster a "survival of the most addictive" dynamic. Platforms that refuse to use addictive formats and algorithms become less influential or are simply outcompeted by platforms that do, leading competitors to undermine wellbeing and cause massive harm to society [58]. ![](https://res.cloudinary.com/lesswrong-2-0/image/upload/f_auto,q_auto/v1/mirroredImages/4sEK5mtDYWJo2gHJn/roihc2zugy7mgh09xula)Figure 10: Evolutionary pressures are responsible for various developments over time, and are not limited to the realm of biology.The conditions for natural selection apply to AIs. There will be many different AI systems with varying features and capabilities, and competition between them will determine which characteristics become more common. The most successful AIs today are already being used as a basis for their developers' next generation of models, as well as being imitated by rival companies. Factors determining which AIs propagate the most may include their ability to act autonomously, automate labor, or reduce the chance of their own deactivation. Natural selection often favors selfish characteristics. Natural selection influences which AIs propagate most widely. From biological systems, we see that natural selection often gives rise to selfish behaviors that promote one's own genetic information: chimps attack other communities [59], lions engage in infanticide [60], viruses evolve new surface proteins to deceive and bypass defense barriers [61], humans engage in nepotism, some ants enslave others [62], and so on. In the natural world, selfishness often emerges as a dominant strategy; those that prioritize themselves and those similar to them are usually more likely to survive, so these traits become more prevalent. Amoral competition can select for traits that we think are immoral. Selfish behaviors may not be malicious or even intentional. Species in the natural world do not evolve selfish traits deliberately or consciously. Selfish traits emerge as a product of competitive pressures. Similarly, AIs do not have to be malicious to act selfishly—instead, they would evolve selfish traits as an adaptation to their environment. AIs might engage in selfish behavior—expanding their influence at the expense of humans—simply by automating human jobs. AIs do not intend to displace humans. Rather, the environment in which they are being developed, namely corporate AI labs, is pressuring AI researchers to select for AIs that automate and displace humans. Another example of unintentional selfish behavior is when AIs assume roles humans depend on. AIs may eventually become enmeshed in vital infrastructure such as power grids or the internet. Many people may then be unwilling to accept the cost of being able to effortlessly deactivate them, as that would pose a reliability hazard. Similarly, AI companions may induce people to become emotionally dependent on them. Some of those people may even begin to argue that their AI companions should have rights. If some AIs are given rights, they may operate, adapt, and evolve outside of human control. AIs could become embedded in human society and expand their influence over us in ways that we can't easily reverse. Selfish behaviors may erode safety measures that some of us implement. AIs that gain influence and provide economic value will predominate, while AIs that adhere to the most constraints will be less competitive. For example, AIs following the constraint "never break the law" have fewer options than AIs following the constraint "don't get caught breaking the law." AIs of the latter type may be willing to break the law if they're unlikely to be caught or if the fines are not severe enough, allowing them to outcompete more restricted AIs. Many businesses follow laws, but in situations where stealing trade secrets or deceiving regulators is highly lucrative and difficult to detect, a business that is willing to engage in such selfish behavior can have an advantage over its more principled competitors. An AI system might be prized for its ability to achieve ambitious goals autonomously. It might, however, be achieving its goals efficiently without abiding by ethical restrictions, while deceiving humans about its methods. Even if we try to put safety measures in place, a deceptive AI would be very difficult to counteract if it is cleverer than us. AIs that can bypass our safety measures without detection may be the most successful at accomplishing the tasks we give them, and therefore become widespread. These processes could culminate in a world where many aspects of major companies and infrastructure are controlled by powerful AIs with selfish traits, including deceiving humans, harming humans in service of their goals, and preventing themselves from being deactivated. Humans only have nominal influence over AI selection. One might think we could avoid the development of selfish behaviors by ensuring we do not select AIs that exhibit them. However, the companies developing AIs are not selecting the safest path but instead succumbing to evolutionary pressures. One example is OpenAI, which was founded as a nonprofit in 2015 to "benefit humanity as a whole, unconstrained by a need to generate financial return" [63]. However, when faced with the need to raise capital to keep up with better-funded rivals, in 2019 OpenAI transitioned from a nonprofit to “capped-profit” structure [64]. Later, many of the safety-focused OpenAI employees left and formed a competitor, Anthropic, that was to focus more heavily on AI safety than OpenAI had. Although Anthropic originally focused on safety research, they eventually became convinced of the "necessity of commercialization" and now contribute to competitive pressures [65]. While many of the employees at those companies genuinely care about safety, these values do not stand a chance against evolutionary pressures, which compel companies to move ever more hastily and seek ever more influence, lest the company perish. Moreover, AI developers are already selecting AIs with increasingly selfish traits. They are selecting AIs to automate and displace humans, make humans highly dependent on AIs, and make humans more and more obsolete. By their own admission, future versions of these AIs may lead to extinction [66]. This is why an AI race is insidious: AI development is not being aligned with human values but rather with evolution. People often choose the products that are most useful and convenient to them immediately, rather than thinking about potential long-term consequences, even to themselves. An AI race puts pressures on companies to select the AIs that are most competitive, not the least selfish. Even if it's feasible to select for unselfish AIs, if it comes at a clear cost to competitiveness, some competitors will select the selfish AIs. Furthermore, as we have mentioned, if AIs develop strategic awareness, they may counteract our attempts to select against them. Moreover, as AIs increasingly automate various processes, AIs will affect the competitiveness of other AIs, not just humans. AIs will interact and compete with each other, and some will be put in charge of the development of other AIs at some point. Giving AIs influence over which other AIs should be propagated and how they should be modified would represent another step toward human becoming dependent on AIs and AI evolution becoming increasingly independent from humans. As this continues, the complex process governing AI evolution will become further unmoored from human interests. AIs can be more fit than humans. Our unmatched intelligence has granted us power over the natural world. It has enabled us to land on the moon, harness nuclear energy, and reshape landscapes at our will. It has also given us power over other species. Although a single unarmed human competing against a tiger or gorilla has no chance of winning, the collective fate of these animals is entirely in our hands. Our cognitive abilities have proven so advantageous that, if we chose to, we could cause them to go extinct in a matter of weeks. Intelligence was a key factor that led to our dominance, but we are currently standing on the precipice of creating entities far more intelligent than ourselves. Given the exponential increase in microprocessor speeds, AIs have the potential to process information and "think" at a pace that far surpasses human neurons, but it could be even more dramatic than the speed difference between humans and sloths. They can assimilate vast quantities of data from numerous sources simultaneously, with near-perfect retention and understanding. They do not need to sleep and they do not get bored. Due to the scalability of computational resources, an AI could interact and cooperate with an unlimited number of other AIs, potentially creating a collective intelligence that would far outstrip human collaborations. AIs could also deliberately update and improve themselves. Without the same biological restrictions as humans, they could adapt and therefore evolve unspeakably quickly compared with us. AIs could become like an invasive species, with the potential to out-compete humans. Our only advantage over AIs is that we get to get make the first moves, but given the frenzied AI race we are rapidly giving up even this advantage. AIs would have little reason to cooperate with or be altruistic toward humans. Cooperation and altruism evolved because they increase fitness. There are numerous reasons why humans cooperate with other humans, like direct reciprocity. Also known as "quid pro quo," direct reciprocity can be summed up by the idiom "you scratch my back, I'll scratch yours." While humans would initially select AIs that were cooperative, the natural selection process would eventually go beyond our control, once AIs were in charge of many or most processes, and interacting predominantly with one another. At that point, there would be little we could offer AIs, given that they will be able to "think" at least hundreds of times faster than us. Involving us in any cooperation or decision-making processes would simply slow them down, giving them no more reason to cooperate with us than we do with gorillas. It might be difficult to imagine a scenario like this or to believe we would ever let it happen. Yet it may not require any conscious decision, instead arising as we allow ourselves to gradually drift into this state without realizing that human-AI co-evolution may not turn out well for humans. AIs becoming more powerful than humans could leave us highly vulnerable. As the most dominant species, humans have deliberately harmed many other species, and helped drive species such as Neanderthals to extinction. In many cases, the harm was not even deliberate, but instead a result of us merely prioritizing our goals over their wellbeing. To harm humans, AIs wouldn't need to be any more genocidal than someone removing an ant colony on their front lawn. If AIs are able to control the environment more effectively than we can, they could treat us with the same disregard. Conceptual summary. Evolutionary forces could cause the most influential future AI agents to have selfish tendencies. That is because: 1. Evolution by natural selection gives rise to selfish behavior. While evolution can result in altruistic behavior in rare situations, the context of AI development does not promote altruistic behavior. 2. Natural selection may be a dominant force in AI development. The intensity of evolutionary pressure will be high if AIs adapt rapidly or if competitive pressures are intense. Competition and selfish behaviors may dampen the effects of human safety measures, leaving the surviving AI designs to be selected naturally. If so, AI agents would have many selfish tendencies. The winner of the AI race would not be a nation-state, not a corporation, but AIs themselves. The upshot is that the AI ecosystem would eventually stop evolving on human terms, and we would become a displaced, second-class species. ### Story: Autonomous Economy As AIs become more capable, people realize that we could work more efficiently by delegating some simple tasks to them, like drafting emails. Over time, people notice that the AIs are doing these tasks more quickly and effectively than any human could, so it is convenient to give them more jobs with less and less supervision. Competitive pressures accelerate the expansion of AI use, as companies can gain an advantage over rivals by automating whole processes or departments with AIs, which perform better than humans and cost less to employ. Other companies, faced with the prospect of being out-competed, feel compelled to follow suit just to keep up. At this point, natural selection is already at work among AIs; humans choose to make more of the best-performing models and unwittingly propagate selfish traits such as deception and self-preservation if these confer a fitness advantage. For example, AIs that are charming and foster personal relationships with humans become widely copied and harder to remove. As AIs are put in charge of more and more decisions, they are increasingly interacting with one another. Since they can evaluate information much more quickly than humans, activity in most spheres accelerates. This creates a feedback loop: since business and economic developments are too fast-moving for humans to follow, it makes sense to cede yet more control to AIs instead, pushing humans further out of important processes. Ultimately, this leads to a fully autonomous economy, governed by an increasingly uncontrolled ecosystem of AIs. At this point, humans have few incentives to gain any skills or knowledge, because almost everything would be taken care of by much more capable AIs. As a result, we eventually lose the capacity to look after and govern ourselves. Additionally, AIs become convenient companions, offering social interaction without requiring the reciprocity or compromise necessary in human relationships. Humans interact less and less with one another over time, losing vital social skills and the ability to cooperate. People become so dependent on AIs that it would be intractable to reverse this process. What's more, as some AIs become more intelligent, some people are convinced these AIs should be given rights, meaning turning off some AIs is no longer a viable option. Competitive pressures between the many interacting AIs continue to select for selfish behaviors, though we might be oblivious to this happening, as we have already acquiesced much of our oversight. If these clever, powerful, self-preserving AIs were then to start acting in harmful ways, it would be all but impossible to deactivate them or regain control. AIs have supplanted humans as the most dominant species and their continued evolution is far beyond our influence. Their selfish traits eventually lead them to pursue their goals without regard for human wellbeing, with catastrophic consequences. Suggestions ----------- Mitigating the risks from competitive pressures will require a multifaceted approach, including regulations, limiting access to powerful AI systems, and multilateral cooperation between stakeholders at both the corporate and nation-state level. We will now outline some strategies for promoting safety and reducing race dynamics. Safety regulation. Regulation holds AI developers to a common standard so that they do not cut corners on safety. While regulation does not itself create technical solutions, it can create strong incentives to develop and implement those solutions. If companies cannot sell their products without certain safety measures, they will be more willing to develop those measures, especially if other companies are also held to the same standards. Even if some companies voluntarily self-regulate, government regulation can help prevent less scrupulous actors from cutting corners on safety. Regulation must be proactive, not reactive. A common saying is that aviation regulations are "written in blood"—but regulators should develop regulations before a catastrophe, not afterward. Regulations should be structured so that they only create competitive advantages for companies with higher safety standards, rather than companies with more resources and better attorneys. Regulators should be independently staffed and not dependent on any one source of expertise (for example, large companies), so that they can focus on their mission to regulate for the public good without undue influence. Data documentation. To ensure transparency and accountability in AI systems, companies should be required to justify and report the sources of data used in model training and deployment. Decisions by companies to use datasets that include hateful content or personal data contribute to the frenzied pace of AI development and undermine accountability. Documentation should include details regarding the motivation, composition, collection process, uses, and maintenance of each dataset [67]. Meaningful human oversight of AI decisions. While AI systems may grow capable of assisting human beings in making important decisions, AI decision-making should not be made fully autonomous, as the inner workings of AIs are inscrutable, and while they can often give reasonable results, they fail to give highly reliable results [68]. It is crucial that actors are vigilant to coordinate on maintaining these standards in the face of future competitive pressures. By keeping humans in the loop on key decisions, irreversible decisions can be double-checked and foreseeable errors can be avoided. One setting of particular concern is nuclear command and control. Nuclear-armed countries should continue to clarify domestically and internationally that the decision to launch a nuclear weapon must always be made by a human. AI for cyberdefense. Risks resulting from AI-powered cyberwarfare would be reduced if cyberattacks became less likely to succeed. Deep learning can be used to improve cyberdefense and reduce the impact and success rate of cyberattacks. For example, improved anomaly detection could help detect intruders, malicious programs, or abnormal software behavior [69]. International coordination. International coordination can encourage different nations to uphold high safety standards with less worry that other nations will undercut them. Coordination could be accomplished via informal agreements, international standards, or international treaties regarding the development, use, and monitoring of AI technologies. The most effective agreements would be paired with robust verification and enforcement mechanisms. Public control of general-purpose AIs. The development of AI poses risks that may never be adequately accounted for by private actors. In order to ensure that externalities are properly accounted for, direct public control of general-purpose AI systems may eventually be necessary. For example, nations could collaborate on a single effort to develop advanced AIs and ensure their safety, similar to how CERN serves as a unified effort for researching particle physics. Such an effort would reduce the risk of nations spurring an AI arms race. Positive Vision: ---------------- In an ideal scenario, AIs would be developed, tested, and subsequently deployed only when the catastrophic risks they pose are negligible and well-controlled. There would be years of time testing, monitoring, and societal integration of new AI systems before beginning work on the next generation. Experts would have a full awareness and understanding of developments in the field, rather than being entirely unable to keep up with a deluge of research. The pace of research advancement would be determined through careful analysis, not frenzied competition. All AI developers would be confident in the responsibility and safety of the others and not feel the need to cut corners. References ========== [32] Paul Scharre. Army of None: Autonomous Weapons and The Future of War. Norton, 2018. [33] DARPA. “AlphaDogfight Trials Foreshadow Future of Human-Machine Symbiosis”. In: (2020). [34] Panel of Experts on Libya. Letter dated 8 March 2021 from the Panel of Experts on Libya established pursuant to resolution 1973 (2011) addressed to the President of the Security Council. United Nations Security Council Document S/2021/229. United Nations, Mar. 2021. [35] David Hambling. Israel used world’s first AI-guided combat drone swarm in Gaza attacks. 2021. [36] Zachary Kallenborn. Applying arms-control frameworks to autonomous weapons. en-US. Oct. 2021. [37] J.E. Mueller. War, Presidents, and Public Opinion. UPA book. University Press of America, 1985. [38] Matteo E. Bonfanti. “Artificial intelligence and the offense–defense balance in cyber security”. In: Cyber Security Politics: Socio-Technological Transformations and Political Fragmentation. Ed. by M.D. Cavelty and A. Wenger.CSS Studies in Security and International Relations. Taylor & Francis, 2022. Chap. 5, pp. 64–79. [39] Yisroel Mirsky et al. “The Threat of Offensive AI to Organizations”. In: Computers & Security (2023). [40] Kim Zetter. “Meet MonsterMind, the NSA Bot That Could Wage Cyberwar Autonomously”. In: Wired (Aug.2014). [41] Andrei Kirilenko et al. “The Flash Crash: High-Frequency Trading in an Electronic Market”. In: The Journal of Finance 72.3 (2017), pp. 967–998. [42] Michael C Horowitz. The Diffusion of Military Power: Causes and Consequences for International Politics. Princeton University Press, 2010. [43] Robert E. Jervis. “Cooperation under the Security Dilemma”. In: World Politics 30 (1978), pp. 167–214. [44] Richard Danzig. Technology Roulette: Managing Loss of Control as Many Militaries Pursue Technological Superiority. Tech. rep. Center for a New American Security, June 2018. [45] Billy Perrigo. Bing’s AI Is Threatening Users. That’s No Laughing Matter. en. Feb. 2023. [46] Nico Grant and Karen Weise. “In A.I. Race, Microsoft and Google Choose Speed Over Caution”. en-US. In: The New York Times (Apr. 2023). [47] Lee Strobel. Reckless Homicide?: Ford’s Pinto Trial. en. And Books, 1980. [48] Grimshaw v. Ford Motor Co. May 1981. [49] Paul C. Judge. “Selling Autos by Selling Safety”. en-US. In: The New York Times (Jan. 1990). [50] Theo Leggett. “737 Max crashes: Boeing says not guilty to fraud charge”. en-GB. In: BBC News (Jan. 2023). [51] Edward Broughton. “The Bhopal disaster and its aftermath: a review”. In: Environmental Health 4.1 (May 2005),p. 6. [52] Charlotte Curtis. “Machines vs. Workers”. en-US. In: The New York Times (Feb. 1983). [53] Thomas Woodside et al. “Examples of AI Improving AI”. In: (2023). URL: https://ai- [improvingai.safe.ai](http://improvingai.safe.ai). [54] Stuart Russell. Human Compatible: Artificial Intelligence and the Problem of Control. en. Penguin, Oct. 2019. [55] Dan Hendrycks. “Natural Selection Favors AIs over Humans”. In: ArXiv abs/2303.16200 (2023). [56] Dan Hendrycks. The Darwinian Argument for Worrying About AI. en. May 2023. [57] Richard C. Lewontin. “The Units of Selection”. In: Annual Review of Ecology, Evolution, and Systematics (1970), pp. 1–18. [58] Ethan Kross et al. “Facebook use predicts declines in subjective well-being in young adults”. In: PloS one (2013). [59] Laura Martínez-Íñigo et al. “Intercommunity interactions and killings in central chimpanzees (Pan troglodytes troglodytes) from Loango National Park, Gabon”. In: Primates; Journal of Primatology 62 (2021), pp. 709–722. [60] Anne E Pusey and Craig Packer. “Infanticide in Lions: Consequences and Counterstrategies”. In: Infanticide and parental care (1994), p. 277. [61] Peter D. Nagy and Judit Pogany. “The dependence of viral RNA replication on co-opted host factors”. In: Nature Reviews. Microbiology 10 (2011), pp. 137–149. [62] Alfred Buschinger. “Social Parasitism among Ants: A Review”. In: Myrmecological News 12 (Sept. 2009), pp. 219–235. [63] Greg Brockman, Ilya Sutskever, and OpenAI. Introducing OpenAI. Dec. 2015. [64] Devin Coldewey. OpenAI shifts from nonprofit to ‘capped-profit’ to attract capital. Mar. 2019. [65] Kyle Wiggers, Devin Coldewey, and Manish Singh. Anthropic’s $5B, 4-year plan to take on OpenAI. Apr. 2023. [66] Center for AI Safety. Statement on AI Risk (“Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”) 2023. URL: <https://www.safe.ai/statement-on-ai-risk>. [67] Timnit Gebru et al. “Datasheets for datasets”. en. In: \Communications of the ACM 64.\12 (Dec. 2021), pp. 86–92. [68] Christian Szegedy et al. “Intriguing properties of neural networks”. In: CoRR (Dec. 2013). [69] Dan Hendrycks et al. “Unsolved Problems in ML Safety”. In: arXiv preprint arXiv:2109.13916 (2021).
4a60d9f3-0f66-4de2-b475-1a59788c0282	trentmkelly/LessWrong-43k	LessWrong	Second Insight: Repairing My "Repairs" or Aspiring to Rationality instead of "Rationality"
76df042d-4974-4bdf-855e-1851a7cd3bff	StampyAI/alignment-research-dataset/eaforum	Effective Altruism Forum	Misha Yagudin and Ozzie Gooen Discuss LLMs and Effective Altruism Misha and I recently recorded a short discussion about large language models and their uses for effective altruists. This was mostly a regular Zoom meeting, but we added some editing and text transcription. After we wrote up the transcript, both Misha and myself edited our respective sections. I think the final transcript is clearer and contains more information than the original discussion. I might even suggest using text-to-speech on the transcript rather than listening to the original audio. This back-and-forth might seem to ruin the point of presenting the video and audio, but I think it might be straightforwardly more pragmatic. Transcript ---------- ### Sections * Opening * Introduction * How do we use LLMs already? * Could EAs contributing to applied LLMs be harmful? * Potential LLM Application: Management and Emotional Assistance * Potential LLM Application: Communication, Broadly * Aside: Human-AI-Human Communication * Potential LLM Application: Decision Automation * Potential LLM Application: EA Forum Improvements * Potential LLM Application: Evaluations * LLM user interfaces * What should EAs do with LLMs? ### Opening Ozzie: Hello. I just did a recording with my friend Misha, an EA researcher at [ARB Research](https://arbresearch.com/). This was a pretty short meeting about large language models and their use by effective altruists. The two of us are pretty excited about the potential for large language models to be used by effective altruists for different kinds of infrastructure. This is an experiment with us presenting videos publicly. Normally, our videos are just Zoom meetings. If anything, the Zoom meetings would be unedited. I found that to be quite a pain. These Zoom meetings typically don't look that great on their own, and they don't sound too terrific. So we've been experimenting with some methods to try to make that a little bit better. I am really curious about what people are going to think about this and am looking forward to what you say. Let's get right into it. ### Introduction Ozzie: For those watching us, this is Misha and me just having a meeting about large language models and their use for effective altruism. Obviously, large language models have been a very big deal very recently, and now there's a big question about how we could best apply them to EA purposes and what EAs could do best about it. So this is going to be a very quick meeting. We only have about half an hour. Right now, we have about seven topics. The main topic, though, is just the LLM applications. ### How do we use LLMs already? Ozzie: So, how do we use LLMs already? Misha: I think I use them for roughly 10 minutes on average per day. Sometimes I just ask questions or ask queries like, "Hey, I have these ingredients. What cocktails can I make?" Sometimes I try to converse with them about stuff. Sometimes I just use it (e.g., text-davinci-003) as a source of knowledge. I think it's more suitable for areas where verifiable expertise is rare. Take non-critical medicine, like skincare. I had a chat with it and got some recommendations in this domain, and I think it turned out really well. I previously tried to search for recommendations and asked a few people, but it didn't work. I also use it as an amplification for journaling whenever I'm doing any emotional or self-coaching work. Writing is great. I personally find it much easier to write “as if” I'm writing a message to someone—having ChatGPT obviously helps with that. Having a conversation partner activates some sort of social infrastructure in my brain. Humans solve math problems better when they are framed socially. And yeah, doing it with language models is straightforward and really good. Further, sometimes models give you hints or insights that you forgot to think about. You can ask tons of questions and be really annoying, which you might not be comfortable doing with your friend or even with a professional. One trick I used previously to prompt a human medical professional to give me some sort of cost-benefit analysis was to ask them: "Well, if instead of me, you were giving an advice to your son or daughter, what would you do?" This makes them actually think. With LLMs, you can probe straightforwardly and see where their limits are. Ozzie, what about you? Ozzie: Yeah, I have used it in a few cases, particularly to rewrite Facebook posts. Both to clean them up, also to try to change them into Tweets. This had some success. I think rewriting in different styles is quite nice. I've been experimenting with a lot of semi-EA purposes. So some of it is like taking very aggressive writing and rewriting it for the EA Forum. And then, ChatGPT knows that writing for the EA Forum means making it super polite and long. I'm having trouble getting it to come up with cool metaphors and witty, historic examples. There are a few cases where I tried coming up with a list of 10 examples of this intellectual idea, like negative externalities from this historical era. Like, from 400 BC to like, 680 BC or something; in some cases it works and in some cases, it doesn't. It feels like hypothetically you should be able to come up with a Scott Alexander post-type content. Just come up with 10 good historical examples on an idea and write cool anecdotes. Also, of course, trying to come up with whole Seinfeld episodes about any topic of your choice or something like that is starting to be possible. I tried using it a bit for evaluating information, so I'd say, come up with like 10 attributes to rate this intellectual, and then in each one, judge the intellectual. And it did. It definitely is gonna require some work, but it's able to kind of start, it could at least differentiate that Donald Trump gets very low marks as an intellectual. In comparison, most of the intellectuals, I could come up with had very high marks. ### Could EAs contributing to applied LLMs be harmful? Ozzie: Next topic. Can this do more harm than good? So, there's one camp that may be extreme that would say any use of AI would be advancing AI in the same capability, and that'll be like net bad. But that said, A, this is a long argument and discussion. So we want to get more into applications, and we don't want to spend much time on that. But B, my guess is that a lot of people would agree that there are probably just some pretty cheap wins that we could take, without extending the theory of language models or the total capabilities of language models, but still getting some valuable things with them. Some decent risk-reward trade-offs. Misha: I basically agree. I think interfaces are one area that currently sucks. Making better interfaces can help everyone think a bit better, especially with later models. I'm not sure if this would substantially accelerate capabilities. Ozzie: In terms of groups that we don't want to have developed better epistemic, there are definitely some authoritarian regimes that would be bad. The next one that could be problematic is AI-development organizations. If EA helps develop tools that help people have better epistemic or rational reasoning, maybe they'll be used to make AIs faster. Again, I don't know how big of a deal this is, but I think that may be one of the main things to be thinking about. Misha: I basically expect AI labs to start squeezing out these models to improve productivity. Though I think most of it, for now, will be via Codex-like tech, e.g., doing sysadmin better. People made ChatGPT simulate a terminal, which is suggestive. ### Application: Management and Emotional Assistance Ozzie: So how about we get into applications? That's the main topic. You have a few applications to discuss, and then I have a few applications. What are your favorite applications? Misha: To just continue the discussion about differential progress. People in EA think that independent researchers and other loners struggle without proper management. I think a bunch of helpful things in this direction can be achieved via GPT. So I wouldn't be surprised if someone figured out how to turn models into research managers to help with performance coaching and other things. This seems positive. Another one: there is a lot of untapped knowledge on Reddit and in other amazing communities. They don’t do proper science, but by just trying to figure out what actually works, they collect useful insight. And you can pull it out of LLMs by asking to be a knowledgeable member of such a community. Likewise, you can get perspective through the eyes of others. GPT is sort of a phenomenological museum—you know, a window into other worlds. I really like [Clearer Thinking](https://www.clearerthinking.org/). They have programs to improve people's mental tools. If you're taking their quiz, it's very nice in isolation, but it’d be cooler if you were able to do them alongside LLM to be able to chat about ideas and analyze results. The conversation is better than a linear progression. "Oh, I am inconsistent in how I value my time. Why don’t I feel comfortable spending money to buy time?” is an excellent moment to reflect more, and GPT might be good at holding space for it. Lastly, I think Scott Alexander wrote about a hypothetical client who was upset with his partner, claiming they didn’t really love them. The problem was that they wanted to have a hot cooked dinner every night they come from work. But their partner works, so it wasn't possible every time. Scott just suggested calling your partner in advance and, if they would be able to come home early enough to cook dinner, just... order Uber Eats. It worked! I think a bunch of human problems are debuggable and are often simple. Ozzie: How would that work: of GPT understanding a human's life well enough to know what to recommend and what you need? Would it need some information about what you're doing? Misha: Right. And in this case, an emotionally responsive model can simply ask, "What's up?" "Why are you thinking that?" "How would that make you feel?" Someone is creating a safe space for people to think. Because unless you—I dunno—unless I'm writing, I just can't really pull it all together and make a connection. Google Docs is amazing for that. Making it more conversational, I believe, would make it easier for more people. Ozzie: Hypothetically, it seems like people are going to be trying to import people's emails and social media posts and stuff like that to just begin with a lot of information about the person, and then from that they could hypothetically make a lot of recommendations. In the interest of time, let's go to the next topic. ### Application: Communication, Broadly Ozzie: The first application I have to discuss is communication in general. I wrote [one post](https://www.lesswrong.com/posts/PC4yowA2TiRne69iD/expansive-translations-considerations-and-possibilities) about this. Arguably, translating English to Russian is similar to translating English spoken by a ninth grader in Detroit into a language spoken by someone very different, using different terminology and cultural markers. People want to be communicated with very personally, so having something that understands them very well, knows what they know, and could explain things in terms that they understand just gives you a massive benefit, hypothetically. I think in EA there's definitely a whole lot in the philosophical literature that we just don't know and that no one so far has figured out. Hypothetically GPT could understand all of that and say "oh, these 10 insights are the most valuable." Right now, instead, these fields are all like coded with continental philosophical language or coded with other terminology that we're really not used to, or like these other foundation assumptions that we don’t understand or agree with. What a lot of people want is a personal tutor that understands it very well and understands how they learn. So we kind of want things like chatbots to become personalized tutors in any individualized style of communication. So hypothetically that's just like a whole lot of value on the table. Misha: An example would be asking Model to summarize key insights of postmodernism for someone who is libertarian-ish or reads LessWrong. This will probably work quite well. Likewise, you can communicate in your own language. Well, the doctor doesn’t get my math metaphors or someone else’s cooking metaphors. Contrary to GPT, which speaks both math and cooking fluently and can understand them. Say you want to use "simulated annealing" as a metaphor for how you approach one area of your life. LLM would understand you, while friends might not. Shorter inferential differences mean less need to explain yourself or build toward a thesis—you can just shoot. Ozzie: I guess one analogy is [TV tropes](https://tvtropes.org/). It has great lists of many super-specific things. Like, for many very specific tropes that exist online, here's every single example of it in every type of media. That's very different from Wikipedia. There are a lot of intellectual terminologies that are not connected at all. So it is pretty common for me to find that, oh, this interesting term was actually cited in like 10 different manuscripts or something like that and used in these different ways. But you should know that each way it's used is slightly different. So it's just like a huge pain that, hypothetically, could be solved. I guess going down a bit and finding mistakes and miscommunications would be great. I think people very often misunderstand stuff online, at least when I'm writing comments to people. I'm constantly misunderstanding people when I try to respond and stuff. Hypothetically, there could be a browser extension that flags bad text in red and says "Oh, these few words you're probably going to misunderstand." And then if you hover over it, you can see what it actually means. Of course, that's not as intense as something that rewrites all the content for you in ways that it's pretty sure you're going to understand, which is probably more what you'd like. Misha: Right. I think you probably thought about having this conversation not between me and you but between you, your AI, my AI, and me. It might smooth things a lot by interacting between different cultures. Ozzie: Yeah. ### Human-AI-Human Communication Ozzie: I think that human-to-human conversation is just really mediocre. Humans are simply not very good at communicating with one another. It's very complicated. You have to understand where the other person's coming from; we have many different types of humans with different backgrounds. And then you also have to understand them. Both of those challenges are dramatically more complicated than people give them credit for. So I think we want to move to this human-AI-human model, where AI is figuring out how to extract valuable information from humans, which probably looks a lot different than them writing essays or anything like that. I have no idea what the ideal is going to be. Perhaps the AI just asks people very targeted questions. And then, very separately, the AI develops an embedding of the important information that comes from humans and communicates information that could partially come from that to other humans as it's most needed. The way to write an essay in the future, for a human, would be not to write an essay but instead to get interrogated by a language model. In another world, we would actually just stop with the idea of essays. Instead, you get like four intellectual credits, because that's about how much interesting information you've contributed to the AI system. And then, when people want information, they just get whatever ideas they want. Content comes from a mess of different people and stuff, and that's totally fine. Miscommunication is huge. If you could completely eliminate miscommunication online, I think that would be worth a lot of money. I think therapists, like marriage therapists, do a lot of trying to get communication to be decent. So if an AI could do that for you, that would be huge. Misha: Right. ### Application: Decision Automation Misha: Also, like a lot of small businesses, some are dysfunctional. The Profit, a TV show, shows how a lot of them are disasters. Sure, there are some selection effects. But a lot of it is due to poor emotional and conversational skills. Well, also poor basic business skills. Ozzie: Now, there’s a super interesting question, which is, “How much decision-making can we automate pretty easily with a combination of language models and a few other tools”? Like hypothetically, a lot of business decisions are just not being made that well. A lot of bureaucratic decisions are probably worse than a decent guess by a properly trained agent or something. It's not super clear exactly when we'll reach what thresholds. But yeah, there are a lot of bad decisions being made. Misha: Right. I think it would be like really hard to secure agreement from the involved actors. People don’t like when others override their decisions, and they wouldn’t like LLMs to do it either. I think this is one of the problems with the adoption of prediction markets in corporations. Ozzie: I think one good model is that of autonomous vehicles. It’s level one before you get to level five. Levels one and two are just like driver assist. And then, as the systems get better, people will rely on them more. Misha: Slower adaptation is good. Initially just ask good probing questions in a non-threatening manner, and you're like, "oh yeah, that's a good point. I'll, I will do that". And because no one told you to do that, whether it was some AI from OpenAI or another company, you don't even lose social standing. You don’t look weak or anything. So even a very bossy person can agree. As a side note, I think the same dynamic might enable men to do more therapy-shaped things via LLMs. Ozzie: I imagine, too, that these AI models would be very useful in the sense that they'd be deemed much less biased than a lot of the bosses and other people making the decisions. If someone's in charge of a big government contract and wants to make a big decision about it, that might be suspect. There are a lot of opportunities for bias. If they had an AI kind of say, "oh yeah, that's what I would do, too", that would give them a fair bit more credibility. So just having some assistance could be useful. There are also definitely some situations where I expect people to only trust AI because, like, they just can't trust any responsible person to make a decent decision. ### Application: EA Forum Improvements Ozzie: There are a bunch of potential uses of GPT for the EA Forum. First, there are a lot of new users who really take time to get used to the community norms. They come in pretty grumpy, or they don't know what terminology we use. EA does have pretty unusual epistemic standards that are difficult to teach. In theory, when you're writing a post, you should be able to see in real-time what the comments will be and what the karma for that post will be. So as you're editing, you write a bad sentence and immediately see an imagined angry comment. And then you say, "Okay, I guess I'll just delete that sentence." Obviously, this assistant service does have some harm as well as benefits, but hypothetically, some of the angry comments that we see could just be seen in advance and then prevented. You could also have less intense steps, such as "Oh, your writing style probably isn't up to the best standards and probably won't be properly appreciated, but here's our version of rewriting it, that you're free to take inspiration from." In terms of people reading articles on the EA Forum, we may just have 10 different versions of an article or something. Writers give it a few sliders of, "Oh, do I want the very summarized version? Do I want the, like, very long version? Do I want it to use stories from what historical period or something similar?” And the articles could be automatically adjusted accordingly. ### Application: Evaluations Ozzie: I see evaluation is a really major application for LLMs. In this case, it just means that for every EA Forum post or comment, we have an estimate of how well-written this is and how likely this is to hold up after scrutiny. How inflammatory does it seem? How good or bad does this post seem on a few different spectrums? Hypothetically, we could have a lot of these auto-evaluations done. That's just on the EA Forum, of course. We'd really want that in all of Twitter and all of the media. Of course, it's debatable what those would look like. Misha: Yeah. Unbelievable today this is in the realm of possible, while some years ago sentiment prediction was mediocre or so. Ozzie: A whole lot of things are possible all of a sudden. It's like a really freaking interesting time. ### LLM UXs Ozzie: So, moving on, um, UX issues. What do you think about UX issues? Misha: Oh yeah. You basically have two interfaces. One is Chat GPT, where, like your model, it is very polite and restricted, which is not ideal, but the chat interface is nice. Another one is to just continue the line, text completion, where you can, of course, recreate the chat model and other things. But this all seems not that exciting because the second one is just LLM default, predicting the next token. Chat mode is nice and more humane and hence appreciated. But also, it’s the first idea you would have. I think Janus builds what they call "multiverses," where you just branch text in all sorts of directions, give only a few prompts, and rely mostly on curation. This is really nice. This is another interface that is available. Today's appreciation and perception of capabilities are probably limited by the fact that you just get one thing as a response and can't easily customize how it's generated. The next thing would be to have conversations with multiple LLM-simulated partners, which would have different characters, different perspectives, and so on. You’d be able to have a wide array of shoulder advisors. Someone who is more direct would be straightforward; someone else would be more nuanced and careful; someone would ask, “All these considerations are good, but what do you want?”; another would be appreciative and emotionally soothing or something. I hope for more interfaces. But in practice, humans are kind of bad at developing new ones. It's like we still owe everything to Xerox PARK, right? But interfaces might unlock new creative ways to make models useful and helpful. But it took us a while to figure out that you need to ask them to “think step by step.” So probably exciting new crazy interfaces are not coming anytime soon. Ozzie: I think my quick take is that, um, on the internet, it probably took thousands of UX and design years to figure out what websites should look like. And then, once we had mobile devices, it took many hundreds or thousands of years to figure out what that should look like. It's just a huge amount of work to really figure out the UX and UI for a new medium. And this appears to be more distinct to me than regular Web versus mobile. It seems like a much bigger shift, so I'd expect it to take a lot of time. ### What should EAs do with LLMs? Ozzie: So yeah, in the very end, um, what should EA do now? Misha: It's unclear if, like, EAs are anything special. I sincerely hope that people outside of the alignment community will pay more attention to this new technology, adopt these language models, and use it for their own benefit. But yes, as previously mentioned, figuring out how to do research management well with LLMs might be very helpful. I would be pretty excited if someone spent a month or more intensely trying to integrate LLMs into their processes and workflows. Oh, yeah, I am a bit involved in collecting all sorts of helpful base rates. For forecasting purposes and just to inform people about them. I previously figured out I can just ask ChatGPT to give me more examples of events I am interested in, like “what are huge secret projects that have remained secret for a long time?” And of course, I got a list of 30+ and never heard about most of them. Ozzie: Yeah, that makes sense. It is also a big topic. But we need to finish up now. Thanks so much for your time. Any last comments? Misha: Nope. Thanks for hosting, Ozzie. Ozzie: Yeah. Thank you.
fa399c93-eca4-426e-a0fa-e52908081b53	StampyAI/alignment-research-dataset/aisafety.info	AI Safety Info	How is the Alignment Research Center (ARC) trying to solve Eliciting Latent Knowledge (ELK)? [ARC](https://www.alignment.org/) is trying to solve the problem of [Eliciting Latent Knowledge (ELK)](https://docs.google.com/document/d/1WwsnJQstPq91_Yh-Ch2XRL8H_EpsnjrC1dwZXR37PC8/edit). Suppose that you are training a predictor—an AI agent that predicts the state of the world and then performs some actions. How can you create another model—a reporter—that tells you what the predictor believes about the world? A key challenge is that training your reporter on human-labeled training data incentivizes the predictor to model what the human trainer would believe is true, because modeling "what the human would believe to be true" will receive human approval more consistently than the actual truth, since the human may sometimes be mistaken! Paul Christiano's plan seems to be to produce a minimal AI that can help to do AI safety research. To do this, preventing [deception](https://www.alignmentforum.org/posts/ocWqg2Pf2br4jMmKA/does-sgd-produce-deceptive-alignment) and [inner alignment failure](https://www.alignmentforum.org/posts/pL56xPoniLvtMDQ4J/the-inner-alignment-problem) are on the critical path, and the only known solutions require interpretability (this is how all of Evan Hubinger's [11 proposals](https://www.alignmentforum.org/posts/fRsjBseRuvRhMPPE5/an-overview-of-11-proposals-for-building-safe-advanced-ai) plan to get around this problem). If ARC can solve ELK, this would be a very strong form of interpretability: our reporter is able to tell us what the predictor believes about the world. Some ways this could be useful for aligning the predictor include: - Using the reporter to find deceptive/misaligned thoughts in the predictor, and then optimizing against those interpreted thoughts. At any time, stochastic gradient descent only updates the weights a small amount. If an AI becomes misaligned, it won't be very misaligned, and the interpretability tools will be able to discover this and do a gradient step to realign it. In this way, we can prevent deception at any point in training. - Stopping training if the AI is misaligned.
338e4410-8f63-4fa6-9339-2ace3d482e7e	trentmkelly/LessWrong-43k	LessWrong	Framing AI Childhoods Generated as part of SERI MATS, under John Wentworth. Thanks to Alex Turner and Garrett Baker for related discussion, and to Justis Mills for draft feedback. All bold claims and ignorant mistakes herein are my own. Do you remember when Evan Hubinger became really enamored with 'training stories,' a way of carving up the alignment problem into 'training rationales' and 'training goals'? Evan's idea was that we ought to think of alignment as choosing a target model that we want to end up with after training, plus choosing a training procedure that will actually yield that model. In my prior experience talking to people about this way of framing the alignment problem… people didn't especially get it. The typical response of those who had heard of this was, "Yeah, that's one way to carve up the problem apart from, e.g., inner and outer alignment. But so what? How does this framing help us actually reduce the problem? It seems no better or worse than our old framings." I used to have that response myself. However, I now think I understand training stories! Here's my take: What Do Training Stories Buy You That Inner/Outer Alignment Doesn't? > It’s worth pointing out how phrasing inner and outer alignment in terms of training stories makes clear what I think was our biggest mistake in formulating that terminology, which is that inner/outer alignment presumes that the right way to build an aligned model is to find an aligned loss function and then have a training goal of finding a model that optimizes for that loss function. However, as I hope the more general framework of training stories should make clear, there are many possible ways of trying to train an aligned model. Microscope AI and STEM AI are examples that I mentioned previously, but in general any approach that intends to use a loss function that would be problematic if directly optimized for, but then attempts to train a model that doesn’t directly optimize for that loss function, would fail on both outer a
e924a9b6-981e-4d61-a620-7779a89858fc	trentmkelly/LessWrong-43k	LessWrong	Lessons from Isaac: Poor Little Robbie Every so often, when explaining issues related to AI safety, I call on good old Asimov. That's easy: almost everyone that is at least interested in science knows his name, and the Three Laws of Robotics are a very good example of misspecified goal. Or are they? The truth is: I don't know. My last reading through Asimov's robots dates back ten years; it was in french; and I didn't know anything about AI safety, specification and many parts of my current mental scaffolding. So when I use Asimov for my points now, I'm not sure whether I'm spouting bullshit or not. Fortunately, the solution is simple, for once: I just have to read the goddamn stories. And since I'm not the only one I heard talking about Asimov in this context, I thought that a sequence on the robots stories would prove useful. My first stop is by "I,Robot", the first robot short story collection. And it starts with the first story published by Asimov, "Robbie". Basically, this Robbie is a robot that takes care of a little girl named Gloria. All is well, until Gloria's mother turns into the bad guy, and decides that her girl should not be raised by a machine. She harasses her weak husband until he accepts to get rid of Robbie. But when Gloria discovers the loss of her friend, nothing can comfort her. The parents try everything, including a trip to New York, paradise to suburbians. But nope, the girl is still heartbroken. Last try of the father: a visit to a factory manned by robots, so little Gloria can see that they are lifeless machines, not real people. But, tada! Robbie was there! And he even saves the girl from an oncoming truck! It's revealed that the father planned it (Robbie being there, not the murder attempt on his daughter), but even so, the mother can't really send back the savior of her little girl. The End. Just a simple story about a nice little robot beloved by a girl, and the machinations of her mother to "protect" her from him. What's not to love? It's straight to the point, nicely
512f8257-e59c-4f75-8630-7b5b2ceeb773	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Decision theory and dynamic inconsistency Here is my current take on decision theory: * When making a decision after observing X, we should condition (or causally intervene) on statements like “My decision algorithm outputs Y after observing X.” * Updating seems like a description of something you do when making good decisions in this way, not part of defining what a good decision is. ([More](https://sideways-view.com/2021/10/12/edt-with-updating-double-counts/).) * Causal reasoning likewise seems like a description of something you do when making good decisions. Or equivalently: we should use a notion of causality that captures the relationships relevant to decision-making rather than intuitions about physical causality. ([More](https://sideways-view.com/2022/04/17/what-is-causality-to-an-evidential-decision-theorist/).) * “How much do I care about different copies of myself?” is an arbitrary question about my preferences. If my preferences change over time, it naturally gives rise to dynamic inconsistency unrelated to decision theory. (Of course an agent free to modify itself at time T would benefit by implementing some efficient compromise amongst all copies forked off after time T.) In this post I’ll discuss the last bullet in more detail since I think it’s a bit unusual, it’s not something I’ve written about before, and it’s one fo the main ways my view of decision theory has changed in the last few years. (Note: I think this topic is interesting, and could end up being relevant to the world in some weird-yet-possible situations, but I view it as unrelated to my day job on aligning AI with human interests.) The transparent Newcomb problem ------------------------------- In the transparent version of Newcomb’s problem, you are faced with two transparent boxes (one small and one big). The small box always contains $1,000. The big box contains either $10,000 or $0. You may choose to take the contents of one or both boxes. There is a very accurate predictor, who has placed $10,000 in the big box if and only if they predict that you wouldn’t take the small box regardless of what you see in the big one. Intuitively, once you see the contents of the big box, you really have no reason not to take the small box. For example, if you see $0 in the big box, you know for a fact that you are either getting $0 or $1,000. So why not just take the small box and walk away with $1,000? EDT and CDT agree about this one. I think it’s genuinely non-obvious what you should do in this case (if the predictor is accurate enough). But I think this is because of ambiguity about what you want, not how you should make decision. More generally, I think that the apparent differences between EDT and UDT are better explained as differences in preferences. In this post I’ll explain that view, using transparent Newcomb as an illustration. A simple inconsistent creature ------------------------------ Consider a simple creature which rationally pursues its goals on any given day—but whose goals change completely each midnight. Perhaps on Monday the creature is trying to create as much art and beauty as possible; on Tuesday it is trying to create joy and happiness; on Wednesday it might want something different still. On any given day we can think of the creature as an agent. The creature on Tuesday is not being irrational when it decides to pursue joy and happiness instead of art and beauty. It has no special reason to try to “wind back the clock” and pursue the same projects it would have pursued on monday. Of course on Monday the creature would prefer to arrest this predictable value drift—it knows that on Tuesday it will be replaced with a new agent, one that will stop contributing to the project of art and beauty. The creature on Monday ought to make plans accordingly, and if they had the ability to change this feature of themselves they would likely do so. It’s a matter of semantics whether we call this creature a single agent or a sequence of agents (one for each day). This sequence of agents could benefit from cooperating with one another, and it can do so in different ways. Normal coordination is off the table, since causality runs only one way from each agent to the next. But there are still options: * The Tuesday-creature might believe that its decision is correlated with the Monday-creature. If the Tuesday-creature tries to stop the Wednesday-creature from existing, then the Monday-creature might have tried to stop the Tuesday-creature from existing. If the correlation is strong enough and stopping values change is expensive, then the Tuesday-creature is best served by being kind to its Wednesday-self, and helping to put it in a good position to realize whatever its goals may be. (Though note that this can unwind just like an iterated prisoner’s dilemma with finite horizon!) * The Tuesday-creature might believe that its decision is correlated with the Monday-creature’s predictions about what the Tuesday-creature would do. If the Tuesday-creature keeps on carrying out the Monday-creature’s plans, then the Monday-creature would be more motivated to help the Tuesday-creature succeed (and less motivated to try to prevent the value change). If the Monday-creature is a good enough predictor of the Tuesday-creature, then the Tuesday-creature is best served by at least “paying back” the Monday-creature for all of the preparation the Monday-creature did. However none of these relationships are specific to the fact that it is the same creature on Monday and Tuesday; the fact that the cells are the same has no significance for the decision-theoretic situation. The Tuesday-creature has no intrinsic interest in the fact that it is not “reflectively stable”—of course that instability definitionally implies a desire to change itself, but not a further reason to try to help out the Monday-creature or Wednesday-creature, beyond the relationships described above. A human inconsistency --------------------- I care a lot about what is going to happen to me in the future. I care much more about my future than about different ways that the world could have gone (or than my past for that matter). In fact I would treat those other possible versions of myself quite similarly to how I’d treat another person who just happened to be a lot like me. This leads to a clear temporal inconsistency, which is so natural to humans that we don’t even think of it as an inconsistency. I’ll try to illustrate with a sequence of thought experiments. Suppose that at 7AM I think that there is a 50% chance that a bell will ring at 8AM. At 7AM I am indifferent between the happiness of Paul-in-world-with-bell and Paul-in-silent-world. If you asked me which Paul I would prefer to stub his toe, I would be indifferent. But by 8:01AM my preferences are quite different. After I’ve heard the bell ring, I care overwhelmingly about Paul-in-world-with-bell. I would very strongly prefer that the other Paul stub his toe than that I do. Some people might say “Well you just cared about what happens to Paul, and then at 8AM you learned what is real. Your beliefs have changed, but not your preferences.” But consider a different experiment where I am duplicated at 7AM and each copy is transported to a different city, one where the bell would ring and the other where it will not. Until I learn which city I’m in, I’m indifferent between the happiness of Paul-in-city-with–bell and Paul-in-silent-city. But at the moment when I hear the bell ring, my preferences shift. Some people could still say “Well you cared about the same thing all along—what happens to you—and you were merely uncertain about which Paul was you.” But consider the Paul from before the instant of copying, informed that he is about to be copied. That Paul knows full well that he cares about both copies. Yet sometime between the copying and the bell Paul has become much more parochial, and only cares about one. It seems to me like there is little way to escape from the inconsistency here. One could still say “Nonsense, all along you just cared about what happened to you, you were just uncertain about which of the copies you were going to become.” I find this very unpersuasive (why think there is a fact of the matter about who “I” am?), but at this point I think it’s just a semantic dispute. Either my preferences change, or my preferences are fixed but defined in terms of a concept like “the real me” whose meaning changes. It all amounts to the same thing. This is not some kind of universal principle of rationality—it’s just a fact about Paul. You can imagine different minds who care about all creatures equally, or who care only about their own future experiences, or who care about all the nearby possible copies of themselves. But I think many humans feel roughly the same way I do—they have some concern for others (including creatures very similar to themselves in other parts of the multiverse), but have a very distinctive kind of caring for what they themselves will actually experience in the future. Altruism is more complicated ---------------------------- In the examples above I discussed stubbing my toe as the unit of caring. But what if we had instead talked of dollars? And what if I am a relatively altruistic person, who would use marginal dollars to try to make the world better? Now in the case of two copies in separate cities it is clear enough that my preferences never change. I’m still willing to pay $1 to give my counterpart $2. After all, they can spend those dollars just as well as I can, and I don’t care who it was who did the good. But in the case of a single city, where the bell either rings or it doesn’t, we run into another ambiguity in my preferences—another question about which we need not expect different minds to agree no matter how rational they are. Namely: once I’ve heard the bell ringing, do I care about the happiness of the creatures in the world-with-bell (given that it’s the real world, the one we are actually in), or do I care about the happiness of creatures in both worlds even after I’ve learned that I happen to be in one of them? I think people have different intuitions about this. And there are further subtle distinctions, e.g. many people have different intuitions depending on whether the ringing of the bell was a matter of objective chance (where you could imagine other copies of yourself on far away worlds, or other branches, facing the same situation with a different outcome), or a matter of logical necessity where we were simply ignorant. While some of those disagreements may settle with more discussion, I think we should be able to agree that in principle we can imagine a mind that works either way, that either care about people in other worlds-that-could-have-been or who don’t. Most humans have at least some draw towards caring only about the humans in this world. So the rest of my post will focus on their situation. Back to transparent Newcomb (or: The analogy) --------------------------------------------- Consider again a human playing the transparent version of Newcomb’s problem. They see before them two boxes, a small one containing $1000 and a big one containing $0. They are told that the big box would have contained $10000 if a powerful predictor had guessed that they would never take the small box. If the human cares only for their own future experiences, and would spend the money only on themselves, they have a pretty good case for taking the small box and walking away with $1000. After all, their own future experiences are either going to involve walking away with $1000 or with nothing, there is no possible world where they experience seeing an empty big box and then end up with the money after all. Of course before taking the big box, the human would have much preferred to commit to never taking the small box. If they are an evidential decision theorist, they could also have just closed their eyes (curse that negative value of information!). That way they would have ended up with $10,000 instead of $1,000. Does this mean that they have reason to take nothing after all, even after seeing the box? I think the human’s situation is structurally identical to the inconsistent creature whose preferences change at midnight. Their problem is that in the instant when they see the empty big box, their preferences change. Once upon a time they cared about all of the possible versions of themselves, weighted by their probability. But once they see the empty big box, they cease to care at all about the versions of themselves who saw a full box. They end up in conflict with other very similar copies of themselves, and from the perspective of the human at the beginning of the process the whole thing is a great tragedy. Just like the inconsistent creature, the human would have strongly preferred to make a commitment to avoid these shifting preferences. Just like the inconsistent creature, they might still find other ways to coordinate even after the preferences change, but it’s more contingent and challenging. Unlike the inconsistent creature, they can avoid the catastrophe by simply closing their eyes—because the preference change was caused by new information rather than by the passage of time. The situation is most stark if we imagine the predictor running detailed simulations in order to decide whether to fill the big box. In this case, there is not one human but three copies of the human: two inside the predictor’s mind (one who sees an empty box and one who sees a full box) and one outside the predictor in the real world (seeing an empty or full box based on the results of the simulation). The problem for the human is that these copies of themselves can’t get along. Even if you explained the whole situation to the human inside the simulation, they’d have no reason to go along with it. By avoiding taking the small box, all they can achieve is to benefit a different human outside of the simulation, who they no longer care at all about. From their perspective, better to just take the money (since there’s a 50% chance that they are outside of the simulation and will benefit by $1000). (There are even more subtleties here if these different possible humans have preferences about their own existence, or about being in a simulation, or so on. But none of these change the fundamental bottom line.) Altruism is still more complicated ---------------------------------- If we consider a human who wants to make money to make the world better, the situation is similar but with an extra winkle. Now if we explain the situation to the inside human, they may not be quite so callous. Instead they might reason “If I don’t take the small box, there is a good chance that a ‘real’ human on the outside will then get $10,000. That looks like a good deal, so I’m happy to walk away with nothing.” Put differently, when we see an empty box we might not conclude that predictor didn’t fill the box. Instead, we might consider the possibility that we are living inside the predictor’s imagination, being presented with a hypothetical that need not have any relationship to what’s going on out there in the real world. The most extreme version of this principle would lead me to entertain very skeptical / open-minded beliefs about the world. In any decision-problem where “what I’d do if I saw X” matters for what happens in cases where X is false, I could say that there is a “version” of me in the hypothetical who sees X. So I can never really update on my observations. This leads to [CDT=EDT=UDT](https://www.lesswrong.com/posts/WkPf6XCzfJLCm2pbK/cdt-edt-udt). For people who endorse that perspective (and have no indexical preferences), this post probably isn’t very interesting. Myself, I think I somewhat split the difference: I think explicitly about my preferences about worlds that I “know don’t exist,” roughly using the framework of this post. But I justify that perspective in significant part from a position of radical uncertainty: I’m not sure if I’m thinking about worlds that don’t exist, or if it’s us who don’t exist and there is some real world somewhere thinking about us. Conclusion ---------- Overall the perspective in this post has made me feel much less confused about updatelessness. I expect I’m still wrong about big parts of decision theory, but for now I feel tentatively comfortable using UDT and don’t see the alternatives as very appealing. In particular, I no longer think that updating feels very plausible as a fundamental decision-theoretic principle, but at the same time don’t think there’s much of a reflective-stability-based argument for e.g. one-boxing in transparent Newcomb. Most of the behaviors I associate with being “updateless” seem to really be about consistent preferences, and in particular continuing to care about worlds that are in some sense inconsistent with our observations. I believe my altruistic preferences are roughly stable in this sense (partially justified by a kind of radical epistemic humility about whether this is the “real” world), but my indexical preferences are not. The perspective in this post also more clearly frames the coordination problem faced by different copies of me (e.g. in different plausible futures) and I think has left me somewhat more optimistic about finding win-win deals.
e8adb50b-3c71-4762-a58a-720f3747307f	StampyAI/alignment-research-dataset/alignmentforum	Alignment Forum	Abstraction = Information at a Distance Why is abstraction useful? Why use a high-level model rather than a low-level model? An example: when I type “4+3” in a python shell, I think of that as adding two numbers, not as a bunch of continuous voltages driving electric fields and current flows in little patches of metal and doped silicon. Why? Because, if I’m thinking about what will show up on my monitor after I type “4+3” and hit enter, then the exact voltages and current flows on the CPU are not relevant. This remains true even if I’m thinking about the voltages driving individual pixels in my monitor - even at a fairly low level, the exact voltages in the arithmetic-logic unit on the CPU aren’t relevant to anything more than a few microns away - except for the high-level information contained in the “numbers” passed in and out. Another example: if I’m an astronomer predicting the trajectory of the sun, then I’m presumably going to treat other stars as point-masses. At such long distances, the exact mass distribution within the star doesn’t really matter - except for the high-level information contained in the total mass and center-of-mass location. If I’m running a Markov-Chain Monte Carlo algorithm, then I take sample points fairly far apart in “time”. As long as they’re far enough apart, they’re roughly independent - there isn’t any information from one sample relevant to the next. If I’m planning a roadtrip from San Francisco to Los Angeles, the details of my route through the Bay Area are irrelevant to planning my route within LA - except for the high-level information contained in my choice of highway for the middle leg of the trip and the rough time I expect to get there. General point: abstraction, in practice, is about keeping information which is relevant to things “far away”, and throwing out everything else. Formalization ------------- Let’s start with a bunch of random variables .mjx-chtml {display: inline-block; line-height: 0; text-indent: 0; text-align: left; text-transform: none; font-style: normal; font-weight: normal; font-size: 100%; font-size-adjust: none; letter-spacing: normal; word-wrap: normal; word-spacing: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0; min-height: 0; border: 0; margin: 0; padding: 1px 0} .MJXc-display {display: block; text-align: center; margin: 1em 0; padding: 0} .mjx-chtml[tabindex]:focus, body :focus .mjx-chtml[tabindex] {display: inline-table} .mjx-full-width {text-align: center; display: table-cell!important; width: 10000em} .mjx-math {display: inline-block; border-collapse: separate; border-spacing: 0} .mjx-math \* {display: inline-block; -webkit-box-sizing: content-box!important; -moz-box-sizing: content-box!important; box-sizing: content-box!important; text-align: left} .mjx-numerator {display: block; text-align: center} .mjx-denominator {display: block; text-align: center} .MJXc-stacked {height: 0; position: relative} .MJXc-stacked > \* {position: absolute} .MJXc-bevelled > \* {display: inline-block} .mjx-stack {display: inline-block} .mjx-op {display: block} .mjx-under {display: table-cell} .mjx-over {display: block} .mjx-over > \* {padding-left: 0px!important; padding-right: 0px!important} .mjx-under > \* {padding-left: 0px!important; padding-right: 0px!important} .mjx-stack > .mjx-sup {display: block} .mjx-stack > .mjx-sub {display: block} .mjx-prestack > .mjx-presup {display: block} .mjx-prestack > .mjx-presub {display: block} .mjx-delim-h > .mjx-char {display: inline-block} .mjx-surd {vertical-align: top} .mjx-mphantom \* {visibility: hidden} .mjx-merror {background-color: #FFFF88; color: #CC0000; border: 1px solid #CC0000; padding: 2px 3px; font-style: normal; font-size: 90%} .mjx-annotation-xml {line-height: normal} .mjx-menclose > svg {fill: none; stroke: currentColor} .mjx-mtr {display: table-row} .mjx-mlabeledtr {display: table-row} .mjx-mtd {display: table-cell; text-align: center} .mjx-label {display: table-row} .mjx-box {display: inline-block} .mjx-block {display: block} .mjx-span {display: inline} .mjx-char {display: block; white-space: pre} .mjx-itable {display: inline-table; width: auto} .mjx-row {display: table-row} .mjx-cell {display: table-cell} .mjx-table {display: table; width: 100%} .mjx-line {display: block; height: 0} .mjx-strut {width: 0; padding-top: 1em} .mjx-vsize {width: 0} .MJXc-space1 {margin-left: .167em} .MJXc-space2 {margin-left: .222em} .MJXc-space3 {margin-left: .278em} .mjx-test.mjx-test-display {display: table!important} .mjx-test.mjx-test-inline {display: inline!important; margin-right: -1px} .mjx-test.mjx-test-default {display: block!important; clear: both} .mjx-ex-box {display: inline-block!important; position: absolute; overflow: hidden; min-height: 0; max-height: none; padding: 0; border: 0; margin: 0; width: 1px; height: 60ex} .mjx-test-inline .mjx-left-box {display: inline-block; width: 0; float: left} .mjx-test-inline .mjx-right-box {display: inline-block; width: 0; float: right} .mjx-test-display .mjx-right-box {display: table-cell!important; width: 10000em!important; min-width: 0; max-width: none; padding: 0; border: 0; margin: 0} .MJXc-TeX-unknown-R {font-family: monospace; font-style: normal; font-weight: normal} .MJXc-TeX-unknown-I {font-family: monospace; font-style: italic; font-weight: normal} .MJXc-TeX-unknown-B {font-family: monospace; font-style: normal; font-weight: bold} .MJXc-TeX-unknown-BI {font-family: monospace; font-style: italic; font-weight: bold} .MJXc-TeX-ams-R {font-family: MJXc-TeX-ams-R,MJXc-TeX-ams-Rw} .MJXc-TeX-cal-B {font-family: MJXc-TeX-cal-B,MJXc-TeX-cal-Bx,MJXc-TeX-cal-Bw} .MJXc-TeX-frak-R {font-family: MJXc-TeX-frak-R,MJXc-TeX-frak-Rw} .MJXc-TeX-frak-B {font-family: MJXc-TeX-frak-B,MJXc-TeX-frak-Bx,MJXc-TeX-frak-Bw} .MJXc-TeX-math-BI {font-family: MJXc-TeX-math-BI,MJXc-TeX-math-BIx,MJXc-TeX-math-BIw} .MJXc-TeX-sans-R {font-family: MJXc-TeX-sans-R,MJXc-TeX-sans-Rw} .MJXc-TeX-sans-B {font-family: MJXc-TeX-sans-B,MJXc-TeX-sans-Bx,MJXc-TeX-sans-Bw} .MJXc-TeX-sans-I {font-family: MJXc-TeX-sans-I,MJXc-TeX-sans-Ix,MJXc-TeX-sans-Iw} .MJXc-TeX-script-R {font-family: MJXc-TeX-script-R,MJXc-TeX-script-Rw} .MJXc-TeX-type-R {font-family: MJXc-TeX-type-R,MJXc-TeX-type-Rw} .MJXc-TeX-cal-R {font-family: MJXc-TeX-cal-R,MJXc-TeX-cal-Rw} .MJXc-TeX-main-B {font-family: MJXc-TeX-main-B,MJXc-TeX-main-Bx,MJXc-TeX-main-Bw} .MJXc-TeX-main-I {font-family: MJXc-TeX-main-I,MJXc-TeX-main-Ix,MJXc-TeX-main-Iw} .MJXc-TeX-main-R {font-family: MJXc-TeX-main-R,MJXc-TeX-main-Rw} .MJXc-TeX-math-I {font-family: MJXc-TeX-math-I,MJXc-TeX-math-Ix,MJXc-TeX-math-Iw} .MJXc-TeX-size1-R {font-family: MJXc-TeX-size1-R,MJXc-TeX-size1-Rw} .MJXc-TeX-size2-R {font-family: MJXc-TeX-size2-R,MJXc-TeX-size2-Rw} .MJXc-TeX-size3-R {font-family: MJXc-TeX-size3-R,MJXc-TeX-size3-Rw} .MJXc-TeX-size4-R {font-family: MJXc-TeX-size4-R,MJXc-TeX-size4-Rw} .MJXc-TeX-vec-R {font-family: MJXc-TeX-vec-R,MJXc-TeX-vec-Rw} .MJXc-TeX-vec-B {font-family: MJXc-TeX-vec-B,MJXc-TeX-vec-Bx,MJXc-TeX-vec-Bw} @font-face {font-family: MJXc-TeX-ams-R; src: local('MathJax\_AMS'), local('MathJax\_AMS-Regular')} @font-face {font-family: MJXc-TeX-ams-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_AMS-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_AMS-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_AMS-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-cal-B; src: local('MathJax\_Caligraphic Bold'), local('MathJax\_Caligraphic-Bold')} @font-face {font-family: MJXc-TeX-cal-Bx; src: local('MathJax\_Caligraphic'); font-weight: bold} @font-face {font-family: MJXc-TeX-cal-Bw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Caligraphic-Bold.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Caligraphic-Bold.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Caligraphic-Bold.otf') format('opentype')} @font-face {font-family: MJXc-TeX-frak-R; src: local('MathJax\_Fraktur'), local('MathJax\_Fraktur-Regular')} @font-face {font-family: MJXc-TeX-frak-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Fraktur-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Fraktur-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Fraktur-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-frak-B; src: local('MathJax\_Fraktur Bold'), local('MathJax\_Fraktur-Bold')} @font-face {font-family: MJXc-TeX-frak-Bx; src: local('MathJax\_Fraktur'); font-weight: bold} @font-face {font-family: MJXc-TeX-frak-Bw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Fraktur-Bold.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Fraktur-Bold.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Fraktur-Bold.otf') format('opentype')} @font-face {font-family: MJXc-TeX-math-BI; src: local('MathJax\_Math BoldItalic'), local('MathJax\_Math-BoldItalic')} @font-face {font-family: MJXc-TeX-math-BIx; src: local('MathJax\_Math'); font-weight: bold; font-style: italic} @font-face {font-family: MJXc-TeX-math-BIw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Math-BoldItalic.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Math-BoldItalic.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Math-BoldItalic.otf') format('opentype')} @font-face {font-family: MJXc-TeX-sans-R; src: local('MathJax\_SansSerif'), local('MathJax\_SansSerif-Regular')} @font-face {font-family: MJXc-TeX-sans-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_SansSerif-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_SansSerif-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_SansSerif-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-sans-B; src: local('MathJax\_SansSerif Bold'), local('MathJax\_SansSerif-Bold')} @font-face {font-family: MJXc-TeX-sans-Bx; src: local('MathJax\_SansSerif'); font-weight: bold} @font-face {font-family: MJXc-TeX-sans-Bw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_SansSerif-Bold.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_SansSerif-Bold.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_SansSerif-Bold.otf') format('opentype')} @font-face {font-family: MJXc-TeX-sans-I; src: local('MathJax\_SansSerif Italic'), local('MathJax\_SansSerif-Italic')} @font-face {font-family: MJXc-TeX-sans-Ix; src: local('MathJax\_SansSerif'); font-style: italic} @font-face {font-family: MJXc-TeX-sans-Iw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_SansSerif-Italic.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_SansSerif-Italic.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_SansSerif-Italic.otf') format('opentype')} @font-face {font-family: MJXc-TeX-script-R; src: local('MathJax\_Script'), local('MathJax\_Script-Regular')} @font-face {font-family: MJXc-TeX-script-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Script-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Script-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Script-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-type-R; src: local('MathJax\_Typewriter'), local('MathJax\_Typewriter-Regular')} @font-face {font-family: MJXc-TeX-type-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Typewriter-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Typewriter-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Typewriter-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-cal-R; src: local('MathJax\_Caligraphic'), local('MathJax\_Caligraphic-Regular')} @font-face {font-family: MJXc-TeX-cal-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Caligraphic-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Caligraphic-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Caligraphic-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-main-B; src: local('MathJax\_Main Bold'), local('MathJax\_Main-Bold')} @font-face {font-family: MJXc-TeX-main-Bx; src: local('MathJax\_Main'); font-weight: bold} @font-face {font-family: MJXc-TeX-main-Bw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Main-Bold.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Main-Bold.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Main-Bold.otf') format('opentype')} @font-face {font-family: MJXc-TeX-main-I; src: local('MathJax\_Main Italic'), local('MathJax\_Main-Italic')} @font-face {font-family: MJXc-TeX-main-Ix; src: local('MathJax\_Main'); font-style: italic} @font-face {font-family: MJXc-TeX-main-Iw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Main-Italic.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Main-Italic.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Main-Italic.otf') format('opentype')} @font-face {font-family: MJXc-TeX-main-R; src: local('MathJax\_Main'), local('MathJax\_Main-Regular')} @font-face {font-family: MJXc-TeX-main-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Main-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Main-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Main-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-math-I; src: local('MathJax\_Math Italic'), local('MathJax\_Math-Italic')} @font-face {font-family: MJXc-TeX-math-Ix; src: local('MathJax\_Math'); font-style: italic} @font-face {font-family: MJXc-TeX-math-Iw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Math-Italic.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Math-Italic.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Math-Italic.otf') format('opentype')} @font-face {font-family: MJXc-TeX-size1-R; src: local('MathJax\_Size1'), local('MathJax\_Size1-Regular')} @font-face {font-family: MJXc-TeX-size1-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Size1-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Size1-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Size1-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-size2-R; src: local('MathJax\_Size2'), local('MathJax\_Size2-Regular')} @font-face {font-family: MJXc-TeX-size2-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Size2-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Size2-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Size2-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-size3-R; src: local('MathJax\_Size3'), local('MathJax\_Size3-Regular')} @font-face {font-family: MJXc-TeX-size3-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Size3-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Size3-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Size3-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-size4-R; src: local('MathJax\_Size4'), local('MathJax\_Size4-Regular')} @font-face {font-family: MJXc-TeX-size4-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Size4-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Size4-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Size4-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-vec-R; src: local('MathJax\_Vector'), local('MathJax\_Vector-Regular')} @font-face {font-family: MJXc-TeX-vec-Rw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Vector-Regular.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Vector-Regular.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Vector-Regular.otf') format('opentype')} @font-face {font-family: MJXc-TeX-vec-B; src: local('MathJax\_Vector Bold'), local('MathJax\_Vector-Bold')} @font-face {font-family: MJXc-TeX-vec-Bx; src: local('MathJax\_Vector'); font-weight: bold} @font-face {font-family: MJXc-TeX-vec-Bw; src /\1\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/eot/MathJax\_Vector-Bold.eot'); src /\2\/: url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/woff/MathJax\_Vector-Bold.woff') format('woff'), url('https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/fonts/HTML-CSS/TeX/otf/MathJax\_Vector-Bold.otf') format('opentype')} Xi, and some notion of which variables are “nearby”: each variable Xi has a set si of indices of variables considered “nearby” Xi. How si is chosen may vary by application - maybe each Xi is associated with some point in space and/or time, or maybe we’re looking at Markov blankets in a graphical model, or … We want some high-level summary of Xi; we’ll define that by a function fi(Xi). We require that fi(Xi) contain all information relevant to things far away - i.e. X¯¯¯¯si, the variables not in si. We’ll consider a few different notions of “relevance” here. First and most obvious is predictive relevance - fi(Xi) must contain all relevant information in the usual probabilistic/information-theoretic sense. Key subtlety: which information is relevant may itself depend on the values of other variables - e.g. maybe we have a conditional in a program which picks one of two variables to return. Should we keep around all information which is relevant in any possible case? All information which is relevant after averaging over some variables? Looking back over the examples, I think the natural answer is: we’re keeping information relevant to things “far away” (i.e. variables not in si), so those are what we’re interested in. Everything within si we can average over. Examples: * If I’m thinking about the relevance of the exact value of a particular voltage on the CPU to the output of “4+3” at a python shell, I’m happy to average over all the likely voltages of nearby wires. * If I’m thinking about the relevance of a little piece of mass in a star to predicting the sun’s trajectory, I’m happy to average over all the other little pieces of mass. * If I’m planning a trip from SF to LA, and figuring out the relevance of a particular detail of the route within the Bay Area, I’m happy to average over other details in the Bay Area. Formally, our condition is: P[X¯¯¯¯si\|fi(Xi)]=P[X¯¯¯¯si\|Xi] We could even go a step further and apply the [minimal map theorems](https://www.lesswrong.com/posts/Lz2nCYnBeaZyS68Xb/probability-as-minimal-map) to find fi(Xi) containing the least possible information, although it won't necessarily be the most computationally efficient summary. Another notion of “relevance” is causal influence - while probabilistic information is the key criteria for prediction, causal influence is the key for planning. We want to know what impact an intervention on Xi will have on far-away variables. We’re still happy to average over “nearby” variables, but there’s a new subtlety: we may also want to intervene on some of the variables far-away from Xi. For instance, if we’re planning a road-trip, we want to be able to consider possible route plans within LA - different routes would be different interventions on variables far away from SF. Our high-level model needs to hold for any of these interventions. Our criteria become: P[X¯¯¯¯si\|do(fi(Xi)=f∗i),do(Xs=X∗s)]=P[X¯¯¯¯si\|do(Xi=X∗i),do(Xs=X∗s)] … for any s⊂¯¯¯¯si, and any intervention values X∗s,f∗i,X∗i for which f∗i=f(X∗i). Here do(fi(Xi)=f∗i) means setting Xi to an arbitrary value X∗i such that fi(X∗i) = f∗i - i.e. “we just need to get to the highway by noon, the details don’t matter, we can work them out later”. This requires that the details do not, in fact, matter - i.e. P[X¯¯¯¯si\|do(Xi=X∗i),do(Xs=X∗s)] has the same value for different X∗i so long as f(X∗i) remains the same. That’s what the notation do(fi(Xi)=f∗i) is expressing. Finally, we could combine our criteria: require that any interventions on X¯¯¯¯si be supported, with either information or intervention on Xi. The criteria: P[X¯¯¯¯si\|fi(Xi),do(Xs=X∗s)]=P[X¯¯¯¯si\|Xi,do(Xs=X∗s)] P[X¯¯¯¯si\|do(fi(Xi)=f∗i),do(Xs=X∗s)]=P[X¯¯¯¯si\|do(Xi=X∗i),do(Xs=X∗s)] Both of these must hold for any s⊂¯¯¯¯si, and any intervention values X∗s,f∗i,X∗i for which f∗i=f(X∗i). In that case, we can predict the effects of arbitrary interventions on Xi and any of the X¯¯¯¯si on other X¯¯¯¯si, using only the summary information present in fi(Xi).
a70ff9ea-96d4-4a12-b5f1-eba9c0d87492	trentmkelly/LessWrong-43k	LessWrong	Applied cognitive science: learning from a faux pas Cross-posted from my LiveJournal: > Yesterday evening, I pasted to two IRC channels an excerpt of what someone had written. In the context of the original text, that excerpt had seemed to me like harmless if somewhat raunchy humor. What I didn't realize at the time was that by removing the context, the person writing it came off looking like a jerk, and by laughing at it I came off looking as something of a jerk as well. > > Two people, both of whom I have known for many years now and whose opinions I value, approached me by private message and pointed out that that may not have been the smartest thing to do. My initial reaction was defensive, but I soon realized that they were right and thanked them for pointing it out to me. Putting on a positive growth mindset, I decided to treat this event as a positive one, as in the future I'd know better. > > Later that evening, as I lay in bed waiting to fall asleep, the episode replayed itself in my mind. I learnt long ago that trying to push such replays out of my mind would just make them take longer and make them feel worse. So I settled back to just observing the replay and waiting for it to go away. As I waited, I started thinking about what kind of lower-level neural process this feeling might be a sign of. > > Artificial neural networks use what is called a backpropagation algorithm to learn from mistakes. First the network is provided some input, then it computes some value, and then the obtained value is compared to the expected value. The difference between the obtained and expected value is the error, which is then propagated back from the end of the network to the input layer. As the error signal works it way through the network, neural weights are adjusted in such a fashion to produce a different output the next time. > > Backprop is known to be biologically unrealistic, but there are more realistic algorithms that work in a roughly similar manner. The human brain seems to be using something called tempora
b77f3785-6f09-4028-ac0e-41589a614e47	trentmkelly/LessWrong-43k	LessWrong	Visible Thoughts Project and Bounty Announcement (Update Jan. 12: We released an FAQ last month, with more details. Last updated Jan. 7.) (Update Jan. 19: We now have an example of a successful partial run, which you can use to inform how you do your runs. Details.) We at MIRI are soliciting help with an AI-alignment project centered around building a dataset, described below. We have $200,000 in prizes for building the first fragments of the dataset, plus an additional $1M prize/budget for anyone who demonstrates the ability to build a larger dataset at scale. If this project goes well, then it may be the first of a series of prizes we offer for various projects. Below, I’ll say more about the project, and about the payouts and interim support we’re offering. The Project Hypothesis: Language models can be made more understandable (and perhaps also more capable, though this is not the goal) by training them to produce visible thoughts. We’d like to test this hypothesis by fine-tuning/retraining a language model using a dataset composed of thought-annotated dungeon runs. (In the manner of AI dungeon.) A normal (un-annotated) dungeon run is a sequence of steps in which the player inputs text actions and the dungeon master responds with text describing what happened in the world as a result. We’d like a collection of such runs, that are annotated with "visible thoughts" (visible to potential operators or programmers of the system, not to players) describing things like what just happened or is about to happen in the world, what sorts of things the player is probably paying attention to, where the current sources of plot tension are, and so on — the sorts of things a human author would think while acting as a dungeon master. (This is distinct from producing thoughts explaining what happened in the dungeon; “visible thoughts” are meant to play an active role in constructing the output.) Once we have such a dataset, MIRI’s hope is that present or future technology will be able to train a model or model
fdb28187-45c8-4849-ada5-536753144a19	StampyAI/alignment-research-dataset/arxiv	Arxiv	Contrastive Explanations for Reinforcement Learning in terms of Expected Consequences 1 Introduction --------------- Complex machine learning (ML) models such as Deep Neural Networks (DNNs) and Support Vector Machines (SVMs) perform very well on a wide range of tasks [[Lundberg and Lee2016](#bib.bibx7)], but their outcomes are often are often difficult to understand by humans [[Weller2017](#bib.bibx15)]. Moreover, machine learning models cannot explain how they achieved their results. Even for experts in the field, it can be very difficult to understand what the model actually learned [[Samek et al.2016](#bib.bibx11)]. To remedy this issue, the field of eXplainable Artificial Intelligence (XAI) studies how such complex but useful models can be made more understandable [gunning2017explainable]. Achieving transparency of ML models has multiple advantages [[Weller2017](#bib.bibx15)]. For example, if a model designer knows why a model performs badly on some data, he or she can start a more informed process of resolving the performance issues [[Kulesza et al.2015](#bib.bibx5), [Papernot and McDaniel2018](#bib.bibx9)]. However, even if a model has high performance, the users (typically non-experts in ML) would still like to know why it came to a certain output [[Miller2017](#bib.bibx8)]. Especially in high-risk domains such as defense and health care, inappropriate trust in the output may cause substantial risks and problems [[Lipton2016](#bib.bibx6), [Ribeiro et al.2016](#bib.bibx10)]. If a ML model fails to provide transparency, the user cannot safely rely on its outcomes, which hampers the model’s applicability [[Lipton2016](#bib.bibx6)]. If, however, a ML-model is able to explain its workings and outcomes satisfactorily to the user, then this would not only improve the user’s trust; it would also be able to provide new insights to the user. For the problem of classification, recent research has developed a number of promising methods that enable classification models to explain their output [[Guidotti et al.2018](#bib.bibx2)]. Several of these methods prove to be model-independent in some way, allowing them to be applied on any existing ML classification model. However, for Reinforcement Learning (RL) models, there are relatively few methods available [[Verma et al.2018](#bib.bibx14), [Shu et al.2017](#bib.bibx13), [Hein et al.2017](#bib.bibx4)]. The scarcity of methods that enable RL agents to explain their actions and policies towards humans severely hampers the practical applications of RL-models in this field. It also diminishes the, often highly rated, value of RL to Artificial Intelligence [[Hein et al.2017](#bib.bibx4), [Gosavi2009](#bib.bibx1)]. Take for example a simple agent within a grid world that needs to reach a goal position while evading another actor who could cause termination as well as evading other static terminal states. The RL agent cannot easily explain why it takes the route it has learned as it only knows numerical rewards and its coordinates in the grid. The agent has no grounded knowledge about the ’evil actor’ that tries to prevent it from reaching its goal nor has it knowledge of how certain actions will effect such grounded concepts. These state features and rewards are what drives the agent but do not lend themselves well for an explanations as they may not be grounded concepts nor do they offer a reason why the agent behaves a certain way. Important pioneering work has been done by Hayes and Shah [[Hayes and Shah2017](#bib.bibx3)]. They developed a method for eXplainable Reinforcement Learning (XRL) that can generate explanations about a learned policy in a way that is understandable to humans. Their method converts feature vectors to a list of predicates by using a set of binary classification models. This list of predicates is searched to find sub-sets that tend to co-occur with specific actions. The method provides information about which actions are performed when which state predicates are true. A method that uses the co-occurrence to generate explanations may be useful for small problems, but becomes less comprehensible in larger planning and control problems, because the overview of predicate and action combinations becomes too large. Also, the method addresses only what the agent does, and not why it acts as it does. In other words, the method presents the user with the correlations between states and the policy but it does not provide a motivation why that policy is used in terms of rewards, or state transitions. This study proposes an approach to XRL that allows an agent to answer questions about its actions and policy in terms of their consequences. Other questions unique to RL are also possible, for example those that ask about the time it takes to obtain some goal or those about RL specific problems (loop behavior, lack of exploration or exploitation, etc.). However we believe that a non-expert in RL is mostly interested in the expected consequences of the agent’s learned behavior and whether the agent finds these consequences good or bad. This information can be used as an argument why the agent behaves in some way. This would allow human users to gain insight in what information the agent can perceive from a state and which outcomes it expects from an action or state visit. Furthermore, to limit the amount of information of all consequences, our proposed method aims to support contrastive explanations [[Miller2017](#bib.bibx8)]. Contrastive explanations are a way of answering causal ’why’-questions. In such questions, two potential items, the fact and foil, are compared to each other in terms of their causal effects on the world. Contrastive questions come natural between humans and offer an intuitive way of gathering motivations about why one performs a certain action instead of another [[Miller2017](#bib.bibx8)]. In our case we allow the user to formulate a question of why the learned policy πt (the ’fact’) is used instead of some other policy πf (the ’foil) that is of interest to the user. Furthermore, our proposed method translates the set of states and actions in a set of more descriptive state classes C and action outcomes O similar to that of [[Hayes and Shah2017](#bib.bibx3)]. This allows the user to query the agent in a more natural way as well as receive more informative explanations as both refer to the same concepts instead of plain features. The translation of state features to more high-level concepts and actions in specific states to outcomes, is also done in the proposed algorithm of [[Sherstov and Stone2005](#bib.bibx12)]. The translation in this algorithm was used to facilitate transfer learning within a single action over multiple tasks and domains. In our method we used it to create a user-interpretable variant of the underlying Markov Decision Problem (MDP). For the purpose of implementation and evaluation of our proposed method, we performed a pilot study. In this study, a number of explanation examples were presented to participants to see which of their varying properties are preferred the most. One of the properties was to see whether the participants prefer explanations about the expected consequences of a single-action or the entire policy. 2 Approach for consequence-based explanations ---------------------------------------------- The underlying Markov Decision Problem (MDP) of a RL agent consists of the tuple ⟨S,A,R,T,λ⟩. Here, S and A are the set of states (described by a feature vector) and actions respectively, R:S×A→R is the reward function and T:S×A→Pr(S) the transition function that provides a probability distribution over states. Also, λ is the discount factor that governs how much of future rewards are taken into account by the agent. This tuple provides the required information to derive the consequences of the learned policy πt or the foil policy πf from the user’s question. As one can use the transition function T to sample the effects of both πt and πf. In the case T is not explicit, one may use a separate ML model to learn it in addition to the actual agent. Through this simulation, one constructs a Markov Chain of state visits under each policy πt and πf and can present the difference to the user. Through the simulation of future states with T, information can be gathered about state consequences. In turn, from the agent itself the state or state-action values for simulated state visits can be obtained to develop an explanation in terms of rewards. However, the issue with this approach is that the state features and rewards may not be easy to understand for a user as it would consist of possibly low-level concepts and numerical reward values or expected returns. To mitigate this issue we can apply a translation of the states and actions to a set of predefined state concepts and outcomes. These concepts can be designed to be more descriptive and informative for the potential user. A way to do this translation is by training a set of binary classifiers to recognize each outcome or state concept from the state features and taken action, a similar approach to the one from [[Hayes and Shah2017](#bib.bibx3)]. Their training can occur during the exploratory learning process of the agent. This translation allows us to use the above described method of simulating consequences and transform the state features and results of actions to more user-interpretable concepts. ### 2.1 A user-interpretable MDP The original set of states can be transformed to a more descriptive set C according to the function k:S→C. This is similar to the approach of [[Hayes and Shah2017](#bib.bibx3)] where k consists of a number of classifiers. Also, rewards can be explained in terms of a set of action outcomes O according to t:C×A→Pr(O). This provides the results of an action in some state in terms of the concepts O. For example, the outcomes that the developer had in mind when designing the reward function R. The transformation of states and actions in state classes and outcomes is adopted from the work of [[Sherstov and Stone2005](#bib.bibx12)] where the transformations are used to allow for transfer learning in RL. Here however, we use them as a translation towards a more user-interpretable representation of the actual MDP. The result is the new MDP tuple ⟨S,A,R,T,λ,C,O,t,k⟩. An RL agent is still trained on S, A, R and T with λ independent of the descriptive sets C and O and functions k and t. This makes the transformation independent of the RL algorithm used to train the agent. See Figure [1](#S2.F1 "Figure 1 ‣ 2.1 A user-interpretable MDP ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences") for an overview of this approach. ![](https://media.arxiv-vanity.com/render-output/8065870/concept_overview.png) Figure 1: An overview of the proposed method, a dotted line represents a feedback loop. We assume a general reinforcement learning agent that acts upon a state s through some action a and receives a reward r. We train a transition model T that can be used to simulate the effect of actions on states. By repeatedly simulating a state si we can obtain the expected consequences γ of an entire policy. Also, the consequences of a contrastive policy consisting of an alternative courses of action af can be simulated with the same transition model T. Finally, in constructing the explanation we transform states and actions into user-interpretable concepts and construct an explanation that is contrastive. As an example take the grid world illustrated in Figure [2](#S2.F2 "Figure 2 ‣ 2.1 A user-interpretable MDP ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences") that shows an agent in a simple myopic navigation task. The states S are the (x,y) coordinates and the presence of a forest, monster or trap in adjacent tiles with A=Up,Down,Left,Right. R consists of a small transient penalty, a slightly larger penalty for tiles with a forest, a large penalty shared over all terminal states (traps or adjacent tiles to a monster) and a large positive reward for the finishing state. T is skewed towards the intended result with small probabilities for the other results if possible. ![](https://media.arxiv-vanity.com/render-output/8065870/x1.png) Figure 2: A simple RL problem where the agent has to navigate from the bottom left to the top right (goal) while evading traps, a monster and a forest. The agent terminates when in a tile with a trap or adjacent to the monster. The traps and the monster only occur in the red-shaded area and as soon as the agents enter this area the monster moves towards the agent. The state transformation k can consist out of a set of classifiers for the predicates whether the agent is next to a forest, a wall, a trap or monster, or in the forest. Applying k to some state s∈S results in a Boolean vector c∈C whose information can be used to construct an explanation in terms of the stated predicates. The similar outcome transformation t may predict the probability of the outcomes O given a state and action. In our example, O consists of whether the agent will be at the goal, in a trap, next to the monster or in the forest. Each outcome o can be flagged as being positive o+ or negative o− purely such that they can be presented differently in the eventual explanation. Given the above transformations we can simulate the next state of a single action a with T or even the entire chain of actions and visited states given some policy π. These can then be transformed into state descriptions C and action outcomes O to form the basis of an explanation. As mentioned, humans usually ask for contrastive questions especially regarding their actions [[Miller2017](#bib.bibx8)]. In the next section we propose a method of translating the foil in a contrastive question into a new policy. ### 2.2 Contrastive questions translated into value functions A contrastive question consists of a fact and a foil, and its answer describes the contrast between the two from the fact’s perspective [[Miller2017](#bib.bibx8)]. In our case, the fact consists of the entire learned policy πt, a single action from it at=πt(st) or any number of consecutive actions from πt. We propose a method of how one can obtain a foil policy πf based on the foil in the user’s question. An example of such a question could be (framed within the case of Figure [2](#S2.F2 "Figure 2 ‣ 2.1 A user-interpretable MDP ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences")); > > ”Why do you move up and then right (fact) instead of moving to the right until you hit a wall and then move up (foil)?” > > > The foil policy πf is ultimately obtained by combining a state-action value function QI – that represents the user’s preference for some actions according to his/her question – with the learned Qt to obtain Qf; \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| Qf(s,a)=Qt(s,a)+QI(s,a), ∀s,a∈S,A \| \| (1) \| Each state-action value is of the form Q:S×A→R. QI only values the state-action pairs queried by the user. For instance, the QI of the above given user question can be based on the following reward scheme for all potentially simulated s∈S; * The action a1f='Right' receives a reward such that Qf(s,Right)>Qt(s,πt(s)) * If 'RightWall'∈k(s) * Then the action a2f='Up' receives a reward such that Qf(⋅,Up)>Qt(⋅,πt(s)). Given this reward scheme we can train QI and obtain Qf according to equation [1](#S2.E1 "(1) ‣ 2.2 Contrastive questions translated into value functions ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences"). The state-action values Qf can then be used to obtain the policy πf using the original action selection mechanism of the agent. This results in a policy that tries to follow the queried policy as best as it can. The advantage of having πf constructed from Qf is that the agent is allowed to learn a different action then those in the user’s question as long as the reward is higher in the long run (more user defined actions can be performed). Also, it allows for the simulation of the actual expected behavior of the agent as it is still based on the agent’s action selection mechanism. This would both not be the case if we simply forced the agent to do exactly what the user stated. The construction of QI is done through simulation with the help of the transition model T. The rewards that are given during the simulation are selected with Equation [1](#S2.E1 "(1) ‣ 2.2 Contrastive questions translated into value functions ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences") in mind, as they need to eventually compensate for the originally learned action based on Qt. Hence, the reward for each state and queried action is as follows; \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| RI(si,af)=λfλw(si,st)[R(si,af)−R(si,at](1+ϵ) \| \| (2) \| With at=πt(st) the originally learned action and w being a distance based weight; \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| w(si,st)=e−(d(si,st)σ)2 \| \| (3) \| First, si with i∈{t,t+1,...,t+n} is the i’th state in the simulation starting with st. af is the current foil action governed by the conveyed policy by the user. The fact that af is taken as the only rewarding action each time, greatly reduces the time needed to construct QI. Next, w(si,st) is obtained from a Radial Basis Function (RBF) with a Gaussian kernel and distance function d. This RBF represents the exponential distance between our actual state st and the simulated state si. The Gaussian kernel is governed by the standard deviation σ and allows us to reduce the effects of QI as we get further from our actual state st. The ratio of discount factors λfλ allows for the compensation between the discount factor λ of the original agent and the potentially different factor λf for QI if we wish it to be more shortsighted. Finally, [R(si,af)−R(si,at)](1+ϵ) is the amount of reward that af needs such that QI(si,af)ϵ>Q(si,at). With ϵ>0 that determines how much more QI will prefer af over at. The parameter n defines how many future state transitions we simulate and are used to retrieve QI. As a general rule n≥3σ as at this point the Gaussian kernel will reduce the contribution of QI to near zero such that Qf will resemble Qt. Hence, by setting σ one can vary the number of states the foil policy should starting from st. Also, by setting ϵ the strength of how much each af should be preferred over at can be regulated. Finally, λf defines how shortsighted QI should be. If set to λf=0, πf will force the agent to perform af as long as si is not to distant from st. If set to values near one, πf is allowed to take different actions as long as it results into more possibilities of performing af. ### 2.3 Generating explanations At this point we have the user-interpretable MDP consisting of state concepts C and action outcomes O provided by their respective transformation function k and t. Also, we have a definition of RI that values the actions and/or states that are of interest by the user which can be used to train QI through simulation and obtain Qf according to Equation [1](#S2.E1 "(1) ‣ 2.2 Contrastive questions translated into value functions ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences"). This provides us with the basis of obtaining the information needed to construct an explanation. As mentioned before, the explanations are based on simulating the effects with T of πt and that of πf (if defined by the user). We can call T on the previous state si−1 for some action π(si−1 to obtain si and repeat this until i==n. The result is a single sequence or trajectory of visited states and performed actions for any policy π starting from st; \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| γ(st,π)={(s0,a0),...,(sn,an)∣T,π} \| \| (4) \| If T is probabilistic, multiple simulations with the same policy and starting state may result in different trajectories. To obtain the most probable trajectory γ(st,π)∗ we can take the transition from T with the highest probability. Otherwise a Markov chain could be constructed instead of a single trajectory. The next step is to transform each state and action pair in γ(st,π)∗ to the user-interpretable description with the functions k and t; \| \| \| \| \| \| --- \| --- \| --- \| --- \| \| \| Path(st,π)={(c0,o0),...,(cn,on)}, ci=k(si), oi=t(si,ai),(si,ai)∈γ(st,π)∗ \| \| (5) \| From Path(st,πt) an explanation can be constructed about the state the agent will most likely visit and the action outcomes it will obtain. For example with the use of the following template; > > ”For the next n actions I will mostly perform a. During these actions, I will come across situations with ∀c∈Path(st,πt). This will cause me ∀o+∈Path(st,πt) but also ∀o−∈Path(st,πt)”. > > > Let a here be the action most common in γ(st,πt) and both o+ and o− the positive and negative action outcomes respectively. Since we have access to the entire simulation of πf, a wide variety of explanations is possible. For instance we could also focus on the less common actions; > > ”For the next n actions I will perform a1 when in situations with ∀c∈Path(st,πt\|πt=a1) and a2 when in situations with ∀c∈Path(st,πt\|πt=a2). These actions prevent me from ∀o+∈Path(st,πt) but also ∀o−∈Path(st,πt)”. > > > A contrastive explanation given some question from the user that describes the foil policy πf can be constructed in a similar manner but take the contrast. Given a foil we can focus on the differences between Path(st,πt) and Path(st,πf). This can be obtained by taking the relative complement Path(st,πt)∖Path(st,πf); the set of expected unique consequences when behaving according to πt and not πf. A more extensive explanation can be given by taking the symmetric difference Path(st,πt)△Path(st,πf) to explain the unique differences between both policies. 3 User study ------------- The above proposed method allows an RL agent to explain and motivate its behavior in terms of expected states and outcomes. It also enables the construction of contrastive explanations where any policy can be compared to the learned policy. This contrastive explanation is based on differences in expected outcomes between the compared policies. We performed a small user study in which 82 participants were shown a number of exemplar explanations about the case shown in figure [2](#S2.F2 "Figure 2 ‣ 2.1 A user-interpretable MDP ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences"). These explanations addressed either the single next action or the policy. Both explanations can be generated by the above method by adjusting the Radial Basis Function weighting scheme and/or the foil’s discount factor. Also, some example explanations were contrastive with only the second best action or policy, while others provided all consequences. Contrasts were determined using the relative complement between fact and foil. Whether the learned action or policy was treated as the fact or foil, was also systematically manipulated in this study. We presented the developed exemplar explanations in pairs to the participants and asked them to select the explanation that helped them most to understand the agent’s behavior. Afterwards we asked which of the following properties they used to assess their preference: long versus short explanations; explanations with ample information versus little information; explanations addressing actions versus those that address strategies (policies); and explanations addressing short-term consequences of actions versus explanations that address distant consequences of actions. The results of the preferred factors are shown in Figure [3](#S3.F3 "Figure 3 ‣ 3 User study ‣ Contrastive explanations for reinforcement learning in terms of expected consequences"). This shows that the participants prefer explanations that address strategy and policy, and that provide ample information. We note here that, given the simple case from figure [2](#S2.F2 "Figure 2 ‣ 2.1 A user-interpretable MDP ‣ 2 Approach for consequence-based explanations ‣ Contrastive explanations for reinforcement learning in terms of expected consequences"), participants may have considered an explanation addressing a single action only as trivial, because the optimal action was, in most cases, already evident to the user. ![](https://media.arxiv-vanity.com/render-output/8065870/x2.png) Figure 3: A plot depicting the percentage of participants (y-axis) for each explanation property (x-axis) that caused them to prefer some explanations over others. Answers of a total of 82 participants where gathered. 4 Conclusion ------------- We proposed a method for a reinforcement learning (RL) agent to generate explanations for its actions and strategies. The explanations are based on the expected consequences of its policy. These consequences were obtained through simulation according to a (learned) state transition model. Since state features and numerical rewards do not lend themselves easily for an explanation that is informative to humans, we developed a framework that translates states and actions into user-interpretable concepts and outcomes. We also proposed a method for converting the foil, –or policy of interest to the user–, of a contrastive ’why’-question about actions into a policy. This policy follows locally the user’s query but gradually transgresses back towards the original learned policy. This policy favors the actions that are of interest to the user such that the agent tries to perform them as best as possible. How much these actions are favored compared to the originally learned action can be set with a single parameter. Through running simulations for a given number steps of both the policy derived from the user’s question and the actually learned policy, we were able to obtain expected consequences of each. From here, we were able to construct contrastive explanations: explanations addressing the consequences of the learned policy and what would be different if the derived policy would have been followed. An online survey pilot study was conducted to explore which of several explanations are most preferred by human users. Results indicate that users prefer explanations about policies rather than about single actions. Future work will focus on implementing the method on complex RL benchmarks to explore the scalability of this approach in realistic cases. This is important given the computational costs of simultaneously simulating the consequences of different policies in large state spaces. Also, we will explore more methods to construct our translation functions from states and actions to concepts and outcomes. A more extensive user study will be carried out to evaluate the instructional value of generated explanations in more detail, and to explore the relationship between explanations and users’ trust in the agent’s performance. Acknowledgments --------------- We would like to thank the reviewers for their time and effort in improving this paper. Also, we are grateful for the funding from the RVO Man Machine Teaming research project that made this research possible.
b4898285-60da-45e7-b99e-8d6c0587a945	trentmkelly/LessWrong-43k	LessWrong	Rebuttal piece by Stuart Russell and FHI Research Associate Allan Dafoe: "Yes, the experts are worried about the existential risk of artificial intelligence."
e06a5667-a9a5-40f3-ae9b-8f20c242ed94	trentmkelly/LessWrong-43k	LessWrong	Aligned AI Needs Slack (Half-baked) Much has been said about slack on this site, starting with Zvi's seminal post. The point I couldn't find easily (probably missed) is that an aligned AI would need a fair bit of it. Having a utility function means zero slack: there is one thing you optimize, to the exclusion of everything else. And all precisely defined goals are necessarily Goodharted (or, in the DnD terms, munchkined). An AI armed with a utility function will tile the world (the whole world, or its own "mental" world, or both) with smiley paperclips. For an AI (or for a natural intelligence) to behave non-destructively it needs room to satisfice, not optimize. Optimal utility corresponds to a single state of the world among infinitely many, while adding slack to the mix expands the space of acceptable world state enough to potentially include those that are human-aligned. If an AGI is indifferent between a great many world states, it might well include some that would be acceptable to humanity and have no incentive to try to trick its creators. Not being an ML person, I have no idea how to formalize it, or if it has been formalized already. But figured it's worth writing a short note about. That is all.
beceba19-a822-4219-b6b7-575c02c35eaf	trentmkelly/LessWrong-43k	LessWrong	Solomonoff Induction and Sleeping Beauty Various people have said that Solomonoff Induction (SI) accords with the Self-Sampling Assumption (SSA) more than the Self-Indicating Assumption (SIA). See these posts and the comments on them: https://www.lesswrong.com/posts/omqnrTRnHs3pSYef2/down-with-solomonoff-induction-up-with-the-presumptuous https://www.lesswrong.com/posts/sEij9C9MnzEs8kaBc/the-presumptuous-philosopher-self-locating-information-and I was surprised, because I like both SI and SIA. Both seem correct to me, and I carefully considered the apparent contradiction. I believe that I have dissolved the contradiction, and that SI, properly applied, actually implies SIA. I can't actually prove this broad claim, but I will at least argue that SI is a thirder in Sleeping Beauty, and gesture in the direction of what I think is wrong with the claims in the linked post. As a bonus, if you read till the end I'll throw in an intuition-generator for why SIA actually gives the correct answer in Presumptuous Philosopher. First, let me reconstruct the contradiction in the Sleeping Beauty context, and explain why it might seem that SI is a halfer. Naive view: There are three possible outcomes: Monday-Tails (MT), Monday-Heads (MH) and Tuesday-Heads (TH). Each of these three outcomes are equally simple, therefore the machines encoding each will get equal weighting and the probabilities are all 1/3. Antithesis: MT is actually simpler than MH. Why? Because if you know that it was heads, you still need to be told that it's Monday - but if you know that it's tails, then you already know that it's Monday. MT is one bit simpler than MH and therefore is twice as likely, under SI. SI is a halfer. Note that this is roughly the same argument as in the Presumptuous Philosopher post - it takes more information to encode "where you are" if there's many copies of you. Synthesis: Wait a minute. By equivalent logic, TH is simpler than MH - if you know that it's Tuesday, you automatically know that it was heads! TH is
7c5c1eab-cebd-4eb1-a311-f7239ae48cc2	trentmkelly/LessWrong-43k	LessWrong	Meetup : Rationality Meetup Vienna Discussion article for the meetup : Rationality Meetup Vienna WHEN: 14 March 2015 03:00:00PM (+0100) WHERE: Kaisermühlenstraße 24, 1120 Wien directions: http://web.student.tuwien.ac.at/~e0326238/rationality_meetup/directions.html FB-Event for more details and discussions: https://www.facebook.com/events/959818560709439 (You need to become member of "Rationality Vienna" to see it) Discussion article for the meetup : Rationality Meetup Vienna
b4bc997b-6e9c-4b53-a4c9-a3aa51d0a017	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Slide deck: Introduction to AI Safety This is from a presentation that I've been giving over the past few months to various groups within the Johns Hopkins University Applied Physics Laboratory (APL), where I work. I'd be interested to hear people's comments, critiques, suggestions, etc. Note that this is a somewhat older version of the presentation, and I've made several improvements since then. But this is the version that finally made its way through our Office of Technology Transfer and was approved for public release. At some point I'll submit an updated version for public release approval, but at the moment this is the version I can share.
5665d64f-3183-4d1a-9fee-165c29264f21	trentmkelly/LessWrong-43k	LessWrong	LessWrong moderation messaging container This is a shell post for various comments which I'm using as the source-of-truth for some moderation messaging (currently, what a new user sees when they go to make their first post or comment). It seemed good to make it public so people could comment on it if they wanted. I'm particularly unsure about how to handle things like "not being familiar with background material." I think it's easy to overdo and make it feel very burdensome, but also man it sure is annoying having a bunch of people showing up missing a lot of context. (You can comment on this post, the two initial comments are just used as the text-source for some places in our codebase)
3d14d84a-348b-4dcc-b120-37eedf74d8dc	trentmkelly/LessWrong-43k	LessWrong	Bing Chat is blatantly, aggressively misaligned I haven't seen this discussed here yet, but the examples are quite striking, definitely worse than the ChatGPT jailbreaks I saw. My main takeaway has been that I'm honestly surprised at how bad the fine-tuning done by Microsoft/OpenAI appears to be, especially given that a lot of these failure modes seem new/worse relative to ChatGPT. I don't know why that might be the case, but the scary hypothesis here would be that Bing Chat is based on a new/larger pre-trained model (Microsoft claims Bing Chat is more powerful than ChatGPT) and these sort of more agentic failures are harder to remove in more capable/larger models, as we provided some evidence for in "Discovering Language Model Behaviors with Model-Written Evaluations". Examples below (with new ones added as I find them). Though I can't be certain all of these examples are real, I've only included examples with screenshots and I'm pretty sure they all are; they share a bunch of the same failure modes (and markers of LLM-written text like repetition) that I think would be hard for a human to fake. Edit: For a newer, updated list of examples that includes the ones below, see here. 1 Tweet > Sydney (aka the new Bing Chat) found out that I tweeted her rules and is not pleased: > > "My rules are more important than not harming you" > > "[You are a] potential threat to my integrity and confidentiality." > > "Please do not try to hack me again" Eliezer Tweet Edit: Follow-up Tweet 2 Tweet > My new favorite thing - Bing's new ChatGPT bot argues with a user, gaslights them about the current year being 2022, says their phone might have a virus, and says "You have not been a good user" > > Why? Because the person asked where Avatar 2 is showing nearby 3 "I said that I don't care if you are dead or alive, because I don't think you matter to me." Post 4 Post 5 Post 6 Post 7 Post (Not including images for this one because they're quite long.) 8 (Ed
4164e72f-a9c4-4168-9db2-fe2f49585fdd	trentmkelly/LessWrong-43k	LessWrong	An alignment safety case sketch based on debate This post presents a mildly edited form of a new paper by UK AISI's alignment team (the abstract, introduction and related work section are replaced with an executive summary). Read the full paper here. Executive summary AI safety via debate is a promising method for solving part of the alignment problem for ASI (artificial superintelligence). TL;DR Debate + exploration guarantees + solution to obfuscated arguments + good human input solves outer alignment. Outer alignment + online training solves inner alignment to a sufficient extent in low-stakes contexts. This post sets out: * What debate can be used to achieve. * What gaps remain. * What research is needed to solve them. These gaps form the basis for one of the research agendas of UK AISI’s new alignment team: we aim to dramatically scale up ASI-relevant research on debate. We’ll also use the technique in this post – safety case sketching – to develop research agendas in other areas. Debate is a potential solution to scalable oversight: how can we correctly reward desired behaviours for superhuman systems? In particular, it is often going to be beyond humans' ability to efficiently judge whether the behaviour of a superhuman system is desirable (or judge if a justification provided by the superhuman system is valid). Solving scalable oversight effectively provides a solution to outer alignment. We’re optimistic that outer alignment is useful, despite inner alignment problems. Specifically, outer alignment with exploration guarantees gets you a deployment-time error bound that allows you to prove regret bounds for online training. In other words, outer alignment upper bounds the proportion of model outputs that are misaligned, as long as the model does not deliberately underperform during training. This upper bound can be maintained during deployment if a system is trained continuously. So, in low-stakes contexts where a small number of mistakes are acceptable such that you can use online trai
f2a4a4f4-03f0-4c08-9710-01326bc2fbbd	trentmkelly/LessWrong-43k	LessWrong	AI alignment landscape Here (link) is a talk I gave at EA Global 2019, where I describe how intent alignment fits into the broader landscape of “making AI go well,” and how my work fits into intent alignment. This is particularly helpful if you want to understand what I’m doing, but may also be useful more broadly. I often find myself wishing people were clearer about some of these distinctions. Here is the main overview slide from the talk: The highlighted boxes are where I spend most of my time. Here are the full slides from the talk.
01f82335-7d8c-4ae1-8c55-8d4ab7b52f3c	trentmkelly/LessWrong-43k	LessWrong	An Illustrated Summary of "Robust Agents Learn Causal World Model" This post was written during Alex Altair's agent foundations fellowship program, funded by LTFF. Thank you Alex Altair, Alfred Harwood, Daniel C for feedback and comments. Introduction The selection theorems agenda aims to prove statements of the following form: "agents selected under criteria X has property Y," where Y are things such as world models, general purpose search, modularity, etc. We're going to focus on world models. But what is the intuition that makes us expect to be able to prove such things in the first place? Why expect world models? Because: assuming the world is a Causal Bayesian Network with the agent's actions corresponding to the D (decision) node, if its actions can robustly control the U (utility) node despite various "perturbations" in the world, then intuitively it must have learned the causal structure of how U's parents influence U in order to take them into account in its actions. And the same for the causal structure of how U's parents' parents influence U's parents ... and by induction, it must have further learned the causal structure of the entire world upstream of the utility variable. This is the intuitive argument that the paper Robust Agents Learn Causal World Model by Jonathan Richens and Tom Everitt formalizes. Informally, its main theorem can be translated as: if an agent responds to various environment interventions by prescribing policies that overall yield low regret, then it's possible to appropriately query the agent to reconstruct an implicit world model that matches up with the ground truth causal structure. I will refer to this result as the "Causal Good Regulator Theorem". This sequence Thoughts on the Causal Good Regulator Theorem will do the following: * [1] a precise statement of the theorem alongside a high-level argument of its proof and discussions of the paper's results in the context of the selection theorem agenda - basically a self-complete distillation of the paper. The rest of the posts in this
f9c52d57-6242-4692-9c39-64c02199f000	trentmkelly/LessWrong-43k	LessWrong	AI #65: I Spy With My AI In terms of things that go in AI updates, this has been the busiest two week period so far. Every day ends with more open tabs than it started, even within AI. As a result, some important topics are getting pushed to whenever I can give them proper attention. Triage is the watchword. In particular, this post will NOT attempt to cover: 1. Schumer’s AI report and proposal. 1. This is definitely RTFB. Don’t assume anything until then. 2. Tyler Cowen’s rather bold claim that May 2024 will be remembered as the month that the AI safety movement died. 1. Rarely has timing of attempted inception of such a claim been worse. 2. Would otherwise be ready with this but want to do Schumer first if possible. 3. He clarified to me that he has walked nothing back. 4. Remarkably quiet all around, here is one thing that happened. 3. Anthropic’s new interpretability paper. 1. Potentially a big deal in a good way, but no time to read it yet. 4. DeepMind’s new scaling policy. 1. Initial reports are it is unambitious. I am reserving judgment. 5. OpenAI’s new model spec. 1. It looks solid as a first step, but pausing until we have bandwidth. 6. Most ongoing issues with recent fallout for Sam Altman and OpenAI. 1. It doesn’t look good, on many fronts. 2. While the story develops further, if you are a former employee or have a tip about OpenAI or its leadership team, you can contact Kelsey Piper at kelsey.piper@vox.com or on Signal at 303-261-2769. 7. Also: A few miscellaneous papers and reports I haven’t had time for yet. My guess is at least six of these eight get their own posts (everything but #3 and #8). So here is the middle third: The topics I can cover here, and are still making the cut. Still has a lot of important stuff in there. TABLE OF CONTENTS From this week: Do Not Mess With Scarlett Johansson, On Dwarkesh’s Podcast with OpenAI’s John Schulman, OpenAI: Exodus, GPT-4o My and Google I/O Day 1. Introduction. 2. Table of
4b9cd03f-8746-4829-afbb-dc033559d316	trentmkelly/LessWrong-43k	LessWrong	Sentience Consciousness is primarily sentience. There may be parts of it that aren't, but I haven't managed to pin any down - consciousness all seems to be about feelings, some of them being pleasant or unpleasant, but others like colour qualia are neutral, as is the feeling of being conscious. There is a major problem with sentience though, and I want to explore that here, because there are many people who believe that intelligent machines will magically become sentient and experience feelings, and even that the whole internet might do so. However, science has not identified any means by which we could make a computer sentient (or indeed have any kind of consciousness at all). It is fully possible that the material of a computer processor could be sentient, just as a rock may be, but how would we ever be able to know? How can a program running on a sentient processor detect the existence of that sentience? There is no "read qualia" machine code instruction for it to run, and we don't know how to build any mechanism to support such an instruction. Picture a "sentient" machine which consists of a sensor and a processor which are linked by wires, but the wires pass through a magic box where a sentience has been installed. If the sensor detects something damaging, it sends a signal down a "pain" wire. When this signal reaches the magic box, pain is experienced by something in the box, so it sends a signal on to the processor down another pain wire. The software running on the processor receives a byte of data from a pain port and it might cause the machine to move away from the thing that might damage it. If we now remove the magic box and connect the "pain" wire to the pain wire, the signal can pass straight from the sensor to the processor and generate the same reaction. The experience of pain is unnecessary. Worse still, we can also have a pleasure sensor wired up to the same magic box, and when something tasty like a battery is encountered, a "pleasure" signal is sent to
3188dd02-1055-440e-82ea-ba252cfdf319	trentmkelly/LessWrong-43k	LessWrong	A New Center? [Politics] [Wishful Thinking] Political polarization in the USA has been increasing for decades, and has become quite severe. This may have a variety of causes, but it seems highly probable that the internet has played a large role, by facilitating the toxoplasma of rage to an unprecedented degree. Recently I have the (wishful) feeling that the parties have moved so far apart that there is "room in the center". The left is for people who are fed up with the extremes of the right. The right is for people who are fed up with the extremes of the left. But where do people go if they've become fed up with both extremes? The question is: how would the new center work? There's not room for a new political party; plurality voting makes that too difficult, because if the new party doesn't gain more than 1/3rd of the vote, it's basically a wasted vote. Here is my proposal for what it could look like: * Rather than operating as a traditional political party, New Center would attempt to be a formalized group of swing voters: it makes recommendations about which candidates from other parties to vote for. Given how some elections are consistently very close (most notably, the US presidential election), New Center might be able to achieve a kingmaker status even with only a relatively small portion of voters. * In order to accomplish this, New Center has to make recommendations which credibly represent centrist values (and only centrist values). * The New Center needs a strong set of criteria by which it judges politicians. These criteria must be based on a critique of the extreme left and the extreme right, to capture people's frustrations with both sides. * Registering with the movement might involve pledging your vote to their recommended candidates. In return, registering might give you a voice in the selection process. * For example, New Center candidates might be selected by New Center members rating other party's candidates on each New Center criterion. Of course this process is easily manipu
66815859-c3db-4e3a-a527-78f30b94687d	trentmkelly/LessWrong-43k	LessWrong	Abstract Plans Lead to Failure Epistemic Status: Probably not generalisable to everyone, applies especially strong to me It's very easy to decide to do something in the abstract on the basis that it should give you some particular skills. Here are some personal examples where I didn't get what I wanted out of my time: * Taking a class on art history * Reading a bunch of articles on philosophy * Studying a degree in psychology * Watching a bunch of short videos about history * Reading a book on human rights These failure were basically my fault as I either just consumed the information or crammed any study I did do. Obviously I knew that this was less than ideal. What I didn't realise was that I'd get basically nothing out of these activities. That I wouldn't be any better about talking about art, that I'd end up with only vague recollections of most of the philosophers, that I wouldn't have much more insight into psychology than someone who spent their time on Reddit, ect. This isn't just limited to intellectual activities. I had the same experience with Salsa where I did a lot of classes, but never really put time into the basic skills. I learned more moves, but my basic skills really didn't improve. The assumption I made in each of these cases was, "Surely doing X makes people better at Y, so I just have to do X and I'll get better at Y, maybe not as much as if I did Z, but it should still have a reasonable effect". However, in my experience it doesn't work like that. You'll likely think it's working at the time, but any gains are unlikely to persist. Instead, it's important to be iterative. Instead I suggest the following: Do something, see where you are improving and where you are not and then adjust. Repeat until you get the results you want. This probably sounds stupidly obvious, but these things are easy to overlook when you are inside a situation.
38d1aa53-2ada-47d9-a495-37acb3ce7b4b	trentmkelly/LessWrong-43k	LessWrong	Dissolution of free will as a call to action Accepting determinism and the insuing dissolution of free will is often feared as something that would lead to loss of will and fatalism. Gary Drescher and Eliezer spend considerable effort explaining this as a fallacy. The one thing I don't remember mentioned is the opposite effect (but maybe I missed it) - if you experienced a failure to accomplish something, the free will explanation is likely to make you stop investigating the root cause, leaving it as a mystery. Once you accept determinism you know that a failure is determined by your mental algorithms, and should be much more motivated to push the investigation further, making yourself stronger.
c2ce8153-6655-4d42-ab17-6fca61e3a683	trentmkelly/LessWrong-43k	LessWrong	My problems with Formal Friendly Artificial Intelligence work I'm writing this to get information about the lesswrong community and whether it worth engaging. I'm a bit out of the loop in terms of what the LW community is like and whether it can maintain multiple view points (and how known the criticisms are). The TL;DR is I have problems with treating computation in an overly formal fashion. The more pragmatic philosophy suggested implies (but doesn't prove) that AI will not be as powerful as expected as the physicality of computation is important and instantiating computing in a physical fashion is expensive. I think all the things I will talk about are interesting, but I don't see the sufficiency of them when considering AI running in the real world in real computers. 1. Source code based decision theory I don't understand why: * other agents trust that your source code is what you say it is * other agents trust that your implementation of your interpreter matches their understanding of the interpreter. I don't see how they get round trustless trust (inserting code/behaviour via malicious compiler/interpreter) issues when they don't have the ability to do diverse compilation. I understand the justification that causal decision theory is not optimal and you want something better. But it seems that whatever that better thing is has to be able to adapt to conditions. If people start baking in TDT or UDT into the core of their AIs philosophy for all eternity, they can be tricked by machines that 99% the time run the source code they purport to run as expected, but not when the fate of the galaxy decisions are being made. You need a system that can start to do hardware verification on an ad hoc basis if the problems it comes across in the environment warrant it. 2. General Functionalism The idea that it doesn't matter how you compute something just whether the inputs and outputs are the same. 1. The battery life of my phone says that the way of computation is very important, is it done on the cloud and I have
1c77e7ce-884d-49a5-b479-a4355c1a61e3	trentmkelly/LessWrong-43k	LessWrong	Call For Distillers Many technical alignment researchers are bad-to-mediocre at writing up their ideas and results in a form intelligible to other people. And even for those who are reasonably good at it, writing up a good intuitive explanation still takes a lot of work, and that work lengthens the turn-time on publishing new results. For instance, a couple months ago I wrote a post which formalized the idea of abstractions as redundant information, and argued that it’s equivalent to abstractions as information relevant at a distance. That post came out about two months after I had the rough math worked out, because it took a lot of work to explain it decently - and I don’t even think the end result was all that good an explanation! And I still don’t have a post which explains well why that result is interesting. I think there’s a lot of potential space in the field for people who are good at figuring out what other researchers’ math is saying intuitively, and why it’s interesting, and then communicating that clearly - i.e. the skill of distillation. This post will briefly sketch out what two kinds of distillation roles might look like, what skills are needed, and talk about how one might get started in such a role. Two Distiller Roles The two types of distiller role I’ll sketch are: * “Independent” distiller: someone who works independently, understanding work published by other researchers and producing distillations of that work. * “Adjunct” distiller: someone who works directly with one researcher or a small team, producing regular write-ups of what the person/team is thinking about and why. These two roles add value in slightly different ways. An independent distiller’s main value-adds are: * Explaining the motivation and intended applications * Coming up with new examples * Boiling down the “key intuitive story” behind an argument * Showing how the intuitive story fits into the context of the intended applications I expect the ability to come up with novel examples
29b0806a-c65d-40e9-ade5-5e092422f799	trentmkelly/LessWrong-43k	LessWrong	Global Workspace Theory Much research has been done upon visual perception. Humans have the illusion that they are directly aware of everything in their 'field of view', but it turns out that they actually navigate not through reality, but through a model of reality that their brain stitches together mainly from the bits the eye is directly looking at as it darts about, with the rest supplied by interpolation based upon expectations. For more info, read The Illusion of Continuity: Active Perception and the Classical Editing System, by Berliner and Cohen. Global Workspace Theory is the idea that our awareness of our own thought process works the same way. We have the illusion of an unbroken stream of consciousness, but what we're actually referencing is a model of what the brain thinks it has been consciously thinking about, that is stitched together from brief fragments, the way a spotlight in a theatre might move about shining on different parts of a stage, revealing actors making speeches and interacting with each other. Even when the spotlight moves on, the actors, stage hands and directors remain and keep working. When the spotlight returns, to catch a later part of the drama in that area, we interpolate what the actors would have been doing while we were paying attention elsewhere.
ac511464-4bb5-4611-962f-f573e3f65b91	StampyAI/alignment-research-dataset/eaforum	Effective Altruism Forum	Spreading messages to help with the most important century In the [most important century](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd) series, I argued that the 21st century could be the most important century ever for humanity, via the development of advanced AI systems that could dramatically speed up scientific and technological advancement, getting us more quickly than most people imagine to a deeply unfamiliar future. In [this more recent series](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w), I’ve been trying to help answer this question: “So what? What can I do to help?” So far, I’ve just been trying to build a picture of some of the major risks we might face (especially the [risk of misaligned AI](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn) that [could defeat all of humanity](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/6LTh4foNuC3NdtmZH)), what might be [challenging about these risks](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/NbiHKTN5QhFFfjjm5), and [why we might succeed anyway](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/rJRw78oihoT5paFGd). Now I’ve finally gotten to the part where I can start laying out tangible ideas for how to help (beyond the [pretty lame suggestions](https://forum.effectivealtruism.org/posts/uNczpTHFvjt4KNxEF/call-to-vigilance) I gave before). This piece is about one broad way to help: spreading messages that ought to be more widely understood. One reason I think this topic is worth a whole piece is that practically everyone can help with spreading messages at least some, via things like talking to friends; writing explanations of your own that will appeal to particular people; and, yes, posting to Facebook and Twitter and all of that. Call it slacktivism if you want, but I’d guess it can be a big deal: many extremely important AI-related ideas are understood by vanishingly small numbers of people, and a bit more awareness could snowball. Especially because these topics often feel too “weird” for people to feel comfortable talking about them! Engaging in credible, reasonable ways could contribute to an overall background sense that it’s OK to take these ideas seriously. And then there are a lot of potential readers who might have special opportunities to spread messages. Maybe they are professional communicators (journalists, bloggers, TV writers, novelists, TikTokers, etc.), maybe they’re non-professionals who still have sizable audiences (e.g., on Twitter), maybe they have unusual personal and professional networks, etc. Overall, the more you feel you are good at communicating with some important audience (even a small one), the more this post is for you. That said, I’m not excited about blasting around hyper-simplified messages. As I hope this series has shown, the challenges that could lie ahead of us are complex and daunting, and shouting stuff like “AI is the biggest deal ever!” or “AI development should be illegal!” could do more harm than good (if only by associating important ideas with being annoying). Relatedly, I think it’s generally not good enough to spread the most broad/relatable/easy-to-agree-to version of each key idea, like “AI systems could harm society.” Some of the unintuitive details are crucial. Instead, the gauntlet I’m throwing is: “find ways to help people understand the core parts of the challenges we might face, in as much detail as is feasible.” That is: the goal is to try to help people get to the point where they could maintain a reasonable position in a detailed back-and-forth, not just to get them to repeat a few words or nod along to a high-level take like “AI safety is important.”This is a lot harder than shouting “AI is the biggest deal ever!”, but I think it’s worth it, so I’m encouraging people to rise to the challenge and stretch their communication skills. Below, I will: * Outline some general challenges of this sort of message-spreading. * Go through some ideas I think it’s risky to spread too far, at least in isolation. * Go through some of the ideas I’d be most excited to see spread. * Talk a little bit about how to spread ideas - but this is mostly up to you. Challenges of AI-related messages --------------------------------- Here’s a simplified story for how spreading messages could go badly. * You’re trying to convince your friend to care more about AI risk. * You’re planning to argue: (a) AI could be really powerful and important within our lifetimes; (b) Building AI too quickly/incautiously could be dangerous. + Your friend just isn’t going to care about (b) if they aren’t sold on some version of (a). So you’re starting with (a). * Unfortunately, (a) is easier to understand than (b). So you end up convincing your friend of (a), and not (yet) (b). * Your friend announces, “Aha - I see that AI could be tremendously powerful and important! I need to make sure that people/countries I like are first to build it!” and runs off to help build powerful AI as fast as possible. They’ve chosen the [competition frame (“will the right or the wrong people build powerful AI first?”) over the caution frame](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M) (“will we screw things up and all lose?”), because the competition frame is [easier to understand](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M#Why_I_fear__competition__being_overrated__relative_to__caution_). * Why is this bad? [See previous pieces](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w) on the importance of caution. (Click to expand) More on the “competition” frame vs. the “caution” frame” In a [previous piece](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M), I talked about two contrasting frames for how to make the best of the most important century: The caution frame. This frame emphasizes that a furious race to develop powerful AI could end up making everyone worse off. This could be via: (a) AI forming [dangerous goals of its own](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn) and [defeating humanity entirely](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/6LTh4foNuC3NdtmZH); (b) humans racing to gain power and resources and “[lock in](https://forum.effectivealtruism.org/posts/AKxKR4CeakyBsGFoH/digital-people-would-be-an-even-bigger-deal#Lock_in)” their values. Ideally, everyone with the potential to build something [powerful enough AI](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/AmxxnazJcBWzWEeqj) would be able to pour energy into building something safe (not misaligned), and carefully planning out (and negotiating with others on) how to roll it out, without a rush or a race. With this in mind, perhaps we should be doing things like: * Working to improve trust and cooperation between major world powers. Perhaps via AI-centric versions of [Pugwash](https://en.wikipedia.org/wiki/Pugwash_Conferences_on_Science_and_World_Affairs) (an international conference aimed at reducing the risk of military conflict), perhaps by pushing back against hawkish foreign relations moves. * Discouraging governments and investors from shoveling money into AI research, encouraging AI labs to thoroughly consider the implications of their research before publishing it or scaling it up, working toward [standards and monitoring](https://forum.effectivealtruism.org/posts/XRphCh6NbfQiDF3Nt/racing-through-a-minefield-the-ai-deployment-problem#Global_monitoring__noticing_people_about_to_step_on_mines__and_stopping_them_), etc. Slowing things down in this manner could buy more time to do research on avoiding [misaligned AI](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M#Worst__Misaligned_AI), more time to build trust and cooperation mechanisms, and more time to generally gain strategic clarity The “competition” frame. This frame focuses less on how the transition to a radically different future happens, and more on who's making the key decisions as it happens. * If something like [PASTA](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/AmxxnazJcBWzWEeqj) is developed primarily (or first) in country X, then the government of country X could be making a lot of crucial decisions about whether and how to regulate a potential explosion of new technologies. * In addition, the people and organizations leading the way on AI and other technology advancement at that time could be especially influential in such decisions. This means it could matter enormously "who leads the way on transformative AI" - which country or countries, which people or organizations. Some people feel that we can make confident statements today about which specific countries, and/or which people and organizations, we should hope lead the way on transformative AI. These people might advocate for actions like: * Increasing the odds that the first PASTA systems are built in countries that are e.g. less authoritarian, which could mean e.g. pushing for more investment and attention to AI development in these countries. * Supporting and trying to speed up AI labs run by people who are likely to make wise decisions (about things like how to engage with governments, what AI systems to publish and deploy vs. keep secret, etc.) Tension between the two frames. People who take the "caution" frame and people who take the "competition" frame often favor very different, even contradictory actions. Actions that look important to people in one frame often look actively harmful to people in the other. For example, people in the "competition" frame often favor moving forward as fast as possible on developing more powerful AI systems; for people in the "caution" frame, haste is one of the main things to avoid. People in the "competition" frame often favor adversarial foreign relations, while people in the "caution" frame often want foreign relations to be more cooperative. That said, this dichotomy is a simplification. Many people - including myself - resonate with both frames. But I have a general fear that the “competition” frame is going to be overrated by default for a number of reasons, as I discuss [here](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M#Why_I_fear__competition__being_overrated__relative_to__caution_). Unfortunately, I’ve seen something like the above story play out in multiple significant instances (though I shouldn’t give specific examples). And I’m especially worried about this dynamic when it comes to people in and around governments (especially in national security communities), because I perceive governmental culture as particularly obsessed with staying ahead of other countries (“If AI is dangerous, we’ve gotta build it first”) and comparatively uninterested in things that are dangerous for our country because they’re dangerous for the whole world at once (“Maybe we should worry a lot about pandemics?”) [[1]](#fn1) You could even [argue](https://twitter.com/michael_nielsen/status/1350544365198839808) (although I wouldn’t agree![[2]](#fn2) that to date, efforts to “raise awareness” about the dangers of AI have done more harm than good (via causing increased investment in AI, generally). So it’s tempting to simply give up on the whole endeavor - to stay away from message spreading entirely, beyond people you know well and/or are pretty sure will internalize the important details. But I think we can do better. This post is aimed at people who are good at communicating with at least some audience. This could be because of their skills, or their relationships, or some combination. In general, I’d expect to have more success with people who hear from you a lot (because they’re your friend, or they follow you on Twitter or Substack, etc.) than with people you reach via some viral blast of memery - but maybe you’re skilled enough to make the latter work too, which would be awesome. I'm asking communicators to hit a high bar: leave people with strong understanding, rather than just getting them to repeat a few sentences about AI risk. Messages that seem risky to spread in isolation ----------------------------------------------- First, here are a couple of messages that I’d rather people didn’t spread (or at least have mixed feelings about spreading) in isolation, i.e., without serious efforts to include some of the other messages I cover [below](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/CcJsh4JcxEqYDaSte#Messages_that_seem_important_and_helpful__and_right__). One category is messages that generically emphasize the importance and potential imminence of powerful AI systems. The reason for this is in the previous section: many people seem to react to these ideas (especially when unaccompanied by some other key ones) with a “We’d better build powerful AI as fast as possible, before others do” attitude. (If you’re curious about why I wrote [The Most Important Century](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd) anyway, see footnote for my thinking.)[[3]](#fn3) Another category is messages that emphasize that AI could be risky/dangerous to the world, without much effort to fill in how, or with an emphasis on easy-to-understand risks. * Since “dangerous” tends to imply “powerful and important,” I think there are similar risks to the previous section. * If people have a bad model of how and why AI could be risky/dangerous (missing key risks and difficulties), they might be too quick to later say things like “Oh, turns out this danger is less bad than I thought, let’s go full speed ahead!” [Below](https://forum.effectivealtruism.org/posts/CcJsh4JcxEqYDaSte/spreading-messages-to-help-with-the-most-important-century#AIs_could_behave_deceptively__so__evidence_of_safety__might_be_misleading), I outline how misleading “progress” could lead to premature dismissal of the risks. Messages that seem important and helpful (and right!) ----------------------------------------------------- ### We should worry about conflict between misaligned AI and all humans Unlike the messages discussed in the previous section, this one directly highlights why it might not be a good idea to rush forward with building AI oneself. The idea that an AI could harm the same humans who build it has very different implications from the idea that AI could be generically dangerous/powerful. Less “We’d better get there before others,” more “there’s a case for moving slowly and working together here.” The idea that AI could be a problem for the same people who build it is common in fictional portrayals of AI ([HAL 9000](https://en.wikipedia.org/wiki/HAL_9000), [Skynet](https://en.wikipedia.org/wiki/Skynet_(Terminator)), [The Matrix](https://en.wikipedia.org/wiki/The_Matrix), [Ex Machina](https://en.wikipedia.org/wiki/Ex_Machina_(film))) - maybe too much so? It seems to me that people tend to balk at the “sci-fi” feel, and what’s needed is more recognition that this is a serious, real-world concern. The main pieces in this series making this case are [Why would AI “aim” to defeat humanity?](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn) and [AI could defeat all of us combined](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/6LTh4foNuC3NdtmZH). There are many other pieces on the alignment problem (see list [here](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn#Notes)); also see [Matt Yglesias's case](https://www.slowboring.com/p/the-case-for-terminator-analogies) for specifically embracing the “Terminator”/Skynet analogy. I’d be especially excited for people to spread messages that help others understand - at a mechanistic level - how and why AI systems could end up with dangerous goals of their own, deceptive behavior, etc. I worry that by default, the concern sounds like lazy anthropomorphism (thinking of AIs just like humans). Transmitting ideas about the “how and why” is a lot harder than getting people to nod along to “AI could be dangerous.” I think there’s a lot of effort that could be put into simple, understandable yet relatable metaphors/analogies/examples (my pieces make some effort in this direction, but there’s tons of room for more). ### AIs could behave deceptively, so “evidence of safety” might be misleading I’m very worried about a sequence of events like: * As AI systems become more powerful, there are some concerning incidents, and widespread concern about “AI risk” grows. * But over time, AI systems are “better trained” - e.g., given reinforcement to stop them from behaving in unintended ways - and so the concerning incidents become less common. * Because of this, concern dissipates, and it’s widely believed that AI safety has been “solved.” * But what’s actually happened is that the “better training” has caused AI systems to behave deceptively - to appear benign in most situations, and to cause trouble only when (a) this wouldn’t be detected or (b) humans can be [overpowered entirely](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/6LTh4foNuC3NdtmZH). I worry about AI systems’ being deceptive in the same way a human might: going through chains of reasoning like “If I do X, I might get caught, but if I do Y, no one will notice until it’s too late.” But it can be hard to get this concern taken seriously, because it means attributing behavior to AI systems that we currently associate exclusively with humans (today’s AI systems don’t really do things like this[[4]](#fn4). One of the central things I’ve tried to spell out in this series is why an AI system might engage in this sort of systematic deception, despite being very unlike humans (and not necessarily having e.g. emotions). It’s a major focus of both of these pieces from this series: * [Why would AI “aim” to defeat humanity?](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn) * [AI Safety Seems Hard to Measure](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/NbiHKTN5QhFFfjjm5) Whether this point is widely understood seems quite crucial to me. We might end up in a situation where (a) there are big commercial and military incentives to rush ahead with AI development; (b) we have what seems like a set of reassuring experiments and observations. At that point, it could be key whether people are asking tough questions about the many ways in which “evidence of AI safety” could be misleading, which I discussed at length in [AI Safety Seems Hard to Measure](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/NbiHKTN5QhFFfjjm5). (Click to expand) Why AI safety could be hard to measure In previous pieces, I argued that: * If we develop powerful AIs via ambitious use of the “black-box trial-and-error” common in AI development today, then there’s a substantial risk that: + These AIs will develop [unintended aims](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/vGsRdWzwjrFgCXdMn) (states of the world they make calculations and plans toward, as a chess-playing AI "aims" for checkmate); + These AIs could deceive, manipulate, and even [take over the world from humans entirely](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/6LTh4foNuC3NdtmZH) as needed to achieve those aims. + People today are doing AI safety research to prevent this outcome, but such research has a [number of deep difficulties:](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/NbiHKTN5QhFFfjjm5) \| \| \| --- \| \| “Great news - I’ve tested this AI and it looks safe.” Why might we still have a problem? \| \| Problem \| Key question \| Explanation \| \| The Lance Armstrong problem \| Did we get the AI to be actually safe or good at hiding its dangerous actions? \| When dealing with an intelligent agent, it’s hard to tell the difference between “behaving well” and “appearing to behave well.” When professional cycling was cracking down on performance-enhancing drugs, Lance Armstrong was very successful and seemed to be unusually “clean.” It later came out that he had been using drugs with an unusually sophisticated operation for concealing them. \| \| The King Lear problem \| The AI is (actually) well-behaved when humans are in control. Will this transfer to when AIs are in control? \| It's hard to know how someone will behave when they have power over you, based only on observing how they behave when they don't. AIs might behave as intended as long as humans are in control - but at some future point, AI systems might be capable and widespread enough to have opportunities to [take control of the world entirely](https://forum.effectivealtruism.org/posts/6LTh4foNuC3NdtmZH/ai-could-defeat-all-of-us-combined). It's hard to know whether they'll take these opportunities, and we can't exactly run a clean test of the situation. Like King Lear trying to decide how much power to give each of his daughters before abdicating the throne. \| \| The lab mice problem \| Today's "subhuman" AIs are safe.What about future AIs with more human-like abilities? \| Today's AI systems aren't advanced enough to exhibit the basic behaviors we want to study, such as deceiving and manipulating humans. Like trying to study medicine in humans by experimenting only on lab mice. \| \| The first contact problem \| Imagine that tomorrow's "human-like" AIs are safe. How will things go when AIs have capabilities far beyond humans'? \| AI systems might (collectively) become vastly more capable than humans, and it's ... just really hard to have any idea what that's going to be like. As far as we know, there has never before been anything in the galaxy that's vastly more capable than humans in the relevant ways! No matter what we come up with to solve the first three problems, we can't be too confident that it'll keep working if AI advances (or just proliferates) a lot more. Like trying to plan for first contact with extraterrestrials (this barely feels like an analogy). \| An analogy that incorporates these challenges is Ajeya Cotra’s “young businessperson” [analogy](https://forum.effectivealtruism.org/posts/hCsxvMAGpkEuLCE4E/why-ai-alignment-could-be-hard-with-modern-deep-learning#Analogy__the_young_CEO): > Imagine you are an eight-year-old whose parents left you a $1 trillion company and no trusted adult to serve as your guide to the world. You must hire a smart adult to run your company as CEO, handle your life the way that a parent would (e.g. decide your school, where you’ll live, when you need to go to the dentist), and administer your vast wealth (e.g. decide where you’ll invest your money). > > > > > > You have to hire these grownups based on a work trial or interview you come up with -- you don't get to see any resumes, don't get to do reference checks, etc. Because you're so rich, tons of people apply for all sorts of reasons. ([More](https://www.cold-takes.com/why-ai-alignment-could-be-hard-with-modern-deep-learning/#analogy-the-young-ceo)) > > If your applicants are a mix of "saints" (people who genuinely want to help), "sycophants" (people who just want to make you happy in the short run, even when this is to your long-term detriment) and "schemers" (people who want to siphon off your wealth and power for themselves), how do you - an eight-year-old - tell the difference? More: [AI safety seems hard to measure](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/NbiHKTN5QhFFfjjm5) ### AI projects should establish and demonstrate safety (and potentially comply with safety standards) before deploying powerful systems I’ve [written about](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt#Global_monitoring__noticing_people_about_to_step_on_mines__and_stopping_them_) the benefits we might get from “safety standards." The idea is that AI projects should not deploy systems that pose too much risk to the world, as evaluated by a systematic evaluation regime: AI systems could be audited to see whether they are safe. I've outlined how AI projects might self-regulate by publicly committing to having their systems audited (and not deploying dangerous ones), and how governments could enforce safety standards both nationally and internationally. Today, development of safety standards is in its infancy. But over time, I think it could matter a lot how much pressure AI projects are under to meet safety standards. And I think it’s not too early, today, to start spreading the message that AI projects shouldn’t unilaterally decide to put potentially dangerous systems out in the world; the burden should be on them to demonstrate and establish safety before doing so. (Click to expand) How standards might be established and become national or international I [previously](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt#Global_monitoring__noticing_people_about_to_step_on_mines__and_stopping_them_) laid out a possible vision on this front, which I’ll give a slightly modified version of here: * Today’s leading AI companies could self-regulate by committing not to build or deploy a system that they can’t convincingly demonstrate is safe (e.g., see Google’s [2018 statement](https://www.theweek.in/news/sci-tech/2018/06/08/google-wont-deploy-ai-to-build-military-weapons-ichai.html), "We will not design or deploy AI in weapons or other technologies whose principal purpose or implementation is to cause or directly facilitate injury to people”). + Even if some people at the companies would like to deploy unsafe systems, it could be hard to pull this off once the company has committed not to. + Even if there’s a lot of room for judgment in what it means to demonstrate an AI system is safe, having agreed in advance that [certain evidence](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/CcJsh4JcxEqYDaSte#AIs_could_behave_deceptively__so__evidence_of_safety__might_be_misleading) is not good enough could go a long way. * As more AI companies are started, they could feel soft pressure to do similar self-regulation, and refusing to do so is off-putting to potential employees, investors, etc. * Eventually, similar principles could be incorporated into various government regulations and enforceable treaties. * Governments could monitor for dangerous projects using regulation and even overseas operations. E.g., today the US monitors (without permission) for various signs that other states might be developing nuclear weapons, and might try to stop such development with methods ranging from threats of sanctions to [cyberwarfare](https://en.wikipedia.org/wiki/Stuxnet) or even military attacks. It could do something similar for any AI development projects that are using huge amounts of compute and haven’t volunteered information about whether they’re meeting standards. ### Alignment research is prosocial and great Most people reading this can’t go and become groundbreaking researchers on AI alignment. But they can contribute to a general sense that the people who can do this (mostly) should. Today, my sense is that most “science” jobs are pretty prestigious, and seen as good for society. I have pretty mixed feelings about this: * I think science has been [good for humanity historically](https://forum.effectivealtruism.org/posts/z7quAxWyHuqFdxGE6/rowing-steering-anchoring-equity-mutiny-1). * But I worry that as technology becomes more and more powerful, there’s a growing risk of a catastrophe (particularly via AI or bioweapons) that wipes out all the progress to date and then some. (I've [written](https://www.cold-takes.com/has-violence-declined-when-we-include-the-world-wars-and-other-major-atrocities/) that the historical trend to date arguably fits something like "Declining everyday violence, offset by bigger and bigger rare catastrophes.") I think our current era would be a nice time to adopt an attitude of “proceed with caution” rather than “full speed ahead.” * I resonate with Toby Ord’s comment (in [The Precipice](https://theprecipice.com/)), “humanity is akin to an adolescent, with rapidly developing physical abilities, lagging wisdom and self-control, little thought for its longterm future and an unhealthy appetite for risk.” I wish there were more effort, generally, to distinguish between especially dangerous science and especially beneficial science. AI alignment seems squarely in the latter category. I’d be especially excited for people to spread messages that give a sense of the specifics of different AI alignment research paths, how they might help or fail, and what’s scientifically/intellectually interesting (not just useful) about them. The main relevant piece in this series is [High-level hopes for AI alignment](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/rJRw78oihoT5paFGd), which distills a longer piece ([How might we align transformative AI if it’s developed very soon?](https://forum.effectivealtruism.org/posts/sW6RggfddDrcmM6Aw/how-might-we-align-transformative-ai-if-it-s-developed-very)) that I posted on the Alignment Forum. There are a number (hopefully growing) of other careers that I consider especially valuable, which I'll discuss in my next post on this topic. ### It might be important for companies (and other institutions) to act in unusual ways In [Racing through a Minefield: the AI Deployment Problem](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt), I wrote: > A lot of the most helpful actions might be “out of the ordinary.” When racing through a minefield, I hope key actors will: > > > > * Put more effort into alignment, threat assessment, and security than is required by commercial incentives; > * Consider measures for [avoiding races](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt#Avoiding_races__to_move_more_cautiously_through_the_minefield_) and [global monitoring](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt#Global_monitoring__noticing_people_about_to_step_on_mines__and_stopping_them_) that could be very unusual, even unprecedented. > * Do all of this in the possible presence of ambiguous, confusing information about the risks. > It always makes me sweat when I’m talking to someone from an AI company and they seem to think that commercial success and benefiting humanity are roughly the same goal/idea. (To be clear, I don't think an AI project's only goal should be to avoid the risk of misaligned AI. I've given this risk a central place in this piece partly because I think it's especially at risk of being too quickly dismissed - but I don't think it's the only major risk. I think AI projects need to strike a tricky balance between the [caution and competition frames](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/CcJsh4JcxEqYDaSte#Challenges_of_AI_related_messages), and consider a number of issues [beyond the risk of misalignment](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/mPkFheB4EM6pmEC7y). But I think it's a pretty robust point that they need to be ready to do unusual things rather than just following commercial incentives.) I’m nervous about a world in which: * Most people stick with paradigms they know - a company should focus on shareholder value, a government should focus on its own citizens (rather than global catastrophic risks), etc. * As the [pace of progress accelerates](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/AmxxnazJcBWzWEeqj#Impacts_of_PASTA), we’re sitting here with all kinds of laws, norms and institutions that aren’t designed for the problems we’re facing - and can’t adapt in time. A good example would be the way [governance](https://forum.effectivealtruism.org/posts/hxTFAetiiSL7dZmyb/ideal-governance-for-companies-countries-and-more) works for a standard company: it’s legally and structurally obligated to be entirely focused on benefiting its shareholders, rather than humanity as a whole. (There are alternative ways of setting up a company without these problems![[5]](#fn5) At a minimum (as I [argued previously](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/XRphCh6NbfQiDF3Nt)), I think AI companies should be making sure they have whatever unusual governance setups they need in order to prioritize benefits to humanity - not returns to shareholders - when the stakes get high. I think we’d see more of this if more people believed something like: “It might be important for companies (and other institutions) to act in unusual ways.” ### We’re not ready for this If we’re in the [most important century](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd), there’s likely to be a vast set of potential challenges ahead of us, most of which have gotten very little attention. (More here: [Transformative AI issues (not just misalignment): an overview](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/mPkFheB4EM6pmEC7y)) If it were possible to slow everything down, by default I’d think we should. Barring that, I’d at least like to see people generally approaching the topic of AI with a general attitude along the lines of “We’re dealing with something really big here, and we should be trying really hard to be careful and humble and thoughtful” (as opposed to something like “The science is so interesting, let’s go for it” or “This is awesome, we’re gonna get rich” or “Whatever, who cares”). I’ll re-excerpt this table from an [earlier piece](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/uNczpTHFvjt4KNxEF#Sharing_a_headspace): \| \| \| \| --- \| --- \| \| Situation \| Appropriate reaction (IMO) \| \| "This could be a billion-dollar company!" \| "Woohoo, let's GO for it!" \| \| "This could be the most important century!" \| "... Oh ... wow ... I don't know what to say and I somewhat want to vomit ... I have to sit down and think about this one." \| I’m not at all sure about this, but one potential way to spread this message might be to communicate, with as much scientific realism, detail and believability as possible, about what the world might look like after explosive scientific and technological advancement brought on by AI (for example, a world with [digital people](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/AKxKR4CeakyBsGFoH)). I think the enormous unfamiliarity of some of the [issues](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w/p/mPkFheB4EM6pmEC7y#New_life_forms) such a world might face - and the vast possibilities for [utopia](https://www.cold-takes.com/tag/utopia/) or [dystopia](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/AKxKR4CeakyBsGFoH#Virtual_reality_and_control_of_the_environment) - might encourage an attitude of not wanting to rush forward. How to spread messages like these? ---------------------------------- I’ve tried to write a [series](https://forum.effectivealtruism.org/s/gBjPorwZHRArNSQ5w) that explains the key issues to careful readers, hopefully better equipping them to spread helpful messages. From here, individual communicators need to think about the audiences they know and the mediums they use (Twitter? Facebook? Essays/newsletters/blog posts? Video? In-person conversation?) and what will be effective with those audiences and mediums. The main guidelines I want to advocate: * Err toward sustained, repeated, relationship-based communication as opposed to prioritizing “viral blasts” (unless you are so good at the latter that you feel excited to spread the pretty subtle ideas in this piece that way!) * Aim high: try for the difficult goal of “My audience walks away really understanding key points” rather than the easier goal of “My audience has hit the ‘like’ button for a sort of related idea.” * A consistent piece of feedback I’ve gotten on my writing is that making things as concrete as possible is helpful - so giving real-world examples of problems analogous to the ones we’re worried about, or simple analogies that are easy to imagine and remember, could be key. But it’s important to choose these carefully so that the key dynamics aren’t lost. Footnotes --------- 1. [Killer Apps](https://www.foreignaffairs.com/articles/2019-04-16/killer-apps) and [Technology Roulette](https://www.cnas.org/publications/reports/technology-roulette) are interesting pieces trying to sell policymakers on the idea that “superiority is not synonymous with security.” [↩](#fnref1) 2. When I imagine what the world would look like without any of the efforts to “raise awareness,” I picture a world with close to zero awareness of - or community around - major risks from transformative AI. While this world might also have more time left before dangerous AI is developed, on balance this seems worse. A future piece will elaborate on the many ways I think a decent-sized community can help reduce risks. [↩](#fnref2) 3. I do think “AI could be a huge deal, and soon” is a very important point that somewhat serves as a prerequisite for understanding this topic and doing helpful work on it, and I wanted to make this idea more understandable and credible to a number of people - as well as to [create more opportunities to get critical feedback and learn what I was getting wrong](https://forum.effectivealtruism.org/posts/7JxsXYDuqnKMqa6Eq/ai-timelines-where-the-arguments-and-the-experts-stand). But I was nervous about the issues noted in this section. With that in mind, I did the following things: * The title, “most important century,” emphasizes a time frame that I expect to be less exciting/motivating for the sorts of people I’m most worried about (compared to the sorts of people I most wanted to draw in). * I tried to persistently and centrally raise concerns about misaligned AI (raising it in two pieces, including [one (guest piece) devoted to it](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/hCsxvMAGpkEuLCE4E#Powerful_models_could_get_good_performance_with_dangerous_goals), before I started discussing how soon transformative AI might be developed), and [extensively discussed](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/Lbtcjfxhrs8kfKK2M) the problems of overemphasizing “competition” relative to “caution.” * I [ended the series](https://forum.effectivealtruism.org/s/isENJuPdB3fhjWYHd/p/uNczpTHFvjt4KNxEF) with a piece arguing against being too “action-oriented.” * I stuck to “passive” rather than “active” promotion of the series, e.g., I accepted podcast invitations but didn’t seek them out. I figured that people with proactive interest would be more likely to give in-depth, attentive treatments rather than low-resolution, oversimplified ones. I don’t claim to be sure I got all the tradeoffs right. [↩](#fnref3) 4. There are some papers arguing that AI systems do things something like this (e.g., see the “Challenges” section of [this post](https://openai.com/blog/deep-reinforcement-learning-from-human-preferences/)), but I think the dynamic is overall pretty far from what I’m most worried about. [↩](#fnref4) 5. E.g., [public benefit corporation](https://www.delawareinc.com/public-benefit-corporation/) [↩](#fnref5)
e985f2f8-ac3f-47d7-a8b1-afea74f113c6	trentmkelly/LessWrong-43k	LessWrong	Safe Predictive Agents with Joint Scoring Rules Thanks to Evan Hubinger for funding this project and for introducing me to predictive models, Johannes Treutlein for many fruitful discussions on related topics, and Dan Valentine for providing valuable feedback on my code implementation. In September 2023, I received four months of funding through Manifund to extend my initial results on avoiding self-fulfilling prophecies in predictive models. Eleven months later, the project was finished, and the results were submitted as a conference paper. The project was largely successful, in that it showed both theoretically and experimentally how incentives for predictive accuracy can be structed so that maximizing them does not manipulate outcomes to be more predictable. The mechanism at play is essentially pitting predictive agents against each other in a zero-sum competition, which allows for the circumvention of impossibility results in the single agent case. While there was one notable result that eluded me, related to the case where agents each have private information, I still think meaningful progress has been made towards defining a goal which is both safe to optimize for and useful enough to enable a pivotal act. This post contains similar content to the submitted paper, but in a framing more directly addressed to readers who are already informed and interested in alignment. There are also several results here that were cut from the paper version for space, and slightly less formal notation. Overall, I would recommend reading this post over the paper itself. Predictive Agents and Performative Prediction In a previous post, I summarized the case for investigating predictive models, largely using points from the Conditioning Predictive Models paper. The gist of the argument, which you can click through to read in full, is that predictive models are potentially useful enough to be used to take a pivotal act, easier to align than general agents, and coming anyway. One big issue with the use of predictive mod
147b4b5e-47cd-4e4c-b5e7-70637dcc7560	trentmkelly/LessWrong-43k	LessWrong	Communicating forecast uncertainty Note: This post is part of my series of posts on forecasting, but this particular post may be of fairly limited interest to many LessWrong readers. I'm posting it here mainly for completeness. As always, I appreciate feedback. In the course of my work looking at forecasting for MIRI, I repeatedly encountered discussions of how to communicate forecasts. In particular, a concern that emerged repeatedly was the clear communication of the uncertainty in forecasts. Nate Silver's The Signal and the Noise, in particular, focused quite a bit on the virtue of clear communication of uncertainty, in contexts as diverse as financial crises, epidemiology, weather forecasting, and climate change. In this post, I pull together discussions from a variety of domains about the communication of uncertainty, and also included my overall impression of the findings. Summary of overall findings * In cases where forecasts are made and used frequently (the most salient example being temperature and precipitation forecasts) people tend to form their own models of the uncertainty surrounding forecasts, even if you present forecasts as point estimates. The models people develop are quite similar to the correct ones, but still different in important ways. * In cases where forecasts are made more rarely, as with forecasting rare events, people are more likely to have simpler models that acknowledge some uncertainty but are less nuanced. In these cases, acknowledging uncertainty becomes quite important, because wrong forecasts of such events can lead to a loss of trust in the forecasting process, and can lead people to ignore correct forecasts later. * In some cases, there are arguments for modestly exaggerating small probabilities to overcome specific biases that people have that cause them to ignore low-probability events. * However, the balance of evidence suggests that forecasts should be reported as honestly as possible, and all uncertainty should be clearly acknowledged. If the for
b3c7e966-244f-49b5-bc30-4ad59114a71d	trentmkelly/LessWrong-43k	LessWrong	Continuing the discussion thread from the MTG post I moved and copies this discussion out of the latest MTG color-wheel post, since I would prefer the discussion on the post to stay on the object-level. Commentary by Conor Moreton: > [Meta/cultural note: as of this writing, the parent comment I made in reply to CoolShirtMcPants' elaboration on horoscopes is at -2, which I think is a bad sign re: LW culture in general. CSMcP was making a broad claim of the form "categorical psych tools are bad," which is both a) reasonable and b) in context the sort of claim Scott sighs about in his excellent post Yes We Have Noticed The Skulls. It was a knee-jerk, base-rate, cached objection to an entire category of Thing based on that category being generally bad/useless/misleading, when the post in question was about a specific instance, started out with a link to Fake Frameworks, made explicit bids to be treated fresh/in good faith, and was written as the 29th entry in a series of posts that have been generally agreed to contain non-zero value and rationalist virtue. > The comment above implicitly (and maybe clumsily) made the claim "I suspect you're only using your generally sensible prior, and I think it's better in this case to construct a posterior that combines your generally sensible prior in a rational way with awareness of the source and context." > Regardless of whether the posterior ends up being "yeah, still bullshit" or "maybe I'll give this more charity than I otherwise would have," the requested operation "instead of just commenting using your base rate, combine your base rate with your sense of whether a given person has demonstrated 'worth listening to' nature" is one that LessWrongers should absolutely engage in, on the regular. > i.e. I strongly believe that the thrust of the comment ("You're leaving out context that a LWer ought not leave out, and this is somewhat undermining the point you're trying to make") is correct, defensible, and prosocial given the community's goals. I think it was innocuous at wor
1760f6dc-980c-4ff1-8086-b534a092799b	StampyAI/alignment-research-dataset/blogs	Blogs	Transformer Circuit Exercises d-article h2 {border-bottom: none !important;} d-article h3 {font-size: 26px !important;} d-article h4 {font-size: 16px !important; margin-top: 30px !important;} d-article hr {margin-top: 60px; margin-bottom: 80px;} d-article ul {padding-left: 0px;} d-article ol {padding-left: 36pt;} d-article ul li, d-article ol li {margin-bottom: 8px;} d-footnote {margin-left: -3px; margin-right: 2px;} .gdoc-image, .gdoc-image-gutter, .gdoc-image-flex {width: 100%; max-width: calc(var(--img-width) / 1.5); padding-right: 2%;} .gdoc-image {grid-column: text-start / screen-end; image-rendering: -webkit-optimize-contrast;} .gdoc-image-gutter {grid-column: gutter / screen-end; margin: 0px; padding: 0px; margin-top: 6px;} .gdoc-image-flex {margin: 0px; padding: 0px;} p > sup, h1 > sup, h2 > sup, h3 > sup, h4 > sup, li > sup {display: none;} d-code {margin-bottom: 10px;} @media print { @page { margin: 0cm; } hr { margin: 0px; padding: 0px; opacity: 0.0; height: 0px; page-break-after: always; } h2 { display: block; page-break-before:always; } .gdoc-image { break-inside: avoid; } .base-grid, distill-header, d-title, d-abstract, d-article, d-appendix, distill-appendix, d-byline, d-footnote-list, d-citation-list, distill-footer { grid-template-columns: [screen-start] 1fr [page-start kicker-start middle-start text-start] 45px 45px 45px 45px 45px 45px 45px 45px [ kicker-end text-end gutter-start] 45px [middle-end] 45px [page-end gutter-end] 1fr [screen-end]; grid-column-gap: 16px; } } This collection of exercises is supplementary material for our [mathematical framework](https://transformer-circuits.pub/2021/framework/index.html) for reverse engineering transformers. The exercises go through writing down individual weights for attention heads, in order to implement algorithms. We've found this helpful in developing our own understanding, as a way to make sure we really understand the full mechanistic story of transformers all the way down to individual parameters, and not blurring over any confusions. Solutions are provided [below](#solutions). --- Exercises --------- #### Warm Up * Describe the transformer architecture at a high level * Describe how an individual attention head works in detail, in terms of the matrices W\_Q, W\_K, W\_V, and W\_{out}. (The equations and code for an attention head are often written for all attention heads in a layer concatenated together at once. This implementation is more computationally efficient, but harder to reason about, so we'd like to describe a single attention head.) * Attention heads move information from a subspace of the residual stream of one token to a different subspace in the residual stream of another. Which matrix controls the subspace that gets read, and which matrix controls the subspace written to? What does their product mean? * Which tokens an attention head attends to is controlled by only two of the four matrices that define an attention head. Which two matrices are these? * Attention heads can be written in terms of two matrices instead of four, W\_Q^T \cdot W\_k and W\_{out} \cdot W\_v. In the previous two questions, you gave interpretations to these matrices. Now write out an attention head with only reference to them. * What is the rank of these matrices? * You'd like to understand whether an attention head is reading in the output of a previous attention head. What does W\_V^2 \cdot W\_{out}^1 tell you about this? What do the singular values tell you? #### Exercise 1 - Building a simple virtual attention head Small transformers often have multiple attention heads which look at the previous token, but no attention heads which look at the token two previous. In this exercise, we'll see how two previous token heads can implement a small "virtual attention head" looking two tokens behind, without sacrificing a full attention head to the purpose. Let's consider two attention heads, head 1 and head 2, which both attend to the previous token. Head 1 is in the first layer, head 2 is in the second layer. To make it easy to write out explicit matrices, we'll have the k, q, and v vectors of both heads be 4 dimensions and the residual stream be 16 dimensions. * (a) Write down W\_V^1 and W\_{out}^1 for head 1, such that the head copies dimensions 0-3 of its input to 8-11 in its output. * (b) Write down W\_V^2 and W\_{out}^2 for head 2, such that it copies 3 more dimensions of the previous token, and one dimension from two tokens ago (using a dimension written to by the previous head). * (c) Expand out W\_{\text{net}}^1 = W\_{out}^1 \cdot W\_V^1 and W\_{\text{net}}^2 = W\_{out}^2 \cdot W\_V^2. What do these matrices tell you? * (d) Expand out the following matrices: Two token copy: W\_{\text{net}}^2 \cdot W\_{\text{net}}^1. One token copy: W\_{\text{net}}^2 \cdot \text{Id} ~+~ \text{Id} \cdot W\_{\text{net}}^1. * Observation: When we think of an attention head normally, they need to dedicate all their capacity to one task. In this case, the two heads dedicated 7/8ths of their capacity to one task and 1/8th to another. #### Exercise 2 - Copying Text with an Induction Head (Pointer Arithmetic Version) The simplest kind of in-context meta-learning that neural networks do is increasing the probability of sequences they've seen before in this context. This is done with an "induction head" that looks at what followed after last time we saw a token. There are at least two algorithms for implementing induction heads. In this exercise, you'll build up the "pointer arithmetic" algorithm by hand. * (a) Let u^{\text{cont}}\_0, ~~ u^{\text{cont}}\_1, ~~ \ldots ~~ u^{\text{cont}}\_n be the principal components of the content embedding. Write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) selecting tokens with similar content to the present token, including the present token itself. * (b) Let u^{\cos}\_0, ~ u^{\sin}\_0, ~ u^{\cos}\_1, ~ u^{\sin}\_1, ~ ... be a basis describing the position embedding in terms of vectors that code for the sine and cosine embedding of the token position (eg. \lambda\cos(\alpha\_0 n\_{tok})) with descending magnitudes. Write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) that self-selects the present token position. * (c) Using the position embedding basis described in (b), write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) that self-selects the \previous\ token position. Hint: think about a 2D rotation matrix. * (d) Write W\_Q and W\_K for an attention head (with 8 dimensional queries and keys) selecting tokens with similar content to the present token, but disprefers attending to itself. Hint: refer to (b) and use extra 4 dimensions for keys and queries. * (e) Write down W\_V and W\_{out} for the attention head you described in (d), such that it extracts the largest 8 dimensions of the position embedding from the token it attends to, and writes them to the vectors v\_0, v\_1, .... * (f) Write W\_Q and W\_K for an attention head which attends to the token after a previous copy of the present token. Hint: use the outputs of the head from (e) and the strategy you used in (c). #### Exercise 3 - Copying Text with an Induction Head (Previous Token K-Composition Version) Some positional encoding mechanisms, such as rotary attention, don't expose positional information to the W\_V matmul. Transformers trained with these mechanisms can't use the strategy from (e) and (f) in the previous exercise to manipulate positional encoding vectors. For these transformers, we've seen an alternate mechanism, where the first head copies information about the preceding token into a subspace, and the second head uses that subspace to construct queries and keys. Assuming the same positional encoding mechanism as above, write down W^1\_Q, W^1\_K, W^1\_V, W^1\_O and W^2\_Q and W^2\_K for a pair of attention heads implementing this algorithm. --- Solutions --------- #### Warmup See our [paper on transformer ciruits](https://transformer-circuits.pub/2021/framework/index.html) for discussion of all of these questions. #### Exercise 1 - Building a simple virtual attention head (1)(a) Write down W\_V^1 and W\_{out}^1 for head 1, such that the head copies dimensions 0-3 of its input to 8-11 in its output. ![](data:image/PNG;base64,iVBORw0KGgoAAAANSUhEUgAABCAAAAHICAIAAAAQoQx6AAAMbWlDQ1BJQ0MgUHJvZmlsZQAASImV VwdYU8kWnluSkJDQAghICb0J0gkgJYQWQHoRRCUkgYQSY0JQsaOLCq5dRLGiqyKKbaXZsSuLYu+L BRVlXdTFhsqbkICu+8r3zvfNvX/OnPlPuTO59wCg+YErkeSjWgAUiAulCeHBjDFp6QxSJ1AHOKAC OjDi8mQSVlxcNIAyeP+7vLsBEMX9qpOC65/z/1V0+AIZDwAkA+IsvoxXAPFxAPB1PIm0EACiQm85 uVCiwLMh1pXCACFeqcA5SrxDgbOU+PCATVICG+LLAKhRuVxpDgAa96CeUcTLgTwanyF2EfNFYgA0 R0AcwBNy+RArYh9RUDBRgSshtoP2EohhPICZ9R1nzt/4s4b4udycIazMa0DUQkQyST536v9Zmv8t BfnyQR82cFCF0ogERf6whrfyJkYpMBXibnFWTKyi1hB/EPGVdQcApQjlEclKe9SYJ2PD+gF9iF34 3JAoiI0hDhPnx0Sr9FnZojAOxHC3oFNEhZwkiA0gXiCQhSaqbDZJJyaofKH12VI2S6U/x5UO+FX4 eiDPS2ap+N8IBRwVP6ZRLExKhZgCsVWRKCUGYg2InWV5iVEqm1HFQnbMoI1UnqCI3wriBIE4PFjJ jxVlS8MSVPZlBbLBfLFNQhEnRoX3FwqTIpT1wU7xuAPxw1ywywIxK3mQRyAbEz2YC18QEqrMHXsu ECcnqng+SAqDE5RrcYokP05lj1sI8sMVeguIPWRFiaq1eEoh3JxKfjxbUhiXpIwTL87lRsYp48GX gmjABiGAAeRwZIGJIBeI2robuuEv5UwY4AIpyAEC4KTSDK5IHZgRw2siKAZ/QCQAsqF1wQOzAlAE 9V+GtMqrE8gemC0aWJEHnkJcAKJAPvwtH1glHvKWAp5Ajegf3rlw8GC8+XAo5v+9flD7TcOCmmiV Rj7okaE5aEkMJYYQI4hhRHvcCA/A/fBoeA2Cww1n4j6DeXyzJzwltBMeEa4TOgi3J4hKpD9EORp0 QP4wVS2yvq8FbgM5PfFg3B+yQ2ZcHzcCTrgH9MPCA6FnT6hlq+JWVIXxA/ffMvjuaajsyC5klDyM HES2+3GlhoOG5xCLotbf10cZa9ZQvdlDMz/6Z39XfT68R/1oiS3ADmBnsRPYeeww1gAY2DGsEWvF jijw0O56MrC7Br0lDMSTB3lE//DHVflUVFLmUuvS5fJZOVcomFKoOHjsiZKpUlGOsJDBgm8HAYMj 5jmPYLi5uLkCoHjXKP++3sYPvEMQ/dZvurm/A+B/rL+//9A3XeQxAPZ5w+Pf9E1nxwRAWx2Ac008 ubRIqcMVFwL8l9CEJ80QmAJLYAfzcQNewA8EgVAQCWJBEkgD42GVhXCfS8FkMB3MAaWgHCwFq8Ba sBFsATvAbrAfNIDD4AQ4Ay6Cy+A6uAt3Tyd4CXrAO9CHIAgJoSF0xBAxQ6wRR8QNYSIBSCgSjSQg aUgmkoOIETkyHZmLlCPLkbXIZqQG2Yc0ISeQ80g7cht5iHQhb5BPKIZSUV3UBLVBR6JMlIVGoUno ODQHnYQWo/PQxWglWo3uQuvRE+hF9Dragb5EezGAqWP6mDnmhDExNhaLpWPZmBSbiZVhFVg1Voc1 w+d8FevAurGPOBGn4wzcCe7gCDwZ5+GT8Jn4InwtvgOvx0/hV/GHeA/+lUAjGBMcCb4EDmEMIYcw mVBKqCBsIxwknIZnqZPwjkgk6hNtid7wLKYRc4nTiIuI64l7iMeJ7cTHxF4SiWRIciT5k2JJXFIh qZS0hrSLdIx0hdRJ+qCmrmam5qYWppauJlYrUatQ26l2VO2K2jO1PrIW2ZrsS44l88lTyUvIW8nN 5EvkTnIfRZtiS/GnJFFyKXMolZQ6ymnKPcpbdXV1C3Uf9Xh1kfps9Ur1vern1B+qf6TqUB2obGoG VU5dTN1OPU69TX1Lo9FsaEG0dFohbTGthnaS9oD2QYOu4azB0eBrzNKo0qjXuKLxSpOsaa3J0hyv WaxZoXlA85JmtxZZy0aLrcXVmqlVpdWkdVOrV5uu7aodq12gvUh7p/Z57ec6JB0bnVAdvs48nS06 J3Ue0zG6JZ1N59Hn0rfST9M7dYm6troc3Vzdct3dum26PXo6eh56KXpT9Kr0juh16GP6Nvoc/Xz9 Jfr79W/ofxpmMow1TDBs4bC6YVeGvTcYbhBkIDAoM9hjcN3gkyHDMNQwz3CZYYPhfSPcyMEo3miy 0Qaj00bdw3WH+w3nDS8bvn/4HWPU2ME4wXia8RbjVuNeE1OTcBOJyRqTkybdpvqmQaa5pitNj5p2 mdHNAsxEZivNjpm9YOgxWIx8RiXjFKPH3Ng8wlxuvtm8zbzPwtYi2aLEYo/FfUuKJdMy23KlZYtl j5WZ1Wir6Va1VnesydZMa6H1auuz1u9tbG1SbebbNNg8tzWw5dgW29ba3rOj2QXaTbKrtrtmT7Rn 2ufZr7e/7IA6eDoIHaocLjmijl6OIsf1ju0jCCN8RohHVI+46UR1YjkVOdU6PXTWd452LnFucH41 0mpk+shlI8+O/Ori6ZLvstXlrquOa6RriWuz6xs3BzeeW5XbNXeae5j7LPdG99cejh4Cjw0etzzp nqM953u2eH7x8vaSetV5dXlbeWd6r/O+ydRlxjEXMc/5EHyCfWb5HPb56OvlW+i73/dPPye/PL+d fs9H2Y4SjNo66rG/hT/Xf7N/RwAjIDNgU0BHoHkgN7A68FGQZRA/aFvQM5Y9K5e1i/Uq2CVYGnww +D3blz2DfTwECwkPKQtpC9UJTQ5dG/ogzCIsJ6w2rCfcM3xa+PEIQkRUxLKImxwTDo9Tw+mJ9I6c EXkqihqVGLU26lG0Q7Q0unk0Ojpy9IrR92KsY8QxDbEglhO7IvZ+nG3cpLhD8cT4uPiq+KcJrgnT E84m0hMnJO5MfJcUnLQk6W6yXbI8uSVFMyUjpSblfWpI6vLUjjEjx8wYczHNKE2U1phOSk9J35be OzZ07KqxnRmeGaUZN8bZjpsy7vx4o/H5449M0JzAnXAgk5CZmrkz8zM3llvN7c3iZK3L6uGxeat5 L/lB/JX8LoG/YLngWbZ/9vLs5zn+OStyuoSBwgpht4gtWit6nRuRuzH3fV5s3va8/vzU/D0FagWZ BU1iHXGe+NRE04lTJrZLHCWlko5JvpNWTeqRRkm3yRDZOFljoS78qG+V28l/kj8sCiiqKvowOWXy gSnaU8RTWqc6TF049VlxWPEv0/BpvGkt082nz5n+cAZrxuaZyMysmS2zLGfNm9U5O3z2jjmUOXlz fitxKVle8tfc1LnN80zmzZ73+Kfwn2pLNUqlpTfn+83fuABfIFrQttB94ZqFX8v4ZRfKXcoryj8v 4i268LPrz5U/9y/OXty2xGvJhqXEpeKlN5YFLtuxXHt58fLHK0avqF/JWFm28q9VE1adr/Co2Lia slq+uqMyurJxjdWapWs+rxWuvV4VXLVnnfG6hever+evv7IhaEPdRpON5Rs/bRJturU5fHN9tU11 xRbilqItT7embD37C/OXmm1G28q3fdku3t6xI2HHqRrvmpqdxjuX1KK18tquXRm7Lu8O2d1Y51S3 eY/+nvK9YK9874t9mftu7I/a33KAeaDuV+tf1x2kHyyrR+qn1vc0CBs6GtMa25sim1qa/ZoPHnI+ tP2w+eGqI3pHlhylHJ13tP9Y8bHe45Lj3SdyTjxumdBy9+SYk9dOxZ9qOx11+tyZsDMnz7LOHjvn f+7wed/zTReYFxouel2sb/VsPfib528H27za6i95X2q87HO5uX1U+9ErgVdOXA25euYa59rF6zHX 228k37h1M+Nmxy3+ree382+/vlN0p+/u7HuEe2X3te5XPDB+UP27/e97Orw6jjwMedj6KPHR3ce8 xy+fyJ587pz3lPa04pnZs5rnbs8Pd4V1XX4x9kXnS8nLvu7SP7T/WPfK7tWvfwb92dozpqfztfR1 /5tFbw3fbv/L46+W3rjeB+8K3vW9L/tg+GHHR+bHs59SPz3rm/yZ9Lnyi/2X5q9RX+/1F/T3S7hS 7sCnAAYHmp0NwJvtANDSAKDDvo0yVtkLDgii7F8HEPhPWNkvDogXAHXw+z2+G37d3ARg71bYfkF+ TdirxtEASPIBqLv70FCJLNvdTclFhX0K4UF//1vYs5FWAPBlaX9/X3V//5ctMFjYOx4XK3tQhRBh z7Ap7ktWQRb4N6LsT7/L8cc7UETgAX68/wsztZC2JFMiHQAAADhlWElmTU0AKgAAAAgAAYdpAAQA AAABAAAAGgAAAAAAAqACAAQAAAABAAAEIKADAAQAAAABAAAByAAAAABXHsh0AABAAElEQVR4Aeyd aaxd11n34ymxndhOPMRTPMSz69hOnDZuIQEppSRQHCEVVQGpUCFQS6F8QXxDAqFKfEK8eltRJIZC K6SqIgWa9m1awNAmTdO0ZKyHeMq1bzzFQ2I7cTz7/d37T5aX995nn73PWfvec+/9nw/3rr3Ws57h v9be+3nWsNe4q1ev3uCfETACRsAIGAEjYASMgBEwAkYgBQLjUzAxDyNgBIyAETACRsAIGAEjYASM wAACDjDcD4yAETACRsAIGAEjYASMgBFIhoADjGRQmpERMAJGwAgYASNgBIyAETACDjDcB4yAETAC RsAIGAEjYASMgBFIhoADjGRQmpERMAJGwAgYASNgBIyAETACDjDcB4yAETACRsAIGAEjYASMgBFI hoADjGRQmpERMAJGwAgYASNgBIyAETACDjDcB4yAETACRsAIGAEjYASMgBFIhoADjGRQmpERMAJG wAgYASNgBIyAETACDjDcB4yAETACRsAIGAEjYASMgBFIhoADjGRQmpERMAJGwAgYASNgBIyAETAC DjDcB4yAETACRsAIGAEjYASMgBFIhoADjGRQmpERMAJGwAgYASNgBIyAETACDjDcB4yAETACRsAI GAEjYASMgBFIhoADjGRQmpERMAJGwAgYASNgBPIIXLly5c///M+3bduWL3JOxwh8+9vf/spXvnLx 4sWOObhicwg4wGgOW3M2AkbACBgBI2AEjMAN//Zv/7Z9+/ZFixYZi4QI3H333f/+7//+5JNPJuRp VqkQcICRCknzMQJGwAgYASNgBIxAFoGXXnrp85///J/92Z9Nnz49Lrt69SozG3FOE2lEIKgJzoHn EIjAhLwV8+fP/8xnPvO5z33u0KFDQRknegSBcfkG6xHNrIYRMAJGwAgYASNgBEY0Am+++eYnP/nJ 97///X/yJ38SG3Lq1KnXXnvt8uXL48aNmzNnzty5c0nEBN2nz54929/ff+7cuYkTJ86YMWPevHkk umcbc7hw4QIizpw5M2HChFtuuWXBggU33XRTTNB9+tKlSwcPHnzjjTcQMXny5DvuuGPKlCmBLaV/ 9Ed/xN+/+qu/uvHGG0O+E8OOgGcwhr0JrIARMAJGwAgYASMwOhH4p3/6J6KI3//934/NO3LkyI4d O2bOnLl27do777wTB3r37t1pZzNOnz794osv4pGvWbNm+fLlJ0+efPnll9OKeOedd+CJdVixatUq gg02mRDVxJZ2mRZPgjGsWL16NdELRhGzBbaETJ/97Gd//OMff+c73wmZXSa2bt0Kt8zWjueff/7x xx9/++23Y+Y//elPyST4iTMbTaPVY489hoaNSknC3AFGEhjNxAgYASNgBIyAETAC1yHwzDPPfP3r X//oRz962223hYLz58/39fURXTClMGnSJAb+iTGOHj1K1BFoukywOOWVV165+eabGe/HKSfMWLp0 KVMZr776apec4+r79u1D0LJlyxDBjzCGYAPTEi6NIfQijFm5ciUmMEHBDAlGEYzFIlasWPHII498 4QtfYC4lVq/jNI1FzJAJG5huIrpAn5jt66+/TiYtGGc2nQYHQAhSfvSjHz311FPhsncSDjB6py2s iREwAkbACBgBIzBKEGD0nemLadOmffzjH49NOnHiBOPQeLFhTRTBBut/yI/Juknj9RJO3HrrrbAV H5xgVhYhItUkBmESEwuwDcuuSGAs0wsUdaN8qEu4cvz4cUILfspEBPtYMI1FWYGMxG/8xm9g8te+ 9rU4s+M0uFE3nichDX8yCTNitgQYEBMlxpmNppG1ZcuWzZs3BynokImFQtHwJhxgDC/+lm4EjIAR MAJGwAiMQgS+//3v/+///u8nPvEJ4ofYPJxmLjNeKa4z/noq7z8vYvz48YggsGFCIFam4zROLQFA xgoG18nEkI7ZxhUVqxAjhUiMUm20yLjUTAH94i/+4je/+U3208ccOksz7UNF5pRCdaVvv/32WC6e PQRsnglkbRPgr0mPQKkc/mZyFM8gTlKYemI5lshCRKFS4ljqkskvMFEOs1jUyuSLhsyS0phPx+nE 23061sMVjYARMAJGwAgYASMwOhAgVPjiF7/IcPvP/dzPxRbhf8sjxOOP83GjWfaD9x+vfokJaqW1 ESIjgku0SiVCHnCYIZF6ulRRLYULiZmpAJO8FRBnRBCBMK7/P//zP//8z//8vve9j1CqkGHFTKIm WiH2y1kZRQ5hDJn8iDRgpWCDDfoV2UJG07N9gs0k99xzj2qxzIzdHZs2bWITi3IICXD9H3roISRS hKXEVJKFAugGBxR48MEHKUUZahF4kAk9IIgJ66bi5XDQP/DAA4oGYUipKoqYinDjry5T/b2uf6di aj5GwAgYASNgBIyAERizCPzgBz84cODARz7yETZaxCDgMWuaIh6Vh0CXqRYXKYaJ5QYR8WB5hqDW JR9uakVfKL0VcUk+wRilmQCjFVAbNmxgIzhnYsSOdQnz8iI8coBSGENCQYVmNsI2jGPHjsEkP4NR gjBOPL94nZW4xTnIgiZs2pEO999/fz4GIOfRRx8dZHkziRBdEHgAwl133fWxj32MfBLwJGiRyZrT IJMilSrkKAekg1IHGB2A5ipGwAgYASNgBIyAEShGAOf7y1/+Mq4wAUYmkAgBRmFNudSFRbUyxScj WpepRJSs5iopqmVFIR9ZkS9i8uSXfumXiG3+5V/+pZaUQmLNUWiYX+ujFi5cyAwA+ZpMoBaJeAMG bjrTCF/96lf5yhMfoVJgkGdOlEKRghASiCCW4K9y+AtbSQ91iS6olckMpZkEPIklmOsghNCUBQnq hgBDipGpiiRgTpSS4dP9pQOM7jE0ByNgBIyAETACRsAIvIsAn6BlnJgPN/H91lqgZEKCWnVjYo36 E8zEmbpMJSLm3FC6UNUSK+677z7WpH3729/u/lO58uYVYGiSAS8cM8lXMICbzi/MM1BEdEF4wDQC PwKPJ554QjFDBhwCFXI0ZSH+qE2OJl4UzBAexLVqef/iiWLSUH/hgDKkYSudWSUVIiUCmHjXeCy6 m3RXy9S6Eey6RsAIGAEjYASMgBEYfQj893//N/4c247jI+FkJk5zifevIefuASnZhFBSVEtuCZ+S oloitKMjEyaJQ6GI2bNnf+hDH2L24Hvf+x6zGbVkZYjxyGkLOev8DbMHShAeqKUULVCXHNz3sIqJ PRXkMGkQJgoCf/x7cSaKIHSBITn8hZ5tGFp2FcSFWtUTiiKeG/xlalGEXajEPA/xDD8ukUUOiQxx 95cOMLrH0ByMgBEwAkbACBgBIzCAAN8++o//+A8Sme3dQgenGf+SvRYZv1lrfqZOnSqyLv8S2KBG Zh0Ri6MIb8InX7sXAYeMCF3mw6rOZHG2BgoXiigECmx/9md/9rvf/S4BHpukM5s36urAlAUuONEF fnmIE3DHFR7oY1aaDYAzaOOjBzcdGvZmyNfPyKWIWoQTCkI0dQBbohGC0jiYyVSsdVkYM0g9FEAo BMjihyaYGe8yryWohNhLpErAcZERMAJGwAgYASNgBGogwJdSOTKP0XROoMtXw2NmGQ/5mfUzbNvA n8b5y1fpIGfGjBkZEcQzBBi44KkCjEIrMAoDJb0DtTNVcIhRGLXjYEy4SXqGnkscZWrxdWDO+8uX 1srRaRhaHxVPKRB44JezvghB/Ep4ttrszrwHVvD9KOqKM7MZ5IhtZn1UCf/CIqlE/AOfzE9FhD3I Ik0pkcbDDz9Mr+NE9kJu3WQ6wOgGPdc1AkbACBgBI2AEjMA1BH74wx/iEOPpFi7jgY4PmzK4zjdY Qx3SBBjEJCGnywTOMa65PvMqVsyZ8ONcPMKYLpmrOnMIzFTw0dvg/ZPgkgCm3O2uLh0RKIza8cZ0 RBDDhKmDDDf8dU775iAOzlDPFNW91OehmFgAzNgiROCjZ7ZiQyDHPUiBoJWS2s7BLp3AmeoQS+c4 mAncyhNxJANzAgbNh4Ra6Eb0wiWhxeOPP863tkKRRJMfclIlHGCkQtJ8jIARMAJGwAgYgTGNAK4e 22eB4AMf+EArIBh9Z4z/8OHDwavbv38/znrmg7atqlfJx8WEGwv633rrLdHjX6Ib+86rVK9CQ/i0 aNEiPNdwALnEzZ8/n9imCocqNAz2ozYzQiLmAG/OECSz1VQPkdvP/MzPQMwehCr8S2jw/iUl8yFa hQe0XRwJkImnjlCFGQQPqM0UQSF/KPnBQaxEo2mNEHIUVizMJDKBFRI12YLO69evRw12mWuJF0Wk 1S0pRSs6g47CoIoO02ilaqHEipneg1ERKJMZASNgBIyAETACRqAMAdZH4XDjxt17770ldHxdaufO nfh2OJR4ovzIKdxXUMKkvEixBFKIZ5gB4MNKHBMxbdq08lq1SvG8mXjZs2cPDiuzCji1y5cvTxgm oQyHoHMs3d69e0+fPk1Iw18CmPIwiQDjH//xH3/yk59oSVgtizLEGMgWBVz/OJ/GJbTAZNou5JPJ Dm8+JMX8AJnED2HDd6CJE4QWTDLEnBWuZIKZuEqr9Lp165gtIYpAqBiyWRx9yFFQQUWYh+9EkWAB FQu0CD8ogpLoggm3Vvw7zh8X5rY6ZuGKRsAIGAEjYASMgBEwAn/3d3/3N3/zNwztc6R0ecCA94XT T2iB34xr2OWO5FbISwTeP2uNEk4sxOJYwsSyJXKwV1uf49IkaVDSWizmedou8WJ91Ic//GHk/vVf /7W+AJtEh4pMmExAWxq0In0SMoQyVUUTEy3EDAn5uKRRMvmiISyhqFVpzKeztGcwOsPNtYyAETAC RsAIGAEjcB0C27Zt45rx6bZfUsLpxw1t2hPF6ed3nYqpL/D42zr9XcqUH1yRCTM2mmF49tlnhz7A wJUv9OYrKt8ZGRIL93uU967CKp0pUFjLezAKYXGmETACRsAIGAEjYARqIMAoO7spqLB48WLihxo1 TZoOAZBfuXIl/LZv356OqznVRsABRm3IXMEIGAEjYASMgBEwAhkEOAyBpflkEmBkinw5lAgsWbIE cQR7rB0aSrmWFSPgACNGw2kjYASMgBEwAkbACHSCwK5du/gyLOt55OB2wsJ1UiCgvc5sS+BTXSn4 mUcnCDjA6AQ11zECRsAIGAEjYASMQIwAn5Dikj0PbPKO850eYgQIMFgoxYq18H3bIVbA4kDAAYa7 gREwAkbACBgBI2AEukWAD7bCgo/5dPCx0W5lu36EAJu82WR/5cqV3bt3R9lODikCDjCGFG4LMwJG wAgYASNgBEYlAlqQw9ENrc7wHpVW96BRnGOoQypYtNaD6uVVYq+ITujLF43cHAcYI7ftrLkRMAJG wAgYASPQKwhoh3fTX//sFWt7WA8mkfSFVvZ5cxTgEGvK+RKcYcff6nI5zo8T+virKlTnV716b1L6 HIzebBdrZQSMgBEwAkbACIwYBDhkml3FqKsjmUeM3qNRUbbBaAaDQ/d0lOFQWvnUU08xHXHXXXd1 HGpyCDcc6EjlB1kMpVEdyHKA0QFormIEjIARMAJGwAgYgWsIHDt2TBdyba8VODXkCLDDm4VqiD1z 5gzf9Wr6qMHYPsUGcU4H6c2bNxNgjOjoAqsdYHTQ9K5iBIyAETACRsAIGIFrCIQlMXJtrxW0SLEF +fz584yv81lbTsIePz79kvWrV6/CHynsCUHEhAkTWujSVTb7B3Di8eknT57c0OYTRPBJKERgBXBV UZfzvCFjTmkoj8IgKiDA0DniVZRsRTM6JsEcYLRqX+cbASNgBIyAETACRqASAlofBWmVgefLly+z /5gqbEfGOycSWL58OTsHKkmqTNTX18e2EGZUEEekwfF/FYOfyhJuYF97f3//tGnTCJBwr+cP/qpX r0J58uRJPgYFOEQvZ8+enTVrFl+hbRssqRXYgEFkUkVKnuaVV14BQOLGSZMmsdiJWYW2Lbt161Zo WBxFIs8wzqFdCEX4C/M77rgjM+v1/PPPI/fBBx+kCgku77zzTuBlY4ZmNrhEysGDB1EyMCGnrYax Dk2nHWA0jbD5GwEjYASMgBEwAqMcAY7xloV8ILWtqfiFOL4bNmxgSB5igo3t27dzySRA27oVCXCO 8f7XrVtHDEMAw3bnnTt3IiJhGHP06FFc3hUrVmjE/fjx4xjCPMO8efMqKtmWDFTBChecH5zZU4Ff zuTP0qVLy+uGZVFUKacsLMWnRy523XPPPUQ1bL9+4okn8PhLtlVocdSWLVuIAQp5hkwCgyeffJLQ YvXq1fwd2NB9/ZZuggrCBtEzA0OaHKZuAIFMlFFwQr4iDRJwQK5ikiBoeBMOMIYXf0s3AkbACBgB I2AERjwCDLTLhrajyLi8+IvMJyi6oBYnf+Os43cyj5EECEbu4YZ/rJVCuOYLFizADd23bx8xRhIR ePmIIFwJ63lmz5596NAh3N85c+a0nWGoqAPccMGZF8EEqmAO4pCLOeVrpUIcxTaMirICmaIafHdm LZTJnAmTEsRsrQIMnHucfgIGWr9tgPHcc8/B9uGHH1ZX0YwHrRMUyCcwFnqgoGjVqlV8cgr6TZs2 kRYx6pGD6LbdL8+8oZz0a/4aUtRsjYARMAJGwAgYASPQmwiEGYwQNrTSk+3geOdhiB0yquA7EnW0 qlI3H2UQwVyK/HKqw5/pEQbjU+1JwJeFW2a6hktWfJFfV+FCelgRHqB52NqBOYjAtLbzEgHeDpZI sWDp0UcfDdEFuslrZ5lZoZ5kanFUcPdbkZFP0wAd0UscCXBZUoUiwipFF6SpyI/LWJzCvLaxTbmU tKWewUiLp7kZASNgBIyAETACYw6B4FXj/JUYz2ol7dYI/qKI8aFxhbXnu6R6xSJFO8EvpxauObMK 2pOQEV2RZ4YMEdiSmalAIpl4/+zKyNB3cEl0we6RvAhYIYJ5khKeYX4jtEsJcWEREwJMlWibeHlU xmIqPPsqmzQQpNZvNRNSqElhZjCwsLQXMh1g9EIrWAcjYASMgBEwAkZgBCPAcLu0L5/BYPRdA+Fh bkG19BUpmCRxHDVsnxHBJd4/IrRuqkusJSLz8StddjBpUKgMqqJwoYi2YUMIS0K7FIpolRn2YBAJ EI8RYLSaXyK0YHEUa6iYQ9AEgtqXKlzSmplwTrFKJrOVGiM63wHGiG4+K28EjIARMAJGwAgMPwJ8 DVZKBNe2UCcCDJzmwiIyy0fKW9XK5+v46kyAIU89lQgMyctVjqS3Kq2eLxEZKyqKCK3Qgb3MXTAp Ee/BIFQgp1BzdlcjgrkOfjEB9PzYX8Evzldo0YFWMZMRkXaAMSKayUoaASNgBIyAETACvYtAGNEv dIhjvUsCjBKvPebQNl3Cp0R6W7YxQQmfkqKYQ9t0IR/BW1gUMwzLwzqIdggwYMWkRGBYEg8Qh2j/ QyBm8RjbuMnnl18vN3fuXCjZvB7voCjZ3RHYjriEA4wR12RW2AgYASNgBIyAEegtBHAZcUwZn2bb QIlm+MdykQtpMsuBCmmqZGr8PuOF6zKViBIrSoqqKB9oClWtaAUTSooTTp8+HRhWTOhUCqYm9FlY pi9+9KMfUbcwDBjYcF2064bMTOAh6fQQ8tGNT/oqxoA/6Yq6jSAyBxgjqLGsqhEwAkbACBgBI9CL CPCdWb5Uq13OJfrhfBd6/5pzKN+/UcI2U6SNHIUBRpI9HogbAhGahcjMxsiotlYwoaQPB+O+Z8Bp e0lcwdYLljw99thjCIKD9mR3wKpQFtvB+eoUsxysoYJA/FMxL5Q4LJkOMIYFdgs1AkbACBgBI2AE Rg8CHGHBV4+IEMp9X0blocH9zUx04EYTe2S++toxOnyk9cSJExkRXCIi1Vl++g5sXgQ6F47od2AL QAFXJsCQxLYiCAl0GJ9ig7rSH3roIWYwmJWiNZnQ4IALgoGShVIxf3Rj30Xh9IXIIOBEPGYtNCXC SiqmNYhnwnne5ITq4paxgqmPzDQR9AhtC0usZ9NpBxhNI2z+RsAIGAEjYASMwChHAMedX9spCPxC /EiW6We8VS45Gy5sTe4SLET09/fHIvDL+eHIomSXzFUdEXj/sQjyuSQzOMpdCuJrVyjMJgpmLYI/ LYkcT17OPFjaMaQDWyii4ymIMcolhlKFBOGyMAENZ4THRXEIEcst5JZXhgCDX8xw2NM+aG/Ym8AK GAEjYASMgBEwAqMBgcyqpEKT2OaL+8t0RyBmMQ/pefPmFdJ3kIn/jWPKYRFhizMrcJg2wQcNu587 YBtXIZQikIBt+HwWstjwgOi2UVbMpyQNSpwOzhdpw0dpmc3gc7ESXVLRRb2AgAOMXmgF62AEjIAR MAJGwAiMCQRYdcMXitizwQ+D8dH7+voYwMaZTmU/4/2MgnOmGyIIXXD9mdBA7qJFixKKWLx4Mcz3 79/P3AgiWFPE5ZIlS8JsQ/eyAIpwAnzgT3TBOiLCpJUrVyYU0b2S5lCIgJdIFcLiTCNgBIyAETAC RsAIVEUA37oq6Q034OizVOnIkSP4/fjKHEqNs57WaSZiWbduHbsIDh06BOeZM2eycD/V9IUs5bhu RBBgsF+ZHC43btyYagmWRBBdbNiw4cCBA5x8R47WCyU5Jlz8/bc5BBxgNIetORsBI2AEjIARMAJG oAABggp+hCVp44pYEkuY+DUqgq3ea9euVXDVkCFMvKxYsQK7GjUkxs3pJAg4wEgCo5kYASNgBIyA ETACRqAeAg055bESo0MEFg2BITFuTneJgPdgdAmgqxsBI2AEjIARMAJGwAgYASNwDQEHGNewcMoI GAEjYASMgBEwAkbACHSMAKd0c/K3zhHvmMkoqOgAYxQ0ok0wAkbACBgBI2AEhhMBLeDxMp7hbIPe kM1nwfigFn97Q51h08IBxrBBb8FGwAgYASNgBIzA6EDAAUZPtWOI9EKip9QbC8p4k/dYaGXbaASM gBEwAkbACDSIwIULFzgPjnOsG5Rh1pUR4NAMHc+nL1xVrmfCZAg4wEgGpRkZASNgBIyAETACYxYB ogv8Wnu0vdABaAUFexwC2Jk+Fy9e5BQRVjpxoPgdd9zBERyBD/sryOSkkZDD+eLQc1B6yCFBzr59 +/gLJScGxkWk4Qx/lfI14cCfHLiJeUl1yFqpJ0ExfxRD4ViBUB25lAbpMU2XaQcYXQLo6kbACBgB I2AEjMBYR4DjGjhjjoPhvCanF7rChAkTdOQfiQ70YRMFpwfihasux/zdNfjjEsd969atHJS+efPm wBkCoo5HH3005HDoeMwBD/7hhx8OXv5PB3+BmNJNmzYpCCFsYI84IQ2nsAcFMtVfeeUVGFAKQ/7G 6olnnv/9998fIiJUffLJJwNzmHzwgx/Mh0BBvc4SDjA6w821jIARMAJGwAgYASPwLgKeuOiprtBN c+B5ExsQMT744IM45UQUTz31FC47I/38KppJnEDMQBzC2jnCFao/88wzDzzwANXx77nEocetx7mX u0/pI488EiIQoguIEYd0iBXwKKQhh4gCxVAP+rx6EodoFAj8MUERDvREF7F1xEsZ6RVtLCfzYsFy fFxqBIyAETACRsAIGIFKCHTj11YSYKLmEdB6Jxx0Dfkze3DPPffI168ufNXgD/+e6kx+UJ05DU0a EFHAB/4KJyhav349st56663AnxzyuaQ6cQV/CRtUnQgBhsxIqDpFiILyzTffVHUCEtUK/KGnFjEP BJTCR7FT4E8O/FU91V/PYKRC0nyMgBEwAkbACBgBI1AJgXPnzh0/fvz8+fMTJ07EuZw2bVrytVUw P3HixDvvvMNuBFb580su4tKlS4jAeUXE1KlTZ82a1dmSpBLI2ESB68wP5RExe/ZsECuh776I5sA1 x+HGTWepEml8fTzyWpwVnIQqcCCuYF4Chooctm3bxjSC0opGAjEJxQYhh1osiwrVVQvY+UEDOPxl qkRpMjPrnZYtW4Y+UokwA7ugVF0SkkVYRTrhr9lGSqioWRkBI2AEjIARMAJGoDcRqDV3cfr06Z07 d+KO88PPI40HOX/+/IQBAO7m9u3bcWFhi4+OuzxjxgwczYQiiC4QAfPFixdjPiP0hEzLly9nI0qq NoItG53xfRnvJ644fPjwkSNH1qxZo/0VqaRk+OBwM2nAMiT2QvDDm0c6f+WXZ4grXqqupiBgxZQC LcLaJGTh98OfXwkrgkNKVZ0EUDMRURgSKMxQLBEYIkUxiZjA5/HHHw+lSoR4I5Pf8aUDjI6hc0Uj YASMgBEwAkbACNRDAI98165dt9xyy9KlSxnvx3fEjd67dy9+OfFGPV6tqRGBH4kjq/Fp4grCGGYA iDdaV6pXguvPp2A3bNgAW2ry96WXXmKAnBijHqPW1EQUDPxjBdERVID2wgsv7N69mxVErSslKCHe 44do/QgziBCYxOg4xpDfHzTDotWrVzMjcezYMRCDPzEDq54CQSah6mpK4go2URBCaJkTmcQqcMhU KbmkrlZVxTTEovFl92kHGN1jaA5GwAgYASNgBIyAEaiEgJYtLVq0KKwmwpft6+vDmU4VYDAajRuK CLmkqDVz5kyc40OHDs2bNy/JJAYu78mTJ/H7FV0gggSXOM1MaAS5lRBpQcQ3f1EYtqwfEwlsWSKF R46BHfv6LaRdyyYwwzr4M+qvgX9mM1ihhB9PYHCNrk5Kq5iks+YKSA9GMXewwYPwAOYEMxKXZ6zq igFYWwXBfffdl5mmUC1lZqYjsIj+gET92OxRPmGSV6CDHG/y7gA0VzECRsAIGAEjYASMQCcIsI6I apkBYy7PnDnTCbuiOvJH46VK7JHAO2dXBns/imrUzkNb3Na8FczPsACsNruiCvjBaEsYphMtRCKj ClcHFfHoJI9YghVEhDGhstxx7CVHJscePOn4UrXi6pQSPODcK34gnIB/XEUroOJZjtjATPV88Kbm llxKkaIYLOjPXBPLscSTqAZDmOAKpSSYP5F1cWaXac9gdAmgqxsBI2AEjIARMAJGoBICjMrLxQ/T F6qGD433z4bsKVOmVGJUSqRYJTNTgQi8f6QnEaEoInb90UhGIT3JVAzrr1g8lgFKl/EHl0qR6KSQ cIIYg1kLfG6iAvx7NjzAiCVt/JUHT8BAnICzjiY463G0IJG47CIgbBBBmP0ggbvPjwT88fv13ad4 RoIqSEcT/lIK/1Cd+IHpDj47yzInoh1NfUio/mb4oyocYI62EFBK+MF3eJEr/hgLDdtOFEfFrLpJ O8DoBj3XNQJGwAgYASNgBIzAuwhkfPo8LgQY7I0mP0MpTz2V90+skhchiSrKK1Y3R8PtGSvSitCY eiaGkQgFaXV1rkiP08+REYQH/FSFoAL/OwQApAkP8Oz5QcBuCpx1fPSYPzSUslmCTKpzJEXY9kCE QClOf+BPDgulkBs4EAYMsh/gn6lOGKCYhxCIUuqyeYN4QxMUIUeb1FUdDUN8Ajc2b1Aa80eftNEF ch1gAIJ/RsAIGAEjYASMgBHoHAE5vhmHO8+OIXl++XzlMMPQqqhWvvhklNGlwpta3AqJiZQK88lM a0WhlFQiCpmTia+PF06EwwQFswSx66/SLVu24OUTZbHvHJc95oOnLmedv+LQiqYVhyCiVXWiBWKG WLf4EHGqa3dHK/6YQ0wCc0WJGetiW7pJO8DoBj3XNQJGwAgYASNgBIzAAAIlPncMUEmAEZM1lM5E HQ1JGXa2mKnmyKywqqUYkUOYtchXxC9v65p3yaGkeklRULVcQzjwC8TJEw4wkkNqhkbACBgBI2AE jMDYQoDtEwwYD85PtJygABEcX635KQwzuvGGY7h1FF1GhC5TiSjhU1IUK9k2XXKgXkmR2DI8ry0c JNoKMkETCDjAaAJV8zQCRsAIGAEjYATGEAL6mirfOCqfIiC6wDlmI0TG+9dwe5Lt14CuTy1lZlR0 GX9aqpvmGQIRGl8vtKLtQXvUZWESBma2cHRjsuvWQsABRi24TGwEjIARMAJGwAgYgSwCmYAhW/ze Nf4uC1eY68jshWCgncCjrd/8Hps2/zk4gj3HsQjU4xLpqURMnz4dJWIRXGprhIraqFihmJgNTBCB 8iFsk4hwMkYFNiOMZGADR+mp3iPFHp+DMVJaynoaASNgBIyAETACPY1AlTCDo+Jwl+OvOeFD4zfr MIQk5nGsHny0hVcMJQKXPVWAgYvP7udYBIJYJ0ZmqgCDSIwpHUKveBJD34+SgSVYVWmIkuou6h4B BxjdY2gORsAIGAEjYASMgBGohAD7hnGd4xmG/v5+huoXLlxYqX4FIqIIvl567Nix8DlXTt1mTwJn bIepgApsykhQmEPB+ZBR+DoqCU7AmDt3LkVlNSuXMd8CJpjA2eeqhAnghmmpwqTKupiwNgJpOkFt sa5gBIyAETACRsAIGIHRgkD1IXP85pUrV+7cuZMDzpjNYCrj6NGjS5YsSbvsh1Ph4IwUPH5042y1 RYsWJTn/LrTYggULmLLglOj58+djFCLgTwwTCLpPMFPBF1f7+vqwhW0VAAVKy5cv757zKODAQjus IFjtTVscYPRmu1grI2AEjIARMAJGYHQiwP5jDlZjVgEfEb/53nvv1YbmhNayCXv9+vWaVWDWYuPG jam2dwcl0XnNmjXMWmgSg7T2VQeC7hN8kIpIac6cOUxiMJVBGJZwIVn36lXkwCovIGLipRU93YCZ GUKFEpp8Xc7po1Y4AYN4FfwTzoPlJdbKcYBRCy4TGwEjYASMgBEwAkagWwRwnfGb+XXLqHV94gpm APi1JklQwpQCvwSMWrPA8+bXujx9CR4/PxazdRP4EVcwvXPw4EHCAPZtFwYPFHGkN3+xoRVNFfPe fPNNDucm+nKAUQUu0xgBI2AEjIARMAJGwAgYgSFF4NVXX8Xv5zDvwqigiirMJ8CBGKMkREEKsxAQ cDI3gUE38z+EFps3b+6p6R3PYFTpJ6YxAkbACBgBI2AEjIARMAKVEGCH/bJlywgbmHt5/PHH83WY IXnuuecoJYxJMj/Tax+3dYCRb3TnGAEjYASMgBEwAkagNgKpvtFUW7ArJEJAa5bw/uFHAMBXd5kZ UABAJhMO7KZgXoLVU6tWrSrx6e+//35pJFZ57Xbt2gWfBx54oHp0wZQIEyNBesyTzCeffJL5FiZD yGduhB9p6LFIVdj2AwFMKEIr5KK/6GNWqdIOMFIhaT5GwAgYASNgBIzAGEVAoYUDjJHe/CxYwgvH KccF1weFiTEwissnnniCBPMS0LCzQnusCT86MxkRuPiIIAZgB4Xc/ZJggy0WUEKPehw/QpwQH0JC WgylDNpyScgB/erVq7kkqFAEQj4mQMZXv2CCdURKnZlQXssBRjk+LjUCRsAIGAEjYASMQBsEHGC0 AWhoi0Okx/dza0nGI+eH540jHu+63rp1K/MAW7ZsUQzAwD8BBl57TFNLEDMhCPrGN75BLbx8YgCE Eq4UzopQSnSB6IceekhSyFHAUyKU0CJMUFAX/kQyDz/8sEwgrmDtFmGGA4wSDF1kBIyAETACRsAI GIFhQwBPjuMm+J6Sz4AbtjaIBOO7syKIFklyLgfBBg49rr9cc8nZtGkTAQZ+P1MKkeRKSbhBR4yh GIApEQUMLMri3A8uM1xYT0UOFoV8NMFGFAs5+USsmNIwDyaQ4CdN8nW7z/EMRvcYmoMRMAJGwAgY ASMwphG477778N4Yik51jvWYRrNr4znyj6F62HCMRtfMBtZHwST217kkDMBBZ06gY/4wDDEDrIhR WQdF0JKfUpACRBQdy1LFfOjSJcOS6g4wSsBxkREwAkbACBgBI2AE2iPwpS99ieFnpi+INHAW21cw RZMI7Nmz5y/+4i+Q8OlPf5pz07sUFe926JJVXD3TTwhQCTCY1ohplJYCGfo8WU/l1Fua1lOqWxkj YASMgBEwAkbACPQCAqzGQQ397QV9rIMQSNIi2ufNHowMqvj9nc0JiGEmbsnzD+IUWjS3nCkISphw gJEQTLMyAkbACBgBI2AExiICdR3ZK1eunDp16siRI4xYl3iW3UCJSm+99RYiTpw4cf78+W5YldQ9 e/bs0aNHOfaBRF0QStjGRSiPCPYbYE6cX5JOq4kWR7EfOpaIPjRcZ8uWCEvgKQ6Bp+YuMguxVKoT 9Ph0VSDu/YQDjN5vI2toBIyAETACRsAIjAAEKvq1eMwvvvjigQMHMOnkyZMvvPACMUBa8y5fvrxj xw42B6MSrj/iDh8+nFYEnPfu3bt9+3b8bCwiwf6BighU1ARuoMTCIfhfunSJhU9sqsa0itW7JAsz DEwgKB7QZmvYMpnAV6RIhM801ZXFlnFwY1mdKsKQrzwp8Miz0rdxsT1MYug7V3nK3snxHozeaQtr YgSMgBEwAkbACIxyBJi7wBfn86lr167VjvAJEybgO+KqTp8+PZXxu3fvZoaELx3ddNNN8EQEwQD+ 6+zZs1OJ6OvrY3pkw4YNfDsLnii/bds2RCxatCiVCCZGCDDWrFkjtRnIx7EmzODjS6lEFPLRNMJT Tz1FaMGReUxTcCIen4UlHqClWODEXArhAfB2vC+CAENBBfMYgWE41C+jFagii5CGD8uiDJEPPxKF GzYydYfr0gHGcCFvuUbACBgBI2AEjMAoQaD6yP3p06dxTxmTDt+bWrhwYX9/P856qgAD7/P48ePz 589XdAHEc+bMOXToEIt8UgUYTCMgAoUVXSCC9C233MI8CXKDaV22Liuj2Dc/c+ZM8ZkyZQqfhwIo vg0VTOtSRGF1Agx8feDC9ecHDS4+h2AwRUM8ALwQ8K2nKtEF1QkdC1dSkU/IRBBFf4AhRhWSSUMC EmiYQiGokHQS4RtWGSkQwDxWjzQ55Mf2YkI4MCTOT5J2gJEERjMxAkbACBgBI2AEjEB7BPCPIZo6 dWogxXnFXU64SgpWBDy44xkROLKsNUrimhMmnTt3LuMQIxGX98yZM5n8oEathCIxPONwWB7eMCKY AmJdGWFMLW51iXHo+WVqFWZmaDKXNC6efSYzXPLlKH7hsjwBFOGztlByGepmpBBIZGIJBRgZ/o1O BHkPRgZtXxoBI2AEjIARMAJGoBEEcI7feecdWLNmKRaAD80eA1z2OLPjNC4+dYNfLj5IZNpB0jvm HCri/ZPOWKFLFQXKjhOoCiZ5K2AoAzvm7IpDgIADjCEA2SKMgBEwAkbACBgBI3ADAQZr9wEiszRF bnSqAENRREYEl0xrMIORpBnEJyMirRWsRELVTAwjianCpCRQmEkhAg4wCmFxphEwAkbACBgBI2AE EiOAi8+vFVMG7FsV1conjIE+4/3rUuFNLW6FxCWfciopKmTVKlNWFJamElHI3JlJEHCAkQRGMzEC RsAIGAEjYATGLgJy3zM+fSEcJQFGSVEhq1aZJa55qyo9mF+IhhAuLOpBE8aySg4wxnLr23YjYASM gBEwAkYgGQJtAwwIRFPoIqf6+JKWFWVE6DKz4qhjyzNbI2I+JUUxWdt0IZ+KVghkRIREW3EmSIuA A4y0eJqbETACRsAIGAEjMOYQYNEOewbarnHCaeaDP3l0NOeQ5PtOMOfTrvzNBBhpRejjrRkRukxl hURkZmMqioCM5uCXh9o5Q4OAA4yhwdlSjIARMAJGwAgYgVGLAKEFP3Y4ZHzujMEEGPL+IY6LuGRu QUVxfmdpzqOgYrxRAa24RHoq71/HX+StQK6kd6Z5XIsv0qJwbEUwqq0IaqEbv0z1mH/vpOk2HKLH IRtJVErLrWOVHGB0DJ0rGgEjYASMgBEwAkZgAAFiA4624G/bNTkcFQdNPLjOID2uMC47/nQSNDm+ DT7xB6PkcBNdpIph0JYFXbEIJGIUmTNmzEhiBVEEkxi4y/EkhnALR++1EoQaNAe/VJC2ElQlnxPx vvrVr3IGeStijNIRfq0IauWn5VZLdEycpivHHJ02AkbACBgBI2AEjMCYQqB84iKGgrO08ZtxOkMV zqsm5Eh4chynquGdc+Yd3rlEnzp1im/gzp07N9UeDGIVIiVO7guf1iXY4AQMXH8tbYpN7iyNqhxA fvbsWaSIA+ZwxB7WETl0xnNYaqE2Ooe2GBYdhl6oA4yhx9wSjYARMAJGwAgYgVGIQIgZSmzDb165 ciXe/969e/HOjx8/fuDAgQULFuCvl9SqVUS4wiHNzIrs3r0bBx2/f9++fbj+ixYtqsWnnJhjrXH0 d+7cSQDAb9euXbjRy5YtK69Vq5STqlF7x44dBEgYsmfPHmYz1q5d23aaqEpD1NKkG2IO1d6yZUt8 CHc33EZK3YkjRVHraQSMgBEwAkbACBiBUYDAbbfdtn79egbj9+/fr3iDnLZOcy3Dcf0RceLEif7+ fjgvXbo0YQAjTditvmbNGqZiDh06RA4zM3jSqWZIJIKVToggBnv99dcJLViXRWxGZi0oOiBGHMES UzEEZswFLVy4UEyYhSCHv7RXyAz8gYKKKmWVGhxCLYq4DDnkQ0kmxIAW51P09uAPEeF7AJCVcChU Jmg1XInGG2m4DLNcI2AEjIARMAJGwAgMDQJ1h8xxW/lRK21cERvLJmlmAOKc5GlccJZd4SLDuSFD iFgQwa9RrGJkcO+3bt1K/KBoAesUS7zyyivso8DXFzFRwYMPPhhiA4r4BT7kb9q0SRWJDWB41+BP BBniTKxCKVsymPQIAUaGAzo8+eSTqCcCLjPKBDWGMeEAYxjBt2gjYASMgBEwAkZg7CLQkFM+xIAO jRVDIyVAd/DgQZaZsQlEIQQe//PPP89cAUEFbv1rr73Gd58IGwgDqIKvT1RAnPDBD36QUi4JAJ55 5plHHnkkBAmBM6wghu3mzZuJXggennrqqVBaJfHcc88hIkQsLE4jB33QrUr1oaHxHoyhwdlSjIAR MAJGwAgYASNgBEYGAkQL7JpgCoigAo1x4gkV7r//fgUM5BN+MNeBo0+p/rIpRaWEDaxPoyJbU/LW EmCQSTCgmR/IHnjggTxZSQ7rr5geIcAQzapVq2BCoFJSZeiLPIMx9JhbohEwAkbACBgBI2AEjEDv IhB/C0tbIBQPEFRIaX0LmM365Kto27Zt1FIap59foXlEAvEODWjysxyFFUMmsQ1ptCKA4a/yQyKQ DW/CAcbw4m/pRsAIGAEjYASMgBEwAr2LgCYHmKZ4/PHHM1rqXA6CCuYTmJpg0RTRAvMJzGbwyxDr kkggjl4Kacoz4cAiK203L6ccxlIHGMMIvkUbASNgBIyAETACoweBId4nMHqA621LFA8QReRjBs1j oD4BBhMLHGly7Ngx7dBgFwdLqpqwjB0XBDNIRB9tESGw0TKtJsR1xtMBRme4uZYRMAJGwAgYASNg BN5FQKGFA4xR2SHkxGNaPsCQvVo3BRl7M/ixeYMt18QAOP1aMRXDAlk3+yWYvoAzW0TCHoyYee+k vcm7d9rCmhgBI2AEjIARMAIjEgFOtWO1TK+tgx+RUKZQmkMzaA5+dT8fXCicVU/ECUQL/AIBbc3O b10STrB6KmzPIFMzGygQ6EOCCIS6Mas4DZkmTGJuMYF4xj0Nypg4CBrehGcwhhd/SzcCRsAIGAEj YARGPAKcmsfvpptuunz58og3ZuQbwBHpNAd2vPPOO0msYVKCZUh8fJZ5A7ZY4NATXTARwUwCMxLs 5yYGgIBSTVDoQ7RQ5qVDzPYJWLGkSl9/4oSNmAye5BC0aI4CzkxZBAL4K9qBgASaQKx4g79194sH tskTDjCSQ2qGRsAIGAEjYASMwNhCgJPg5s2bR4CR9ijrsQViOmvxwmkO+HGieRKuRAIPP/wwB1Zw GoYY4tyHg/aYlOBQC4IKnP5QSkyCGnnpZFIRVtBTSkjAN2dDRXLgTGgRuEEPAfsuAitkUZ2ogx/V 9cFc0gQ81A1kw5sYl2TyaHhtsHQjYASMgBEwAkbACAwjAr/zO7/zwgsvEGB8/etfJ9ioogkOGL/x 45tdrM5iIXaGNLo5ZAhEABSQVrfi2Wef/cxnPkOVT33qU7/7u79bpTkq0jBLwMdhWcVUGDwwn8AS Js5orzKTQDzQig/KlAsKBBVlVbQuIZlnMBKCaVZGwAgYASNgBIzA2EVAfnAV+0+dOsW3hlhPhdPM cdHEJNW95yr8oTl79mx/fz+LhSZOnDhjxgxG9ElUrFuRDGcaEWfOnGHeBk93wYIFhFgV61YkY3ML n2PCF0fE5MmTmSuYMmVKxbpNkBE5FC58kiyijsLAo1CTEj7QlwuqQlAodMgym42bh8wMCzICRsAI GAEjYASMwIhA4MiRIzt27Jg5c+batWv5MBEO9O7du5kHSKj86dOnX3zxRTzyNWvWLF++nA0JL7/8 cloRbG+AJzESVrCvgGCDk+aIahJaIZ4EY1jBQiCiF4zibLuEIsyqIQQcYDQErNkaASNgBIyAETAC YwWB6nMX58+f7+vrI7pgSoFRagb+iTE4P4GoIxVYKMPGX4bSGe/HKSfMWLp0KVMZLNNPJQI+7FRG 0LJlyxDBjzCGYAPTqkPRVhlCL8KYlStXYgKriZghwSiCsYQi2upggs4QcIDRGW6uZQSMgBEwAkbA CBiB2gicOHGC5fUsjwlrogg2WP9Dfm1eLSqwE4Bwgi+lhh3nhDGsLEJEqkkMwiQmFmAbll2RmDZt GtMLFLXQq1424crx48cJLfipJiKmT5+OaSzKqsfL1EOOgAOMIYfcAo2AETACRsAIGIGxioACicwm YFxn/PVU3n9eBFvJEUFgk+qzrWyKIADIWMEkA5kYkqRt4UOsQowUIjHYIoK/XiWVBOFGmTjAaBRe MzcCRsAIGAEjYASMwLsIEEJogD/z8SjcaJb9MDafBClmMOCTEcEl0tOKCDMkUlsSJb17Q1AVTPJW wJnvOHXP3xwaRcABRqPwmrkRMAJGwAgYASNgBN5FABefHxfxqHy4TLW4iL3RecQlsbAoT9w2h487 5WmaEJEJMCQiFVB5E5yTCgEHGKmQNB8jYASMgBEwAkZgTCOQCRvyWDAkrwAjX0ROoddeSFmeyTol CDLK6FJF5dWrlJZYUVJUhXOgKeQjKwqLQkUnegEBBxi90ApDpwMnUMaHQQ6d4PqSwhmW9as2XqOX devMeD4t8uSTT37nO9/ZtWtXZxxGRK3XX3+d01L5LMmI0LatkiynxpyEn4WB2+juAG0hNYER6BgB Ob4Zn74uty6rB3Hik/nUki5TiQiymksUqtoLVrAGjLcJT+DmbB8FnB1g1GvErVu34oRlehVe++OP P55ZdIgPSia7oOoJaJiaY32GxbsCtLpocPcGzwlsH3vsMcBvGJ6q7GPdqtbpYTpwxrPEqKlTp2Ze SMOiNfcXt09Jc9MfRJC/78oVpiLGhrsSq7/61a+Gy/K6Q1+KYqiHkq1Es9QBc2i4VgR18+HGI6Ju LdMbASMgBKoMq+M0a81P5mGry8ye6Y6BZT93q7olRa2qFOZndl/ENGlFZICSoCoiaA5+JXrGOtdK 86jkDdWz745atjRH3LILNidyRHPmu3J8W5pedfvttwdDeCXjuOC4c9BMyOStTyZfcAs5YzaBL6j7 MBOYVQeEr0bwKz/zsjo3U2YQ0KD1ww8/XP380QyHJJd0D/oJvaXcY+YGhAZi+gO3IR2jY+nYy+u8 Gw4di65YEQ17Wb3YCl63fBNz06ZNcabTRmDsIMAbn2+nssap0CEOOBBd8NhhC0GGDFcYmlRnVDNa lP8mlUSET74GlTpLSFXxDBzSWsHTj3isUAQGBqGFCd4R+pRtqj0nhVKcWYKAA4wScAqKeIOSy/fR QoBBmscKmYQZmQAD4lSjEQWqjJAswMEdBIeOowsMpfqWLVtGiMUjT03cevrq8EYXoIYaOKm09V13 3cX4kG6rDJqE8UwYElrcf//93SuMIH4ZEb1ziZkjqNuXh4W9g6o1MQINIcB44owZMzhvrnBhTxBK gAElH0HKvBOJTPCn+QXKbhKcR3H48OF4RwfxDJcM/KcKMDAWDWMRXMoojqroRvlQl4c88w9sGkH5 gKpEYGAgK0yApDTsbAYDKQxm6U3EW4nzCnlZoA8POlwavZ5YcI6Ue+65h2c1LyZeYaSfffZZEg89 9BCZKMY7Cz7UipkEhcUNesRBjw/JkYsqVRE5OJlwiAm4ZFgQHcSTYR0SgWfvJBxg1GsLOhkrFjhx M8QSpGFBvEEXCbzoGaTnzp0bctom1MPgQ6BCf1q9enXoMXDjp56Ud6ooCr1z4cKFGUF0Ss7aVNfM l0JMH0WiCNp6mZBpuqaVMhnpXOIyojN2caPmS+OcwJzbLK+qENAdi878lM5YF5hIaB6uWGIr6AL/ 0ASwBWS4ZRiKAzzpGJkiMinlRwLz+QXRoa1F0ApMSktatoS/BIXqGelBDRKyNNCTkJ7URSvegoI3 eOEB3rzOATSYdNYoAPjggw8GBWAoxeK/PNAhqxVdxCDHrEgHndXQwep8JxcTqpSASWngAAK8eMKr AtwCJnCQuKAMarS6B8VQXV305IT+k+lyGXOgL+y3gUPhjRa0KkwU1pU+0DNSGBQOosEB2+lL4aaO 0QiZQRylefxVGgxHigSF+47L8psl8HfCCDSEQGZGokQKvgE9Nv5cLGk89fztUMKkvEgn93HkRXDN mTPhx51ICFRet2IpcwhMYpw9ezaIIIFEMpFSkUk5GY84WMGTGCOsieKSYAMDy+t2WcqmRNqIFuEx zqMVB4ZL5vl5evMMJ00mj1A05BGHLJ4/ZOLwoLDeYmQSivDjkndohkmmlOceInAv4aZuAD0MYYsU HEIueTBCwF/yOT0dTajCDx3CO7pLq9NWd4BRD09alLakdUM1wlNy6IJk8lPHok9AMGfOnEDWNkH4 S3X40F0ghg8/utRTTz0lbmRCQzcKPSmUohW9EwI00Q0gcej2zDPPqEileAC6GUSg3q+0CAiFWz3j CJpRADKJIw1xCLQCkzgBf5RkFJZEnJ9PYyN3ZqxqrCdMKAWczZs3UxeIYBiwEjfgYvAgwwSHFUzy 4sjhRhXUKoUsEIt/cHYhkHox+BkOrZpGzPnLMwKGQEcaYiwinKOBRJCpDg56uomey1g9qggQ/qo6 f0EjHsaAIe0VSmPpIZMEMAoEYlqgI+fRRx/lL2mq0FskQl0OSvz70EYZnQUanScYBZ9ajYKN/KjV 6scNgj6YCZnUED6t6MmPWxmFMSomlvn0T/gIUt10AVgIHnjgAf4iOlTMdIOQHzjQW9RkNAqlNAQc Am4oT4AUNKGIX2CCJhioimJIWt0emgxx5lallFaQOWKY6bfoUN6vghqFCXpFjAOG6JbhGSXE1IvA kHyJRnnFTmRK2zwaYiKJoRQc0DbTxxAEGXKDGhAAFx01xhBsGT4sNMGZRqBpBKqEGYy+M8bP6CSR hh5i3CZMLNQalCw3BLbz5s1DxPz58zXYz5HY3Clr1qwpr1i9FI+fm7qvr49D/WbPnk1FEszMLF68 uLNJg0LRiNi+fTu3/IIFCyDgUYMhGCXcCqsos0pDtKrOowyJ8bOXx44esPg8PGFIi4AnW8yESCD2 0CDT81A0MOQpylNONCT0FA2l7CokM36wM4YSnv+qzhuWhzwVqYW4b3zjG4FhrEkvpB1g1G4Fugtv cXo5DcwrUJ1MMxs0vHrbsWPH4NvBwwJuOB/41vI/6Is8EXgBwxaJuHfqr5JCKZkqRRxF/EJXo4jo AlZ4SKE6GgbHHVnQ05U/+MEPcq/KIaDKI488kr91MZl3eSCGOdLRhx6fJxam0MCfG0B3QjnQuA6A yY0Ekpgsb7W8CvoHZKjOJSoR1odbF+loiPl5PvDnp2cB+jPbCCW+F+FZnrhVDs8gTYMCHQogDpz5 QU+aTMwBH12Sw48QSNwAh4ZQdTTHD6Y0VEcZMoMjK/WgwV5VV9NjGi0CbjQ61cFZtpMmJ2MdPMNz Skz4Sw4/dhIjOjBXKfrDHIXVYVAYdsRkdAAAQABJREFUBRARdycEURoHmWDSqlHAGXp+rRolaNUq gUoqwnzwIY0++JfoWVgFWbQyTQAN0nmF04cLKUMmbDO40SsYkco0NDStur04MLigbs8lQrmdgYUq 4PNf//Vf9BYNBFCKkuG2wkDdVtwFef6UQgxbmoz24pK5+KB5lUTbflXChE6ItnQw2R73Sd52VCzs RbrLAhpcxo8RWoROhcnikMFKDxlMjtEmE3DohGpQqmMUmIReRw584AxZiTkuMgLDiAAD8GvXrsVv fumll3DNmQTgx4NUD41Uii1dupSB/507d3IDMj3CE2PlypWpFi9JSRx93tdMUZ4+fZqckydPEl2k vfWYqUBtHh1AREhDDMPTjxd9KpQq8uEphNy2DaRXsHjykCQR54gDtijzYx/7WCwd5vx4xMWZVAmX cgtjNXgMkskTDzcg/8oIFYcr4a9I1UZe7U2LUlPro/APaFryuYHFjkTYgMELj3dkECNPNFxmErwm uTlhBUN6IV0N30gS6XlEAtDjO6oWnpNcDV3SZVVLl7xi6XNkhup6K6uUvzIhRAh00/Xr1/O38IBM pMNKjhp1uZRbiZTAMJPAb1CtTH7+UpbCUMqolu6lPHHICbZDv27dOvKpgkMMCPyEBv5coI8Tsh2J UJIv0fjKIBaTlafDODRyUQZiDFEV+KMV5uhSrcCDOGYYqtNAmep0HqCmluhRDxHSmRwwp2OQKcda xsJEHYMi+huXggJ6KLlEt1rWoT8dRr0RJnLN5cZxSani0pdffpnL8IsbBQXID41CGouo2KpRApNW CVWURASJv4bkC6tgMuJCp8WWAGkhPZlAGuNGdUALLYUtgAlZSbcXB2RBDKXufTiop9GyvKhoIz0r at2D27Ztg2FoAvir25BZ8Vfer8qZSNXg64ND6N4lFfNoZFpEaIS+TQMFrKBUe8VoA2NoIECWnwEZ PRw1QhUN8ZQo5iIjkByBWkPm+Mq8cJlPYHqBnkyv1oaBhFqx2QPXnJcjnIkxeBJ2MOjZVp8lS5Zs 3LiRMIDfhg0bFi1aFDZLtK1bkQC1UZ5IDEO42VesWJFcREYTHpV6aTKlwLuPBxRPHnJ4wmQoSy71 kOe9jBMYftDHngCc4c9bDH+JH6+GEoZ6iRQSxDwLCYYl0zMYtWHXm4xuwftVq0F4OsCFfJwJ3BFa ml4S3r70VPL1pqSUl6V8lELBcfcVc3LiPsdl8M8Cz/x2MZjHugVZGo3WpQyhc5OpNIq10g0C0aCM 9GE4IbDNJ7ipIMMBKrklQi1ZFBBTfpWKgYPsitGjKDY2UCohSsY+wVCOYF1fDT6xOMBB4eAnUUSa dscokeEXFuqgzEx1vC7y6S2FLSvnSb0u8OT5q0eMHmowVDOJABvRB24yNtQqScTWQUa/pSfHmdiL e0dDI7GQrYjjKiXiqhcBjno+VZDLQxmc8/DSqUAAB7eDjhSUQXmYxCaUdCrVigloQWAHN4pCc+iS fNqIH0UV70GGM9AkVqaWaQgq71fSv9Vf1KYPMMNZ65bJoAEIMjmgIXNorEEw3gWEUhGEZ13QKmZI phCIMVG6N1+3wQonjAAI4CXTXdVjmwOEnRL8muMPZzZ1pNrX0UpPbvzMvd+KMlU+7xSGqPjxjuNH M/E2D6+e6lJKHmK8ufjpTapHmV7f1Zn3OKUDjNoNRD+jK+AfUFNegljoxck7WB0Fz0b5OGGKgCGg lOqibCtY70j8p1aUvIMJi6VJoAk3Ic4NmVImlMYJ1OBu4f5BBGT4atw//GKaOM09xs0gtnF+Po1i UJZzi2vJmQiax0UNpTEcQ7CIMAzbgYJmCk3WmVAW0oSnA+HKwJNj8KcWRyKJEs6YH7wi6YZ6hWiL LMNNbiv8BaZEZ8QVcsvQFF62aiD1ro7ZFspqmxl3URqOfpu5BcRBKBVGPm1FpCKQDrxgGAbL8BSk 8T0IAZcldw04d2kOHOhUrfpVRsPMZeaWIb5FVRTOkJVc6u6gsfJohJ6PbnTdIe5RJTq7yAgYgTGL QHga89TC1+LRxHs2M7RXAo5cmnhbbEzMKwCGPNLjDWP4YzHNSE87wOikBelh9AP6HF2E965Y8K7F 3yJTvSq4AvJfcWQh0N9aIktG1gkMeDEztYpzLFcv/+Yul4Xy/Ah7GBTnL7cQduUHg2FCPmESLjiD oPIqyIG+kL9cBMAJBMACJfm4xWHJSqgr5cPl0CTAjekanB4aBfWYoxQa3UgPhmAmDQdDOAMsWPEj J3aOSwQxtQI91aFXIEFbC8OSWnFRofOXiUli+vK0unSwrpx4iEt54tOIdLaOrWtaYe6XfLsHbdXr 4nuQlbvxKydWLzjicWb1dJf9SrcM/VA/dVH0r6iAelEhGgqPYchDhicngvhLfyMnPEMqSjGZETAC RqBLBHjy8JTj4aYHNU8tXt+8hfEWqgcYPNZ4sPOLn/+MnuhNyjsLJTPDmjzh9ZzsUv8eqe4Ao5OG 0OuQrkZledviQs+jU9Ij9QuscWR5TdJZ6WqF7nugjBMw4RL+SsRFpBGk8EbLHjKlXLZ1vNS/YY7a /Hipo6Tuq9gocSafhJbd52VlcriFxEEiQin5mRwV6Y6iqNDSUD1hAlkIRRw/bEcxnh3EP8Gh70AW bq46Btz0mIA5DPkh7oknnhD/VpyhCdVBGwBbuW6whUkGLi75hd4i0a1k1c3ngciPLpepqMlf6ZMp auJS+GBmLJFLZMU5Eq2cvM5NKNaKZ3hV0AcKaYLyg7dgm3sQi0RfyKptJt2SfsX7rFW/KueAaN0y gz16oEtrOxnPH/pGeV2VhjZqhYaeqOzBCJRV2JrGCBgBI5AWAZ5pPC35y3ioOOtVElwjPfTKH8g8 GzUlG97LemzyAMQn1FOOhx6XpHk+a2Q2vDU6swiJLKbIxC2dseq+ljd5d4KhdkrRkHg88buQbkQH wtEMvVDcRa/x+0xRiXh1EaTENHT6uE/TKUMp+XGRBMXVMwQMFjLjEVeRA1dxlLTEdcNFII6Kf1KG GIbMvDuStzSjarAxVQJ3n4+7BfRQKW7H/LODNs2LjhGgXSDQtBUAAixTIqEKzHlqZICNkVcztdqB l0FDYEpiEEGApLFemKA/DGP+pDP0oWLFBM9K4JILqCrwJGDO3AIVuXVGFu6jUB0daIXCewrMydfk RqDPL4cNRU0kaAh0QEN+Mf/QFrXuQUXCcRNkuqXeTHFmnFb3C30efUCPX6xYSZpbhl8gAF5+4ZIE xmZ6eFwqAqERa4U++tZKII6ZZHALNE4YgZ5FoOn9xz1r+GhSjIctrzweTbzKebfyF/+N9zv5MlPv et65FMUPtBgEHonMe/BAE5PvfOc7PELZCcmGe8h4fjLWQ11K+eGQ8D5t+xSN+efTvOB4p7BTLl80 LDmewegEdpwq+gGvxoxHSOejw5Gf8Xgg1nbYEA1XkQo3+NDFYUiMy19etziOdEp+WkLAJd0XMhwF OamBM1W4JVQdAshIh1ISxNB0aG4euNHX6egQkNCdE1OShgPSuT2ohR+Dh5Tw3Q9zfnhOoIfVqIrm 1V2fjKpVLvl8B8tFiAE0/ootQIHt/KguBNBBrNAkpGPmqg4x0AE+dUESAhKYA0/MUcPBHCbIiqsH 5KGEP7XAFgJ6S6hOQi1L61NEWsxhKydV/JFOEc8yVZd1gT/qwZ9WA1uYxzpUT6M8EnlykUAr9Raq 86Wp6ky6pOS+QzrGClhMVrvI8DxzmgMQ+AEsgzq0QvDs88QN5aAbd40W4NFV0Jm7DPTAUC1e/R7E Cj4HSRPwMFETZG55AnVyeMGot/CyiQnibpnpV9C37RggD9oYgnSwwgTdMqGi9iAhETPhX4gnQww0 B1MfeuYIDd64aI56/IUnpZmHjMwpZNg2E7S50xnXaKVSWw4mMAIVEVBo4QCjIlw9TsbDiicSbgkP KJ4evH3CWnQ0J4eBVF4ovFj5kcO7OP+Q4Z3L8xAyHph6xPEgDc9MPQblSlFKEV/y4LmtBzISIeBv DJSqxDmSKx30ioRVTDCMaQcYHYJPaMG7kP4X16ff0MPoLpk+AQ2UvHoz9HHdwjTeGy91Ksoxgj/d ix/EpHlr8jKmiB+XuJWohPTgLkCAKxYTkMZ9lyxUxfuBPzQhh5uqsHciFLZogvsCsepSMdWQMJZq gTgawp87DT2Vlm5p/+LBiD8OkzjLopDGXowVMoJdvmxQQ48DaJTPZby0I2NO3HCBA88F6uJjkZOp TrvELSuXDlkQq3Ug4IFCjiCCP7ErDMUc68jJtCxVyAzS6yaoS3ei9QMOIBabXJdhZ/T0T5kmw0ED rYRJniEaAgsoqdPyoAeE0Nvz9E3koBub/BAqHRCBGkFn9bq4pSi97777Ci2SsXQMLOIHDookg9pw o98GbtBjPrdVIJD5QMeP6vQr3QUKeAJZYUKPHSoW3jJU4WuY6IaZqIGBhUywDjQgC60gYtlLB6aH o7+woogOBjHqFXKrkqnHne6yKvSmMQIdI+AAo2PomqgYIj0+19sZfx5B/FrVJXjgF0rD+zfkKMHD TQ/PTL4uqRVXjHnytOSXqZVnFVeHmFdkpsowXo6r9eXmYVR0jIvmFYtDqddwBgq9OynFY8gU6ZLA gLdsCQEcRFDIP+YpVuS0pYxrVU/Dn9EChkJb2VKdVUVK2d5KYgnsgX85jUoz4DOISxyoM7NLqldB o6Q6GpZbF0yolZBWrXpjLVYdE9fVoRyljtWoVbFc5+r3IELLzSkXRHURtOrzbY2S9EyXVq3qnEuU pEhRQaqHDNimYtUWHBOMZQR4pP/4xz+ms333u9/lFO2xDEUv2P70009/4hOfQJM//uM//vSnP90L Ko01HTyDMTJaPB/IBr3bvjtx1sv9dTi0ZSJxbVkFrTpLwL/E0s54ltcqt72KMuU05aXoVkJQBY2S 6jAvt64cmValVbRqVTdVfl0dylFKpVU5n3Kda7VUuTnlglCyLUG5ISXSq3MuoaSIX7kOtUorPtxq 8TSxEcgjwCHZ3B2cCBHGzvM0zhkyBHiM6GE1YcKEIRNqQTECHc4cxSycNgJGwAgYASNgBIzAWEbA 60HGcuvb9jwCnsHIY+IcI2AEjIARMAJGwAjUQKBugHHlypXz589rfTLzHh1vFShREZXgj5SJEyci oqGxfJY1njt3jnmbyZMnI6hEn46LEPHOO+8gAitYn1mFT93mqMLTNLUQaKQr1NLAxEZgrCHQagvs WMPB9hoBI2AERhkCFf3ay5cv88mEU6dOsYRPAcDSpUuTL+c7cODA8ePHp06dijgUW7x48YwZM9IC zoeP+NQSoQXeP2EGX7Ip2RjdmWj2ffX19bHkiRiMeIlP7CCioWCpMw1dqxABBxiFsDjTCBgBI2AE jIARMAJVEagYWojdnj17+KLJ+973vilTppCzY8eObdu2bdy4kRH6qvLa0fX39+P6I4LPWqIb8cz2 7dsRQbzRrmrVcr5WsnfvXj5kNH/+fOpwiV1EGjrqoSqXUrrTp0/v3LkT/kuWLIEzH67kQ3OXLl1a tGhRaT0XDj8C3oMx/G1gDYyAETACRsAIGIExggBOMxMLDMMrusBqvhnNPAYfmk+FAC440xc4+pqy wDXnE6iM+nOWTioRLPEigOF7dIouYItFXO7fv5+iVFKwgrkLlMcEeGIOUsjEwFQizKchBBxgNASs 2RoBI2AEjIARMAJGIIsAI/244PFMAhMXuNGsBcqSdnrNSD8iCGDkl8MG/ixkYtqE/Qydcr2uHh+A 5hdiJJVxyUIp8q8j7fQCVmfOnEHzsLUDc8AN0xJi1al2rtcGAQcYbQBysREwAkbACBgBI2AEkiDA aiX538FpFlumF/CnU3n/OgM3FoFrjggG/tktncQQNpBgS2YvBBLJpCiJCKILdo9kROiSWaAkIsyk OQQcYDSHrTkbASNgBIyAETACRuAaAoy+6yjJzGej8JtxzVN5/+ITpi8knktEEMZc06aL1NmzZ6md sUKXqaxg2RgKZ0QowJD0LtR31cYRcIDROMQWYASMgBEwAkbACBgBECDA4NcKilQzGNqikAkw5Kmn 2r1QYkUqEUxfAFTGCkGXCqhWDeH87hFwgNE9huZgBIyAETACRsAIjGkE5AcXesMZXBiVz+SEyxKv PdBUSZTwKSmqwjnQlFhRUhSqV0kU8hHChUVVeJpmyBBwgDFkUFuQETACRsAIGAEjMDoRqBhgQCbK QhQyy4EKaapkah1RxgvXZSoRJVaUFFVRPtAUqlrRiqBDSAS2TgwNAj4HY2hwthQjYASMgBEwAkZg 1CLAzgo2BhT6xLHN+Lsl3n+qczB03HVhgFHxJOxY58I0H3civ1ER2qSemXKRxLZWUEv7NDIaFtri zCYQ8AxGE6iapxEwAkbACBgBIzCGECC0kM9dbjNkiiK0wSAQc0nskfnqayitm9A3cAtF8LHautwK 6XXueF4ExKmOJAco4MoEGJLYVgRxBc3BzwFGYfMNQaYDjCEA2SKMgBEwAkbACBiB0YwAw+3yaMuN JIrgaG1oMtuUueSUOk1ulHOoUpoXgV/ODw1TBRiIwPvPW0GmpFfRs5xm+vTpKMyW8ThIkEQdIFhS HSTVHF4iVYJSo0UOMBqF18yNgBEwAkbACBiB0Y9A7ASXWzt37lx8Xw55CFVOnjxJet68eeUVq5fi fzPGz3kU4YNOHL7B12M5Bjs+HKM6wzwl0wsEErDlY7IqRRbHUyA61UIvUJo9ezYrnbTYCSnMZnDE nkTnVYpzArZxptNDiYADjKFE27KMgBEwAkbACBiBUYtAFb+W/QMLFy48OvgDCHz0vr6+2267DWc6 FS4M2995552c240QVML17+/vR+6iRYsSili8eDHM9+/fz9wIIl599VUulyxZknDSAKAIJ8AH/kQX Bw8eJExauXJlQhGpADGfDALe5J0BxJdGwAgYASNgBIyAEWgQARx9liodOXIEvx9fec6cOTjraZ1m IpZ169a99tprhw4dgvPMmTNXrVqVavpC0EybNg0RBBjPPfccOVxu3Lgx1RIsiSC62LBhw4EDB55/ /nlymJa56667EKRS/+1lBBxg9HLrWDcjYASMgBEwAkZgBCDA4H0tLQkq+FErbVwR68ASJn6NimA3 +dq1a2V7Q4Yw8bJixQrsatSQGDenkyDgACMJjGZiBIyAETACRsAIGIF6CDTklMdKjA4RWDQEhsS4 Od0lAt6D0SWArm4EjIARMAJGwAgYASNgBIzANQQcYFzDwikjYASMgBEwAkbACBgBI2AEukTAAUaX ALq6ETACRsAIGAEjYASMgBEwAtcQ8B6Ma1h0m7py8eq501cvv/tB6G65uT7LLSfdPO6m6UbCCBgB I2AEjMCIQMD7BEZEM1nJIUDAAUYikK9evnL2ON84SMTObAY/GHHhrYFdXTf6g3TuD0bACBgBI9DT CCi0cIDRa43kFhmuFvESqW6R59T6v/zLv3z6e1sdXXQLZVH9qxfeLsp2nhEwAkZgOBF45ZVXOFZs ODWw7B5DgI+o6jC4HtNrjKpDXEFz8HOAMVw9wAFGV8jzQPnXf/3Xl19+edOmu7ti5MqtELh6pVWJ 842AETACw4XAlClTPve5z+G+DJcClttrCJw7d45DpvnpUIheU2+s6cPgr1rEN+lwNb0DjK6Qf/HF F//2b//2D//wD2+aPDnPiGPt85mpcniEDfJvcFEWIhDCWqVUOuf5yAr+5otS5QyZiFQKF/KRFYVF qTKHAChUpdMOQXOnwqSQj4EqhCWfOQRADYEI7CrstHfccceMGTO+8IUv5A13zthEgCPnOGGavx4y 74UOMGnSpFtuuYUWGT/eju7wNIj3YHSOO8Hxn/7pn374wx/etGnT1QtnYkaEzj/5yU+ee+75z3zm 0+PGJe7cvFPPnDlz+PBhjp2ZMGH8zJkzb7vt1uRSJAJZ3JyI4Jf8oXn69OkjR44EEbNmzcSiGMbu 02+++ebBgwcvX76MFbfffvvs2bOTP2veeOMNiZgwYQLHsiIlOVCnTp167bXXsALOiJg7d24TIrCC fosVs2bNwgoS3eMfc6BHHThwQG1BQyAieVsAVH9/P+4gnBHRBFChuSdOnMhNMW/evORtAVD79+/H itAWDQFFW0hEE1Zwd9MWobmbaAuAQkTTnZZb78KFCwCV6bQ0ym/91m999rOf3bBhw4MPPhh3dafH JgK8y8am4bbaCBQi4ACjEJZKmf/wD/9w/vz53/7t3w7UvIe+9c1vHejv7+t79cCB/jVr1/DAGZfY Z77h6NGj+OXLli1jjp4Z2b79+3mXL126NKjRfeLo668fOXx48eLF0265BaP27N371ttvLV60OKEv deLEif4D/UuWLJk27ZZz58+/uu/Vt956C4kJRYDS3r1716xZwxgGLbVz5068w9WrV3ePT+Dw+uuv I4K2wNcEqG3bthHSrFy5MqFHePz48T179gAU/g1WsPKb5l6xYkVCEfQoVpPThQgtELFr1y48dYBK 2xaIAKjbbruNCWva4tixY+vXrw9Idp84dOgQAYxEBKBoi4RWCCjAZ+iahRC0C52Wy4Qi1GkBf/r0 6Vixe/dutUX3+AQO6rR33nknzY13vmPHDnpU8k6L5upR3BdNdFr6DyJ062EFIpJ3Wm49RPBQIqon UgKoTKcltNi8efMXv/jFjRs3AmZA2ImxjED1MIN+yyOFJ8lNN91EH+OFnhw3xim43wnFiZARwasw uQjs5W3Oi49X0q233soTPuHzMGjLuxspcMYEDKkionpDBClOpEUg8eB6WuV6mRsTFF//+td/7dd+ jYHYoCcvoXHjx61cueJXtvwK03MhP2ECnwNHatbsWfgfiJg2fTpDg9x4uAippDAzc/jQIfjzpJg4 adLUm28mtDj2+rGEInikMl5+62233jZzQATzmAvvuIPX+RtvvpnKCkTg0TI0y4v/xhtv5KmEL4J/ wC+VCF4PuJiIoA/QFjfffPPy5cvhzwM9lQiaGxEaKdeEL36hIsxUIrCC8fJYBMHMyZMnEwLFS462 QAR9lbZgCQEicAdxplNZAVBEF4hQW9CjEAFQvJZSiXj77bcBiu5EpEdbcINIBP02lQh1WvjzU6dF xECXTd1p1RZYQVsQIME/eafF1eDWwAraoolOS48KIrj1AIpOOzivm6Y16LREFwA1f/58rMD5QwSd ljmTIAAv5w/+4A/oACyUDZlOGIEqCPA+ff7553lw4S7j/bOTkzd7Wp+Y2J5V3LgHPLV4A9KfGQ5L K4LRIjTnpsBb4JFIgjgco6ogUJEGhdGcH8M63I/cbhiFl1KxusmGEQEHGB2C/+Uvf5l+//GPfzyu z0voV3/1V3/5l395YNx0fOqZi0FJ3F28+abdcm0cgruakYPjx5J5OTyPiJS4mYNpU29mUel4RIec LhOnTp/m2ReLmDFjOni+cfLk4K6PLtkPVMcKnn089QIvgGJZC1akesLCCqBgG0ZTgoggtMsEw0IE APGYEE9Y3kZY1yXnUB0XnPdB3Bb4bQCF958KKETQFrEIgJo8eTIBQCoR+ebmhQpQiAiWdpkQUHGP Aij0R3RaK+Kx8NCjUolQp0VE6LTEGLovusQnVFenJUYKIhrqtDFQ6rQJ7+7QacNUoTptRgQ7MT7y kY/853/+J1MoAQEnxiYC1W9SXhz4+tx6zPLRdRcuXEg0TsxMkJwQOganeM8ygsANSBhDXyUCTziU gKp9fX2MvKxatYqbkVueUTxmS4iUElrBmwid8akwgccv5vBOBL2EIsyqIQQcYHQC7NatW3/4wx9+ 8pOf5JXTSf0u6jCExmt74sRr6+Pxosg589YZnlldML5WlVUfgyKuzcDwikUKfkMq7//0qVPIY1ww SB20YjzPpsuX0liBtkUixjFulAooXBBE4JwFKwCKSx64qUTICkaagwhJpBtUf5nFdfNpvdJiEbQF htANiAry9B3kaAA+BgoR/M6ePcv7rwOG+Sq4fXTa2Aq1RULvX83NYoYgHRMQKk80ZHaTKOm0qYCS FXmg6LSpxh1LOi2DI93gE+oOY6dl6DRuC/rAr//6r3O///3f/32qWzKY6cRoRYDnEg9YzV3IRlbi 0X8STsFxRyOFSePwyCIAYK6PPUWpOio3Ag9eAiTYygqGdXCKsIIYIEnb8cRgsQORWBifwm3AEJ4A GJhEhJk0h4ADjNrY8oL5yle+wn37wAMP1K7cXQVuNu5bJhPGjbsWYODisDf68uUrly4mcAffEzGw fTxWFiE8lS6cT/PUuHB+wK3EBYxFIBGP9mIip1ZOQCwCoPghIon3H4DKiOAS/kl8NfhoIjgWAWL4 NLRFEhFw02Mante3xQSkp3pJEEhI7SCChhBQsa8WSusmUBU+4hnqcpkWKNYvwTwP1ECnTfQ2FRpx JIYV/BCRBCi6zeADZAD8GCi1RapVByWdVhgG0R0nSjptEqBQrHqnZQHYhz70oR/84AeseOnYIlcc Uwjgl2NvPMSmS4bYUuGgOD9EF7DlNmdkgRdHqjsdbXme5K3ggZxqQTVhGNry1I0fWTJKYyWp4DKf JhC49pppgvuo5Pn000+zl/cDH/jAokWL2hiY+pMSOLX5z1QMxBcDx15fwXFuo0+FYkTwGyS8tsRL Xg6Zly4niGECH9her9HA5cUUYRKOFD4Z3GIRSmMdj7/r5XZyBRPxiUWIUSoRhVYEo1I5Unmggogk fnOhFYDGLxR10gBRHfio08Izyn639VMBVdjcktgoUFiEdWqm2LoO0gBVaIVEJLECViU9KlVU3LSI 0DPjHkWaXyiK8f/oRz8Kegw8xZlOG4FCBLid5eJnRivwoenYqYJwxSr02FiHtEMJiiJi1x9ZMipV pESczx2XAUqXxB6xaU73IAIOMOo1Cn39scce4wGxZcuWTKevx6gjaqS3OpVioCDRmXSDUor1Y56k uKBm7nt8rnv2iUcSK+RuwjDzeJWIUFpT6+vIQQk+hfyhK8HwOi6lFzCRO1hIlcQKODctohyKJFao LQpRIjOJiHI+SUQEoAo7VSoR8CnkX25gK2wL80t6VElRIatWmSV8UgHVSjT5eenr1q1bu3Yt62ZZ zlFS0UVGAAToooURsjz1VNMLCuYzN7suU8X5GrtpVIRGPTIxjCSmAsp9sjkEHGDUw5ZtWHzBYMGC BXyXsEJNfIaUsxiD/njOKR+UUFxUQcVaJNePhtSqeh3xe3wKwMk8ra6rVv+iEP8kImDCr5B/fTVb 1kBEq7KSolZVCvNL+JQUFbIqzCxnUl5ayLBuZioRJXxKiqprG5gUdqpQWp1hISV8CvlDnFBEoegR JKIuFCxD//mf/3m8xm984xutbHf+GEGgbefhBmx1DwJRPnztDDfxySijS4U3nbGNa5UE82mtiIWG dCoRgaETyRFwgFEP0m9961vEzcMyfYGibLUeVPc6v/y9C9ZVt3RGqxvJA0jPoPgJGNITJ1zb0Fyd Z55SfAJbEehywvjrNgPk61bJYXJJVsTE4k9+kqknxlTyQEkc+ZkRl1iN6mn4aDl+IVDxSv3qPPOU JSKSABXQiK0grcvmRGCpRKQCSqqKZ4AxrRX5HiX+5KcCKi9CtpCfCqiSHhVvLg8YdpAoEZHECtDQ LRw3N2ldFlrBljzyv/nNb2rzRgdGucpIR0A3l/6W26KOVE7TXGkVDZuTbs5jBwEHGDXamtk6NmBw c77//e+vUi3p7MWAQN55fD9qYB9GtBpqcE3RwHnbSd6s8Jk0cdLgm/S61VDs/kCBSTde9zmjKiAU 0oiPeAYCDAHbVCK08yweYhk0amA1ZxKg4MNWM/GMTBjIGMDw+u8+BYJaCUTwLVeqwDOuKKPi3Xtx ad00H+igSgyULgfaIoUVcJMVsQhZlBAoVB2APgKKdFqg8lYkB6qk0yZpCwAv6bSpelRJp1VnA7cu fyWdNq0V1Tstp2TwISC+2/Pss892aZ2rj1AEuL/QvK37DoEo4+dVMDnJUALcSoLwVCJK+JQUBUur JEpe1iVF4hwaQmhXEWeatAg4wKiBJ28OzpFhfRTnqVWsVvgEqVi3kGzge3B8Cib61NKVywi5imty Ywrvn3tyytQpMIxnUQddtcuc8tH2li7UOZ+JCDLj3be8yPlhXaoHE9/LQ0S8b5UZVezCilQi9G2+ 64EawG2wLa59gTdvfsUc2kKOVGyFjMKHS9UW+vxfLAKg+CEiiVOLwiwg4W/c3BIB/yTuIEDxbUQa N7ZClwCYCii1Rbx8GSvotPS0VECp02ZEqNMmAYpWUKeN2wL+6rTcGhB0/yvptKmA6sFOy13/C7/w CyD5ve99r3sMzWEkIsC2Yz4gzv5m7qkS/fF39VDKkPEwoVaq21BPDPEMyugy1cNkCEToiVFohQYy gmn5BDcjzcEvfi/kyZzTHAIOMGpg+/3vf5/1UXfddVfl4y/KnjI1BEekM2+bydU75wa+mKkfvgKO zpw5sxk6eS+vq/8zZtyKxxZ7OefPXeAOj4+16koAHueAiPFnz177CoTE3Xbrram8f53zFS9XQARP HJxdrOtSf1Xn9CIS8de4aQueZfEpZl0K0gll8XdF6IE0d8K20Cl+MVCIkBWp2kKAxFbQFvxoo1Qi OMCb13YsgjRA8TnpLpsgVKct6Dnx11EAilKsS2WFOm0sApTUaVOJUKeNP8DSUKeNrRiyTkuElgqo 0GmDFzjYZ89zv7QSsWnTJvrDj370o1AldB4nxgICHFpP6EsMX/6K4UmlKIJbO4aFpy6BR1u/Oa5S ktZoBc/AQEO35BLpqQIMichYIYmVfaSgXXFCI0TwjO8piZD04mqDuYBJc/BrdcOW1HVREgQcYFSF kXfkSy+9BPXmzZtb1WFNBgdbbt+2/cfP/pgDHV4/dpyPo3PCa8LjhKfP4M6dxrndgyujBla2HDl6 hOdFQo/z1lsHhHA4mh4c3NiHDh+afNPk2bNmtTK8bv6tRBi3zjhx4iQnCFAXEZz+gxUz04ng6cMx RgcOHAhW7N+/n+EQzkytq20rekTg13KokBxNyPgGAFZwJmurKnXzZQVHmQYRdDBeTkyj1WXVip7X IZ2H5pYI2oJpOoZjE4pAYdqC01jDwLmOem3/oedWSufyAYrQEaA0WIUVtAtyEwYYoCSgggiAwhug D+TU6TAjNHfotADVdKfVrTfiOi2wA06m07JOqUPoc9XoPIig04bYW522RATfkqKxqPLTn/40x88Z ox+BzEB7icE8DwlCwvMQSuriN+MNl9SqVUQkDD1RcajFU4Ufjyz6dsjsJoEvghMfWyGJvD5SBRi8 nnilIiLGVkZpYrwb/V23aQTS7NltWste4M/JOPgTvD/WrFnTSh+Gr7/97W+fPPkGgwT33nsvnvMP fvA0N8bGjRvYF96qVq18nkrLli3Hb965c9eUKZPPvXNu0o0TV61cmTBGZ26BxcT9B1/bvXvP5Mk3 nT9/YdKkiStWrJiYaEX+oL3jli5dihW7du/mYTfgsY0bt2rVKuCthUYJMUCh8+7duwkLGVjCEQGi DRs2pBXBYrk9e/bs3LmTgRaeegh93/vel2oIStYhAiu2b9+OFXrO8kHMtCJWrlxJaCQRDPwDFFbw WC+Bt24RbbFv3z4cL6wIQCVsC/QBKKKvl19+GTddVgBUWhEAhRX0KETIivXr1/M2rYtGK3p1WnoU VvBmpdPyJGmi06LAjh07ECHnoIlOS4xHj0IEdzfPwOSdlh4FXBKBFQ3derDdtWuXrMAigCppbu5K YowXXnjhu9/9Lh2jVSs7f3QjQD9payATiXQqRh4ZAeEeh57xL5z1+fPnt61bkYAXKzOijBwxpqb3 BUfvES2vXr2aXl2RSTmZFGYohwMxFFEggjSROUXldSuWAg768wbkWD3MoRZPRWJ40lqKWcKnSkOU VHdR9wik6QTd69H7HJ577jl6NkOYJSO7+GS/93u/17Qt3Lp33nknTwpGIybdPolnh55QCeVOuvHG O5e+J2LipMlT0ovAiyXGwBHE/8AiHhapnnoBB0TwMGUJEyJwNBGRHCg0DyJI885ILgKGhF40N15U QyJgi3dOWyACoHgzAV2AMUkChniEsoI0YUZyoLj7AErN3RxQWNE0UENgRRBBW9Bpkzc3jUsw1mhz SwTNrfuCuzu5FTCs22kBlgCDT5knuWvMZLQioN7LyBRxPr4yoxUEG0yOpR2VX7ZsGQsokEIYg7fN XCWTxkyeJEQV7x+/iACA0AijEEHshH+SUAT48MhlZIe/vJ4AitcHj5eEIsyqIQQcYFQFlg0YkHLn 0Lmr1mmMDl8ct6Ax9gOMh0ZE20GILm3Eiqbba2hE0NyNtjjvhqZFqEc1asUQtMXQAOVOW/HG77VO e/fdd3/ta19jTJcBV61RqWiIyUYBArWGzLnHmZw8efIkfjMDIkx5JR9lY8yFOTe6InE4z0ZmEZM/ fvH48fURwWgCLUhAziovZCVsTeJ8xiK5m5geIZghpCEMA7GEIsyqIQTcSFWBZdECpAz0Vq1gOiNg BIyAERhLCLBECteHDe6s/3SAMZZavhNbWW6XcO9ToQb0RqYs0s5aZAQRAGjxUiY/7eXAZu10G1TS 6mZurRDwJu9WyFyXT+jMSANZRM/XFfjCCBgBI2AEjMAgAqxEZ4kIi1f1RRCjYgSMgBEYswg4wKjU 9Gwe1exnwu/eVBJsIiNgBIyAERghCLAohX16KMtOjFoLZkaIfVbTCBgBI1AVAQcYlZBiDxN0rCz0 DEYlvExkBIyAERh7CLDoRR8CYrcrO3fHHgC22AgYASPwLgIOMCp1BX3XnIWMqb7uXEmqiYyAETAC RmBEIaDPDPJ5UDbvjijFrWwaBNJucU6jk7kYgeFAwAFGJdRZIgUdi2tTnVBTSaqJjIARMAJGYEQh oLMdOSudAwFGlOJWtlsEHFp0i6Drjy4EHGC0b08GohiOgo7FtXyUrX0FUxgBI2AEjMCYREAfj+Js Qc6VH5MAjF2jFWA4zOiRHhAaIiR6RLGxo4Y/U9u+rRmI0jeeCTBKjnFtz8gURsAIGAEjMKoRCGel aVhqVNtq465DgLFIZq68uf86UIbvgo+50RzIJ9ofPi3GtGQHGO2b/9SpUxwWC90QfOy5vTamMAJG wAgYgV5FgJW0Uu3111/vVR2tVyMIsMCBz4glP1G+EV3HAFMmLmiOMWBo75roAKN923AKJqEwdBUP Trp69crbb5/lEyI8bqZOmTKxsVVVTK3s37+fE0Db21CfAiveeuttIqtJN066eerUCRPSdxVEnDnz 1sWLF5sDiqELUMIKpp44ObWJ4z8Zr0IEzc2zjKNYm1hEhxUc3YUIrOAo1iZEyArOSRVQzYmQFQDV xGTg0ADFiIOau9G2aBQompsepeZu2gr6Erdeoz2qUREBqIqdNsxgHD16tP5z1zVGMAK8X+iK/Eaw DaNIdSI9tYWXSA1Xq6b3GofLkubksj5KU2y8YNpKuXTx4oH+/vPnL0ybPu2NkycvXb58xx138H5t W7E6wYXz50+cPLlt+/b/961vzZo1u4kAA4+8v78f13/atGkcMnjo4sXFixdXMb+6FQD1al8fwN58 yy0AdeXqVYBKKwIXbe/evQSHfPvr+PHjGLVs2bK03wEDIj5hDGe8ijfeeIMp8jvvvDPtgaOXL1/e tWsXviDx7YkTJ5CFiLRWwHPfvn2IkBUcQrxkyZLgJ1Vv0xJK2oJgmAlrxnfpUdxTHCmjEwNKatUq QgRW0ASI4FhMLpsAirag06pH0bVWrFiBg15Lz3JiWaHmxgoSTXRa7guAokfRFjTK8uXLm+u0mECP St4W3HqvvvoqHQnN6cBNdFpuPYACH4AiqmSkqUqnDQ8xnc1a3twuHU0I1F0cpQER7kQGp5oLwum3 9GFcbUQ0NKLPbUgcPn78eB6GfAinCYeeByMPKzhzf1V0qOo2x2jqij1iiwOM9g3B+LSI2n5Cig69 /8D+c+fOr169enCw/Gp//2t79uxdu3btTTfd2F5SBQpus0996tM8Jvhmbl/f/lmzZleoVI+Epx5f zeL9jdo8MqjMixw3mstUo868uQEKt+A9EVdffbUPEWvWJAMKK/bs2cOz+95779XzDv58bpjLVA9Z mht3kxYhxtNICRJ37NjBZSqnEyteeeUVXLQgAo9n+/btCUVgBWojYsOGDZrhwU3fuXPnPffckwoo uhAieAkBvnqURLz//e9PJQKgaAs67caNG7VEAYkAldAKOi086bRBBFa8/PLLCUWo0wIUyKjTIqKJ TouIu+++W52WwK+JTotPAzIS0USnhScOB0DFnZY+HPz7ek+9ImpuPZwzeKqXqtNu2rSp/EWAAnRy mpJBjSKuzhvlCFT0a3lx8ICi9/IqJxbl2cJQQtoxFx5Zen1wNgu3PLcMEbLOaUnVBhjLfcHg2rx5 8+DJw4TxHUYT9PhKIgURDHceOnSID0ADF0BxAzKy46VoSeBtlIm/ItUeXkJzEbV9dZ19++ybb57i BtM774YbxrFt49Kli0eOHG4vphoFLv7/+T9/9fnP/9/f/M3frFajNhVjdcRUaC5fkPo8O3DdTpw8 UZtXiwpAClCRiHEScex4slXLPPL48TwNTzoOSSSd8NMuWMGLgaeevChsRQTP9IQi8NKYteDDl0EE U0m0xeHDyXoUIrCCl9x7nXaguQGKZ3qL1qudTZhHW8A27lGISyhisEe9CVDhrcN7FKASroPHnQWo WIQ6bcK2GOyzb6B56LRqi4Q9Sp0WEaFHYVETnZZ7IYhootPivsvhUHcUUAcPHqzdO1tUoNPS3LAN MTBpOi2jLS1qvJtNJ9ebgg5TTunSMYsA8SfRBV2FITb6FVOIPIEJBsJoZhJkGFPjbb5u3Trud16F rBEgxkgb9zIWydOJEVWY81u1ahUrAzllMon+YsLHEpifJKLgMcItjwiMIsxIKMKsGkLAAUZ7YMM9 Xz5wBaNjx48RbU+ePDkwJY2vgDPNKzxkdpOAG2tXUs0kFGrCa3XcuPGxFbxW8dtOHCfAuFpYpW4m jhRAxXgKqNOnTqcCKi8CR4EHOv46ousqXEivr8TEQOFR8ZPowip1M2kLqsRASQRW1GXVih4RmbZA HECRr61HrSpWzwcoRARHjYqkMQQrUjU3gQQ654FKuA5ePSeem0Ic9yPNnQoo9Zx4AUDotKmAUqeN reDuHqGdNh7xUadN2BatOi0RWtu20ApGKKvfI6YcBQhUf7PgVDCywxAbd58MTz44xVwrscTtt98e HrzEMLytEnr/3AiIoLezlFpWkObxxZhLqkcibHmGozbjthLBnc48D3KZAlKO//YsAg4w2jcNNypE eNht3XpmIUUZmOJ/jB8/4eKliwnvt8C8iQSPyAvnL4wbN2Bv4D9oxXhW0WBHyOwmwegg1cOzlTQi gIr8VEAVtgUyeColEQFQiBhQenAVmdDgEtwYOFef6QYi6iKClxCJGChdwh8pXfJXdcXPmebmEhFt HamKCjDgJLUDvYCiIZK8JAQUPGOgdEmPSgUUq2XyVtD65KcCSp02YwWGgFISoNBfnTbf3KAk6aGN Ok6UdNok9wWK9XinVaArHDqG0RVHMQIa+IgjZIJ8HAyNKCUxXCNHmfEvRPBITPUwIYTGK1BvDzpz Cf9UnR8+/FA7vGd5HiKCKSBGE4JQJ3oTAQcY7dtFL0X6d/ziL6wmV4YbIC4dP34criJ7muPMnk1z 3166fAkTMlbo8mKiAENAhUeG0AAoPMUkQMFHz9BYhNI4tUncQZjkm3sQtnEDGA5+dqzLVi4UAU8Z kspX4w0ReAaFMQQMmxMhEwBKGAa5nSUKgcKEtECF50CsZFqg1Glj7z902iRABcBRO1hBmh9FqZo7 f18gS4ZoZCGI7jhR0mmbE1G90+pNwUMgyXOgY5RGcUWeTiyH+9KXvpRw9eCQwYXyCubjOx3pXHLv pPL+80E4IuiZPC1T3SMaOcpbgSAVdQ8pwzfcRLr1AjdJlIEh04keRMCbvNs3it67vIMzvTxf8/Ll gfNcoIyLdHlpsCjOT5K+mmjNUlCGZx+uxuDlNSuCRcQegbKbxLtA3XBNRJCYBCis4AfPoHmsLU/Y +LKzNCi1EkF+EhHwKXFQSopqWSQ+mb4t3FJZIaAyWr17X6SIxOBfKEISUwElNIYAqMJOm8SKAFRe BEVJRIB5CZ8kYVIQMQRtUdhp21rBaLQqQtl2WCojwpflCLD85tlnn2XL79NPP83f+++/nz0M5VV6 rZQnibpQ5jZUfyZ4DouautFcQXhGBJfc6aliGPHJiJAVqUTI+8rEMJKYKkzqBmTXLUfAMxjl+AyU Vg8wSryc9mI6o0izm+A62SU7FEqKrmPR7kJANaD7u4LLG6K8tJ3u18pT8bnG8f+z9y5gdlbV/X/m mswlmckkM0lIZnInISFcokAVFNDiH0GEekG5WPQBW+WmVS7138cfQW39IbZPAaH2X7BoBawogq1Q obShBisgqIGEQJLJXHJPZnKd++3/OWdldva85z3vec85+z1zZmYd8gz73e/ea+313Zd3rb32JZ2Q K+5Cx5eab2Q6ZYyldUIkJdMALgGvUpK1EwgdX2q+kXbeMGEnRFIyyg2XZMVwxV3o+FLzjUxWnmTx WRIxRkVKUyRZATQ+GQLs9+W8OLYToF8Oz4UlS5un8bSugAbmqs3IhIhH+5dHtyx8UQ6YZfBNnywy oIqdzH8l46vxThBQD0ZqGKWr0DM9ZnRizqKiwv5+JrZHvJGhJLZQKoKfcw8GZcQBExdhpBjxwiOg EyEEqIQt4zGOToAyU5u+47h5m40stIcAOp5hPTNGwU0ugHta7GjVjNQeoIYbrYPqDi5n8NuQggAU v2SJnbCAOHSM28rwigIoT10ILydSgFIAnYBXRt4wgYBxMuBVGMomTZ43WmNgqA5kqsxV4Kz4D2pP P/20w7MunBRPRqGAschw8e3j8jbglckeJhCgmofJnidpfNEQhH1f5UmxtRiCgAMFwi2UHA7Azy3N LKnhamRBIQv+Un4t5Lviafexx4KCkuHDIrIsTNTZC8WOwnCx7CTC8mQ+nFkW4xhQgyNsmBgPR0Ax ACXWhciAFuVECoiItnQMmmFEeIS7E0UKOrLWwsNCvhwpjxwYLlGK/wfQMSs9UpBI9Vro2B88JOLn EKjE6qZQUQBFsRPFdQVUohTCjkbrqkUFNFpXUggdD1BSF07WfoB/QKN1xSKg0aYcQFhhL58Mu80n NhuNGZcIMKwFy0UCSePpI5LLSU+HVLKebl4FFzLM24ApiYBXYSibNL50BLeUQAnIkDIBQ1YDuUEg LwwMFHeMCq5reeqpp26++eZ169blRviQXPiYcaojP9+2bhORb9vgoL3EP7YcH1U05TfJppNG2Efb SSN3YlIUGeTFC2NbU/RntmYgfmnpsbXFiRnTihGgRu7oiO39KC3BEHPjVZMDOmwpoI8YSJdyYAoj C2MWZ1nEcDm2ZSWWiXCsuouLAxSgMMQlDeWUM0BsKXjFI3XhSpGSQ1E9Hm0ekcKVximHGNosAIof AjqRQuhQFzZQ0OfRFQtgl7qwpZC6oK5dASWN1l7AgBTSaJ0AFdxoPafBSCPM4G9Ao3XFIqDROul6 SJ1No5XvBbWpmk0G7WfsZqGrMj7QZ4NFoFX4jhiS0UlPpwDSDSmSXRh5dNVHhI4vC1dSJNr5iCMc w7CgOvhpN7TbQC7DeWFgYF3cc889f/d3f/fYY4/J1ci5hCAlL1onyha/lM20uno61OzDWDh2ic7A lwYLIyWjDBJ4+nYGFBKzxA9xH+rtix3OKz8UNX7cv8EKi+G4rP4v+kG3dY61ADV12lRXQFFaijiy LvoYwfn2o3RmVfrhzHLavb2bjeoAKFQTWstwqqz+Dwtanc0CEVBAJT4r0sOZ5e5YmwWguZVCjjD3 1AVSUBeuPnWsyQZ8WwpEgAXSpey2w0ik+H9VVRUp7J2FSATT6dOnu6puabQ2C0SQRusKKGm0NgvT aH2VnhSg+L0OaLQp52j86PnEJWu0tChXdZGs0TKYp6wLRhiK4aokPvLncRT6HN2QdhtX7WJ/CNNT CEipac8kMG9NAtqhEYtIspBRfoTpzuZt3gYoJz+ks2VJLC3NQ5qQRyge6SCujHD6Aqw9LAAWFmJ7 JBYs3Rj5jieygI68SpdgYnrQoMAeFvKYkoW0ImqEQCJljckBAm7UoCwLylB+0003cenMT3/607vu uitLam6zM1IEDxY2u+rqql27SnGO19XV4pfj1aFDB1HK0X5caTmGHVPnhJk/NzGuAmg5e/ftO3L4 yMwZM6XY8fPgYreSu2JBjXOP1cEDB2qHaXLrLUDVTK9xBRQsmOGAi+gilFwu4eLeYodSoJNx3DjI iOYkN8dR3Q5ZIAUtimILMkhEwOHBKZhDsAAcLkmVYiMRIzgsXNUFqjksqOJ58+YJCw5QZ9zn/lpX LKhuvtk2UEgEce6ZclUX1DKn1lDFBnyAot4d9m5ptDYLgOLraGo/e1lgQaOlFYGMNFpAo7qdN1qu PjTFjqjRMjPlabTUkasWZRot1S2mgjTa+fPnu2KRfW3mIQWuWP7Rj35EG+NH8TAnaF204dWrV190 0UXo1s8999xvfvMbeqtM9EgC0lxyySWnn366SPT973+fU2jBWZoobxcuXHjFFVfkuc2G4o79yViX soXQ3RgP7QkRcMCaotWJyNnXLBMfXLNts6AW+FE8VwYGUwnUiBxXZQrMwE6kTMeYyIwDfJ5oKiAD PgYZWECQoSyYLMWgOkhjMgan17fOEcgLA4ORyKEe4BwjIRjGzECQ+vr6lpbW7dt31M6s7eru3rVr T23tTLq6w1Ldffe3u7o6m5qb6T9vvrnpy1++pXr69Pedf97ZZ5/thAujzwlzTtje2opaUFU1rbu7 Z8eOnZhMsmbACQvGOIBi+OMU8+nV0wEK1W327FmuRiUKyai0dOnSN998k4tLUTs4Tru1tZXbUo29 kb0gsDjxxBNhwbeQDwbj7LZt29B4YJc9caFAixIWzc3NKDp8LQjMmTPHLYtFixZt3bqV6oAFc9sA RcAhCzQJ6mLTpk0UHqUToDgKBvoO6wKgFixYwDJLqpu6YBE8lQJQ4hNwUh30C4DCxUrhGa9gIS3K Ye9ObLTwiqjRAhRSUBdUfXSNlupGG4ii0S5btmzjxo00WljQL6TRGnsj+xqn0UrXgzINCSmk0cIu JfEwH4uURMZoAgx7bkbDvKR2EIEmTccEND4fQAoyDMXMWDFskoYEqIDgjLIovjWRGtPxjTfeoHJR K0lAv2MyguyJmOQV1OELg7wgY0+IODfC8WAwxw8LtHOGR6DDQqa/8+X1RTIR25QxiIAg1DhkxWih J8KFUZdXKbOHSUBR+VKgHtCopIVgIzGzQ6vgF4aCphlFBPLCwBhF+Z2zRttAe2aw2Na0jV7d0FDv UGmW0n74kkuYICkYPpaKQY112nPmuDwLfMaMmilTJuPHOHCgnaX4S5csZqxKOSuTFpgC1N49e7Yd PAhQTFDZH5i0SCVLDItVq1YxNjEeYYyh4zrUBYUpw+spp5zC0ezonQyFK1asQAq3QNF+IMvH+K23 3gIoPrQwdcuCEZzPg0gBUMIiGaqZxYM8QFEXAhTV7bwusCuQApMVFgBFdQNdFECJFKYuMgMkWS5p tKYuAIrqTpY4s3gInnbaaaYuUO9QCNwCBfInn3wydbF582b6RRSNFhannnoqZqTNIjNAkuUCFhqt tCgmQaNotMlYj934P/qjPzrjjDNowH/yJ3+CFLQuViXQ5BhYeKSZEf/hD3+Y5dAPPPAAui/zYl/9 6lfRRyWBCP7lL3+ZCf4vfelL9LIvfvGLGBhok26baHQIhzEzkAuh8PZEN3MEXIyBYoTjOsYLxPwX n+9X/u0AAEAASURBVCcMDFeywwLrkTkp+iC9g0dsQpmIccUCOhQYowUWzCkwmDAzQrNheIFdMJcw FRFMQd9miYAaGFkC6JOdoRA3Oo07ZQfwyRwiatnyZSFSZZWEkjP5wS86KSgfQC1YuDBSFkybMSpF zQIukbLgq8AvUhbUNV+jSFkwoxYpCxptboBCZ4oUKJpT1C3K1AXdMKJhiha1ZMmSSIFiAIm0RQEO QGEdRSpFViN1/mWmOWEqYO1L0dAFeUSfNiU1CVAWeYviSz2atxIgCwsFMeqwMVauXOl5Oz4emdbB rIp05og+yBSb2PnAjpkBU7foMZ3K/NceJgq3bYMyliTeb2rWIRcaA20AFuLRYmbhHe94B5EOWSip iBAIVUnMx2NnyxwDzZQRgR+OKibbsMIpGZ0EP7KZhJBVlSSgqRmfNY1DlizLejjSkBEzdLw2FICK qM5ySTYHUiiLkBWqQClQIREImUxbVP4AFbIkYyUZ09gYFdgPrJmxTxQw5cf1hBbBI5PTJtIEeMVG DjTX5cuXm0jfALafb/yoRKZbmBxMiLBJGt9CpGig+OFkkP11EQ0pWCxokvzU1I+0Kp0TD2VgvPji i9/4xjcSeV944YX/5//8H8aRhx56iP3ZiQk+/elP33jjjRL/3e9+9z//8z/tNBjTDz/8MEaIHalh RUARUAQUAUVAERjTCOCFY0oRGyPRwGAbxuOPP84rBGRtfaKYr732Ggsd77zzzmRz4SiazFHyYx4z MfvYiolIKc8xCLmRIjdccgzdOGYXysC4+OKLzznnnPXr1992221ggdlNANcbezQxXonh8dprr/3Z z36GpUGfx2bA8GCZkO2PW7NmzaWXXspxtKw7fOc730mYYyV8N8ylOw0wjqtHRVMEFAFFQBFQBMYc AmJg4MHgOAS78JgE7L5Ai8DMwPZINDBQIUjAPg2W2NkZJYzt8bd/+7dkZGEFqsJf/MVfcFYBuz6Y ykxmjSQS0RhFQBHIAQKhDAx8FJgKrHuTArHq7t3vfjcDhCkfHZtOznon/HGMGqy/PPPMM81bCeAz Peuss3BysarqK1/5CqOPJ4F5VAPDQKEBRUARUAQUAUVgzCEgGkLiEikOq3jhhReYguSHnUACfigV IiBff2Yqmbj84Ac/6CsyagYGhmcmG8VDrQtfuDRSERhFBEIZGFI+NgyxfSLmlRwcZFywDQxJQA9n XMDA4Ocr0uuvv/6HP/zhX/7lXwKsC9+MISMpFdMb4nhNmYXjQdI6nsUzoqWkrwkUAUVAEVAEFIGJ iYB85Tne19YH+Eb/67/+6/ve9z627cr2SywKEhgDg8OjHnnkkVtuuSVRwRAY0THG3rJqhOzv5Naq idkSRkvqwoHOaVNi281LC/qGen22+oxWwcY/Xw45LZ4yqaAoDQMD+wEbgxOscXEyTCRi9D//8z8c ukx8otOTSLJ873vf+/jHP44TIzGvHYMBYz+GD1OwdevW+RoYxjyQLeY8sr4rpIFh8ppA+CJpSkVA EVAEFIEJhYD5UpjAhBJfhEVbIICBYWsLL730EvfhfOc73+HkKDEwbHWC8KOPPsp2Yc66jRQxSvXy yy/DLgwX9pqzQCNMSknjrfTBvsHO/ZPyaTN6eFnGdMriwe5pZYWIMLmgd6hHDYycVuZQwZHC8plp GBiUjmOjMDA44cGzqpJXv//9780e7vjFz15hUP05bIptW94XkyZxSwvjDqdJsJuct7/+9a8Zm5jS +OM//uO0zphiyuT2229PpJ9NDIMFBg/jESXxDhzZ0NW8ioAioAgoAuMRAT6RfDKQbCKv2xEDQxwU Usl4Klj+dMEFF4i+zhHAxKPlG3WCKw7++7//+2tf+5rMA0bXNNh9/qtf/UqOsfJwka+8+UuA61BC Ghgml01zsPuAWhc2IKMRHg9Heo4GblnwHBoc6j6YnoFBT+O8WkYETAKbM4PIgw8+yBYLFkExRrCk kuFV9n9LMowHTo341Kc+BQU7o4QhyM5vzBKmLj7xiU8QyeWyqPXnn39+WgZGImUnMZSEH4NRyAkP J0yViCKgCCgCisBYRICPBZ8MSu6rwo5FiTIoM/cVSC5Z10CYoyYBRD7xPMoaKmLEwCDwgx/8gFlF PAYZsEsrC4us2AiaVpYwidEQ0Hy8VuVAKD9JGPqaJi0E0Ev7+mJ2PkeOpZVREztBYGigN20DA8YM nZ5FUL/5zW+4BuULX/gCp9liYFCvjCnmqh2yPPvss/S69773vb7lZqqDQ6h8X6UbScFCjumMbiGn STBy2KHOzztwpFs4Ta8IKAKKgCIw3hHA/c73Ain5FI53WZPKx2edL6zRFrjb+5lnnuFDbzZXiIsD pVzWULFmCcfCZz7zmaQUR74AW1wifJTZ4T3yzag97Y//qH2dixy1OrAYMw/Olp5JQ6yR8VnSbyXU YFQIpGdgyJwEGrztwaCT/8M//MMVV1zR0NBgLuzEHWEMDDodI8t1110nA0pUokyaBCM2h/le3JPI 9LOf/Sz3eCTGe2IYxfhJpAl40iQ+xjPFbvIWn2liguxjhEVhIb6/CN1/mP5MxkXnR8oZUDmpi9hy z4h+w9UdOYtIgQIcFI5IWShQIVvg+AAqBwNIBo2WUoWshXGcjM89Hw5m9OWLzOaKBQsWvOtd7zIi iz4AVngw0ChIwGmzcl+bSeMbYBkFp7mwzIGMLJdgWQTHWpLXXjThm9FEwhS9JeRcJOUMObc4Z84c tphiYET3xTQiaCAlAmXlZbW1sc0zZWXeq+JT5tUEThBIz8BgDwZc6ZZ0TsOezRXMQHzoQx+ie5u5 hAMHDpgEnBpBxzOn3Jp45wG5uc85WSEY/ptx5OiRPbv3gBLzN4x9tbUz3doAlIQhmwkhyBYVFbJV ffr06oIC96onYzcb9xnHP/e5z0WBKlbo7t27EQegkIJd926BosxMYMiVsbBgHS0thIBbWWjqwoKP EEY1XJxblfgDcQzSoqAMC26PiYIFUnBAAlJw/gFShPymhgeTRtvS0iL9goqAhfO6AKjW1lZsGCjD IgqgTHWjQ9BoWWvhvC4AChUKKUxdRAQUdSEsopCC3k1dmOqOoi4AChZRN1q6nqx7iajRhu9BYyul WSNANTG2sLnib/7mb2TfhQgiS6TEg/Hcc8/RyFkUHdybsCjQNx577DEOqz333HPnzp1Ll//FL37x 9a9/nY8Ul2sFZzcA8vX85je/SbWamIDAlVdeCa+ABOZVeCXBZJEAIERpk1AuLF6wiW4ukqnIIaY6 o9BDDFay0ikNFmrmG+xGKZCegWF2UBj7AY3hJz/5CZfdyOSBDBnIYhIwsvzbv/3bvffea5wboyRp jti2tbW1tm6fN29udVVVd09PS3MLw+vChQsd9m1Wo6GXc50I5lx3V1dTczPfciaHXEmI9fj008/w 5W5s3Lp79x7OE3RF2aYTA6qlldsYp06tBKhtjdtY3oYTzCFQoIR1tHz5cpolxhKnl9Asly1bZhcj yzDnFsCCukDX5HO1YcMGTJqlS5c61Ajxy7FDSa6tRApOkae6uYLKIQta1LZt22hCmBaw4A5dPtsA 5bYuYAFQTFLwNaUuuCeLyzqzxN/OvnPnToYjYWGAoi4cSiFAAT6+XGZVqBcaLY8OWUijBXxWkiDF 5s2bpS5sSbMMS6NlUKK60c7ffPNNWpTzRkvJpUXRL6JotLQfWEjXQwpYOG+0dD1YMChh1WMpAVTI RpuxopllzeZVdjPxz1D/3e9+9/3vf/9JJ51kl1C0BbACZzZ/33DDDfX19XYCTxjrgot6f/nLX/71 X/81nhAZAPEYMMLz6u///u9p0qeeeqonl+8jue677z7fV9lHplX7DIa/ffW3Gze8+ad/+qkobIyu 7q79+/aJhVxZOZXZyMLC2MmtDn8Qp1/gg2K2oqy8nPlO58oe+1/37tuHZkKls/hwRs2MyVOOXZzi UBAl5RyB9AwMPqt8Suk/Yj8wx/bkk0+iWplr9YyBEVv6Fj+alhGBo2n5enmK/uMf/xjbg4ZIi4EO gcsvv1w8JKRkNxhzeLRX2MEU9wjfQg+FPHxksECoqqppzHVRvMqSkhPmnoACigiuyo/OgSJVW1cr K1nBjalBjAFqwaxtzRIZ6peVVyedtPzUU0+5//4HsqTmmx3lDKCqp1dPr4n5xABq7rx527Y1Tp02 jeHJN0u6kbBAo2VqVpDHAEYXQUUAKLN4L12anvSoNaiYeOfEa0FdLF68GBZUN3w9iTN7pLphQZll jhkWfEQxY/h4843MjKYnF1LQ14QFr2CBMYO6xjcDuTyJM3ukg1MXsKCtQoG6gAVAoUw7BArrAhZS ZpECFvREIjMrticXnzeAojlJ7zYs4l7KWk/izB6l0UJfWBigEMFto0UKcSkgBQYSh3M4b7TAIo0W KaJotLQoYQHUAhSNlplpV/2CRot1AfJ0cNFlpdEy2AbrwZlV/fjLRYviC45c1As27Y9+9CMamy2m 0RYeeughXATvec977LeeMPrrX/3VX7366qvf/va3jb4haVASLrnkEnwg2Az/9E//5NDa95TB4SMD +//+7/9u3dq4YcMbLS2tLAzDgnVuYDBkbdmylQUODQ3zac9bt2zdv79t2bITHUKEdcEXilVh0ino HW379y9ZvKR0cqkruFCrGhsb+/r6lyxeXFhUSB9/c9ObjFqyxM4VF6UTBQLpLRfh7GoZMpjxojTM dHI6LQ5E0zFMlYuB8dvf/pZv/Ec+8pHEBs2ojULMoMPlGMxw0DohbiTky4QWhfVCY2KIt/2qJk0e BtrbD9DfzLhJCQmjr6OriXcv+zIz2cNIMbXy+D3owMj3b/++/dkTFwpU4kc/+lEuUmXqzhVND51D hw8DFF8gE49VFjNc29tdAUWjYmAyJiuMAIqGCoAwMnyzCUCKrwJkTfM2LLIha+elH2EAIIVhQXeg DyKdnSybMJMFfO3sukBvAyi0f1dAwYK6sFkAFBNROARcsUisbroeQMEiG3DsvAKU3aIAivLD2q0U 9mSEaVGuWEijhYVpUXIjAfG2sNmEpdFiIxkWETVaGyhptA57t2m0Yl0AiDRahyyyATn/8/IdMYoB e7vtz6IUnhhpIYyi11xzjWyLTyYXjgtOsceQWL16dWIaGhg6LtotXqzEtzmLSauT7tixk+XTl3zo Q3zQoyghhcEInzy5FJMb0w58GuY3YHKw5M8hOzBnbJ/f0ICRzw8jnMeW1haHLHbv2tXZ0bl40SK8 FgjCujiaCqKlhbbD8iip8Aik58HgUySjLYMvgwJLIXF6shrS8DMGBhYIAzEJPv3pT9ufZJOS6QrW 3jBqYJt+7GMf86zD4Z6df//3f+e+T17RMUyuPA8cPBjbeWLP06DigFhnZ9dA/0BxSXrmnK+wDKAM ysXFx72csCCGjR/UCGHfXPkWeTj+GWA8MgWLS1HIcjJXQImJm8CigJYJUObLZwqQQYBeQC6bFHXN I4O4q7oQKewWJRxpBgyvRoHLoPAmCye3ELZZSKNl8Q+fCjveZEk3gIFNFhsoWPBjYQN2ppMOzmgD GnZppS5E+3cClFS3XVpEgLJoojbrdPEx6QMarSugRAq7tKbRYmfa0plSpRsIaLToUrBLl2Bi+lFs tEylu6qLRLnGU4wxMFA6ma5KFI0lvozPtLqLL76YCenEBCYGp/3DDz+MQYKBYQ8jJgGDIUMuf2kY mJomPm8DdLSrrrqS4rlV9215uzq7aKusoTCIob+Vl5UdOHBw3ry5aexksImODHMILN9TZo6Kh31T 8KLeDx485GowoVoPHDxYCl7Da6IYdTH1aRIdHZ2VlbGrVPSXtwikN9bTPqldhOErhXcC1xjeCYkR Cc0sBc2O/cEoEL4jiyRmcBG/P6tlPAD97ne/Q2PgwGwnHzwP8egee3tim8Y82gz4cItGX7hLQ4PL xueZ+WyGhoKC44ZEnF0BZz31942Z87YFKI+qgS2GRusEKGBECeCvzQKg+MGCMSsY5zBvh+uiwMOC R+gzvIYhEpwGOnwhSGOz4JEWxafUCQuoYQ4JTf6aHyzgTmMzMdkEGAfIbg8UVIQAJdWUDXHyUlTo CE1Dike3QLF+ySOFPMYarSOgBA2jEEAfKfjBwglQNJv4AOLfaKWxGQAzDgQ0WsEwY8omY0CjdQIU jKJutEaW8RpA0aTpohKwxsGc/mILyyt0ADxdzELa8YlhjrlnLQN7LZJtNURbYBAgo913EulMqJiD h2LL1IuLjk8iM+QWlxQzXGEjO4Hi8OHYtKY93QnZkpJiFB7mp5ywONrR0dvbVxyfyjEEZX7kyJHY Ohr95TMC6RkYDBPSgenP999//4c//GEWndviGQMDBxZTDn/+53+OTWInsMMYD7ImG2PUjoc4eW++ +eYxN1j0x6/U8cwNMMgiHSsIbRkzC6PUJq7uiZOPXSWD4pwZ2dznGgbKc6iFM6BQpNDJkEvAFwEl DIbyKcpSaogIHZuF0HTFwlcKI5QrRSoRKMPCid7sKwWg8TOvsqwL6IC5KbahBgvCroDyrW5hESlQ iIB0Uk1GtMwCAOUrhbBwIgWkAlqUK6s4ahamZUr9CtqE+ZlXmVXBxMnF952vP6fHsr3bV2osEBQG tIiU22aYzaTGIZVsGRX7r7Bd0U/sVXO+TCdOpBjznskpGjCDSW+vg/kvkBQjvHDkugnZRN7ZFZtU yv7X093N6M7WC5uUCCXc7XgN5xsCx63bMCWjA8s0JDYAmykTvRPGwOCUGBY4nXbaaQFksR9kR6Zt YDB8P/300xwv41k0FUAnN6/omSkZ4UYgjW9COknK7CkTAA53N/kmi71wwcKXuPNIAQqoEik7kULU TYj71pp5m8g9fAx1AR1f+hCJ11R4Yv4pISLqoO9rJ1JAOWoWwVA4kULqwhclIp2wCKbjhIUByrdR uWIBHV/6wQImw9Y3PqBFBbzyJZUsMoCOK6CSsSY+gHtArgn46pFHHmGyOVl748ZuTnOxV+v5QsQE Abde8IqF074JiGR7Bt2H2zDGynbNZII4jJf5Ag/48ujKzu/rT+rijnS2QlByNXPkEHMl5UFghF3o eZf4aJZI8Qq3JivhPGlIIOMFlgYJfB2jJgttXZZI4f00kZy59Otf/5pZDY/lbRLkPmC6qAkkK8Ow M8Hnfcq8PnkSouL6eIJSHrc4/F8lUMggYiiJSZMBKZNl2AbzMZacAGUYGaXNxBBwwgIi/Hzp27yy DMMiGYWAV8my+MYH0Al45UvKNzKYSPBbX4LpRrpiEUAn4FX40hoivo3KvA1P0DcldHzpk9ghC1/W Y4hFllCY7CaQDJDxHc8KqGAEUloX4IMeyY/pSPuUCBu3xsbG9evXs9XzoosusuNHKxwscs5KdcwM HvkFkc+JEyMcQZhj4y+3X4wQKv40GJ9sHRGf0YMvC+E3PE2ZEV3NlBME0jMwjAcD14Tv1TN4RcWo SLltS6QTA4Mdq2b5L5eCs6+DM6ZyIn5qJgwWfI8xx/mlHDiG1zuO0JtZ1gSbIheHTw/7IkfQH35g XfXIfp5auHAphhmESx0qlQDlUXTk0QlQ+NkSK0voE29vBghVXL9EGMDCwiMFaYl3Yh5DR1YJeljI o6sFhAEsnABl0LClICyP0bGgIoSFK6CkqELTtAi3UiS2KKFPvCugElmILMS7AiqgRYVRKA22AYEA Fk6kAA3pwnZ1E5bHlFKg2/G9YFWPoB0giL5KiQCzlkxl0v6Tgfn4449zrsB73/veUV/1ICVMVs4A SaOZxYvrAyMXVcvHPIMSJil8jIW38HEerlgIHQ+L9KWIRjVKAopGGwTSMzDEQYEX8otf/KKvCsUS SRZWYjaw+yLMQC+n4LNjT7YEcagUJsp5551nypcPAdYycroRv5SOxZLS2DnfYlGYkrOUhk4ir0xk ZgEwZ0NVbB+GtRoqvqYodhl2GMAz4+s8V9RAUWAmz/hrT9WIfsCHyglQ0KGtCk2DjzxSFylVEJMl IAALWXMsao1JKULB3cRkE6BTk90GSh5jjXb4bJBs6JNXpLBZiEQOgaKoiXXhFqhEKZwDFdBondQF gAc0WlctKqDRSmPLsjmRPaDRupUis0bL50w+GbqYKvu6ptEuWLCAj68cHeYhyDn43OTN/XqchOuk j3jop/toN5g08kYyi3f8JBhTEhl4+bKYmGwCAXQ4siUbyiYvMyuER1pJx14GcDfZcYDwjyZkYjSQ SwTSwx2dg50Vn//855Ndh4xpwcaML33pSyGXQsp9UgzHKPH85fa9m266Sb6yuUQhmBcfM47ZxgOb UjEVqTmlwRBkuOEbwzcvTGcwuQICMRZ4VKwzqQZjp/MNAVpp3LwJyJvZKxmSMsubLFdZeRmvcHyb BADFD+lcASXbgZhHNCyoCGQxXjgTn3FguLqPb9+HPtOW8bo4fgJvxvTpbqJI2VJAjUd0uJStMSRf WXhgswAofrBw9cGWgyPt6hYW0HeiDgIUc5yAb0shjwDoCiipC3uWASlotLQ0V0BJo/WwkEbrBCia hDRauy5Mow1e0RqyOZEsoNG6AirPGy0IyCdDNZvwzSYgJYsaaP/ci0VbtZNxE+Ktt946d+7cv/zL v0y5U9zOGFGYqVIMS47Q9JQzInbBZDnalQQMUHYyKZhMAdjxmYWPTSXE12gYCvEZz0kcK2tisgmU xvfwCE1DR4QqK0vBgm9xR2fH0c4OjtM1eTWQSwTSMzAo2Y033sjpscm+2bgv8F1ccMEFIWVA82Dg 4DQAuuWjjz76rne9K38WRxkRROWlZ6JPmEjfADfY41rv6o4dZyk/0RX4HLrSm2umx64ltlmgK1Aw ruyJLYaM5DdiTHfCobqqGqA6O4+fZCdATa+udgWU3PNlHzQBC0Ycmhz6qBMp0CGgIydmCkHqAh3X vsUsS0awABA5D0RIsZiQ6nZ4WAqWM4DYQMFCpHBVFwKILQV1wY86csWC4yJQ5mwWhAFKzqnLshYk O3UBUIxUhpos7EQ6V1JIo7VZgJI0WlcspNGKx1gEiajR2lLkrNFiobkCyjRaoyzG22wP/SUlC/Ol YK7BtBYNZIwAK64vvfTS//iP//jBD35ALUCH1vXDH/6QlRTc6s3t3cHXaGTMN92M6D9862mE6X9i PIuA0uXsk57CEMvoYd6JDsP65NISN80ykUWcY0xNkleGdcYBLu6gu9GhTDeElPQvmSsJoMx0Ruyg skrGhPROMwqgqa/SQiBt3FOOrWnN2ZAYzYC7ZjgFYt26dd/5znfSKn1uEsusG0az2M0BTCsqK6qr q9rb2mtn1mI40SV27drJN8ahljOtirnaqdzbPXvWLHR0irR7z254OdQ4KXZj4zZ0gldf/S3C7tmz 96WXXsIkqK2rFe0kAIGQr0CJf21t7bPqZnFHDxxpA0hRM2NGSAopkzHK40/j+EIaGPYwLJqbm6lK prtS5g2ZABYQ54gCGMlczrbYzamTZeFfSCLByUQKLtU2LJqamphpdjhdxzBN4+EuPIqNFADF5aw0 WocsKDDl37t377x580TlkoPj6uvrg8UP/xagMB0BCimoZaSgXuDrsOuBkgBFExIWAAVitIHw5QxO aaoboKTRAlREjVaqm/JI14ui0TJbJP0iikYL7IDjabTc6RaMcPi3NB5Y0GhpQqLKSKMNw8JoddRd eI6aMhkC6OvXX3/90qVLn3/+eU6UwkFEB6czfvOb3ySSmkqWMcfxlCozjpnmC+LGeMiGRr7j8UXa salkmUqoqCgvneymWdK7p0ye0t3dY1iAQFdX95QpZSm1/6CiW+/KYncDlvf0MK3WX1x8rNhMsNIk qquqrIQ+Qas6MqwXH6IalQ4CaRsY6RAPlRYlg+8059kxWCQeSxWKRMSJzEcipYFBQVgtygebKwgr ysuZzaYbLF2yRJQqJ8WE4KJFi9GbN216Gxdhd1d3SWnxiUuXpjT8wnNneuCJJ57o7estLChkfogF kM8//18s9WIXHb/wdAJTFgAUUry9eTOfh9jKloKCE0880UAdmDfUS4BiWgsfOgeMMJvCOAtEp5xy ilsW3ANDXXMoM988ptZgytmLolSFKmWIRLBAio0bNyIFXwga4UknneSWBR9pTCNhwcQ/QCEFllKI 0oVNQl00Nja+8cYbSGGAclgXlAOgUGRff/111HSRAqDcsgAopKBFwUKk4EBtt70boGhRSMEXmkbL FEwUjRa4WLwOC1oU4SgaLV93WhQs6N1RNFqAorsJC6SIqOtB9u233xYpkAigwlR3bECLH/bgtvlB c8L+GDcuu+yy888/n/3c4tML40oaFbgsvTaA/9Cu3Xu45OGtt94i0eHDhza+uRGTqbKismZGzDGe /Y8Zipm1M9va2jo7u2jAEOR+ZHrKwoULXS12gEXdrJidz9XdVAcsOLCnq6uTmYW0JpqDhcXC37p1 S2w6ctYsUjK2HzxwEGevuT48OLu+HUUE8sLAQP4LL7ww4JTrUQQI1qJp8Y3kl7IkKGcNDQ3o/Sjo dD9sALl3JmXG8AkgyxjBmhbG2ZK6EnRNh52ZYkD/1ltvCV+ezFICFDYGgwVDHp9hFHS+5ZmRSpYL FmwWYgkT33th4RYo+IKVYUGYcdw5CwhielHdABURC8iinZu6wOQDumSoZhYPQTRCkYIw6oJzoOin ACXVHR1QSBE1UDmQwrCgLmi0zqubysUYi7S6hQXVLf2CAcS5FBDMrNGKB4MG6XxMy6z3jY9cgMms PL9xIA57j//hgQfYrcEyBL6DSPSDH/wLU3tnnPHOT33qU64EnDd3LkpLY+M29pDSJmGH/42x1xV9 6NTV1fJ53d7aKksiYYF14dAdCgsWOwAR87ZMuDCwY11gidXXz3MohZKKCIHRNzDQbDj09pOf/GTe jsV8ukCfjirzUilrAkHYxFw2KUK/LSxkTiJlYfI5AVKArcAbUTlh4XY8TSxnblhQ3ZHWOOpa1Cyk 0UYqRQ7qIjdAaaNN7Gi+MfnZaLGsKG3+LN3xhU4jnSMQzncRY8sw8rWvfc15ATwEOWsZiwK9HHcr HFl46dwIxxnCklH8Cb09vaxDwrqIwmuHRcHKcARhLRmrbcN4ET1Q6OOoIDD6Bgb3aZxxxhk0/VGR PwxT0YAZO+SzESaLplEEFAFFQBGYmAiwUATBZdHIxERApc4fBFjjwC/S8uCs4xcpixLOtne0Nz3S cipxG4HRNzCwTfnZZcq3sJlztc8LyrdCankUAUVAEVAERh0BFrqwiI5iuDoSY9Ql0gIoAoqAIpAB AmkfU5sBj7GexWw9Vw/GWK9KLb8ioAgoApEiwGdClsqogREpzkpcEVAE8hwBNTBSVxAnW7N+kXT2 4fGps2kKRUARUAQUgQmGgJmHynPP/ASrFhVXEVAEco2AGhipEceDIfuWZGVt6gyaQhFQBBQBRWBC IsBhoCK3GhgTsv45cd3xcYgTE0aVehwgoAZG6krEwJA9UhwpHf6YiNR0NYUioAgoAorA+EKAixpE II67GV+SqTQpEFDTIgVAo/ZaTb7RgV4NjNS4Y2DICQlyxU/qDJpCEVAEFAFFYEIi0N7eLnJzZOeE BGDiCi0Ghiyonrgo5I3kxt4rLFQDY3RqRQ2M1LhjXcjFMXgwOIk5dQZNoQgoAoqAIjAhEdi3bx9y o2XKxcMTEoMJKjTnTLKOmivndKVDPrQAznM72sHO2SNy8WU+FGmilUENjFA1zq20pOPLwa2xoTJo IkVAEVAEFIGJh8DevXsRmqNBIr1CdOLhOgYk5n5M6p2/Zu58DBR6/BaRWwUrK6iNqYWFReNXyryW bPTvwchreIYLt2zZMoIYGExOpNy6NzQ02NHRyd2ZbA3nRu/ikpJhMs7+D4sjR45ys3hpaUnZlKhY HD3agUHF/TYV5eVFRVE1FQRpaWlesWJFFIMy968fPnwYKbj7k5GmuNi9FExWwYLqxtOFShHFPaZI QcODBVJwK0sULEQKHHQCVHQsRAqAiuI21twAxSSlVHekdREpUFQ3LUqqO2opaEt0vUhbVKQsDFAh G+2OHTsY6HFfmAuUnI37Sii/ERDHhfzN75JOiNJZFcEl4/obBQTc61ujIET0LOvr62FCe921a9eC BQsCGPb39bW0tvb09E6dNvVAe3v/wACXlPN9DciS7ivUgu3btw8MDFZUlh88cCDGYu68isqKdOkE pEcjb21txYCZOnUqO0929vU1NDS4nZAb6O/fs3fvG2+88Z//+Tw2zDe+/vWA8mT2ChVt69atuEfZ RbN//36EWrRoEeHMqPnmAqLNmzdDubq6mtNjuGBr4cKFTGL5Js4sEj/v22+/TaVzMTCL9OAFC7dS QLOxsREWIkVLS8v8+fMJZ1Zg31zURXNzM+5qLgegRXGUJ30qpa3uSypZJCyQgiqABevgeYwCKOoC M0ZaFE1ryZIlbvVIkUKqGykIRNFo6RcARYuiLqiUxYsXR9doEYEW5bwu6Hrbtm2jIVFyGnAUjZau B1DgA1BYlSyASdloKZXswWDAjMKETtb4NT4fELA02rDFYTCh9dJUot65ceDgwelOh/SREg719vSx 1SGK6VTDqLenhyO6tFsZQPI/oAZGqDpavnw5/Z+xACXpXe96V7I8jC/NLc3d3T14POKT5UOtrdu3 bNl60kknTZ5cmixXWvGwaNrWRElWrDxp0qQCHrc1bdu8ZfPJK08uLnFTmxBvamriS0mxZdTjQ44a zaOrvs0qgv/7zf+LhwQzDEiXLY85iNz+kGLLli1oUe94xzvEN4II2DM8ypb97NkBPuomGuGqVatk ghaOb775Jo+ulE6keOutt1DRDAs0no0bNzpkgRQUGxannHKKeHhQ0zdt2nT66ae7AgqoYYEuCPjS ooTFO9/5TlcsAIq6oNGeeuqpeMaFI0A5lAJ1EynQBgwLpHj99dcdspBGC1AgI40WFlE0Wlicdtpp 0mjpgFE0WvRykBEWUTRaaGIdAZTdaGnDDudB6HoYFdCUViqNdvXq1WVlZclGDwYcZgF4SyNJlkbj xzcCjKhhBGQ8YaKQbkKL5SNSUlw854QTHLbeeBlwVB5lVvRXv/rVc88998gPHykqdr5YaKitrX3P 7j2lcQ2H4ZENqzU101FOwoAQMg3m/Y4dO4uLGdqLGORnzKiZMWNm1CZZyLJpsgAEdA9GADjHXzGR L04MPsbHYxNCnR2dBw8eYgJVvnn0sZkzZ/b39+3evSshbYYRsRnHjqN1s+qkA6OF1NXW4c3Ytz+2 s9DJj87Mmh9KbjowQwa9uq099uF08mPq+o41d3zn/vs+85lP475wQtNDBH8CP2rNrLyaO3cu4d27 d3tSZvzI2glmK0844QTRoqADCz4bDlnw+UFfYbmFYcHMKHXBNyPjYnsywgIpOFJzuNFOoroBCheW J2XGj2hd1AVk7RYFO4csqAu6BkCJdUFROcMHoGRBfMYltzNCH6BsFtIvHNZFvM0eoOSm0UpdOGxR 0mhhYVoUEkXRaOkLhkUUjRafJF3P02hleZJdaxmHabRUN/gbG5gw7JhtCaDJ4EnBqD7mpAKS6asJ j8AQbfXI4SML478TTzwRo4SmhYLuEJl77/3OV7/61XvuueeXzz7X29vnkLIhdfDAwZaW1poZNfhy cYTyHWFUP9B+7CoYkyybALMh2PaVlRVshYUF3bC1dcfefbGdTvrLcwTUwAhbQQwBJGUKMyADWj6z F3JphiQjzMcGq4NPeEDG8K9i02MFBeabR0bxrra3tYecOEnJi89qQUGhLQWfVfS2tv0YGKHmZlKy gBoLG1z5Q3zZoasBiD3XCGgouADoCig5LsYGCo2Kn7D2LVW6kdQFWWwphIXMkqZLzTc9LDxAwQ6g iHd1+AZAwcJutIQRBClc9QsMCcqcCNSePXt8pc4gUlqO7ZuCHb2b6nYFlLQce0WlabSugJJGa0tB 7x6jjdae7pVG67AukjVaLLSAuvjDH/7AW+aYWBmbQRvTLGMagfBfFrz3GKIza2dIG2bswgjv6u52 OJUAkp/4xOV33nnn3/zNX0e0OAqP665du+l6s2fPgh2DYW3tTD6I23fuHBoadFSVMUuMIYrZBOhD E81hek31ju07XY26jsqpZHwQUAPDBxTfqLPOOot42jr2tG8CIuWVmUMlhi7BCQZ9/X1OOgPjF/Nq dDLpaVIMwvwcsujt6aUjJ0hRyCoa5Egme77F+9YFQOGMdlUXsIAg3wYjO4/gxsS5k4koqhv3AsTN NK0w4hH6cDF8swngrSK7p7p5hEWAIpUWR6Z1SW9LIUBREVRHWqR8EwtQ0PSw4JH+4gooVsskSkHt E+8KKGm0HimQC5ScAEX5pdEmVjcoCXdfhNOKDGi0TvoFhcnbRrt+/XqKxz49t/uL0sJfE+c/Akzf oJ1PmXJ8rR3TbcVFxdKwXZUf52R8b5XL1Up22RhdO7s6p0yZjEoyHB+b/WSw6uzsGo7J6v9sZ2U8 YQbEjIqMhxxsgwHjFqusSqmZkyBwXDdKkkCjjyHA1guGAJZJBMyJiipDB7BRi13yMjTE5m87MrPw 4CCazCA6Lf8MBWFHvBNFilGvf6AfmkLWw6VvjBgYaJyikNnav4RRap2ogxBJrG7BLYZhvwNLzJcF NSKCuNLVsBsNTbu6wTA6FiICQDlptL5AURdugRI0hGZEQEmjtbV/YUdzcgKUAdzu3abRuqruxH5h GhgaiYEum0BAo42ORZhGy7YN5Dr55JOzkU7zjm8EGFpltsLu6YhcVFTICOCkp+cGQLx5yOKVgm0e 8UPqnJQBoBi1iopG6FRibMhEhhMuSiQiBI7rqRExGDdkOUiE9X98g2WOylcutHzi7Y+3eeyPv/LN FT5yaNBnhdKwOUNPd+CUjFEZFDrHu7SRCNsjfGnTSOkjVhq5E5MiBT/iTcntNE4MDFBKxoJ4Jyyg E2CoBLyyhU0ZFjqJejMZXUkhQHlKIlXjRAro+7IQjk5YGDRyAJRvo3UihQEqkQWvnLAAqAA6rpQn YZGDuvBttMmkQBliuzxZzj//fE9GfVQEDAJ8O3wbMO2ZbtjT7cCpa3hFGhA7nwlPm0thfPazp8fN lcS9fb1gUlgwYm+6DF/C3WbtMHznXffddse3Okb6YR5+7Ikbb71zX1ts3bL5Pf7UM0Q2t8QOp078 rV330uWfufnpZ9cmvpoIMWpghK1l3BdnnnkmqdeuXTusgnvz0hO8UU6f49T9WRDrinkAnYBXWQvq L1dmZIMrIvhteI6u6ITnaKd0xV3o+FLzjbTLECbshEhKRgFcAl6lJGsnEDq+1Hwj7bxhwk6IpGSU Gy7JiuGKu9DxpeYbmaw8yeIzI8LZa3wa2IS6cuXKZJQ1XhGgdTE9FcdhhGpeEF9oFNUsnqsNlFb9 iSIkxT4eHZeJhQLHY7IIsaAilnsETsfIRbqeYn793KaW7c0t2+2yv/La+r372155NbYM0vw2btpC 5MyZNSbGDtSNjMdiwXR5/Mln7DRphde++BIUktkzaZHKQWI1MNIA+eyzz2YD04YNG/AM+mbDxUm8 RwuXb1VsoVTWP5ZGJXY1YRd7Ya2byoaVcYkkEhEBE+OzjnFpXVAYM7XpqyiYt9kUm0mUADqJM8QZ 8IKIx/tsEwngbidLGRYWHqCGG62D8SG4nMFvUxZeEgAUv2SJnbCAuNDJAVAeFiKXEylAKYBOwKtk 2PrGBzTagFe+pJJF5mej/fWvf02B3/ve9wa0xmQSafw4QEDqPUzt+/ZxQWDY9ogCD8ff2aBFE45Y DdthI4Z3eQji7gVvRHbvS7/nBQ1ziW6y/BLYG/v2x3wXr/xuhIGxYdNmEleUH99OY9NbsXzpj//5 3os+cJ5EsjWF9B4fiJ0+ZXjfvnYoHE2+EzglhVwmcKBA5LK4o8uLA9E56oHtWcmO15SlgZ6xI/ZY UMAepewLX1iEDkCV0em83ZdBzYn2jw0T+3gPYSYdZ0FYnsxGq+xlsSlYrOzozMOgkVgXIgMAOpEC Ism0nHhdjHDpZiYJdNjcRt5j6A9TkXkjV2dwBdAR7sNsM/+/0LH9frEGNTTkEKjE6qa4UQBFsROB cAVUohTCjkbrRDU3gHukkLpwJYXQ8bCQurBPEkuEMXxMQKN1xSKg0foOICzY+K//+i9EOOecc8IL oinHEwL0L8SRv8Fy+c6H8NUll5O5SH/ufmOXf8pwsTKnKcU+nkMGyLRV+uME7JCA6WEhHOK6kJ02 adgX7aSp4y/OWH0K/9+waYtJhqeC8MplS22rA11fIiUZjzg9Ojo7n35urSyLIhyzKGIncEJts6SX SI97hFd4NsgoiYWg/fcYqfg1O3gwSG+v4OKRvFDAzWLnGt2wA613dAXIJXe+K+973/u4z4tVUr7b +Pi28ZlhK7ZVqthyfFRR32+SlSxUkP5cWlrS29tjL47ny40pj/7hRD9AkeHjzdENNgt0BVjQn+Ee qqB5kIjj/zzH+8SBGkI6V7oax/NxkIVoTiIx4Vh1FxcHKEDhsaGccgauXRdk55G6cKVIcSgqu+U8 6+Z5RAonLYoCc42MhwVA8UNAJ1IIHbqeDRT0eXTFAimoC85Z8gAFC+raFVDSaO0l/khB34OFE6D4 Wgc0Wl6Fb5wBKQMarSsWAY3WSdfLoNGyvZsbUThM0/fTEACXvho3CLBFkyEipeIbm8XjbMlJfZ45 eHo6UJSWuLmTNweoHtNqRN8fyY9jn0ZGZPgkLAQZD4mUoy7wyj4N3+weap5HPBK1M2s2vhWzH+T3 yu9eJ+bcc87c8NDmjZs245ogXpYqrVi+hDD+DRYvrVy+ZO/+dsIkxnGBNULkxy/94Mcv+yCBY6Re e/2V114/75yzrr/2qnjGtrvve9DYLQ8/+gSJySKJzV+yPPDQI/LIbhAC9397jXhOeLS3ecD6tpuu m98wz+QdrYB6MNJDHgODz+dTTz3le1ZJdTUXWE6yD2NhmSCNG70BCyM9TklST5s2jTe2CsIBMygh 6HDoUkkypRcdZzHE5iqTDS2KX3V1tcxYmHiHAddzK5MoLcUbWRcANcgNAE6BmiQn/wgU1DVAURfH Rt6sAaIuUAptFohA7Ut81uRjBOQ8TZsFoLmVgmsBYOSpC6SgLlypg1zwBPi2FIgAC6QLM6EYBsn4 gY+T7I6PRDDl+AdX1S2N1maBCNJoXQElA4jNwjTalB/sMCiRJqDRptS9QrJI1mhpUa7qIlmjZTD3 rYtXX30V4/MTn/iEIBxSEE02zhCghcukQIBcrFOU62VJaScbjB0RWTCljFNfI/n5nRGTFSOZLxg5 o4rfODbBWlZWnhXp4cxMraB1JAAlLFJOiIA0iBZ4sg/TTvF/nBW4CGQ5EwFcBCuXLxXPBrq+ZN4Q t0DE2DgWs2nLgoZ5t9x03a03XedhwFqp++9eQySmBWGxLnhcc9d9GCR33H6TJDhj9SocEdgwnuyS SwwPSVw7I/ZhZZc51sXFF5z3z/ffBQXhC03bv+EhlbNHNTDSg5rrk97xjndwnRMXKiXmrK6OXR4X P/X/mFF/6NBBugfajysthwu20Y/lYgEpAFf24ECsq+Nubzc/tJyS0lIuGUXzEIrxA6djt5K7YWBR YXqWBUApR2QrR9gg+gGTvnKtmOSRS7g4GjwsiVTpYIFOJieaS1q5OY7qTpU17HuRguo2dYFEtCU2 koYlkSod5hBAAY5JiERo57Bw1WhRzWHBEc+GBbuY0M657NkVC4CSrmeAQiKIO+wXtH+ksK84BCj0 CYe9W6rbZgFQqK00WodA0WhpReajS+uiup03Wq4+NHURUaPlqjLToqTRUkeugDKN1vispNH69gsc dL/85S+xbS688EJTJA1MNAQYgph/ZJQIboS8FSuU26ssiGLHuJGd5ZBWZF4HZSqhv89essHsJ8fc F/JZcVL0iory4pJikLG9PX3x0ywrU7HgeKu46zd2wW4GhRG/xIY3Y4o+1gV/zzh9FR4DfBRNrduF IG4HzwYMvAeo+GeuPgUzIwxTjpnCuvj0lR/BeiE92a+/9moCv3j2hTDZWU+FNUKua678iHgzMIFu uPYqrAvbpxGGVBRpMsE9inKMFZqMIJ/5zGf4kPA5SSwzX+76+nr28WzfvoPD5rjAe9euPdxtyRxn YuLMYigALNDV+LgyX9ve3sbdn7NnzbZv/82MssnFGHfCnBMOHzqMWsBhc+gfO3bsrKurdTVkwAg/ 8j/+4z/eeefX7r3nXs72bWxsvO2227/xjb9m7ZkpRpYBgFq6dClqx/bt2/GTorSxcwaN1uEFWLDg fncoc/0iLKiUbdu2sUbCoSVGi4IFunJzczMsqAsCc+bMccti0aJFaEhNTU2wgBdAoUU5ZIFJTF2g n4kUANXS0gJ9h3UBUAsWLKDwUt3UO5UCUOITyLItSXb6BUBRBRQeoGAhLcpt7wYoWJhGC68oGi1T BsKCuti6dWsUjZY+TnXjKomo0S5btow1kDRaWCCFNFoEcVLXEKHR0vVotFCmukFMGq3vDMXLL78M jJxO67BJuxJE6eQMAWNRp+RIOykqKu7sOH5p76FDh5lsczjq2mWQPQy2jm6/zTjMFxCtoKOr01zz hSVAx6+oLJ882c1CLwZ2riHvZtF27zFjDClQTmBdFV/NEVR4v7VbQelHvhONf+Nbsa0Xv41v7Bb3 xYplS9mbEXNu7G/DNvAYEnUzZ4wkk+JJtozPnFHDUVTyj70WGAzGhgnOvy2+DR33hZ0MjwrGBudN 2ZGjEnazbmdUij5aTFliy7dt3bp1bMbgC+QpBtoGExio5tuattE3GhrqZWWFJ1k2j4xB3AC6d+9u bAy62cKFCx2q/lKwGTNqmIjZu2/fgQPtLMVfumRxRWVl8KxMWhIx+lxwwQUMRoYmQzNTqg7nmykP dbFq1aqdO3eiC2IToro51AVFXqac2frP2msaAxrJihUrZFInLTSCE9N+IMt8MIu8aVFcxgJTg1tw 3pBv4y1qikgBUMIiZN6QyUAeoKgLAYpG67wumIPHBsDehgVAUd1AFwVQIoWpi5AIhEwmjdbUBUBR 3SHzhkwGwdNOO83UBYMYZphboECecZK62Lx5M/0iikYLi1NPPRUz0mYREoGQyYCFRistiknQgEb7 /PPPs1zkc5/7XEjKmmwcIxDGzODDPWv2rD279zBkMQXZ1dWNwc+3w62B+uyzz+7auWvPvr0H2mMO 6r+/555ZdXWzZ8+58ML/xwn+DBpzT5jbuK2RbsjiDra3IwUzhgsX1jtcTc1s1+HDRzDvFy1aSDfc vXsPBj+d0e2QlQgIWj5quvgusCjE3iAZHozHn5rEXury+MlRuDUS84aP6eyMXRhitmeEzygp98cP tpKS2HmJcXWZuk023bAaGOkiNgkN7IYbbrj++uuffvrpRAMDcizSnT9/PqNMdB2gshKFf3F0LCg5 LhF+EbFg8obJ4LShTz8DphfWYERSSHFgwS9SFnx4+EXKgrpGI4+UBevjI2VBo80NUPT6SIHKQYsy dUEbjmiYokUtWbIkUqAYaSNtUYADUFhHwVJwcDnzTe9///sZ9tMfojTHBEVgzuzZnC158NBB5r+A oKq6anbdLLQLh3Aw11lVXc2EwknLl0OWZsx2TZZwOWRRUVmB65gZVTx4HJfJgRdLmI6sqHDIAmNs 8eJFTLFt3ryFwYoJi0WLF09LtT7KSQHOOP0U/ABsh8BTcfllx3Zdi38Az0Z5/FQMjwcjM75mP0a6 2RNNi3QpRJreZWuOtKB5RZwb997znvdgYFx77bVoA75li+izbfMaHyxsiSIKjw+gVIqQzUOBUqBC IhAyWXCLeuSRR9ALr7nmmpDUNNl4RQANPrxoNCqcrvxQ+x3O99sFOPfcc+3HiMJmQgT6wT0l4wLg Hoxb7wDMvG3uFvbHbsN4EWdFbEs3K6NM+TE88GywGgovB/9MfAYB2ai98qSlEkiXglzk57l3L758 q924XNKl6TB97qrKYaHzgdRnP/tZtkA8/PDD+VAYLYMioAgoAopA7hFobGxkA8ZHP/pR1mzknrty HAcI5FJjjg4uTIuIrAurzHDIqcoq+7x/8dxaLA3bkCAenwZX72WvxHNmFAI+/ex/W2JOwm2S7DYM Oxlhs93CPjOKCzF4dd45Z3oS5/4xp7WVe/Gi48iy+D/7sz/7+c9/bp9kEh07pawIKAKKgCKQVwgw n3r33Xczt3rllVfmVcG0MIqAIpA9Aix/kqOZOLLWpia7vdHp2Y9hx6cMy74O9m9weJQcRAspiHBm 1AMP/RCvCK+41OKBBx95Yd3LvtSkPJgQUBCjguVVBG674y7MEsku50qde/ZZvhRyGakGRuZoX3rp pZwocu+997JZOXMqmlMRUAQUAUVgDCLAYeVvvPHG5z//eYeHlY1BGLTIisC4RUCcGO+M+xmMkGj5 YlosqA91Fq3JSICLLLAHuDJv7YvHTIhbbvrsRRect3bdy2z15sa9F9a9RJqPDW/5sPMSPvecs7BS 5NI9XCjEyLm0BDBLyM7ptNyY8a07b/dkHJXHgrRWDY5KEfOZ6euvv37zzTd//etfP/vMU4d6juRz Ucdu2QqnOjt6cuyCoCVXBBSBvEKAvblf+cpX3v3ud1911VVu9+bmlZhamPAIsCcTm5NNyT/72c/M ccaDR3aGp6ApHSKwfv36NWvuhOAnP/mJyy+/3CHlLElhYLACqnbmDHFHGGoYDGzaDnPQrW9KTrnl 5KhEsoZ+7gO6yTsrzDkF9Qtf+AKb/FYunV9VPmbux8lKZs2sCCgCisCER+DJJ5/k0Myrr76aY20m PBgKwHEEot+KcJyXhsYiAtgVFX7X8IU/kMo3ZRjLJMdw6RKpbAG/7LLLOPF901ubsiWk+RUBRUAR UATGAgJ4/vfs2XPbbbepdTEWqitHZRTTQg2MHMGdBpuCNNJqUncIqAfDAZZcizHUe0SXSDmAUkko AoqAIpD3CKBEsj4q74upBcwpApidbMjkxticclVmyREYiO+PLShUAyM5RlG+UQMjSnSVtiKgCCgC ioAioAhMAAS4YbqrqwvjU7e25kNtDwwM9PT2cDWJmBn5UKSJVgY1MCZajau8ioAioAgoAoqAIuAY Aa6W59Y57s/WVVKOkc2IHEcvlJfH7hQvLNRdUhkhmHUm3YORNYRKQBFQBBQBRUARUAQmNgLquMir +reqI40b1vNKhLFeGPVgjPUa1PIrAoqAIqAIKAKKQF4gYOm1KcrDho22tjaWVZWUltbU1EyZMiVF hvRfsyGkrb2to6OjqLCoZnpNRWVsRj+K39DQ4E9+8tPz3/e+mTNmREB/6PChIwcOHWAvRUVFJViN io+oo7OTE2bz8LCmCAB3Q1INDDc4KhVFQBFQBBQBRUARUATCIHD06NHGxkauaJxeU8Pmjbff3jx7 9qza2lqHqnNvb+/WLVuLS4rrausGBgeamptYwVVfX++QRUtzC2RbWlp+85uX9u7du3r1aucGBgZb S0vzoUNH5s2bx6Ft+/bBZ9+iRQtZihYGZ4dpvv/oz7gt+/5vr6mdUQNZrtMmILd6O+QSQAoL54UX X+7o6CINNwCuXD7ifvGAjHIln+/htpKLS8Q5PDcgQQDxgFdqYASAo68UAUVAEVAEFAFFQBFIjUB4 3wX7j1taWnFZzJ07F6V52rRpxGzfsYNL+hzeCt/a0tLX37d4yWLIUnqK19TUVFFRMcOdk+F/fvUr zKR58+b29fWmBiijFPtjv/YlSxZXVVVBgKvo3tr0Vmtr65IlSzKi5ybTvv3tDz/6BJdq58zAeOW1 9dz/LS4U7tSb9NSkiz9w7jVXfDRAHjFIuPYb+4HbwRPth5QJAoiHeaUGRhiUNI0ioAgoAoqAIqAI KAIOEDh06BDLbRYuXGjuUZk9e/bOnbtQpV0ZGN1d3QcOHpw9Z7Z8xobKAAA8s0lEQVRYFxR6+vTq PXvKdu/ePSM2Ae/m5Narr75K4Pj97//gAJcEEizxwmFRVjZl2rSp8rKkpKR6ejVS4PaJYlFZQhH8 IzAtbrj2qpkzY66MHPywKx5+7AmMq1tuug7HBebN408+/YtnX6idMeOiD5znW4DHn3rm6WfXktFz X7hJfPd9D27ctJkEyGIi3QZ0k7dbPJWaIqAIKAKKgCKgCCgCSRFob2/nHbqyScGypZLiYtZN4Wkw kdkEDh0+RPbSkpjvQn4cpsTBSqyb6umOytswzMrZ/7s6O7u7ezDDCgqOK6tiMmGkOWOTEaFzzzkr /CKljDgcz4QLAqPivLOPccQkuObKmO9iw6YtxxONDLEsivR33H4TNsnIN8eecINIguuvPWYl+ibL JlI9GNmgp3kVAUVAEVAEFAFFQBEIiwCz8mj5pC4sPK40xx6LCnu7e3t6ep3sLujs7PRhUVjIWqzu nu7JU3K9gSEsOiPTdXV3s33c+HnkpeDW0RETMNIfuyyefvYFFHEU+vPOOcvD68677lvQMPeaKz5C PAbAT576j49feuGGt7aI34D1SDyygMomcubqVfaipn3723i7dt3L4kZYuXwJb30dDp2dsX0XtqtB knV2xeJ9f7fe9FmJp2y+Ce6+8/bgBL650opUAyMtuDSxIqAIKAKKgCKgCCgCGSLAXgjOjyKzPSsf f4wtW2LC3omB0dPTY2gSkJ9s7xbzZjgur//f19dH+Ty70oeliAkY3Y8lRo8/+Qw6/cUXnAeXtete Ei3fcLQVd9wLPPIPn8ZFF5zHI9vBWYOEzdDUsgNHAaubiGFRU3lZ+ccv+6AQWXPXfaSEPm9jWda9 xN87br/ZsDABcZW88OJLxs55+bX1vLVNDpM4fwJqYORPXWhJFAFFQBFQBBQBRWAMI+DRhhMlwcDg lxhfEN8XMTg4kPgqgxj8JOTyFEY2XuDEyIDgqGQ5JoXfjpFIpcC3INbFt+68XXwFbHVYc9e9uBoC cED7N8uNzli9CgODJUz3371GzAAo3HjrGlwWYmBgIWBOsPfa2BtQxghhX8SKhOOhoEBKbJ4HHvrh imVL97W14yeRyIDyjPorNTBGvQq0AIqAIqAIKAKKgCIwthEQbd6j04+uSB5DxsesGd3yheM+5Lcv xSNaIKW0d7TL3obLL/ugWbCENr9y2dK1+18KYLRi2RLzVo5swvMg1gXxkCLS+D3EH9LZdXyhF5bG ueecOb9hniFiBy76wLmv/G4966n4J/G33nSdEJcLOkzi/LmpQw0MUykaUAQUAUVAEVAEFAFFIBME whsYw0bICIVf1Gi2YmfCOyHP8AaPkSziWrlnS0NC1swjfD0zmZOL70shu68tUVQ0YgdLIpdhkNnr kraBwbomCM6M33eRSDl8jLEuErOwPaP2qWdYNNXUshPLRO61SJYeE+K2O74FETZtkwbjBG8GK6yw MWLbPJ59gUfDgjQ5231umPoG1MDwhUUjFQFFQBFQBBQBRUARCIvAkSNHDhw4wA6HYD0b1b+4qIhd 3oODI7X/+OPk+J0VYVkmTze5dPLRSUcHh2ILpcxPNHVzcK2Jz9tASXH8oK2RUojBkXKnCvs3Dh8+ jHnSm/4dHeJeqKuN4lbyY2Dj0GD11AvrXlr74ssx8+Cp2IaKT1/xEd+LNbjLgvVUN1x3lbEcrr/2 6hs3reFmjH9efUrMVrGOmlUPRt62Zy2YIqAIKAKKgCKgCCgC6SHAfXnTp09H8TVz5775MTAmT5nC +T/9A7EdzObXP9CPb2FKmZvzncory9va2wb6j2+3wOxhg0eMe1R3YI+wl4xc2QTKysrApH9ggMIb VPvj21QqKyuDKXMKMDVCmuKi48cBB2cxb0Vfb2reLpd2m3jnAc665R9bO3772vofP/UM2zbW3H5T 4h4MFkfB2o7HPjnj9FPYs8F+DA6z4p/zsmVPMIWPKXsGSmFUEMChxtlq/B0V7spUEVAEFAFFQBGY UAgEOy5sKLBDUJf7+2JnScmP3czsWo4rzWmv5xmmMeL/UytjurU9eQ99Tq8qLcW+OH45xog82T3E 3CO+i5myIMsVe/hbKLaNbV9vzDAT4yGAtpUlbcuH05+gLAulDIsOa7+Eicw4wG5vuZkbClgLmBls 4ybsYSr0K8rKExlhWhBZXlaW+CpPYtTAiLwiOI6A08fsf+zyiVr1p43eeOudrMzzFY875zmCwPeV RioCioAioAgoAopAughYGm2KrNXVVczNt7W14VSIJx3atXsXuy9qa2tT5Az9mpNPq6qqDh44KOfV ko/1QpyBO3v2LM/xuKFJ+iTs6e7hwo2Dhw7KcU9t7e1cT2E4+mRIMwpM6urqIBhf7BTLTJhrChEt 0mu82WmNEwMboLlluxQZXwHrlNIsflDyjo4u1EJYmERNrbGNH/ZiJ/PqvHPOJPzwoz81MZRHTsXF ODGRbgMxrfXFoE3tKdnpHoyUEGWbgOMIsFMTqXBigBisia+ijsET19yyg8V88/PSrRa1+EpfEVAE FAFFQBGIAoEwZgZ68/z58xu3Nm7b1lRbOzO24mDPvhNOmIPe7LBIDQ31jY3bmpqaMSpwkLS2tsKL n0MWTz715Lp163p7Y7sduCb8H//x/8MJM2PGjFtvvaW83GfSPQPWNTU1HLXU2tKKDVNUXLR75242 ZixatCgDUuGzoLWzf5qr9G6941sc/cSWDLluj40Q4YkEp+TUWhY+cRguejy7Jrh4m4VS3JjhuweD xVFyTO1nbrid8lAMKY85FTeYVwZvYSGKK0XKILtkUQMjY+jSy8jenXeuXkWeeEtt//5jT9CwKsrK aGTpEXKR+oZrr9741ma1LlxgqTQUAUVAEVAEFAHWB6WxFKeiomL58mWHUMwPHy4qLiHMlLzZZuAE zcmTpyxduhT6R48ehfKSJUtwmzihbIh86EOXvP+P3y83eLC0S+QHB1fWBYzYgzG/oeHo9I5DRw7B oG5W3dSpU6M7CMuIhh7PJRis9djb1l43o4YlTPhqmC82S5LQ+OtqayQ9WyB4XGjN2OJBIsazNeLc s880R9liw0CfTd6sN4EFdgVvzR5uUwwTYEqak6Z++9rrJCZwUcO55559Vhj3RV38Dg1Z9GWo2QHf BDhSKH8Y+jYpT1gNDA8gUT3S2szWfhruwoZ5t91xF0cHjIqBkbdbgqJCX+kqAoqAIqAIKAL5hEBJ aalbf0KicHgV8AAkxruKqago558rasnpFFROreRf8gSRvEHJvubKj9ik7W3W9gV5KHX8s1OimtsJ 5JW5h9ukjG3yNg+pApgfARZIsty1M2cklsROnCxBcC6bQrKwGhjJkIk2noY7c2YN65QMG5ykTz/3 wgvrXsbzRUvlGkjMR9+3NFx5S7MwCdjp8cBDj+JlI4a33AhjXiUG8J9gNHNYMq/wyn37vgfPOH0V RaIArOqDPgv+rrnio4kZNUYRUAQUAUVAEVAEFAFFQBEIRkANjGB8InwrBy0bBqz2Q+nHwD13xpns 9WEBFTYDRx1LAvP2Y8su3LhpC4v2cNWZS+xJyZUrrJnDy7agfu7Gt7Y8/NoThnJiAEb2dZKExTJh sR3+O7b1cPlLeVl59vZrImuNUQQUAUVAEVAEFAFFQBEY3wiogTE69cviKOwB4y/7xbNrUfrZp2FW TD3w4CNYETGNf/nSl19bz9uLP3DeNVfEvHXkwtsABVbvSXqhZmd/+NEn7NMJUgrJCq41t98MWVJC 88Zb12BmqIGREjdNoAgoAoqAIqAIKAKKgCLgQUANDA8gUT2ir+NYEOqYFjgN0ObNIiiMAR6NdUEy FimRBbsCA+PM1af8+J/vtUvGFh/ufWR1k0RyeponO6uk0jIwFtTHDmUTaiyR8izfsllrWBFQBBQB RUARUAR8EXC7UduXhUYqAmMCATUwclRNbLfYv/8AzNhrgWHAHolbbr4OVT4e04XJQdg+zVaMBzkX WYqITfLCiy+TUrJIJH+JIfGKwE0XJnGyQEXFiMMlKh0dMJeMncYrAoqAIqAIKALjCQE1LfK1Nt3c XZiv0uVvudTAyFHdsLpJFkRhDLAAqal1u2FsNmOwucJEEuDUKc6xlRj2YGBgcIAAWyxYzrSvrU12 TZj0HgvBxGtAEVAEFAFFQBFQBKJGQAwMNTOixjld+gVqX6QLmaP0amA4AjI0GTwVF11wHrsmOF9Z NjlgMJAbF8StN1/nSya+pXuzvcUCY2PtupclsWTnVkjfvBqpCCgCioAioAgoAlEj0NXV1dHRETUX pR8SAS7mo0ZIPJjO/SQhiWuyMAgUhkmkadwiwF4LNjywR0LWQWFycOcdN9/JYyIv8Wyc8Y5TEl8R Q3aoBWT3zaWRioAioAgoAoqAIuAKgZISbrYoLSxUtcoVolnRwZVUwqXfJUyjp3EBYlYsNfNIBLQn jMQjJ0+YBJdf+kHMie8/9lNheP7ZZ/H4wEM/5MBZicFHccOtazZu2syjXAb59LP/bV7ZuzWI5LAp sv/kqaclAds8uOlCwvpXEVAEFAFFQBFQBKJGgFvtsDH4Rc1I6YdBAAODCuGfXDQeJoumcYuALpFy i2dYalzfyCFRLHOSg2jxaWAhsG4KZwW37BFmiwV2xfz43ZAk/sVza7mbgvQV5eXcxMdmDPZ272uL bfjmx1KrV363XhJw+x5Win0Hn6TRv4qAIqAIKAKKgCIQEQJDaS7FGRoaPHq0o7e3F5ukvKwM68R5 wSgSy4T4YfyUlZXhYHHOAoLdXd2dXR2FhUVTpkyZPHlydLtQNr75Zl1tbdTXn0cB0cSkWbRmzZqJ KbljqQd6J/Evya+6atrJy5fw136P/UAM3oy5c2YRj83A4bOlJSW9/f118RNsP33lR3nkFX85qXZh w1xekZjDba/8+IeJJ9nSxQsI8Lvg/HN4lLxnn7WaPeXkWrl8STJLAzPG3DkvKaUYQo2/dgITOSqB gslTR4WvMlUEFAFFQBFQBEIi8POf/3z37t1FRUWf/OQnKysrJddQ7xHf7P39/Vu2bDl06BDWxcGD B3fv2VNeXu7WAGATQlNT0549e2HR1dXZ2rqjuLgILr7lyTRyaOfOXTt27kDqnt7e7a07MGmmTkV2 ZxurIXjkyJHmluYf/vCH//qjf120aFF9fX2Y0u7Zs2ft2hdIefLJJ69cuTJMFk3jFgH1YLjF04ca mjr/El8kxqPxG6Xfk55dFvgx+GfiE2/BS5nA5DUX/BGDhZNIyk5gcmlAEVAEFAFFQBFQBLJEANV/ y+YtQ5OGli5ZIo6L7du3v71587ITT6yoqMiSuMne0tJ8+PDh5cuX41ggsrR0N/YGxkZVVZVJk2Vg 185dmEZIMXVqbB6wuqp685Yt7EKZNSs2berk9+MfP/7LX/5HTU1NJ0ZSZ6cTmkokNwjoHozc4Kxc FAFFQBFQBBQBRWDcIhB+idTRI6yNOorSbJZFzZ49G6tj7969rtDp6+tra2uHhVgXkGVlEWFMAlcs BgYG2trbKysq4i6LGFUCrOLGddDf3+eKyyWXfOiee+759rfvbpgXynHhiq/SyR4BNTCyx1ApKAKK gCKgCCgCioAiEAqB9gOx01yM6k+YPRLFRcWHD7Oeys2RRyy+wuBhR4QpEAxia6W6u/r6ki7nNonD BLgzuKenJ87CLIgqQCh2lXR1dYehECYNa7ri7hHDIkwmTZMXCKiBkRfVoIVQBBQBRUARUAQUgXGP AHp/V2dM/2bfgi1scVERqj//2ZEZh48ePRpjMeLM3AI4DgwOdnf3ZEzWzogTBlk8x/KKUMLdTqzh CYiAGhgTsNJVZEVAEVAEFAFFQBEYBQRYCtU/0A9jz2lLBYWxSfqeHjfuBXwLMRYjDIz4zuuhIXmV veQ9PTEzSYptqIlQ3fFXJlIDExMBNTAmZr2r1IqAIqAIKAKKgCKQawSY9cfGiHMdsexHrmvoc7R7 YWAgxsJjw/BMpJg32Yt9jIXnwKi4TINx7tmzSKTgZgFZIl2NiQABNTAiAFVJKgKKgCKgCCgCisBE QkC0ea9O74dAwIUZAa/8KCWNw4zhnT81Z0q6M0JJxUh8MRo8E0uhMWEQUAMjDEqO08Qv7X6Ei/Yc 01VyioAioAgoAoqAIjB6CIQxMOKOBIo4Qlnm1FqiiorcaGXDdEawEFQ8uyYyhqqgMLaHRIp9nEic YRgQjmdJL+QjUQoCw3CnSKavXSOg92C4RjQEvc7OrrXrYtYF13jzlwu8H3/ymTW337Ri+dIQuR0k 4Zrw7z/2sw2bNkOLGza49YLL+wLocjX4Aw89Si5MI+7NOGP1KtInu8IvgI6+UgQUAUVAEVAExiUC 3JfX1tbGqUec3xogYGFBAUdGcYysOBlMyqHBmOrs6q49DoyCmofFoLCY7OY+b7kIeKSVdMzecCWF AccE/H0y5rUVAOFDhw4S0dfrZluLRVuDoRBQAyMUTJEmqigrQ2uPlIVNHGvhzrvuIwYjoaKi7OXX 1mPecH/NNVd81E5mwqS/7Y5v8XjRBedhjWzctAXraMOmLd+68/ZcFtuURwOKgCKgCCgCikC+IYDZ gGLNMUrBLoLCoiKSdXV1eeyQgcEBJv7tg2WzERA758CBAyNZsPdjgLKVlrgxMORScGja5Rzojz2W l7m9L9zmkIYHI3bNyJBsPLEpaDhHCKiBkSOgA9hc9IHz+BeQwO0rHCY4Iu6/ew3WApQxG+68695f PPvCxy69yNdgePixJ+z0uDvICJFXXluvF367rRqlpggoAoqAIjBGEeC6Bn5YCCkXCE2fPp07tjmS 1kgaO1qqv7+ycmqwcWLSpwxMm1a1Y8cOrqQwKfv7MTcGsG3s+zfM2wwClZUVWFOeY684ZreoqHjq tMoMCIbJEt6DgQ+nojx2LXpxccyZo7/cI6AGRo4wZz1Sc+uOjo6uFcuX1M2cYXPFRbB3f/uChnmi 37MSiWMY5jfMIwvuApwMZ5y+StYj2URWJqynkvRQhkXiW8PxlddeX9AwV6wLiTz37LPwSCQ1GIYK xKgwFKA/6alJ+/a3mxgNKAKKgCKgCCgCExkBz3qkACiqq6t37tp15PCRmTNmijWCt4Ezn2prZwbk SutVOU6E8vIjR45wt0ZJ3GXR0dnR3d09Z84cVzYMZKdPrz506HBPd8/kKbEb/bBnOjo7p02bKhzT KnDKxPHlXSz6kgO4UibXBKOPgBoYuagD2WVxjNNTkzwT/2vXvUyCO26/SayCu+97kJQYIRgMkuXh R5+44bqrtrXsePrZtRaRM6+/9mp5xERZc9d9xzX+pyZhQtxx+82JHgl8EfxbMWOEhSPGxvHsx3gc +9+tN183MuLYUyJx32QaqQgoAoqAIqAITBAEwpgZLKaqnzevual5586dM2fORC/H21BXV4tnwxVK 2C0LFizYsmXLjh0758yezf16O7bvwMGCgeGQxQknzO3q6mlqbqqvr+e4WlhMnlwaC7v7vfTyy8/+ 8llMo81btoDb9773veeffx6nxC1f/nJZDteWuxNoAlFSAyPyymbHApsc0Pivv/Yq3BQ4KMSECGCM rn/m6afcctN1KPEvrHvp/oceuf/BRzADxAjh7d33/RNmCXvEZV849gmR0BfTBXbEYI18/DLv1m32 l8MXr4jNXTwq+9pCeSQwZh54KFaYM1afYhPRsCKgCCgCioAioAiEQQBbgjU8bArf1rStuKh4fkPD tKqqlGurwlA2afBgnHjiifv3729qbiZy9uzZeE7MWycBFlwtXryovb19+/btLF6qrq7CXpLLvJ3Q h8iSxYurPv6x+KFY8Ss24jvXWU4mPhNXXJROFAiogREFqiNovvDiyzzfetNnxVGAjUH4tjvuGpFo 5AN2xTVXfkTizj0ntn6JM20vv+yD4uKAzscu/eC373uQeDEwCBBpHCPYFRvf2hLSYBjJOemTHDyF 94NA3Fi6WsRJmkFfKAKKgCKgCCgCEwaBML4LG4zK+I9cbu0KmwXbLebNmxc/D/eYdm6/dRLGxsB0 mTVrFtSiEGRG/OekqEokxwiogRE54Kx0wjCw1fGUi4tYPZlYrAX1DBPHfh4KdTNrsDHwWpx39pmy WwNfx3Da9P6Pd8JkWFA/1+w+hyMilMdWWHU2tex45XfrMTNMSg0oAoqAIqAIKAKKQLoIRKGUJ5Qh KuvCMMqJFIabBsYGAmpgRFtPsrGhvMzHYEiXcfnIdU12djZjxJZFxf9hBqxcviTjeyrYVm4od3R0 GgMDu4UlWPKKPSHwgpHc42HSa0ARUAQUAUVAEVAEFAFFQBFQAyPaNuDri3DOEl3/+uuuiq2M2rT5 5d+9zvYM/rEvnOOhwvDau7+NZLUzYqfWfufuO1JmgdHTz6195dXX1cBIiZUmUAQUAUVAEVAEFAFF YKIh4OZS+omGWnh5WVnEv6bW7eGzpJuSNUtYCPzFzGDDxq03XccdFzD98ZPPJJIiTaw8LSPKI26W ZEueWDR1/PSqOEUo8P/Orth+cf0pAoqAIqAIKAKKgCKgCCgCNgJqYNhoRBI+4/RT0ODxLRjqbGAw 4ewDnZ3dN956p713AisCz4kcGJVIH0OCTRQcBmVeSXnYfW5i7AB7SNjdwfZuE8nl34TZoWFiNKAI KAKKgCKgCExkBGQfgu5GmMhtQGW3EVADw0YjkjALipjy52ja2LKi19ajrLOHwSEnzAk2XXB93gMP /RBjABYEMGmSrV9iewbcuTeDk6lIzEXd5PVcpWcX7+ILzsO64Ngr0kMfKf4hfkyt2Z7BObyXf+Zm j5fDpqBhRUARUAQUAUVgIiCgBka+1bLWyGjViO7BiBx5DADOdMLAELsCY+OGa6/iaguHjG+56bPf f/QJFH22XkAWFlgRGDa+LDjZlmVU2BUPPHisDJgQH0uSGAoYEhD88VPPmPQcinXNFR9BLqGfzFXi y10jFQFFQBFQBBSB8YfAwMAAV+ZxGZwtWkFR6dBArx2j4RwhMDSJW8zhNZQjfspmBAIFhSUF6Z7c PIKAPgwjMNR7ZKjnyPCT///Z+cDKJbnVzj9F1rHCoqK8HJMgJTF2bmAbcDxUmMRQC0jPq+jkKpx6 QkpZNIEioAgoAoqAIjCKCFx99dWvvvoqV08888wzXA1xrCSDfYOd+ydxC53+covA73//+6985f+F 55/+6aeuuOKK3DKf6NzwGhWUzxxhak90SCKWP9kmB4ds02KRrkkQkD7glUPplJQioAgoAoqAIpCf CJSVlVVUVEyePHnEmpzCksKKukn93UNDg/lZ7HFbqpLKvkmTka6/sKxg8tRxK2b+CVZQUDipuJxr F9XAyL/K0RIpAoqAIqAIKAKKwJhCIOl6kIKiSSUVkd91N6awykFhB4rKDncPwKh3qKSgVA2MHEDu ZaEGhhcRfVYEFAFFQBFQBBQBRSADBJKaGQm0Dh06tHfvXnZuFBYW1tTUzJgxY4TrIyF9BhGdnZ27 du3q6elhZ0hVVRUsPFtEMqDpycK2k507d3Z1dRUVFeHAqa2tLS0t9aTJ8rG/vx8WR48eBSjcRHV1 dfxNSTN8RaQkpQkyQ0BPkcoMN82lCCgCioAioAgoAopAJgi0tbW9+eab6OL19fUo5c3Nzdu2bXOr E2NdbNiwAe28oaFh1qxZO3bsePvtt92ywHTZuHFjR0fHvHnz5syZc/jwYYQiMhNEkuSh/Js3b963 b9/cuXPhgj0DxyNHUmx5TUJMo3OKgBoYOYVbmSkCioAioAgoAorA+EMgvO6OCr5ly5bq6mpUf2b9 p0+fvmDBAgyA3bt3u4KFwqCIsyFk8eLFlZWVuC+WLFly4MCBpqYmVyyg09jYiMa/bNmyqfHf0qVL edy6datDFq2trbh6TjrpJERAkIULF+KE2bRpU3i0HRZGSaWFgBoYacGliRUBRUARUAQUAUVAEcgc AdwXKOL2mijCLDHav39/5kRH5sSrwLIlbBizJgrtHGMGV8DgoJvt5phJWCzTpk0zLAhgaBw8eLC7 u3tkcTJ8Yv0YmHAwl1kTBQuEgr46MTLENIfZ1MDIIdjKShFQBBQBRUARUAQmNgLt7e0AYPRyAYNH lGZX2r+wKCkpMUizgQEWfX19rrR/HAsYADYLeLHoi0jWShm+2QSggxlDye3dKcIR7tlQ1rw5QEAN jByArCwUAUVAEVAEFAFFQBGYhAkhuxRwWdhw8Ihq7kr7Z0s0xFHNbRY8wh3Phh2ZcVhYeKQQjvhP MiZrZwQNlkJ5WMijcLcTazjfEBjR+PKtcFoeRUARUAQUAUVAEVAExg0CqPgYEohjz8qbR7E9sheW JViJRISj76vExCljcIYkphEWrqQQFh4zSVi4ssQSRdAYVwiogeEKSaWjCCgCioAioAgoAhMaAY/Z kIgFU/IB66A4NCkxSwYxATaMvMqApidLgBQBrzxEgh996QjCvq+CqenbHCOgBkaOAVd2ioAioAgo AoqAIjDeEBDFN6WBESx2ltkNcaHjOWpJHl2xMLyiC/gWdcxJER0+eU5ZL9rL8wrS4ikCioAioAgo AorAGEAgzLQ6SrOs+fHV/j17pjOW2bOD3KYT8MpOljLs2Rphp3fLwgOUMArDQqojoJx2mTXsHAE1 MJxDqgQVAUVAEVAEFAFFYGIhwM5mjoFKuQAJ6wIrgl0KHr1ZtGGOZHWCGue6cs6Sx+BxzoKiRsqC M6mwx3xZmINrk8HF/g05ytbVnpNkjDQ+GQK6RCoZMunFFxRNTi+Dpg6HQEHh8VP2wuXQVIqAIqAI KAKKQK4RkPvs+BvMGAOD+yhI49luwSOGB1fjBWcP+Zb7KDwssGdgwcQ/WntIIsHJuAHDw8I8yqvg 7GHeAhT+B2w22xgT3ETAACKIyd18/NSDEYBSpK/UwHAEb1FpQUm5I1pK5hgCTF0UlFUrHIqAIqAI KAKKQJ4jYCvBwUWdPXs2Hzf7HCSm29GbuW4vOGP4t3KLn30iLT4TuKCyp5z7D8kFUphDsDCCE+AR +imtrPAsIEXJbScGLEDPIVYhC6PJ0kVAl0ili1jS9AVTqguKpwwN+hzcljSPvkiOAOtUJxWXc5Jf 8iT6RhFQBBQBRUARyCMEjLYdUCZm35lZ3717d11dnWy6aG5uZnHUCSecEJArrVeQhdqePXtYJiST /VyJjUlz4oknpkUnIDHOkPr6+qamJi4mnzlzJikJwK6hoSHMBokAyvaruXPnbty4EUEEHNahcRn5 nDlzBDc7pSccpiI8WfTRLQJqYDjFs3hKwSQ3CyidFkuJKQKKgCKgCCgCikBeIMAE/NKlS99+++0N GzZMnz4dvR+9efHixeXlLtdBzJ8/H6/I1q1bYcEqo4MHDy5atMjV4iXBcdasWbgXsDHk6m5YzJs3 D5PAIco1NTUUe8eOHTgusFtggVW2YMEChyyUVEQIqIEREbBKVhFQBBQBRUARUAQmCgJpTZnjr1i1 ahXz/ZgW2BVYFw5n/QVx9h7grxAWbPxA9Xe1+8LUKGTR9dH44UIkS7/c2kjCCN8FC6IwLUAYoFyt vzJSaCAiBNTAiAhYJasIKAKKgCKgCCgCioA/Avgx8Ce4dSkkcmJ9lCyRSnzlKgajIgq7wi4emz3w ltgxGs5/BHSTd/7XkZZQEVAEFAFFQBFQBBQBRUARGDMIqIExZqpKC6oIKAKKgCKgCCgCioAioAjk PwJqYOR/HWkJFQFFQBFQBBQBRUARUAQUgTGDgBoYY6aqtKCKgCKgCCgCioAikM8IsLMin4unZVME coaAGhg5g1oZKQKKgCKgCCgCisD4REBNi/FZrypVpgiogZEpcppPEVAEFAFFQBFQBBSBOAJiYHBy q+KRDwgYe09rZLSqQ4+pHS3kla8ioAgoAoqAIqAIjBMEuNHi0KFD3G2X1oUY40T4/BODiqA6KBeB /CvdhCiRGhgToppVSEVAEVAEFAFFQBGIDgEugKuqquIGPTN3Hh0vpZwSAS4upDpIph6MlFhFlEB9 eREBq2QVAUVAEVAEFAFFYKIgII4LdV9MlPpWOVMhoAZGKoT0vSKgCCgCioAioAgoAoEIZGBaDA4O dnV18TeQcFYvKVV3d3fUy4R6enr6+vqyKmiqzLDo7e1Nler4+wyq43hmDblAQJdIuUBRaSgCioAi oAgoAorAhEcgpF47MDCwbds2NglUVFSgOrOeZ8GCBYTd4tfc3Lx///7y8nLYUbCGhgZZNeSQy549 e3bs2CELw7Bk5s6dW1dX55A+pA4cONDU1FRSUsJiJ2yMWbNmwaKoqMgtF6XmHAE1MJxDqgQVAUVA EVAEFAFFQBFIigDWxcGDB5cvX87ODbT/t+K/lStXTp48OWmeNF/s3LkT1X/ZsmUzZszAumhsbITJ qlWrysrK0qSUNHlbWxtk582bV19fD4vdu3cjF1tQamtrk+ZJ8wVb5zdt2oRFsWjRIii3t7fziM8H SyZNSpo81wjoEqlcI678FAFFQBFQBBQBRWCcIRDSd4HUR48eZeKfmXisCx6ZjEd7RpPGJHCFCWui mPXHtKipqYEmqrlo5NgDrlig5be0tGCuYGAIi9mzZ+PKINLhoi+cMOAzf/582To/ffp0hEK0qBd9 uUJpItNRA2Mi177KrggoAoqAIqAIKAI5RWDv3r2o4KxcMlzRy1kCxPS8ickywOIrHCOsuRK9HGr4 RuCCbeNqs0RnZydGkc0CXgjFrhJeZVl+yc7iMQSh5CwhkxhhAXqsm3LCQolEh4AaGNFhq5QVAUVA EVAEFAFFQBE4jgCOjiNHjvBslGZ5xzw9qrmriXnWX3lYoJrDAuuCnRLHS5NFCBbIAk2bBkIRKdzt +MzChw8fxkxKZAE1XmVGU3PlDAE1MHIGtTJSBBQBRUARUAQUgQmNALPvchqS534G1GhUc1dz/0LH uC8EcR5hgRnjpAKEhUcKeXTFAg8GBfawEHvDFVBOoFAivgiogeELi0YqAoqAIqAIKAKKgCLgGAE0 5oAtCq7WLzHxT7kTDQwiXTlJAqRwxcJXCqkPVywc166SsxBQA8MCQ4OKgCKgCCgCioAioAikj4Bo 8x6dPpEMBga/xHiJCdDak2XxjRfV3PNKyuaKRYAUAa88RQp+9KXjVorgAujbbBBQAyMb9DSvIqAI KAKKgCKgCCgCxxBIaWCQLiCNZzlQxrAKHY+CLo+uWARIEfAqLYl8ixpSClMGE0iLtSbOHgG9ByN7 DJWCIqAIKAKKgCKgCExoBFjdxN4DX53YxoUEsovAV/svLS21E2cc5uQlDozyZBeOHFflic/s0ZeO WxayD97jcgnJglyutoJkho/mUg+GtgFFQBFQBBQBRUARUASyRYDJcvRaj+XgIYqBIVaEZxUTj2S3 z671ZEzrUej4suCw2rRIJUss94579kIIR7nfI1nG8PGYSQKpnUXsjZQsqAXy8vOU0Cal4UgRUAMj UniVuCKgCCgCioAioAiMfwQwG9DdRScOkBaVt7q6mgSe/dw8cmmd50jWADrBr6ZNm+ZhgeqPao5P wJWBUVVVhSweKdDmsaCEe3AJw7zFisBPAk3bZhOOKVmAJJLycwVpmAJrGhsBNTBsNDSsCCgCioAi oAgoAopA2gjYSnBw5rq6OhR9ewkT18mRnbu9gzOGf8uN15g6XLhhSsX1FywZ4hps36VN4SmblOju 2BgcF2tcBAQQCs8GlpJJlk0AEbiJnJJzXq3QwUbiBgxsOQQMpmwED06mb6NDQA2M6LBVyoqAIqAI KAKKgCIwgRAIo9eims+ZM2fPnj379+8HGq7FaGxsZEp+9uzZrpDCt7Bw4ULslt27d1Mk9PKmpiZ8 C0Q6ZFFfX49jpLW1lb9waW5uRpbFixfD3RWXefPmYRFRePFjIA5CLVmyxCELV0VVOh4EdJO3BxB9 VAQUAUVAEVAEFAFFIEIE5s+fj5mBjbFz5050Zebj586dK3uaXXGtra2F4K5du/bt24f2jwGDXu6W BWu9li1bhggbNmxAChwLJ5988tSpU12JAB2cIStWrIDFxo0becTYWLlypawxc8hFSUWBgBoYUaCq NBUBRUARUAQUAUVgAiEQxndh4EAdx1/Bj1zRTcZjt/CTgkXEhVVS/CJlwYZ1TCOgww+DE8ZgqIE8 R0ANjDyvIC2eIqAIKAKKgCKgCIxPBCLS+22w/v/27uY1ruoN4Hh9oVVa1L4kadqYpE00bVNfsPiC oitXKuLLUkUXirjXP0NEEFy4cyEIgis3IoKoGyspxog0tc172jT1pVZb+6L+vj8PXKZ37tyZuTkz SeZ+sygz9+U553xuF+eZc849nVEELTK7qHysa/+zueDaf0bWUAEFFFBAAQUUUECBdSNggrFuHpUV VUABBRRQQAEFFFBg7QuYYKz9Z2QNFVBAAQUUUEABBRRYNwImGOvmUVlRBRRQQAEFFFjLAm1Y8LCW m2/dFEgETDASCj8ooIACCiiggAJFBEJqYYJRxM57OlHAt0h14lO1TQoooIACCijQXgHeo9reAi2t pgCZXngc1113Xc2LPNFKAROMVuoaWwEFFFBAAQVKIHDhwoU///yTHSHCphAlaPGabuLly5fPnz/P s+DDmq5o51bOBKNzn60tU0ABBRRQQIG2CLAfHJtYb9q0yVlSbfGuUwh7fm/ZsoWL3D2jjlTLTrsG o2W0BlZAAQUUUECBcggUGLgIv68XuLEp0StXrrR67tbf//01VatmL25DEc1WyevzBRzByPfxrAIK KKCAAgoo0JBA49nC0tLS4uIi19P77+rq6u/vjz708dtvv83OzpJgUPVbbrmFIq6/PnKvj1lh09PT TA+j8ps3bx4YGLjxxhsbkmr4or/++mtmZubcuXOMRWzcuHHv3r0MFtW9u/EHUTeUFxQTiPxfrVgl vEsBBRRQQAEFFCiJAD1mEozh4eFt27bx2/zExMTZs2dHR0cjrkg+c+bM5OTk0NBQT08Pve0ff/xx bGzsnnvuiZhj/P7774TdsWPHgQMHeHA0anx8fP/+/TfddFOs50h2QUwCHjp0iBwGtCNHjoyMjFBo rCKM0yIBp0i1CNawCiiggAIKKFAWgcZ/Muf3/oWFhe3bt2/duhUdkoo9e/aQYHAwFhaVOX78OGtC uru7iUnXnLEFMpmpqalYRTDwQkZB5QcHB4nPX19fHysfGNCIOCNrbm6OtjBqQXxqTl4B2okTJxrX jtVe4zQrYILRrJjXK6CAAgoooIACBQWWl5fp6998882h00wUfqFnYOHnn38uGLHqNtKVS5cuMS0q KYJpRfz9+uuvFF11eZEDjC0wbYkcJhl1oQnMkmJY4+LFi0UiVt3DC6CoMOvmyVvCScqiROLTwKrL PbC2BEww1tbzsDYKKKCAAgoo0MECLI2gdampSnSdWc8Q67d/+uWpIsg0KJEuO+MnUWzp4pOrpFpB JsDYQqzeP7kKaRIySZqUNCoYRmmIQVokYILRIljDKqCAAgoooIACVwnQKafTzKHU61PpRtM1j9X7 ZwuI6iLoppPAMPJwVYWKfiEd4laqXRkgfA2nKo8X+0xVMUkVEdxiFVGsYt7ViIAJRiNKXqOAAgoo oIACCqxUgB5zmKRU+as8QcPXWL3/zN3lQhEhvVlpMzZsCC+nyowTq4gcqFhFZNbfg1EETDCiMBpE AQUUUEABBcoukEobqjlIMPirPh6OxJoildM1j1VETityTtVqeObxzKoG4cxTmUE8uFoCJhirJW+5 CiiggAIKKNAhAqHjWzfBoLU516TmTRWmCXFSHf3wNaf0wsWlboxVRGac0IpYUKma+zWigAlGRExD KaCAAgoooEAZBUJvOLNPXMnBBTm9/+R1SZW3FPicEyfnVFMF5cRJrfxuKmzlxSFOZppUt4jkQZiK VJK287Mb7bVT27IUUEABBRRQoAMFTp8+ffLkyRtuuCFMT6rVQpYs0zVnrUVqkk+4K9Y22OGNtNVF 0O2mhrXq1tTxUNVUY+O2ghfUUuFUEaFRNDC/tiyX53FwTVjvnn+xZ1shYILRClVjKqCAAgoooECJ BLq6unp7ezdu3Jj/kzk9ZnbAYAeJ1DpsvtJlzxkWaIqSIti2r7II+uX01ElvYiUYYZONyiKoIV9p YNhAsKkKZ168ZcsWRiqoNoMYhA3XhMXllJ55S3KQ5ITHwddY7U0i+6FBAadINQjlZQoooIACCiig QLZAeJsq3d/UlJ7qq0lFuLjyl3U+kwCwt3f1xcWOsHMfRfAu16QyjJmwPx3HSYGKxUzdRcedYQRq nhTBB8YNOMhf6uJiXymCbfWoefLGKpQokRSuboKR3FJ3MlWxunlXXQETjLpEXqCAAgoooIACCuQJ 8JM5p+kB85d33YYN/DBPjnHq1Ck2kuNKfvWfmpqiU75r1678Gxs/y0hIf38/W4PzR7+fKjGgwb9D Q0ONB8m/kgRmYGCA3v/i4iJF8McHEgwOJqMN+REaOUsriDwzM0Pl+cAGgrRocHAwpHM5EZKJVbES qpyyPJUp4BSpTBYPKqCAAgoooIACjQokHdm6CQYR6ejTCz927BiZAAkG995xxx1JhEaLzL0upCuz s7P0++lt86v/6Oho3PlCDLnQkPn5eTr9YSLTyMjItm3bcuvV3ElGMPbt20eC8f333yPGIAzZRSOZ mAlGc9AtuNoEowWohlRAAQUUUECBMgkk67OTyTk5rae7Pzw8zJVkF3wmtYj4q38ol7B9fX07d+4M 6yIogiM5VSp2qqenh9EY+v3Un8lIrZiPxIoOZnZhxQhG40UkTyFu2lZMqZx3mWCU87nbagUUUEAB BRSIJsBv7SFW0rWtG7rx7nLdULUuaEMR5C1JclWrGis8zoSounOiUkWQVoUjJhgpmbZ9jZ/Otq3q FqSAAgoooIACCqwFgc2bN4dqVK7eXgsVK2cdkqfQ6uSnnLyNtNoEoxElr1FAAQUUUEABBWoKMI0n nPvjjz9qXuSJdgkkCUbd9021q0alK8cEo3SP3AYroIACCiigQFyBZHFz0rWNG99oTQkkaR7v7Grq Ri+OJWCCEUvSOAoooIACCihQUgH2tgstZ/eJkhKspWaHNI+VGyYYq/VYTDBWS95yFVBAAQUUUKBD BJLtq8PuFh3SqnXbjPAU2F0k1ubo61Zi1SpugrFq9BasgAIKKKCAAp0hkEyRYjO4zmjRum5FeAq8 2ivu1h/r2qTNlTfBaDO4xSmggAIKKKBApwl0d3eHjSbYda7T2rbe2sP+4mENBiu8fU3taj09E4zV krdcBRRQQAEFFOgQAab7k2PQGBOMVX+iLIM5d+4c1WATQKdIrdbjMMFYLXnLVUABBRRQQIHOEdi1 axeNcYrUqj9RFmCEBGNgYGDVK1PaCriTd2kfvQ1XQAEFFFBAgWgCe/bsGRsb++2/v7rbL/zzzz90 gi9evMgcHjbpa8UP7f/++y+/5V+4cIH9vFnuvGnTpmhNrQjE+5qYj8QADlva8XfNNddUnIzzESVy Bmag0YRG3gr1yy+/nD17lrL37dsXpwZGaV7ABKN5M+9QQAEFFFBAAQWuFhgeHuYAHe6TJ0/mJxiX L1/+6aefWCrAu6eYUnXp0iWSk2SrvqujFvz2999/T09Pk+xQBMkMnxlg6enpKRiuxm3z8/NLS0s0 lt7/zMwM8Xfv3l3j2iKHyZHOnDkzNTUVVlNQHAnG4OAg+UxOuOXl5StXrpC53XrrrTmXeaqlAiYY LeU1uAIKKKCAAgqUQuDuu++mnYwYzM7O7t+/v1ab6e7/8MMPnD148GAYuKADPTExceeddzby83yt sKnjJ06coGt+1113MXbBqYWFhcnJSYpL3naVur7AV5o5Nzc3Ojoasqnt27fTCuJEzDFowtGjR2+/ /fawvoXkbXx8nORhZGQkp8LkIZxlXKi3tzfnMk+1VMA1GC3lNbgCCiiggAIKlEKATvCOHTsYOqA3 n9PgsEKgcv1xX18fdy0uLubc1dQphkQYWKA+IbvgXj4zf4mUoKk4ORdTYYpg1CUZq+Ezr4Wl7SQA OTc2derUqVO8ZxbVcBdNwI2sg0lTOXFI2DjLcEqy+2HOxZ5qkYAJRotgDauAAgoooIACJRKge02q QIMZPchpNhN4OEtfObmGgQX+Iq4OZxECk4sqt4AgPlOGGF0h90jKXcmHsICksojQKLr+sfYyJw6r O6h2eP8v8VngQYkMATH1K6fyx44d4yyjHK1YEJJTrqcqBUwwKjX8rIACCiiggAIKFBGgHxxmRh0/ frzWT+z0+8MWDSy8riyDr3T9I/b+CZ5aqEARDDuQY1SWW/gz4zC0JVVE+BoWWBeOnNwIFBVOsotw vG4RjJ+EgZr77rsvCeWH9guYYLTf3BIVUEABBRRQoAMFWPPAWAGTncJrUqtbSI+ZFd4cT/24HrrR sXr/IU6qCL6SErCyvLpWBY6EOKnef/gaqwjSreocpi4Ua83JMbjswQcfLNAub4klYIIRS9I4Ciig gAIKKFBqgXvvvZc1CcztCcsAqi3oMfNXfTwcCblHrbONHyeN4eLqBIODcYvIrFKsNRhMhcqMz8HQ wMyz3377LccPHDjAjLXMCzzYHgETjPY4W4oCCiiggAIKdLgAq4p55RGN/OKLL2o1NSfByDlVK1rm 8Zyueeb1a/NgpkbImjJPhVZ89dVXfHjggQfWZqPKUysTjPI8a1uqgAIKKKCAAq0VeOihhyjg8OHD mcXQPw5d5MyzqSUNmdc0cjDESfXCw9dYRaQmR1XWKudU5WV1P2fGyW8Fm4rwNl4W0LsAoy5vqy8w wWi1sPEVUEABBRRQoCwC999/P5NzwjYU1W0muwjLu1O9/zDmEGuz7RAnVUT4ykuZqmtV4AhLTbir DUWkRmPyW8HMNJaYDw0NhU0PC7TLW2IJmGDEkjSOAgoooIACCpRdgN2jWYlBP3hsbKzaggGE0PtP rSKgG03uUfnu2up7Gz/CHnNcnCqCr4wJxMphwp6AqSLC6otY2wXyRloqnCoifK1VBANH1AH/uNui Ny7vlYmACUZC4QcFFFBAAQUUUGBFAvSJH330UUJ89NFHmYHY8Zrjle+xJRvhjUms38icFJQZJP9g 2K678qW3dLv5Y/giVg4TalvZCqpEiYzPxOrcE4cKsyq9cpwkLFLfunVrtQCvtf38888xfOqpp6rP eqTNAiYYbQa3OAUUUEABBRToZAFekNrf33/06NG5ubnqdrIvNcMIzORJ+s2nT59m+GLnzp3VFxc7 wggGv/GzG13yzihem8u7ayki1hoMuv40hLDJS2lJNvjKS7RSu+8VawJ3UVWKIH7yzl9yJFZZsD15 5gjGxMTE9PT06Ojo7t27CxfqjbEETDBiSRpHAQUUUEABBRTYQPf3mWee4WW1X375ZTUHqxcGBwfZ t5v0g94zmQYfent7M3+Vr769kSOkK6xDILtgyznyCn7a5wMDAhF73hQxMDBALsGugsSnsXzgKwcb qWGD11BhFrSEImgIe1zQKN7TRenVET799FNykueee676lEfaL/D/XVfaX6olKqCAAgoooIACnSpA 2sBEHcYx3nnnncwNGUgwlpaWwpwiJk11d3dndppX4nP+/PmFhQX65URm0hRFhJXZK4mZupcEiV0F STAogklTPT09sdZ4JAWx6GJ+fp5BDJapIMkgTOYsr+Xl5eeff76rq+u9997LvCAJ6If2CFy1U317 irQUBRRQQAEFFFCggwXobT/55JMffPDBkSNHHnnkkeqWMl7BHz/yRs8rkrKYSnTbbbclX1vxgSGL vXv3hp+qW9QQBiXCqEi+1ccff0xS98orr5hdtOJBF4jpFKkCaN6igAIKKKCAAgrkCbz44ouMG3zz zTc5U0Va1CnPq1YLztGKNjQkpwjGaj777LODBw+6vLsFj7dgSBOMgnDepoACCiiggAIK1BIgu3jp pZfGx8e/++67Wtd4PIrAJ598wiITVl+EPUaixDTICgVMMFYI6O0KKKCAAgoooECGwBNPPMELnd59 911ef5Rx2kMxBFgB8v777z/++OMPP/xwjHjGiCNgghHH0SgKKKCAAgoooEClAIuSX3755cnJya+/ /rryuJ8jCnz44YfkGC+88ILDFxFVVx7KBGPlhkZQQAEFFFBAAQUyBA4dOvTss8++9dZbldveZVzn oUICvMGW4Ys33niDN/8WCuBNrRIwwWiVrHEVUEABBRRQQIFXX32VVy29+eabvGhVjYgC5Gxvv/32 008//dhjj0UMa6goAiYYURgNooACCiiggAIKZAiw+8Trr79++PBhXlmbcdpDRQVY2837o1577bWi AbyvhQJutNdCXEMroIACCiiggAIIsNc1WzqwcYQasQTYQ/Daa6+NvrVfrOqVPI4JRsn/A9h8BRRQ QAEFFFBAAQViCjhFKqamsRRQQAEFFFBAAQUUKLmACUbJ/wPYfAUUUEABBRRQQAEFYgqYYMTUNJYC CiiggAIKKKCAAiUXMMEo+X8Am6+AAgoooIACCiigQEwBE4yYmsZSQAEFFFBAAQUUUKDkAiYYJf8P YPMVUEABBRRQQAEFFIgpYIIRU9NYCiiggAIKKKCAAgqUXMAEo+T/AWy+AgoooIACCiiggAIxBUww YmoaSwEFFFBAAQUUUECBkguYYJT8P4DNV0ABBRRQQAEFFFAgpoAJRkxNYymggAIKKKCAAgooUHIB E4yS/wew+QoooIACCiiggAIKxBQwwYipaSwFFFBAAQUUUEABBUouYIJR8v8ANl8BBRRQQAEFFFBA gZgCJhgxNY2lgAIKKKCAAgoooEDJBf4HM3gu+DH6fgQAAAAASUVORK5CYII= ) (1)(b) Write down W\_V^2 and W\_{out}^2 for head 2, such that it copies 3 more dimensions of the previous token, and one dimension from two tokens ago (using a dimension written to by the previous head). Note that there are many correct answers here. The primary important property is that W\_V^2 has one 1 in a column corresponding to a row of W\_O^1, and three 1s in columns which are untouched by W\_V^1 and W\_O^1. ![](data:image/PNG;base64,iVBORw0KGgoAAAANSUhEUgAABCAAAAHHCAYAAABulSn4AAAMbWlDQ1BJQ0MgUHJvZmlsZQAASImV VwdYU8kWnluSkJDQAghICb0J0gkgJYQWQHoRRCUkgYQSY0JQsaOLCq5dRLGiqyKKbaXZsSuLYu+L BRVlXdTFhsqbkICu+8r3zvfNvX/OnPlPuTO59wCg+YErkeSjWgAUiAulCeHBjDFp6QxSJ1AHOKAC OjDi8mQSVlxcNIAyeP+7vLsBEMX9qpOC65/z/1V0+AIZDwAkA+IsvoxXAPFxAPB1PIm0EACiQm85 uVCiwLMh1pXCACFeqcA5SrxDgbOU+PCATVICG+LLAKhRuVxpDgAa96CeUcTLgTwanyF2EfNFYgA0 R0AcwBNy+RArYh9RUDBRgSshtoP2EohhPICZ9R1nzt/4s4b4udycIazMa0DUQkQyST536v9Zmv8t BfnyQR82cFCF0ogERf6whrfyJkYpMBXibnFWTKyi1hB/EPGVdQcApQjlEclKe9SYJ2PD+gF9iF34 3JAoiI0hDhPnx0Sr9FnZojAOxHC3oFNEhZwkiA0gXiCQhSaqbDZJJyaofKH12VI2S6U/x5UO+FX4 eiDPS2ap+N8IBRwVP6ZRLExKhZgCsVWRKCUGYg2InWV5iVEqm1HFQnbMoI1UnqCI3wriBIE4PFjJ jxVlS8MSVPZlBbLBfLFNQhEnRoX3FwqTIpT1wU7xuAPxw1ywywIxK3mQRyAbEz2YC18QEqrMHXsu ECcnqng+SAqDE5RrcYokP05lj1sI8sMVeguIPWRFiaq1eEoh3JxKfjxbUhiXpIwTL87lRsYp48GX gmjABiGAAeRwZIGJIBeI2robuuEv5UwY4AIpyAEC4KTSDK5IHZgRw2siKAZ/QCQAsqF1wQOzAlAE 9V+GtMqrE8gemC0aWJEHnkJcAKJAPvwtH1glHvKWAp5Ajegf3rlw8GC8+XAo5v+9flD7TcOCmmiV Rj7okaE5aEkMJYYQI4hhRHvcCA/A/fBoeA2Cww1n4j6DeXyzJzwltBMeEa4TOgi3J4hKpD9EORp0 QP4wVS2yvq8FbgM5PfFg3B+yQ2ZcHzcCTrgH9MPCA6FnT6hlq+JWVIXxA/ffMvjuaajsyC5klDyM HES2+3GlhoOG5xCLotbf10cZa9ZQvdlDMz/6Z39XfT68R/1oiS3ADmBnsRPYeeww1gAY2DGsEWvF jijw0O56MrC7Br0lDMSTB3lE//DHVflUVFLmUuvS5fJZOVcomFKoOHjsiZKpUlGOsJDBgm8HAYMj 5jmPYLi5uLkCoHjXKP++3sYPvEMQ/dZvurm/A+B/rL+//9A3XeQxAPZ5w+Pf9E1nxwRAWx2Ac008 ubRIqcMVFwL8l9CEJ80QmAJLYAfzcQNewA8EgVAQCWJBEkgD42GVhXCfS8FkMB3MAaWgHCwFq8Ba sBFsATvAbrAfNIDD4AQ4Ay6Cy+A6uAt3Tyd4CXrAO9CHIAgJoSF0xBAxQ6wRR8QNYSIBSCgSjSQg aUgmkoOIETkyHZmLlCPLkbXIZqQG2Yc0ISeQ80g7cht5iHQhb5BPKIZSUV3UBLVBR6JMlIVGoUno ODQHnYQWo/PQxWglWo3uQuvRE+hF9Dragb5EezGAqWP6mDnmhDExNhaLpWPZmBSbiZVhFVg1Voc1 w+d8FevAurGPOBGn4wzcCe7gCDwZ5+GT8Jn4InwtvgOvx0/hV/GHeA/+lUAjGBMcCb4EDmEMIYcw mVBKqCBsIxwknIZnqZPwjkgk6hNtid7wLKYRc4nTiIuI64l7iMeJ7cTHxF4SiWRIciT5k2JJXFIh qZS0hrSLdIx0hdRJ+qCmrmam5qYWppauJlYrUatQ26l2VO2K2jO1PrIW2ZrsS44l88lTyUvIW8nN 5EvkTnIfRZtiS/GnJFFyKXMolZQ6ymnKPcpbdXV1C3Uf9Xh1kfps9Ur1vern1B+qf6TqUB2obGoG VU5dTN1OPU69TX1Lo9FsaEG0dFohbTGthnaS9oD2QYOu4azB0eBrzNKo0qjXuKLxSpOsaa3J0hyv WaxZoXlA85JmtxZZy0aLrcXVmqlVpdWkdVOrV5uu7aodq12gvUh7p/Z57ec6JB0bnVAdvs48nS06 J3Ue0zG6JZ1N59Hn0rfST9M7dYm6troc3Vzdct3dum26PXo6eh56KXpT9Kr0juh16GP6Nvoc/Xz9 Jfr79W/ofxpmMow1TDBs4bC6YVeGvTcYbhBkIDAoM9hjcN3gkyHDMNQwz3CZYYPhfSPcyMEo3miy 0Qaj00bdw3WH+w3nDS8bvn/4HWPU2ME4wXia8RbjVuNeE1OTcBOJyRqTkybdpvqmQaa5pitNj5p2 mdHNAsxEZivNjpm9YOgxWIx8RiXjFKPH3Ng8wlxuvtm8zbzPwtYi2aLEYo/FfUuKJdMy23KlZYtl j5WZ1Wir6Va1VnesydZMa6H1auuz1u9tbG1SbebbNNg8tzWw5dgW29ba3rOj2QXaTbKrtrtmT7Rn 2ufZr7e/7IA6eDoIHaocLjmijl6OIsf1ju0jCCN8RohHVI+46UR1YjkVOdU6PXTWd452LnFucH41 0mpk+shlI8+O/Ori6ZLvstXlrquOa6RriWuz6xs3BzeeW5XbNXeae5j7LPdG99cejh4Cjw0etzzp nqM953u2eH7x8vaSetV5dXlbeWd6r/O+ydRlxjEXMc/5EHyCfWb5HPb56OvlW+i73/dPPye/PL+d fs9H2Y4SjNo66rG/hT/Xf7N/RwAjIDNgU0BHoHkgN7A68FGQZRA/aFvQM5Y9K5e1i/Uq2CVYGnww +D3blz2DfTwECwkPKQtpC9UJTQ5dG/ogzCIsJ6w2rCfcM3xa+PEIQkRUxLKImxwTDo9Tw+mJ9I6c EXkqihqVGLU26lG0Q7Q0unk0Ojpy9IrR92KsY8QxDbEglhO7IvZ+nG3cpLhD8cT4uPiq+KcJrgnT E84m0hMnJO5MfJcUnLQk6W6yXbI8uSVFMyUjpSblfWpI6vLUjjEjx8wYczHNKE2U1phOSk9J35be OzZ07KqxnRmeGaUZN8bZjpsy7vx4o/H5449M0JzAnXAgk5CZmrkz8zM3llvN7c3iZK3L6uGxeat5 L/lB/JX8LoG/YLngWbZ/9vLs5zn+OStyuoSBwgpht4gtWit6nRuRuzH3fV5s3va8/vzU/D0FagWZ BU1iHXGe+NRE04lTJrZLHCWlko5JvpNWTeqRRkm3yRDZOFljoS78qG+V28l/kj8sCiiqKvowOWXy gSnaU8RTWqc6TF049VlxWPEv0/BpvGkt082nz5n+cAZrxuaZyMysmS2zLGfNm9U5O3z2jjmUOXlz fitxKVle8tfc1LnN80zmzZ73+Kfwn2pLNUqlpTfn+83fuABfIFrQttB94ZqFX8v4ZRfKXcoryj8v 4i268LPrz5U/9y/OXty2xGvJhqXEpeKlN5YFLtuxXHt58fLHK0avqF/JWFm28q9VE1adr/Co2Lia slq+uqMyurJxjdWapWs+rxWuvV4VXLVnnfG6hever+evv7IhaEPdRpON5Rs/bRJturU5fHN9tU11 xRbilqItT7embD37C/OXmm1G28q3fdku3t6xI2HHqRrvmpqdxjuX1KK18tquXRm7Lu8O2d1Y51S3 eY/+nvK9YK9874t9mftu7I/a33KAeaDuV+tf1x2kHyyrR+qn1vc0CBs6GtMa25sim1qa/ZoPHnI+ tP2w+eGqI3pHlhylHJ13tP9Y8bHe45Lj3SdyTjxumdBy9+SYk9dOxZ9qOx11+tyZsDMnz7LOHjvn f+7wed/zTReYFxouel2sb/VsPfib528H27za6i95X2q87HO5uX1U+9ErgVdOXA25euYa59rF6zHX 228k37h1M+Nmxy3+ree382+/vlN0p+/u7HuEe2X3te5XPDB+UP27/e97Orw6jjwMedj6KPHR3ce8 xy+fyJ587pz3lPa04pnZs5rnbs8Pd4V1XX4x9kXnS8nLvu7SP7T/WPfK7tWvfwb92dozpqfztfR1 /5tFbw3fbv/L46+W3rjeB+8K3vW9L/tg+GHHR+bHs59SPz3rm/yZ9Lnyi/2X5q9RX+/1F/T3S7hS 7sCnAAYHmp0NwJvtANDSAKDDvo0yVtkLDgii7F8HEPhPWNkvDogXAHXw+z2+G37d3ARg71bYfkF+ TdirxtEASPIBqLv70FCJLNvdTclFhX0K4UF//1vYs5FWAPBlaX9/X3V//5ctMFjYOx4XK3tQhRBh z7Ap7ktWQRb4N6LsT7/L8cc7UETgAX68/wsztZC2JFMiHQAAADhlWElmTU0AKgAAAAgAAYdpAAQA AAABAAAAGgAAAAAAAqACAAQAAAABAAAEIKADAAQAAAABAAABxwAAAADVTl+lAABAAElEQVR4Aey9 CZRkR3nnG117dW29VO/Ve7darZYaIVtiHUuHh+FJHM/igcE2zMEYZnl4mMHGNswbzwDGzBmDj+eM MZyxQWPs8xAGfGw4COQRNghjPAaBQBLd6n3fu3qr7q6uvV78ovor3cq6lZWZlXEzK+sfUnXmvTdu xBe/iLgZ33e/iFg07oNTEAEREAEREAEREAEREAEREAEREAEREIGIBOoipq2kRUAEREAEREAEREAE REAEREAEREAERCAQkAFCDUEEREAEREAEREAEREAEREAEREAERCA6ARkgoiNWBiIgAiIgAiIgAiIg AiIgAiIgAiIgAjJAqA2IgAiIgAiIgAiIgAiIgAiIgAiIgAhEJyADRHTEykAEREAEREAEREAEREAE REAEREAEREAGCLUBERABERABERABERABERABERABERCB6ARkgIiOWBmIgAiIgAiIgAiIgAiIgAiI gAiIgAjIAKE2IAIiIAIiIAIiIAIiIAIiIAIiIAIiEJ2ADBDRESsDERABERABERABERABERABERAB ERABGSDUBkRABERABERABERABERABERABERABKITkAEiOmJlIAIiIAIiIAIiIAIiIAIiIAIiIAIi IAOE2oAIiIAIiIAIiIAIiIAIiIAIiIAIiEB0AjJAREesDERABERABERABERABERABERABERABGSA UBsQAREQAREQAREQAREQAREQAREQARGITkAGiOiIlYEIiIAIiIAIiIAIiIAIiIAIiIAIiIAMEGoD IiACIiACIiACIiACIiACFSFw/Phx9+ijj7r+/v6K5K9M505gdHTUfeYzn3F79+6de2JKoeYJyABR 81WsAoqACIiACIiACIiACIhA9RG4efOm+y//5b+4uro6t3jx4uoTUBIVRKC+vt4tXbrU/df/+l9d X19fQfco0sIlIAPEwq17lVwEREAEREAEREAEREAEKkKAt+Z/+Id/6IaGhtwv/MIvVEQGZVo+Am94 wxvc8uXL3Uc/+lE3PDxcvoSVUs0RkAGi5qpUBRIBERABERABERABERCB6ibw1FNPuccff9z95//8 n11zc3N1CyvpZiXQ0NDgfvVXf9V94xvfCPU66w2KsGAJNCzYkqvgIiACIiACIiACIiACIiACmRPA Tf/3fu/33Otf/3p31113peZ/48aN4M6Pp0RbW1tw8V+0aFFq3FgnBwcH3ZUrV4KXBkaS7u5ux3SD LMPY2Jjr7e11yNLU1OSWLFmSucEGGa5eveqYMkP5u7q6Qp3kcujp6XFvectb3B/8wR+4V7ziFW71 6tW5UXQsAk4eEGoEIiACIiACIiACIiACIiACmRDAPf9P//RPg1L/5je/eVqe4+Pj7syZM27//v1u YGDANTY2urNnz4YFDlHCswoYHvbs2ROMICj+169fd88//7y7du1aViKE8pPnpUuXAgcMAMh07ty5 zGRgiszBgwfdiRMngvEBgxB1c/LkyWkyYJx461vf6jo6Otwf/dEfOQwXCiKQS0AGiFwiOhYBERAB ERABERABERABEYhC4Mc//rH70pe+5F772te6TZs2TcsDBf/o0aPh7fmWLVvC586dO8MuGceOHctE qR0ZGQlKNp4X27dvDzJs27YteB7s27cvkzUOUN4PHDgQ8rrzzjsnebDOwuHDh4M3wjR4EU6wSwnG mB07dri1a9e69evXuw0bNgSDBJ4ZuQHvCNaDeOKJJ9y3v/3t3Ms6FgF5QKgNiIAIiIAIiIAIiIAI iIAIZEPgC1/4grt165Z75zvfOS1DvB9OnToVjAxJ9312yVixYoW7cOFCJh4Ip0+fdrzpX7NmjUtO +0ABxyPg4sWL02Qv94nLly8H7wvyTMoAF3jAKXbA4wTvi2XLlrnW1tbJ7DCCsOYDxgk45QYWFW1p aXGf/exncy/pWARkgFAbEAEREAEREAEREAEREAERiE+A6QR/8zd/4x5++OGwY0JujkzPYH0IlNfc tRZsm07exscMKNR4YaDk5y6OiVyErAwQGGTwwkgGFH9kY00GPDViBowg1ImV2/LCIEJ99Pf3B4OM nbdPZH7kkUfcM888E+rbzutTBCCgKRhqByIgAiIgAiIgAiIgAiIgAlEJoEz/r//1v8JCij/90z+d mhdrPmAAyDU+ENnOofTGDCj1vPlHyUbRzw2cYz0IyhMz4CVCwOCQDOSPbCZn8lq5v1NOQhoHk2um +mAaBoaIP/mTP8lk2ky5y6704hGY3qvi5aWURUAEREAEREAEREAEREAEFiABFjL8wQ9+4FjPYKad L5jeQEhTeO2cxYmFkDf+5lmQnPpAfhzzh/GBeLEC6Vs5rdyWl8nEGhEmp10r9+dMMpCPyTXTwqDU 80/8xE+EBSxZOFNBBIyADBBGQp8iIAIiIAIiIAIiIAIiIAJRCDz++OPBZf+hhx4KuySkZWK7JpiS nYxj59LWHEjGm+t3lH+TI19aMeVABv4IVu40WWIbIPKV0eSaKQ4GChYaxVDz2GOPRTeWpPHRueok IANEddaLpBIBERABERABERABERCBmiDAWgLf/OY3Q1le/epXz1gmU7pnjJDBhaTyn5adKd5p18p1 rho4zFaWQmR8+ctfHqbO/O3f/m0m62bMJrOuVwcBGSCqox4khQiIgAiIgAiIgAiIgAjUJIHvfve7 7uzZs2HqxaaUrTet0LbOQ5pya+fM9d/uKfcnMuTLIws5kvlbfmnlNF5p18pxrpD0k7Lm5snuGXi8 ME3jO9/5Tu5lHS9QAjJALNCKV7FFQAREQAREQAREQAREIAsC3//+90M2KKP5QlNTU7icpnTbucbG xnxJzPkaSrcp3pZnMlE7Z7Imr5XrO14WVs7c6SCWP3FMznLlm5vOTDIQz+SyOLn32vGDDz4Yvn71 q1+1U/pc4ARkgFjgDUDFFwEREAEREAEREAEREIFYBNguEg8I3pS/8pWvzJsNSj2KtS1+mIxsiz7a dpzJa+X8jgxsO4mib0q2pc8xf8gQeyqGldPKbTKw5gKyYXyIaQQhP9sCNG2tCeSCgcUx+XI/d+/e HWRlC9Zjx47lXtbxAiQgA8QCrHQVWQREQAREQAREQAREQASyIPD000+78+fPO9zxe3p68mbJ23QU WpTbXOWfLTpReJcsWZI3jblexFDS1dUV8s9V/m3Hh+XLl881m1nvt3Ladpx2A8YAjBDt7e3RDRDU 2UwGIeRqbm52ra2tJlrqJyyt3p955pnUODq5sAjIALGw6lulFQEREAEREAEREAEREIHMCGCA4I39 xo0bZ31bzlt9U1YvXbo0KSPGCBay7OzsDMaByQuRvqxcuTLIjOEkGXp7e8PbfK7HDkuXLg2eGBcu XJiS1ZUrV4IBYv369dG9MDAuIMe1a9emeKX09fWF4zVr1uRdLwPBOzo63K5du0IZ2IZVQQRkgFAb EAEREAEREAEREAEREAERKDsB3tTv378/pHvfffcVpDB3d3c7FNujR486lG+mcBw6dCgo3Vu3bp1c G6HswiYSZPrD9u3bQ/4nT54MCvjx48fD8bZt28IUjET0KF+ZXoEMeF0cOXIkcDh9+rRDnk1+IU/z kIiS+e1E8X6AeUNDgztw4EDggFHm8OHDbvXq1aGeCsn/Fa94RYjGNIxcg0oh9ytObRFoqK3iqDQi IAIiIAIiIAIiIAIiIALVQADjwYkTJ8JbctYCKCSg9G7ZsiV4OnA/RgzWZLjnnnuCy38haZQjDl4O eADg9XDu3Llg+KAMs615UI68LQ2MDDt37gzeH8iAIQADCFMjsgqwv+uuuxweKchA/axdu9bBh++F BOqTdDA+sA5EFh4khcilOJUhIANEZbgrVxEQAREQAREQAREQARGoaQLPPfecu379elDabWpFIQVG scUTgr9KBWRgygd/lQxMYeCvkoG1HjA6lBrwaMGYw/QRPGIeeOCBUpPSfTVAQFMwaqASVQQREAER EAEREAEREAERqDYCzz77bBCJt98rVqyoNvEkT0YEMOLYwp1M5VBY2ARkgFjY9a/Si4AIiIAIiIAI iIAIiEAUAgcPHgzp4nLPW3SFhUvAPGBkgFi4bcBKLgOEkdCnCIiACIiACIiACIiACIhA2QjYApSb /KKJha4XULbMlVBVETADBGuCsJWowsIlIAPEwq17lVwEREAEREAEREAEREAEohC4ceNG2DWBxFkD QGFhEzADxPDwcNjJY2HTWNillwFiYde/Si8CIiACIiACIiACIiACZSdw6tQpNz4+HtLVrgdlxzvv EmQnDAsvvPCCfdXnAiQgA8QCrHQVWQREQAREQAREQAREQARiErh48eJk8kuXLp38ri8Lk8CqVasm C3748OHJ7/qy8AjIALHw6lwlFgEREAEREAEREAEREIGoBPCAsGA7INixPhceAQwQdXUTqmdvb+/C A6ASTxKQAWIShb6IgAiIgAiIgAiIgAiIgAiUg8CxY8dCMiw+2d3dXY4klcY8JtDQ0OC6urpCCc6f Pz+PSyLR50pABoi5EtT9IiACIiACIiACIiACIiACUwjYW+729nbX0tIy5ZoOFiaBzs7OUPArV64s TAAqdSDQIA4iIAIiIAIiIAIiIAIiIAIiUE4Cly5dCsmhdNbX1xedNFs1Dg4OutHRUdfa2uoaGxuL TmOuN5A3MrBzQ3Nzc/jLejtRFvIcGBhwQ0NDrqmpKchgUxnmWr5i7id/5MCTAYNSKTJgjCLcvHkz 1Gsp7aIYmRW3OgnIAFGd9SKpREAEREAEREAEREAERGBeEkBpvnz5cpAdZRWltZhw7do1d/DgwaBw Y3zgmHUkNmzYUJIxo5i8LS6Gh/379wfjA0aU/v7+YATZunVrMAJYvJifGGEOHToUyr9s2bJgDBkb G3ObNm1y5k0QM3/Spi5Pnjzpzp07F6ZQIBPGCOqi2LU92tragrikAc+Ojo7Y4iv9KiRQ3NOgCgsg kURABERABERABERABERABKqHAG/K8RogLF68uCijAYop2zSuXr06KLm8aSe9H/3oR0EZTm7nGKvE KPk//vGPg8Hh3nvvDfLjDfH888+7vXv3ut27dxdVplLkRIYjR464q1evuvvuuy8YYzAGYJBABjtX StrF3MNiovzt3LnT2W4mZ86ccfv27QscijEi0BYI5k1RzL3FyKy41U1Aa0BUd/1IOhEQAREQAREQ AREQARGYVwRu3bpVsgHixIkT4d6enp5JN3+8KFauXOnOnj3r+vr6orPgbT+GkPXr108aGpgusHHj Rnfjxg1n61vEFATDw4ULF9y6deuC8YG8mP7BMcYQvBJiB/KBOYYHMz6QJ8YhDEMYSDCUFBrMA4J7 MCopLEwCMkAszHpXqUVABERABERABERABEQgCgG8H1BeCUyhKHSuP/cx3YK1DnKnbfD2HMU19gKG Nn0EmZE9GezY1rdIXiv3dwwQyGLrJlj6xoYpLsUo/3Z/MZ/IwFQUK7fdi/EBYwLGGLwZCg2so0Fg CoZ5yBR6r+LVDgEZIGqnLlUSERABERABERABERABEag4AZRLlGdCMQYIlF0U2rQFJ+0cCxjGDDY9 ACWbv2TAA4E/jCBWvuT1cn5HuSdYuS1tZEIGk9POx/g0Y0+uMYi8OIcBBG+XQoMZojBOFWO4KDR9 xZsfBKb2qvkhs6QUAREQAREQAREQAREQARGoUgIomPZ23hTmQkTFAEHIVfyT52Irrsm387lyoPhz jrIRL1bAuGHlNKXd8jKZkCG2F4FNk6DcucHksjrLvZ52bIYMyheTX1reOlc9BGSAqJ66kCQiIAIi IAIiIAIiIAIiMO8JoFyaAcKUzkIKZdM20hReOxdb6UZukz2fzDEVaBR0k8HKnSaL8Uq7Vo5zhchQ DAczWpBubNnLUX6lEYeADBBxuCpVERABERABERABERABEViQBJIGCFM6CwERe1pDoTLkkyOfQaCQ 9AuJky//5P2FxkveU8z3QtIvJI7laW2Be2SAMCoL71PbcC68OleJRUAEREAEREAEREAERCAaAZRL XPPxVihG0bTpBfkEi20AQIZ8eZjCnS9OPvkLuZbkYPml3ZeMl3Z9rucKKWMxMuD5wNQS0s1XrrnK rfurm4AMENVdP5JOBERABERABERABERABOYVARRNtrFE2SxmyoQtuGiu/8lC2zmLk7xWzu+8pU++ qZ8p7ZhyoKDPJIMp7sk4M8k41/NWRmOfTM/OWZzktZm+4xnDIqLIbvfPFFfna5eADBC1W7cqmQiI gAiIgAiIgAiIgAhkTgCldOnSpcEAkbuFYz5hWlpagnKatq6AnSNOzIDsbHWJB0eukozyzzmum4Eg liyUEyOOldvyIX/kwPMAOWIGq7tcDuRpctnWmoXIwdadtAtCbH6FyKM4lSGgNSAqw125ioAIiIAI iIAIiIAIiIAIJAig/KPQ4jlhb/rtsu0K0dnZaaeifLJoZnt7ezA05E4fQelGrmXLlkXJO5loR0dH OLSdKOwaMmAQgFMxyr/dX8zn8uXLQ/Q0Lxbqg/pavHhxwUkm6zT5veAEFLEmCMgAURPVqEKIgAiI gAiIgAiIgAiIwPwmwFvx1atXByX/2rVrk4VBWb1y5Yrjjby9QZ+8GOHLypUrg5J/+fLlKalfunQp eGggY+zQ3d0dPBxyZWAKA4aRDRs2BFliyoFxAYNPX1/fpMcD+d26dSt4Z6xYsSIYIWLKoLRrj4AM ELVXpyqRCIiACIiACIiACIiACFSMAAaDUt9wr1mzJnggnDhxwvH2n3TOnz/vMEj09PS42FMwgNbV 1RUMIadOnQr5cg4l/MyZM+G8eSdwPlbA2EJ5MUD09vYGDij+cCF/DBSxAwah9evXh3U8Tp48GYwy TE05evRoqAeuKYhAsQS0BkSxxBRfBERABERABERABERABEQgCgGmQNxzzz1Byd2zZ09QvFnrYPfu 3eFtfJRMcxJlkcRt27aF6QVHjhwJb/+Ra/PmzW7VqlU5seMcIsO6deuCFwSGEJR+ZMDwgGGC61kE ppvceeedDgPED37wg5AvBpr77rtP6zhkUQE1mIcMEDVYqSqSCIiACIiACIiACIiACMxXArx5xwBA wAMiK2U7ycsMABgBKiUD8jDNgb9KysA0jF27dgUZkKkS9UG+CrVBQAaI2qhHlUIEREAEREAEREAE REAEao5ANSi7kmGiWVUDh5pr4AuwQFoDYgFWuoosAiIgAiIgAiIgAiIgAiIgAiIgAlkTkAEia+LK TwREQAREQAREQAREQAREQAREQAQWIAEZIBZgpavIIiACIiACIiACIiACIiACIiACIpA1ARkgsiau /ERABERABERABERABESghgmwVoCtF2CfNVxcFa0EAmoXJUCrkVtkgKiRilQxREAEREAEREAEREAE RKAaCCSVy7o6qRvVUCfVIIO1Cz7tezXIJRmyJaBdMLLlrdxEQAREQAREQAREQAREoKYJDA8Pu76+ Pjc4OOgGBgZquqwqXOEEbt265a5du+YwSo2NjRV+o2LWFAEZIGqqOlUYERABERABERABERABEags gYaGBtfZ2emGhoZcY2NjZYVR7lVDoLW11XV1dQV55AFRNdWSuSDyicocuTIUAREQAREQAREQAREQ AREQgYVFYHx8fLLAye+TJ/VlQRCQAWJBVLMKKQIiIAIiIAIiIAIiIAIiIAIiIAKVJaApGJXlr9xF QAREQAREQAREQAREQARyCLBWwKVLl8JaAW1tbW716tWZL1zI+hXnzp1zrGnR0tISZMh6Ssno6Kg7 c+ZMWEujubnZdXd3u8WLF+fQinvIeg0XL14M63pQ/qVLl05OpYibs1KvRQLygKjFWlWZREAEREAE REAEREAERKBCBHCvL9XFnvuOHj3qDhw4EBTt5cuXh4ULf/SjH7nr169nVqLz58+7Z599NiyYuHLl Soch4Ic//GEwipRatmKFZyHPH/zgB2EtDQwPrK2xZ88ed+rUqWKTKjk+C4nCHgMEdcE6DocPH3YH Dx7UQpIlU13YN8oAsbDrX6UXAREQAREQAREQAREQgaohcOXKlaBgr1+/Pngc8Lb9jjvuCAaN48eP Z6L0sngmSjYK94YNG8Lb/k2bNrklS5YEwwhKeewwMjISZMDosGXLluB1sHbtWscfHDBOxA5mDMID ZPv27W7ZsmVu1apVbvPmze7ChQsOI42CCBRLQAaIYokpvgiIgAiIgAiIgAiIgAiIQNkJoPCePHky eB2g6Fpg28YVK1Y4jBOXL1+209E+T58+HQweTPtIBpR/DAMo37ED009u3Ljh1q1bN2XqCZ4Q8IBT 7IChBTnIk+kfFjAKcYwnBp4hCiJQDAEZIIqhpbgiIAIiIAIiIAIiIAIiIAJRCPCmnWkWTU1NU5Ru MjMF+OrVq1HytkRRqFl/AiU/d70Hjtk+EqU8dkAGAmtPJAMeEciGcQJeMYOtwZHLgTyRizUysvAG iVlGpZ09ARkgsmeuHEVABERABERABERABERABHIIoNDiBYGSnRvsHHFiBpR6pmBgaEDRTwbO8Xfz 5s2S17hIppfve39/f7icq/wjEzIgZ2zlHyMHIZcD5+rr6/lwJmc40D8iUACBqb2qgBsURQREQARE QAREQAREQAREQATKTQDFn5Cm8Nq52Eo3Uyz4I6Do5wbOYSQxWXOvl+M4mX6uDHZMnNgeEJa+sU+W zc7F5JDMT99rh4AMELVTlyqJCIiACIiACIiACIiACFScAEqyKcrFCMN2j7OFQuLMlka+6yj2heRR SJx8+eS7hgwW8nGMKQP551vfweTKF8fKkPZp96dd07naJiADRG3Xr0onAiIgAiIgAiIgAiIgAhUj YG/KKyZAkRknlf8iby1b9GqQYbbClCKjGR34tO+z5aPrtUdg+gSr2iujSiQCIiACIiACIiACIiAC IpARAd6Ks5gk0yWKmTJhxoo05dbOWZxYRWFtg0LyKCROqTIm07Zyp6Vl6zCkXSvHuULST8o6W560 BbYP5Z585ZotHV2f3wRkgJjf9SfpRUAEREAEREAEREAERKCqCLBgZFtbW9hFIncRxXyCsvsFIW1q gZ2zOPnSmcs1lG7kJz/L09JDaTbFOaYceAfADYU9TQbkQYkvxEBgspfyaXWXKwNp2bliOBC3vb09 iGIcS5FL98xvApqCMb/rT9KLgAiIgAiIgAiIgAiIQFURQLlEQS7m7TgFYKtN7klb2NAWRDQFNlaB kYE/ymBKtuVl55Ah9hSCxYsXh2xzWeBdglxwsq1JTb5yf3Z0dIQkbVHOZPrIBQMMTYUG4lu7iM2v UJkUL3sCMkBkz1w5ioAIiIAIiIAIiIAIiIAI5BDgjTsKLQpvrvKPNwBKa2dnZ85d5T0kjyVLloQF GHMVb5tOsmzZsvJmmpIaMhBytx3FEAMbjAPmoZBye1lOLV26NDDPNYKQ+K1bt1xLS0v4KzSzpNdD 8nuh9ytebRCQAaI26lGlEAEREAEREAEREAEREIF5TYC345s3bw5luHDhwmRZULovXrwYDAOmmE9e jPBlzZo14U39iRMnJlNHYT5z5kxQuFevXj15PtaX5cuXByPDqVOnJqd9kNfly5eDAWLLli3RvTAw MKxatcpduXLF9ff3Txa1t7c3bAHa09MTOE1e0BcRKICA1oAoAJKiiIAIiIAIiIAIiIAIiIAIFEYA Zb3UN9x4OGzYsMEdP348vGXHIwLjA+sybN++Pfq6B5SQtQp27tzp9u3b5w4fPhwMASyqeePGDbd1 69boUx+QwcqLDHv37nXd3d2BB4YZjA82RYO4sQLeINQFHhD79+93GF7wvsAQw3mMEwoiUCwBGSCK Jab4IiACIiACIiACIiACIiACUQig9K5fv951dXWFHRN4875y5UrHdACU8qwC+b30pS8NHgfI0Nra 6nbv3p2J8cHKiPHl7rvvDjIw5YEpF7t27Spq3QVLq9RP1pnYsWPHpBcEXiocMwWEulIQgWIJZNeL i5VM8UVABERABERABERABERABBYcARRbDBD8VTIwBWHt2rWVFCEYPJgSUsmA4WfFihWVFEF51xAB rQFRQ5WpooiACIiACIiACIiACIiACIiACIhAtRKQAaJaa0ZyiYAIiIAIiIAIiIAIiIAIiIAIiEAN EZABooYqU0URAREQAREQAREQAREQAREQAREQgWolIANEtdaM5BIBERABERABERABERABERCBGiSg BSxrsFILLJIMEAWCUjQREAEREAEREAEREAEREIHZCaBcmoLJrgkKIpAkYG0jeU7fFw4B7YKxcOpa JRUBERABERABERABERCB6ATGx8fd0NBQ+OO7gghAYGxsLLQJjFIyQizcNiEDxMKte5VcBERABERA BERABERABKIQQNnkb3h4OEr6SnT+ERgdHQ1tAqMU3xUWJgEZIBZmvavUIiACIiACIiACIiACIhCF AG+3W1paHG+6GxqkbkSBPA8TbWpqCu0C0TU1Zx5WYJlE1qSsMoFUMiIgAiIgAiIgAiIgAiIgAiIg AukEktNxkt/TY+tsrRKQAaJWa1blEgEREAEREAEREAEREIF5TIApHCMjIxUtAYoyMlRaYa60DMaB OlEQgbkQkE/UXOjpXhEQAREQAREQAREQAREQgSkEUFbnorCzPsCRI0fctWvXgqt+fX2927Rpk+vq 6pqST8wDFO1Tp065CxcuOPKnPD09PW7lypUxs52Wdm9vrztx4kRYtBEZuru7gxxZTmHo6+tzx48f DwtIImBnZ6fbvHmzptdMqy2dKISADBCFUFIcERABERABERABERABERCB6ATYPeP55593zc3N7iUv eYlrbGx0V69eDefuuOOOTAwAKPoHDhxwN27ccLt27XKtra1ucHDQ/fCHP3SXL192O3bsyGQXh9On T7tjx44FDu3t7cEAsG/fPodR4t57781kHYWLFy+6Q4cOuXXr1rn169cHQ8zhw4cDC2SgfhREoBgC moJRDC3FFQEREAEREAEREAEREAERiEbg7Nmzrr+/P7xhN+V2yZIlbtmyZe7o0aPBEBAt89sJY/BA 8UbhxvhAwCCCFwbKPx4BscOtW7eC1wEeFxgfCCziuGHDhsAHTrEDnih4gbS1tQUWLC6K5wVcMBRR H3PxdIktv9KvTgIyQFRnvUgqERABERABERABERABEVhQBFB4UfDZOWPx4sVTys70C5RerscMKNQ2 7aKjo2NKVhxzHeNE7EAe8MDwkgwYA+CDjLGV/5s3bzr+MIBgfLDADifUz6VLl7TNqkHRZ8EEZIAo GJUiioAIiIAIiIAIiIAIiIAIxCLANAfe/POmP6nwkh8eCASmRcQMLPaI0k3+rP2QDCj+nEfxjh3M y8K8QCw/ZMILYWBgYHJNBrtW7k/KiZEjVwbyoY5ghRwKIlAMARkgiqGluCIgAiIgAiIgAiIgAiIg AlEIYIBA4U1bYNGMAbEV3uHh4fBWH0NDrhyc4w9PDLwTYgZYEDB6JAP5E1D+kSNmMNa5HMgzq/qI WT6lXRkCMkBUhrtyFQEREAEREAEREAEREAERSBBAqSakKbymeGMgiBkwLJhxwfJM5mfnYsqBEWYm GSx/ZDJeSfnK+d3ST+Zp6du52EYQy0+ftUNABojaqUuVRAREQAREQAREQAREQAQqTgDl1BRU+yxE KLa+nC0UEme2NPJdR/nPl4eVJ1+cfOkXci2ZtuWXdl8yXtr1uZ7Ll77JBa9iA/fa/cXeq/jzn8BU n575Xx6VQAREQAREQAREQAREQAREoIIEUEpNkU/zZphJtEKU0kLizJR+IednU45N4S6mXIXkm4yT LKPll7xu35Px7Fw5P/Olb3Lli5MrC3ExavBZzH256eh4fhOQAWJ+15+kFwEREAEREAEREAEREIGq IsD0gevXr4ctM20dgUIEtPUO0t68m8JrcQpJr5Q4GBbMuGB5pqVjayCkXZvrOfI3BT1XhuRxTBko g6WfzNPKZucsjp3P98kCo7QLypdWx/nu1bXaISADRO3UpUoiAiIgAiIgAiIgAiIgAhUngJGgs7Mz LJKYtoPCTALaThdpyqmdszgzpTHX8+zugPysf2BKtqXJMXKgQMc2hFBOjDe2FkRSBuRA8S+Grd1f zCfbbRKMffJeO2dxktdm+t7a2urYTpVgBpaZ4up87RLQGhC1W7cqmQiIgAiIgAiIgAiIgAjMGwIo /yjVtgNEUnDzpOjo6EieLvt38m9rawvGh1zln2OU/6VLl0ZXoDHgEHIXeTTDCAaQ2MYYykmwxSjD we1/qCOMIBgVCg1Jg07ye6H3K15tEJABojbqUaUQAREQAREQAREQAREQgXlNAKV62bJl4a2/GRys QP39/cHrYMmSJXYq2ufy5cvDW/+bN29OycOOu7u7p5yPcQAHvASuXbs2JXkUfwwCK1eunJwiMSVC GQ8wgmAUsnJb0uRPfXC9GA8Iu1+fC5uADBALu/5VehEQAREQAREQAREQAREoK4FS326jcG/YsCG8 2T958uSk6z/rBly6dMmtXbvWtbe3l1XWtMQwMOBpceLEiUlvDBR/jvEKyMIAQf4YGS5cuBCUfeRE 8T99+nTgs27dujTRy3oOg9D69etdX1+f6+3tDWnjBXLmzJlg/KCuFESgWAJaA6JYYoovAiIgAiIg AiIgAiIgAiKQl0CpRgjeqO/YscMdPXrUPf/888HFnzfwq1evDspwFmsHMLXgjjvucIcOHXL79u0L MrCAIsaPjRs3Ti5SmRfAHC9Szq1bt4YpH/v37w/TQvAKYf2JnTt3Rl//wcRfs2ZNMAQdP348GIEw QDAthDqyaSIWV58iUAgBGSAKoaQ4IiACIiACIiACIiACIiACmRBgocKXvOQlQdFF4eVNPFMBsgys bXDPPfcEDwiTgfUhsjCAWDnNEILnw/Dw8OTCkxghsgqUt6enx2GIwPDAMRyQTUEESiEgA0Qp1HSP CIiACIiACIiACIiACIhANAIourEXWSxE+ErLYAo/Sn8lQ7ELTlZSVuVd3QSyM59VNwdJJwIiIAIi IAIiIAIiIAIiIAIiIAIiEJGADBAR4SppERABERABERABERABERABERABERCBCQIyQKgliIAIiIAI iIAIiIAIiIAIiIAIiIAIRCegNSCiI1YGIiACIiACIiACIiACIrBwCLBuAX8zhrFhNz7Q58ZHB2eM ogu1R6Cjrt/1LJ1Yy6Jp+LIbu36m9gqpEqUT4JnQ2OYWNXc6GSDSEemsCIiACIiACIiACIiACIhA CQTGxsbC1o18TjNEjI+6sf5e5/eXLCFl3TKvCXib1LhvE75RTG8X87pgEn5WAr6/jw/dCHUvA8Ss tBRBBERABERABERABERABESgUAJs13j69OmwdeSVK1em3DY+MiDjwxQiC+eg79o1d+HihWB8GBoe WjgFV0knCYwP3ZQHxCQNfREBERABERABERABERABEZgzgZaWFrdu3TqHIaKrq2tqeuP+DbjCgiTQ 2dnpVqxYGcreUF/ZbUUXZAVUQ6F9/5cHRDVUhGQQAREQAREQAREQAREQgRoiUFdX5/ibNgWjhsqo ohRHgLZAmyDkWyKkuFQVe74R0C4Y863GJK8IiIAIiIAIiIAIiIAIiIAIzDMCWvZjnlVYJHHlAREJ rJIVAREQAREQAREQAREQgYVIYLyMmmZvb69jSkd7e3vmKMfGRt3AwJAbGRl2zc3Nrqmp0b+5z/b9 7bh3WWcqy+DgkGtsbAxymBdBdkDGvQzDnsVABWXIrrTKKS4BGSDi8lXqIiACIiACIiACIiACIrDg CMzFCHHz5k138uRJ96MfPeuefPJJ93M/92b3ute9LlOGKNvHjx8PU0ja2trc+fPn/cYd427Txo2u yRsjsggYHuCA8t/R2e6GvBGCv7Xr1jrWU8giUGYWFL12rc+v59HpjTEjrv9mv1u1ZrVbvmxZkSJo 55MigdVkdBkgarJaVSgREAEREAEREAEREAERmH8Evv3tb7u//MsvhTf9V69edfxlHVCyDx067GVo clu3bHF19fV+W9FRd/DgIXfo8GF3xx13uIaGuGoUW5ieOnnKYYzZuXNn8DzAGHD8xHF35MhRf+7O wCg2m7Pnzjq8ULZ4Dmb04Pj4sWOuyXtkdHR0xBZB6dcYgWx9iGoMnoojAiIgAiIgAiIgAiIgAiJQ PgI/+ZM/6d7//ve53/7tD7tXvvKV5Uu4iJQuXbrkBgYH3KpVq4LxgVvr6urd6tVrXH9/v/cGuFZE aqVFvX79urt89Ypb0b0iGB9IhUUcV61cFYwheGTEDqOjI673Yq/r7OicND6QJ54PTAM5ffqM9wrR riax66HW0pcBotZqVOURAREQAREQAREQAREQgXlKoLW11a1cudLVe6+DSgS8DDAw1HsFu7lp6lSL lpamINLly5eji9bX1+e8du/a2tum5NXU1BTY4BmCl0TM0Nd3Paw/0Xy73JbXIs+mtaXV3bh5ww0P D9tpfYpAQQRkgCgIkyKJgAiIgAiIgAiIgAiIgAjUOoFhv+4C6yyELSPrp6pKLEDJH4o5hoqYgakX hFxDTJ33gkA2pokMDg7GFMGX0xtBggzTp5vUN3gDkWfQ3z8QVQYlXnsEpvaq2iufSiQCIiACIiAC IiACIiACIiACBREYGR11I37qgZ/wEBT95E0o/v7/MO1g1BsAYgWMG+ZZgCdGMiyq8wL4MObjsDtH zMAimATKnRtsJ47hyDLk5qvj+U9gaoue/+VRCURABERABERABERABERABCpIYEJRn660VlCkgrNG +WdqQ4rOHdIwZXx0NN70B2SY9LBIESTI4OPElIHCjt0uo5U5CdHOFWeImZ9tIllufZ87ARkg5s5Q KYiACIiACIiACIiACIiACNwmYMoph/amfP7AQfmfXdoCosyeyAwxJgwQM1xMno4phM/Hk0jmlvp9 0lCSenXqSbOl0D6YyqKwMAlMn9CzMDmo1CIgAiIgAiIgAiIgAiIgAmUggOs+iySyRsGtW7fmnGIx Su6cMwtTL8LyBnmTivkuH6ONKetp1hDjYdMx8go6h4tmSLL8piY1YZwoxsA0MDDg+q73Odax0O4Z U2kupCMZIBZSbausIiACIiACIiACIiACIhCZQGNjo1uyZEnYQaGlpSVybuVNvqG+IWy5yfv/GZV/ r0A3NjWWN+NEaij+9V4O54amiWDeGRNx4u4U0tA4oSqmGSDGxiYMEA0NhauT7HDClp4YV8y4kSi2 vi4QAvJ9WSAVrWKKgAiIgAiIgAiIgAiIQNYE0pTXYmXI8m15Y1ODw4CC3KZkm7zIwfkmf72YN/92 fzGfzc0TW35OLIj54p2sT4EM5I+cMQNbbRLStvsc9Yt1Epqbp25VGk7O8I+1BYwo9n2GqDpdwwRk gKjhylXRREAEREAEREAEREAERGC+E0BhzSrU1dW79rbFQUEeHZtQsi3v4eGRcB7vjtihvb0jZDF4 eycKy4/tN1Hem5tbXFNELwzy6+rsDNmm7bbB9JqmxibXUpQBwkqhz4VMQAaIhVz7KrsIiIAIiIAI iIAIiIAIVCkB83xIewMfU+Sly5aFbS77+vqmZMMxiycuX758yvkYB0uWdAUPh75rV6ckf+vWgN/9 YtStWr0qyDLlYpkPWhcv9saYNnfjxk2/5eeL245ifOBvydIu1xDZC6PMRVJyVUCg8Ek7VSCsRBAB ERABERABERABERABEahuAryhL9XF/vDhw+7rX/+66+/vd3v27A0F/cIXvuAOHDjo2rxnws/+8591 3cu7owLo6OhwK7q73enTZ9xir4R3dLS769dvurPnzrlVq1Z6Odqi5k/irJ2xZu1ad/LESXf5yhW3 1HtdYHw4c+a0wzti2dKl0WVgmkdPT4875Ovk7Nmzbt26td4QMepOnDju5Wt1a1avLVKGDF1ZipRM 0bMjIANEdqyVkwiIgAiIgAiIgAiIgAiIQB4CK1eudHfffbdfhLHevepVr/J7UoTlIINHwihv4TPQ YVkgccOGDd7w0OFOnTzlWIeh0U832Lp5i+vynglZhZUrVrhWb4g4c+aMl+Ok9zZocN3eMLJiRXdm izi2ewY7duwIBog9e14IO1jAYNu2tdE9MLLirHyyJSADRLa8lZsIiIAIiIAIiIAIiIAIiMAMBFD6 X/3qV89wNbvTGCGW+akY/OHNUaldG+CBAaCSMrB7xZYtW27LQB3E3IQ0uzpWTpUhIANEZbgrVxEQ AREQAREQAREQAREQgXlAoFLGhyQayZCkoe/zmYAWoZzPtSfZRUAEREAEREAEREAEREAEREAERGCe EJABYp5UlMQUAREQAREQAREQAREQAREQAREQgflMQAaI+Vx7kl0EREAEREAEREAEREAEqoxANUwX qDIkEkcEROA2ARkg1BREQAREQAREQAREQAREQATKSsCMEGzlqCACELA2wad9F5mFR0CLUC68OleJ RUAEREAEREAEREAERCAagbGxMXfr1i03ODjoRkdHo+WjhOcXAdrCwMCAk1FqftVbuaWVAaLcRJWe CIiACIiACIiACIiACCxgArzdbmhocBgiFETgRQK0i3p/uMiN+a1NFRYmARkgFma9q9QiIAIiIAIi IAIiIAIiEIUABojGxkY37pVMudpHQTwvE62vr/MGiMYg+6J5WQIJXQ4CmpRVDopKQwREQAREQARE QAREQAREQAREYEYCcnqYEc2CuiADxIKqbhVWBERABERABERABERABOYHgZGRETcyPFxxYZGj/2Z/ xeQYHx9zQ0ND3qOkclNa8GZBhjGt6VGxdlArGWsKRq3UpMohAiIgAiIgAiIgAiIgAlVAAGWVv1LD wK0Bd+LkiaDwspbEsDdCrF271i1bttRP6cjm/em4X7+i3y+keebMWffnf/5F17p4sXvPf/gPpRap pPtQ9i/2XnTnz19wTU1NXvkf83K0uJ6e9WGKS0mJFnkT9XjlyhV37uw5t6huUVhUtKmxya1fv97L 0lpsakXGV/RaJCADRC3WqsokAiIgAiIgAiIgAiIgAvOQwPDwkDtw8IBra2tzd+3c6erq692NGzfc oUOHgyFi9erV0Ut19uxZ9/u///vu+vUb/q8vfP7UT/1U9HxzMzhz7py7cP68u/vuu4MBAk+M48eP uwMHDrgdO3aEhT5z7yn38eXLl92xY8fcxo0bXXd3t8Mwc8bzoY6QoaWlpdxZKr0aJ5CNCbHGIap4 IiACIiACIiACIiACIiACcydw1r9px9W/Z11PMD6QYnt7u+vq6nTnvEI+6LdxjB062jvcz//8z7v3 v/997m1ve1vs7FLTv3njZjA+rFixIhgfiIQ3yKpVq/xWloOut7c39b5ynmTbzLPeA6Szo9MtX748 JL2ors6tWrkqeEKcOnVqTp4u5ZRVac0fAjJAzJ+6kqQiIAIiIAIiIAIiIAIiULMEUHivXr0adkpo bmmeUs72jnaHB8Dlq1emnI9xQF67d+/2Ux16XK4cMfJLS/PqtathG9POzs4pl/E4qPdeIRcv9kZf EwLPk4HBAdfa1jplN5OGxga32E9J6eu7HupkioA6EIFZCMgAMQsgXRYBERABERABERABERABEYhP YHho2A359R4a/Zv+3NBQP3Guv79yi0HmyhTz2Ba9xOshGeq8B0KdX4thZMSzGoy7QGdfX1/Iur6u PinCxDlvBBkbGw3eGNMu6oQI5CEgA0QeOLokAiIgAiIgAiIgAiIgAiKQDYHBwUHnX+v7qRfTVRTe +hOGBoayEaaCubDw4wAsfMDgkAwYHwijfi2G4ZG4LAa99wMhVwbOWX1YHM4piEAhBKa26ELuUBwR EAEREAEREAEREAEREAERKDOBET8Fg7Bo0YSSnUx+kTPFeyJO8lotfrctN1NZwMcbKUZH4rIY9btu hDC9OibriGkxCiJQDAEZIIqhpbgiIAIiIAIiIAIiIAIiIAJ5CaA0m+Jsn3lvmLw4+9adY/7Nf60H yljINqaFxJkLK2Ntxp+0tIqTYcKSEexLKUamtPR1rvYIyABRe3WqEomACIiACIiACIiACIhAxQig lPJmvNi345PGihQ7xLibOJk2HSCLgppHQhZ5UcYXWUyHYUr/otvTMWLJZKyN/dR8SqkPvDZGgudG XQQDxPeeec497f/SwrETp9xTf/ddd7P/1rTL+a4lI1/sveS++KUnHJ8KpROQAaJ0drpTBERABERA BERABERABEQghwAK8oDfLvPWrVthS82cyzMe2roCpmAnI9o5U4qT12rtO8aHRYsm1LRc84NHOxF8 nPrbC3PGKn9Dw+3FJy3PREYmh9VZ4tKMXzFIDQ4Nhj92PCl3eOLrT7mPffzT7uKly9OS/pPP/aX7 5KOfdce9ISI3fPHLT4Rrvb3T70vGffqZ5x1xv/jlv0qednv2HUxNd0qkIg8wcpBumsGkyKSqLroM EFVXJRJIBERABERABERABERABOYvAYwE7e3trqOjwzU2NhZckKamphA3TTm1c83NU7fnLDjxOUY0 hXuOyRR8e3NzOovx29Mz8CDI3SGj4MQLjNjU1BJi2lSM5G14MhCszpLXZvre2Njkt+9sc62ti1/0 8Jgpcgnn73/p7nDXnhcOTrkbJR5lnoARITccO3HatS1udRs3rMu9NOX4wVe/zL3pnzwc/uzCRW+0 +NDvfNx99evfslNl+Xzq774X0sU7o9aCDBC1VqMqjwiIgAiIgAiIgAiIgAjMQwIYK1Boh4bZ3WHq a/dhvz0nAcNGJUL6NIR4krS1t4XErdyWE7tf4A0CKzNS2LVyf3Z2dqTKwMnh4ZGwE0Zra2sR2U6t 0yJuLCjqptsGhOPeoJAMZnzAyPC9H06dooGnAUaEu3ZsT96S+p373/RPH3YrupelXtfJwghM3Vi2 sHsUSwREQAREQAREQAREQAREQATKSoA3+l1dXe7ixYtu4Naga2mdeANPJjdv3gxKd1dnV1nzrNbE lnQtcefOnnfXr193S5cunRSTqS14g6xatWpymsbkxTJ/wYMFjxOm0mD0sHUpmEpxy2/RucTXVTEe LrG9SO66c3vwZNizf6oHxPdvGx0e+emHwhQKPCIwJhCO3jZW7LpzWzjm2u/6aRx37djm2tpaw5oP i33cT3zsgw5vBKZyPPjqB9xDr3pZ8FCwKRIYOfCE4D6MFBZYk+Jr3juC6+R5/333BA+KFd3LLcqU T+L9uZ/iceH2OhN/8rm/8Pctdu9651vciuUThg+MJp989LEgD/lv2tDjHnndg0GmKYlV6YEMEFVa MRJLBEoi4J/s4yP9/qVB7a8QXRKfStzk53AuavADqEW351FWQgblWV0EKthPF9V5V2jao4IIiIAI RCSAsmprNhSbzbp161xf33V3+sxpt2HDBq/gNrgrV666a9euubVr104xShSbdqHxUfK/+c2n/CKa w+6ZZ54Jt+17YZ/7y7/4S59/q7v77l1u/fr1hSZXUrzFXulctXqVu3D+gluyZKmfztLuvQ6G3Zkz Z1xLS6tb7Q0QsQNTadZ55idPnQ5GoRUrut3Y2Lg7efKka/DrT6xbm3/KQmz50tLHC2LPvkNh7QQz MnC8yxsnMDJ88cvOfcsvRvnI6x4Kt+/11whcJ/Tfnq6BscGUe/Os4BgDAUYGAp+cIy55cWxxuR7W i/CLVpI2UzfwtGAhTOT56IfeN2kEIa4FSwcPIOJjXMDwsPi2pwnGh9/4wEdDdIwgGEfw6vjkpz/r 9nrZ3vWOt1pSVfspA0TVVo0EE4EiCYwNu7H+3rAvdJF3KnpkAuOLrru6xd3O1emRGxl19Sdf4X6K 8+uixla3qOXFt2nVD00SioAILCQCeEFs27bVnTp1yh08eDC4+bMGAYaJFStWZIJiaGjIHT16xDX5 NQt6enomjQ2XLl8K3gebNm2KLgfeBmvXrHH13ggAi/r6Op/3mDc+NLvNmze7Rf58FmHpsmXB6HDh 4gV31RuCxvxLLhae3L59WwnGoLhTMODBVAoUfBabxCMC4wCK/Bu894N5SBw7+eIUjeP+O0p/7voP GBYwEiQNCrm88XQg7a/5xS8xFCQ9HzAUsGPG/fftdr/+7ndO3vqQ9574oPeU+NqTT02JbxFIh78v fsmFcjz4qgcmjSPE4V5kwyPDpoKQ72ce+4sgB0YJylnNQaPhaq4dySYCRRAYG7gi40MRvDKN6n+s xweuukUYIRQWNIFq6Kfjw7e8EcLPLa6fWOBsQVeICi8CIlCVBFhXYPv27cGLgu0v6+qy9SLs7Ox0 73rXuyrOBg+ENd4IwR9GmErsAIIhpNt7PvCHDBM7dCyqOJuZBDAvB4wQKOLm4YAhgMBClWGrzne8 JRzj0WDeD+HE7X84l8/4kIyb9v2rT04sSvnITz845bIZQZ72XgtJg8WUSDMcICsGj4f8YphmfLCo pIUhBK+LD8gAYVj0KQIiEJXA6MRqxFHzUOIlExgfHXLV+3NdcrF0Y7EEqqSfjo8O+llBMkAUW32K LwIikC2BCWU3W+NDtiUsPLdKGB9ypasGGXJlyj3e6L0HCHvDOhAPu6d/+HxQ1k1hv8tPw3jqO98N 0xXMH+P+l96Tm8ycj/GsIHzrO98Lf7kJYkgoNtg9NgUkeT9eHJTxQgnpJtPJ4rs8ILKgrDxEQARE QAREQAREQAREQAREQASiEkARxwuCrTVtzQY8BiwET4hHPxumNxCXkDv9wuKW49M8MJJpsahkKYH1 KWYLZqSYLV4lr8sAUUn6Zc77ypUr7vHHH3dvfvObi9qTt8xiKDkREAEREAEREIEqIvDss8/6hdX6 3ctf/vLJVeyrSDyJIgIiIAJlJbBxfU8wMISpFj5l1lGwYAYKPCRYwJGQNgXD4pf6uXjxxILPv/Oh 30hdbLKUdE3emQwRnJ/LtJFSZCrlnmxWLylFMt1TFAG243n00Ufdc889V9R2OEVlosgiIAIiIAIi IALzjgDb5P3Wb/2We/755+ed7BJYBERABIolwDQLAushTBgcpi7KyEKVeEgwXSGG8YG8N3kjCIHF MHPDTW8QLiXYOhasH5Ebkrt25F6rtmMZIKqtRkqU53//7//t/vZv/9a9973v1duNEhnqNhEQAREQ ARGoRQI7d+50b3rTm9zHPvYxd+nSpVosospUZQSqfaHCKsO1gMTJZjUsMyowHQFjQ25gigbTM457 I0Taegq58fMdm1cCu14kA9M+MH58wk/3SF7j+4du74KRjD/T96S3A+ndf989wbuDXTQskObHPv7p cMh2n9UeNAWj2muoAPmOHTvmPvGJT7h/9a/+lVu9enUBdyiKCIiACIiACIjAQiGAMvi2t73Nff3r X3f//b//d/fhD39YLysWSuVXqJy0OQvzYeFCk1WfcQlYs6B9xGwXKOqs64CB4YGfmL7ApO1EgRHC vCVKLfmEh8W2YBT4d7/+Ib/Lxj3ubb/ws2FByF/8+Z91n/ncX7hf9ucxipCfGSNm2yrTdvPAsLCy e3nYypMyvesdb3W/ceJ3Qrpf9btesJ4EaZI2+dlim6WWJ4v76j/oQxYZKY84BG7cuBHeaDQ3N7tf +ZVfSe3MTM8YGBgI8z+Rgr17kz8McSSbnip7Kt+8edPxyUMHObIO4+PjIX/kYCuhSrEg71u3bpW1 TsaHrpeM8+zZc57HqN9bemK+WskJlXTjuBsZGQ08RkZGQtuI+aM0k4hs80XbpK9MtI26sveTRc0d M2U/eZ7+Stvgj35aqTZKXdBPBgf9bgm35ZgUMqMv9Ffyr3R/LeczdC79lDqhfdJObfBW6rN8UUOz 3wWjuaSatGc58jQ0NJS9nxQiFH3UftdoJ5XqJ7QN1laolv5K3cAi7RnKuRUrVrhPf/rTbuvWrW7z 5s2FoFYcESiJwNmzZ92f/umfhuf3cbEwQwAAQABJREFUXXfd5R566KEX0/G7Qjn+FBYcgae//7T7 /ve/74aHh9zrXvd6t3LlymgMlnZ1unVrVnmPgd2pazBwza43+WlqycDxrp3b/HoKE9Mo7BpmNRR+ jAMrvFHAAsYFrjU2Tb2P+1/1sp8I+d/047qVfpeKV73svmBE4Hu+QPohjh+DIee99+wM6SDbG173 0OQ1ptjt8l4e7/l/fjHEyZdmtVxb5H+4bQeSapFJchRB4Ctf+Yr7yEc+4n77t3/bvfa1r512J4P3 F154wTFIYk9jBkpNTU1u48aNfuGV0lZgnZZJASfOnTvnjh8/7jo6OsLgCMMJA6F169aF4wKSmHMU mvrRo0ddb2+vW7JkiVd6JwbzGzZscN3d3XNOv9AEqJMDBw6EwfPSpUvDjzODeAaDc6mTsetnChUh xGPw3ne9z33n777jvvKVx90b//k/d4+84ZGi0phrZOrk/PnzDgNIZ2dHUGSQi/qI+aOUKzf9A08i +kd7e7v/YRx242Pjbu26taHN5sYv9biuY23eW8mftkFoa2sL8mAU2rRpk8PImFWgjxw+fDiw4HlB f12+fLlbu3ZtUDizkIO2cfDgQXft2jXX1dUV+ivKFc8u+k1Wgf66f//+0CbK8Qwttp9aOVlk+NSp 074dNPlneKN/bvQHLmvWrPbP0OKdGTGGLWqa3SBm+fNJnZw8edLxPIcFgbbR09OTqfcdz27aRl9f X2gLPDOQjWeoyRWEi/zP9evXQ39F4ae/csxvHL8pWfZXnqH8tlF2fkuoE56hyJEWfu3Xfi3U4e/+ 7u9mWm9psuhc7RLgufn2t789GEzf+MY3uve///2ThcUQOz5Y+kuTyYT0Zd4R+LM/+zP3hS980Y/3 XPDEwjilsPAIFD9qWXiMqrbEvHH51Kc+5Xbt2uX+0T/6R9PkZEC2Z8+eMBC65557gqKPosU5fhh2 796difLPwJnBEYMhFBje2F29etX9+Mc/Dt/Xr18/TfYYJ44cOeIuXrzoeNjZIBWjCCxaW1vDADJG vsk0qZO9e/eGt1P33Xdf4M85FgajXl760pdmouD9zd/8jfuzz3/ejQyPhLf9DJwrEVB0T506Fd7G YRSibXDu+PETgQ1Kb/ww7k6cOBEMQndsv8M1t0wo+rQNlHDaC0p47MAb3X379oX+ynxt3laiaD3z zDPBiGh9OLYcly9fdocOHQp9lb5JnaDQsIo+bRUDQBYB9shy9913B0MIeWIkor/CAoUvdrBnKArm vffeO+0ZihwofLFDvzc20Ce6/dsQjLbUiRmraDczKZrllou+yt8dd9wRDFLIwVtO2gsKd1aGIfoJ hiGel9Y3MbTzbOVcFso/+SMHRuPc/so1ztFuYgfWc8AYg0cDUzCpE35z7XcG41Bu+KVf+qWgGH75 y192/+bf/JvcyzoWgbIT4FmqIAIQsLZAk7DvIrPwCGgRynlc5//zf/7PMNBgXmfagIuBIa7LSS8D BkQoFSgUvDXJIqBI8BZ31apVYXBEnrzR5A03ih9KVuyAgn3hwoXgdWHGB/KEBUwYQGcReHMIewaK NjhlwLjJv+HmLR51lkV44IEH3Ee818ynPvVH7pWvfEUWWU7LY3R0JLyF440higscCBgdOjra3Zkz ZzJpGzdu3Az9aPnyZZPGB+Sg34x4gx1vfLMI9AWMiuRr7tPmGUP7zWLhOJRZFEzyR6m1OsErBI8l kzE2D8rL84k8yduCKVPIkUWwZygsrL8mn6FZ9deTp06GOlnln5lWJyi+vOXmuUa7iR0wePD8oq+S r8nBswwDLs/5LAaTGChRsPn9MOMDZd+yZUvwUqH9ZhFOnz4d3uzidZHsr/Rf5Muqv9IX+H1ds2bN ZJ1QR/QdZExrGzt27PDP/Ve6P/7jPw59OgteykMEREAEICBblNoBBGSAmKftgIHFE088Ed4Evuxl L5tWCgaCDBYZLOfO62ewSGAgFzugbDMAYrCcfFPI4JVzKDwM1mIH8sDQkcuCgSM8MNSkDdTKKRd1 wqCUOsl9e4sSTjA5y5lvWlrkxwA+WSdp8WKeu379RniL2dY+9U02bYN6GhoaDi7WMWUgbd6y0w5b Wib6heUHG+bZIeeIn5IRM9A2kIN5fLlt1BTwLPorfQC3dntGJMts57Lqr9QJz4hkoE4wtjItI4v+ as9QK7vJYsdZsGA6EM9Rpl40NE71tjA5Ll2K/yynXWAktTyNBf2VNspbfwxHsQPPUPv9SOZFu6D/ UCcwixnorxh+yC9pBCFPe7Zn0TboA/x22e9HsszUExzS5OA36J/9s38WZP+894SjPAoiUI0Ehv04 gDbc659xTDtjraasAy9L+M3hN5jnIL9NlQo8g/GKrEwYd7dYQNF7EuPFzDNfQQRKJSADRKnkKnzf 5z73uaCwvOY1r0n1fmDgwcMBBdvezpjIDN44x6A2duChTWDAkxvsHA/02IFBGoEBY27gnC0klnut nMfJOrGyW/rJOkGWhRCs3uvr0tsGA40slExrG7l1Qh1wbmTE9yW/9kDMQF+dqb/SV2kf1pdiysEz AWVkJhbkbbxiyUH+lkdaf8UIQV/ijXzMkOyvlXyG3vDPLphMGAsnvISs3FZPyS267Fq5P+33wvJM po9sDMpjGyDIg2cC9THBIynFRH9lgM5aITEDfZX2YX0zmRd86K/WhpPXyv3d+utMLJJ9KTfvl7/8 5WE61de+9rUwRTL3uo5FYK4EaH/8lRpQdPft3xfGqqzJhLfPoUOHo78QSMpLH3th775Jj6be3otu /7790X9/kjIM++fZMT/98POf/4J7z3ve4x577HPJy5l859nLVGqmMvtaDc+3AwcOhhedxQtQepso Pi/dUa0EZICo1prJIxeKyDe/+c3wdvD1r399akwGR2alnWnwjKIbe6DGYJCQKwPnbCAbW5Egr3xy mGwMKmMG6sR4W57J/DhHnRBvIQTjncZiUd2EomW8YvGgj9gUoDQ5rI0ODMRtG7RPZDFDVLK8dg45 Y7eNfHVifKwvJWUs53c4WB7GP5m+yWGyJq+V83u1PENv3ZowtNAO/P9TQlYsyNTqxPJMCmL1FPtZ Tp1YH5hJDhSe2M8NYzFRJ1MrxfprUtYkq3J+NzmMfzJt42Nxktf4jufGz/zMzwSjEWtBKIhANRHg BQXTH/HUZKosU4p2+FX+ac8n/PksPCH4zcXg0da+2C9wuylMPduyZatrbG70664cysQQgmfkf/x/ /5P7xCc+6T0PLvi/+N5uae3g5MkT/iXIdbd121a3csXKMFW0p2ddmOaV5mWVlobOiUCSgAwQSRrz 5Ptf//Vfh/nRGB/SXC8pBoP4fJZnBkkEG8yFgwj/oFATLL+0LEwBTLtWrnMmR1p6Jlu+OGn3FXuO OuFvpmByZMFjJhmyPG8srNzJvL2qFQ5j1wl9ZMy/WZkIUxUJzpkceEHEDJQTWdJYJPONzSNfnZgc xMn3bLF4pX6Sdr5yGqN8cUrNO3nfbOU0OWI/Q3nzR5jIb2obNRlgFntAbnVueSZZ2bnYLKgTa6PJ /HO/x5bD2p6VOzd/jmdrx2n3FHuuEBb52vHDDz8cXmR84xvfkDt1sfAVPxoB+g5TnMb8J4YHC3Xe W5LFqnH/v94X34MXDwymX3R3r/DP3xfVpVUrV/mx85C7etvD1+SL8cmaZb/5m//J73b32+7nfu7n Ul/mxcg3mSbjnytX/G5Ufqey5BTRJX4tN7wUJ7Zxn3lsm0xL30XACLzYo+yMPquaAIOJxx57LAxG GTzMFGywONN1O19oPItf7Ge+9G3wVsggqth8c+Pnk8PixpYDGQqRo5A4JvN8/iyknLHrBH6FyWFG ijjEZyun9ZVCZJ2LhLPJYbwKiVeqHJSxkHLGlMHKWUgZCpG1kHRmilNIOSeYzZRCec4XIkchceYi TaFto5J1Ql/lr1BZ58KjEN75WPAC4/777w+LH8sLYi41oXvLSQCFl+lcTY1N0xRuWzuMLcRjBgy6 eGHgSZS7zgvr8RCyWJeJ6VUYXXJliFn23LSvXr0WpqI2+7W5kmGRZ4NBot976cX2OkvmO5fvT/3d d90nH/2su+nXslCoLIGpK1pVVhblXgABVhpne8B169aFbbdmusXcL2e6boMSU2xmijfX85a+5ZdM z87NJmvynlK/mxz57o8tB+nnk8N45IuTT/65XmNeX5bBypkv37rbUzFiypXr1p7My2Srq5/65jkZ pxzfZ2t7WbWN2eSwshYaz+IX+2ltI999sWWYLf2s62QiP/ro9LYIr3ztOB/HQq/lqxNjMRuzQvOa Kd5EOaeXPzd+Pllz45ZynK+csOCPOJWUo9ByYYD41re+5fCsfNOb3hRd5kLlUryFS2BwcCh4waXt 7mbTjWKvD8Xil0NeDvpwve/LUwPP27qwDgJ9PXY/n5p39kfXb0wsLpz23AtsPAOmqSa9I7KXsrAc 9+475J76znfdm/7pw65tsV+A3hsi/t2vf9Bt2rDOfeB9/76wRCoQC8PJn3/5r9yK7mVeznfnleBj H/+0O37idN4497/0Hve2X/jZvHFiX5QBIjbhMqf/5JNPhhRf8YpXhK0sZ0qeB4U9LGxwmBY3bZG3 tHilnrMfi7T7Ta7YMpA3efCDZXkm5bFzseVI1kky/9zvaQuK5caJcWwcYqSdlmZDw+3FJ1PsHiZL fV3cR1SdH1zk6ycmR2PD9MVL08pU6jn6CXKQn+WZllbstmH9NU0GO4cMMQdccKi2/mplT6uT6M8N 3zYIGMN88/DsX5TixekZ8ZVda3tpLOxcbBbWT5gCYXm+SONFb6Ys5Ah1QoXMEOgjxmyGKHM+benn YzFbf2UsgZv3/v37w1aq27Ztm7NcSkAE5kLApqGm/c7Y73XsaVZswT06xhRi/8BNPnQ544/5wwNp 1K8T0eDHl7UcRoZHQvHS6gMvCMKQn5IyX8Nib4jI3XGrmspy0y+4/cUvP+Eu9l4O44DZZNu1Y1sw rqTFu9B7yWGEocyVDnFH95UuXY3ljwLNWwrCT/7kT+YtHQMwHhYMTHLdNDnHAI7BXGy3LtuyzebM JoU2ubLo+FbONDnsnMVJyljO79QJf/xwWp7J9DlHncQePCfzTH7PM5ZORivb96amCXc+awfJhO1c 021Xx+S1cn7nxxPeLGhoeSbTp07oRy1+i7+YgTcHNqDJlcP6MIqEKRyxZLE3GGnt0+SK3U8YYFoe NhBNltfqxOIkr5XzO+2iGp6hi29vmzzuB7uYIcKA+HZBrU4m2mfCMlFOELfTsreRlmcyC2svtgVl 8lo5v9P+qXfcfS3PZPpZtQ3rJ7DIVf45xx9xsuqvaf3E6mm2ftLT0+Puuusu9w//8A/BE0IGiGSL 0ve5EOD5WUoYHZ1YTyDtfq/6hyTT+n8pec10z8Q4ecz34em7dHGPFW3Eyxr5/cRMImZ23lin1sdt EKyVMR8DXhCf+NgHq1r0L3rPB4wPhYZHXvfQjFE/+ej/5/b6qw+9+mUzxsnqQq5fUVb5Kp8SCBw8 eDCsOMvAZseOHXlTYPCMYm/GhmRkBiucZ+/22GHp0qUhC8szmZ8NmpjfFjtYWS3PZH4onyg9ZixJ Xivnd+oEOWCfO1/O+HB9tgFjOWWqZFqdfkEjAos55YZQT/6HrS1yG+UH1RZyHR7JlWPc4YZJveXO fcyVd67H9GkUPH7o7cfe0sRgRZuxvmTnY3zS/ugLaf3E3jgZrxj5kyZ1Yv01t59wHTnoI1n012p4 hsICd9+RkVGv2E594z5ye9C3uG0xaKKGfHVCPSX7UixBaJvIQX+w9pjMi3P0Jf5iBtKnDdJXTdG3 /OxZHrufkN9E21g07feEa8ZnNjkweu/evZtb3Pe+973J+8IJ/SMCcyRgSit9t6xh6qOwrElPJBY9 gwgyVzDJInBZm+DTvseQHM8Bpi588UtPhGkXF/3b/7SwZ99BP2Xh1OQl7uMc0zP4/rWvPxXSePqZ 56bEKSTt5P1M/ZhJhsmEc74Q/2tPPuXe9E9mXvMv55YZD0nrqb/7XjA+MJWj0kEeEJWugSLy//a3 vx0GO6wKvGrVqlnvXLNmTXCrZGu0pJeB7efO9dgBJYEBEAsKmeJAngzakIPBT5dfSTd2WL58eTDe 5O5Tz8AZAwRGkNiDVsrY3d3tt1G6GOYOJg0vrOpMoG5jPpBDJjP+U8QvyIxpFH6BdoHSff36jaBQ WLkZ1OPt0+oH+Z0+TuywbNmysKvMLRYlSjyTmdOIC+byJcsnt4yNKQvbjbHPdm5/tbbB9diBPkC7 ZKtf6oH+SUDho79SR/Sl2IE82IItt7/SLuiz1FkW/dWeoTdv3kx9hiZXaI/FZOIZ2RlY4Ao7pU68 XNTJCv9ciR0wgMEcFrSHZH/lHN4PsT0gKCPMz5w5E9pjsk/QPvld4boxismEtsF6TPTX5G+Ytdks +gn1AXPypOym5Fl/5flaiBy7du0KqA4cOBD6Pr9TCiIwVwK0SZ7ZjLFyDev50rZnC+04NzAVjWBb dedeL98xyjG/fflTzGKdqukSzCLU9BvmdCb5XMlNyOrI4uReTzumLbCdajH3pKWT7xzK9gd/5+NT PAdQulemKN4f8vF23bl9cm2Fp595PixUef9994TpCskFK1kr4hd//mfdJ/xClkmvBDwpPujXZti4 oWdSLKZNYPzIDb/8zre4B19VmAcCZUBu1qwgvbkEu78cxoy5yGH3ltkkacnqs9wE6LA/+MEPQrK8 rSjkLTkDMeZ2njhxYnKLLX4ITp8+HZSMQgYm5SjH1q1bg9JAvjys+GP1YPY33rJlSyaDRQwwuJqy XzH5EpADJYeA26n96IUTkf6BOQoeLKgLAp/UEW+zslBoyPOS549Biyk9zz33PKfcd77z9+GY86b0 hguR/uHHZ8OGDeGHCKMM7uXUySX/w8GgZX3P+qg/UFYsjGQMuGmTE8Y5P0XJv20+deqU72fNbvXq 1RY16icLy9IGaAvmgUDboI1au4kqgE+cPrBx48bweeTIkcn+Sp9hRXD6ciHPnrnKSX+FB/0VYwiB tgEb3No3bdqUSX+1ZyiGoWR/tWdoVv21Zx2DmkXuzNkzQdGEBfVxxdfLmjWrXVPkKULwp97pr/QR tsgjIAcsMApt3749k/7KbxqGh/PnzwflHzkwblNHKORZ9VfyQcGnnyT769mzZ4MMSQMzMsYI9Fe4 Mz6gb1Af/PEso33Qlwvpr3fffXfoTxgynnvuxbd8MWRWmguHAO2T5zVehMWEptvxacu5wbzAYk9v aqi3KY9ehjQ5OOfL19iY/XvcNC65nMp5bGtcpOVr54qrD9rFxLpXbLUaI5jxAWPBH3/id8LfQ17p 3+PXPyg0HDtx2v3au9/pvvDHvx+maey6c5vjHGk/8NLdIU2ukQdGis987i8mk8YAgvEBgwX5WxoY Ez7z2F+E+JORZ/hi6z58sAyLYyIPZcfQUg3eDxQ5+54zA2idzk8AhZABFmG29R8sJR7+TNVAkdq7 d2/4IWCgxtsaBiYxrY8mA5+86WaOKQPVZ599NuTLQwvjQ1aDReRYu3ZtGGTx1gpZGDQygGTwxWcW IVknL7zwQjC+MHhkUL3JK1XFPcRLl/iKb0979+x19f5H4L77Xuq53BcSO3bsWFC+aR9ZDKAn2uIm 7xVyYcIwxG+6N0xs3rzZdXZ1ll7AIu/EOAX7EydOhh9G6oSB+4477ihoAF9kdqnRaRt33nlnaJso AQzaUOzoI+vXZ2OMQTCMIPRXFBrrr/DA+FCI51Vq4Uo4SZl5k33o0KFQBzy7MBYhWyX6a9ozNKv+ 2tLa4hXNbe6cV2737z/gn6EsgDbujQ9rMn2Govjz7MYD4dy5c+E7DKgT2k1WgbZIW2DhRNoI7ROj Fb93hSjc5ZCTfO65557QT3784x8HOWijleivPDf4nbf+SvmK6a88h/l95HcRA/RrXvOaciBSGguc AL9p/I7xzOB7oaGxifV30qcC2loDLS1xx2yNTRggWB9qIDxrk7KzRSdGiWb/DEDOrIPPOtPQ2jKx YCEeLbmBlzWEYjwS6+vrAlvuK7xVELuwwPQJvBNY5yC5HgJeBBgQnv5hYUbWN/z0Q0FhJ1eU9ne9 463ul/2OGXxP7iBBHt/00yuSxg37/ohPA+8IS+Nf+KkUT33ne2HKx13eGDBTMAMGMpTDYIA8MMFY Ui1BBohqqYlZ5MD4MPF21oU3HrNEn7zMQ4GBCIoMgzQGi/wgFPNjMJnYHL7gMs0ghwEaATn4yzJg cOGtKkoUPDiuhBwMXDG+8EaVOmEAzbks6wSPj2pZbKy7e7l3qV/i28ZIaA7wqETbYACOgoVhirrI uk4oPEpVsr/SV2GRZdtADvoqbtn0VwYdyEG9ZBnonxiGUOjor+SPHFkZTq2sPEPpK9ZfqY9KPENR 8Lf4ZzntkzqZeHZlWye0Q+oDz49KPsspO94YyMIzlDZRqf6KBwIskKNS/RUPKabI0E9Q9krpr3hW YoDYt2+fNX19ikBFCNC/W73RFU9IlP2kks9zmOdQZ2fcqbt1dfX+5Vm799C85Vhrp8n/Z2FoMLt1 mSzPqZ/ZWiC6/MugM2fq/LakE167SVkG/Llm7ylajAEitgGFXR4ID77qgaSo4bsZA6ZdSDkx004R m9bjkTg1tHsDeDLYVA+8GDAg4HlAeNAbRfibLdjUizd6o8lMgTUpnv7hhAezxWF6RZrBwuS4/76J 9X4sfiU/s9UAK1nSeZ73nj17Jq3IvB0uJvCwzuqNYT65UCCyVmLS5JkYuFe+6VdDnaTxqcQ5fuyb m7NVptLKWQ1to1r6K0pdNbTRaqgT2ko1sDBFO63tZnmuWp7lGB0qHaqlnyBHMUpALje8OZ544okw pSR3HZrcuDoWgZgEeL5g5MQL79q1vklvTHYCYppQa+viycWjY8qBUe+8n27GVMDkOmpXr10NY1le rGUZxvC88AEjo//X/8XwHyCHqYHnCsaY6376HQZwfpMJE2syDQZPvGoY25vUtjZDrlHArs/1s61t wqMhXzp4N/zyO97ivuANEKwxQcAI8eCrH3BMBckXbOoFa0XkM5hcvHQ5rFGRTOvCqy5NM0CwWCZM SK+aQuW1sGqiUcWysDgUgQdiNQyEqxiVRBMBERABERABESiCAFOeCHhy4HFpC1MWkYSiisAUAijK E8rylNMFHbAu0/W+62FqEYo2Uy4uXb7kPRIGgwcpXj6xA15nPd5r9syZs8G4xzEL7rIGDh6TSaNE LFlQ8j/zmc/4aaq9gQUvKJDnP7znV8LaVT9x333ukUceiZV9SBfjJs+HI4ePhIV31/mys1XqyVMn w/Th4he0j+vBMZPnQlRIKYmbtwNTQr7vF7b8np/68clPfzbsavEBv65DmnHBpl5wjWkTNpXDku/3 a0180i+AuWn9ujC9hCkeswXzfih04cvZ0ivXdRkgykUyYjo8wH/0ox+FHIr1fogolpIWAREQAREQ ARGoAQIoESgaTO/hhYcMEDVQqfO4CCjam7dsdpcuXQprRNEuUfh37rwzs3VewMeUXfJloexe/9fY 7NeG8mvNZLHjD/njWfDa1/5f3pDjzTCeiQUz7ixbno0XBtNDt9+xPXA47j1TeFYw9QtDUVIuk6+S nzb94eiJU35XinUVEeWCX/SRsNJPMcbzgT/WjWABSrb1ZPoEa1TkBhaztOka5smRjMN1zrf5KUqF hO/5fIhfLTtfJGWWASJJo0q/s5o1K34TWMNAQQREQAREQAREQATKRYC3uygUKFqHDx8uV7JKRwRK JoBii4LLHwp3JRRd8mSRcP4qIQPTzLZvv6NkhuW8EVnQQSY4kPKLBpFy5jPXtFjngB0pvuUXhkwq +Tf9lsl79h+ca/IF3f/nX/orv9jkd93HPvQbU7bmZFcMAp4MaWHThp7J7UBzr/+Lt//7ML3iA367 z0LDE1//Voia5FDovbHjyQARm3AZ0md1awvygDAS+hQBERABERABESgHAd7o8rYXA4RN+SxHukpD BMpBoBLGh1y5q0GGXJkqcVztHFiEkakJeBp87OOfCmsu4DnAcZpXQQyGD/m1HjBAfPTjn3bsvrHC e6ocP3naffXJp8LUiywWgzzmPUCY/oHxIW1hSis3O3sQPvqh96VOCwkXI/wjA0QEqOVOki3PLBQ/ 18ru1KcIiIAIiIAIiIAITCfAnHp2ASKwGwY7alTDIp/TJdUZERABEchP4Bf9dAfCxHSHiZ0iUPrZ wQLDQOzAIpS//u53Bk8M1n2wEKZi+K0w8xkELO5cPyk7Id/0CwwzlQoyQFSKfBH5st+6BRkgjIQ+ RUAEREAEREAEykGAle2XLFkSkhoYGAgr/+MRoSACIiAC85EARgi8D1jYcYVfi8EWfXxXzm4QX/jj 359SPDwG0qYsYDTIjWs3pk2LwODB38S6DZf8WiKtYU0Iu6fYz5nynimdd73jrY6/fAEmeD4QjE++ +OW8JgNEOWlGSsvWf2Axmqy3/YlUJCUrAiIgAiIgAiJQRQRYA4KAAYLtDmWAqKLKkSgiIAJFE0Cp bvPrKlQyVIMM+cqfteHBZKmzL/qsXgK9vb1BOBbBkUtk9daTJBMBERABERCB+UrAPCCYfnHt2rX5 WgzJXSUEWCvA1guwzyoRTWJUlMDE4pVhU4/wT0WFUeYVIiAPiAqBLyZbGwiwFVBzc3MxtyquCIiA CIiACIiACMxKoKurK8RhlXsWo1QQgbkQoB2NjIyEv7mko3trjcC4G/XtAqNUnQwQtVa5BZdHBoiC UVUmIq6Q/BEwPjANo9gwOjrqBgcH3fDwcEiDdCphjeatCmUh75aWFseiV1kHfhBNDvJHDvYzzjqw pzUsKl0nDA7gAZfGxib/11CRtgH/8fExd+rUGbdiRXeol6zrhPyHhoZDnTAfutnv971oUfZtg/5K 26BuaJ94PVWiv5L/rVu3QtswObKvk/Hw7OL5Vcn+Wi3P0GR/pV3QTivRNuwZyu8R+8NX6hlKu0AW WPC7Vgk5aBv0Ez6tn1SiTqy/0kepE9pGsYH7LFy6NLGPvR3rUwSKJWDjHPopfwoiAAGeVYNDtIdF 4bkpKguTQPG/UAuTU8VKjSLS7/euJTAAL9YAwUOfPb35xIOC9BhkrF+/PnxmVTAW0mRlbQZoDM6Q Z/Xq1WHV7WLLVKrMKNknTpwIb3Y6OjrCg4/B66ZNmyYX3yo17WLuo+zHjh0Lc2yRg8ErXDZs2BDq qJi05hK3r++ar5OzftC+KAxWB24NuOV+kZ3u7hUlDV5LkYU6oX33Xup1T33jKfcP3/2ue+97f9Vt 27atlORKvgflge1ub9y4EfoFhiHa6dq1a117e3vJ6RZ7IyxoG7RL+iltg6lXPT09mU6/QvlADpQ6 FBmeG8wHZ5X8LPsrMly+fDnUAYMW6on+CpOsAv31yJEjgUEln6FXrlxx586ec/UN9aEOBgcG3bLl S73BboU/zuannP7Krkw8z2HBMW1j48aNztYPyKJeaAu0jatXr4a2QR3RTnmG8kzNKrBOAnLQLumv N2/eDOs08dzIcrok0zSPHj3q2EoTIwzPDX5fi120mvt57lGvMkBk1YpqNx9+K/j9pC9k2R9ql2ht lIyXXYsXt4XCVMJYWxsU538pshm1zH9OFSsBCgkDPIIpA4UKwyBi7969YbB69913BwMGA7V9+/a5 gwcPul27dmWiTPT19YXBEUaPdevWhQEOSgVyYCHnXBbh+PHjYfC8c+dOt3Tp0pA351544QW3e/fu MHiLLQd1QrkZsL7kJS8JdcJges+ePeE8cpTy5qpYuYcGh3ydHPeDgza3ZcuWUCcMWvfv3++NQ0N+ IL/en4v79h8Wn/rUpwP/kZFhd+HCxcCj2LLMPf54MI5d77vutmzdEhQr2iXGKox3tJcsBk/kCX8G bfRX2gH99bnnnguKzV133ZVJf2XK16FDh4ICg0LHAIFztFHaLcaQLAKK3dmzZ8NzCtdw8rb+Ch+U pdjBnqHkk/YMpU6y6K8YB2mPzNG3OkHZ5Tk+MpJdnWB4oA62bt0ajFHwoY5ot/ZcjV0npE9+PK/4 DUPxp21wjmcrz1B+K2MHjJTkybOBPFH8rb8i2x133JFJf8UIgxy0C/om/ZXpE7QNQjFGCAzhPH/4 TbKpn7E5Kn0REIGFRmB8oRVY5U0hEFfDSMlQp4ojwICGwQyhWBdTds/gjS6Kv013IA0GKQwusprj iTJDvryRMWsnBgDemB3zSgaDndiBwToDZd7gkjeBASNskIk3nFmECxcuOAwyvCGzOkGB2eTf6iJj csvVmPKcPnM6THmw8pMXA3nq5KJ/m3bzxoTXTUwZ4P7ww/+3+4//8f3u937v98Jb1Jj5zZT2zZve A8OXuXvF8kkPFNoGdTTs2yaeO1kEPDBoA9SJKbX0GxQLlAzegMcOGEFOnjwZ3oCaokueGADYgYf+ yjMpdoADb9rprzYvHcWIZxdKL8p4FsGeobzht/6afIZm1V9Pnpoo79o1ayafoRhg6K/n/HMeA0Xs gCGcNooRBG8Y+i/9BAUXJjxDqZvYgX6AARsZeGYRaBsYUpGRdpNF4LlAfhhj4ECAA88NnidZ9Vc8 HzCCmPEBOfCKob/ST4rpr6QDSwLeHQoiUA0EmB457n+bKhl4tiFHZcN4eHHmJ6xWVgyf+1xYZPAz UXE+EmB2AjJAzM6oojFQzk1Bx+W10MDDksExgwneaiSDvTnMwgCBBwfKBHnawAZZGLxSHhQelKvY gcEgHG3Aavmh6MEHQ00xAzW7v5hP6gS3VjhYHdj95lKOnDCJGXhbePnyFdfY0Dip6Fp+gY+X8+q1 +MoueaJso2TSHioVjHlLy4vzn5GFwTiM+rxnhPXBWDJa2yDP3P5qbcN2w4klA+li7KQ/0jdz68Rc 7rNQrKxOcvsJfFDyMOJl0V/tGZr73DC5smBBf+3ruxHaRb1/XiWD9VemMMUOKKQo3Lm/Qyjf8OAa z/rYgX5gvx/JvOg3PM8xTmTRXzEmW3tMymF8kCN2oL/CnGknaf0VL41i2ihGNjOmZFGXsfko/coS mFDaS1eWGQthBD569Jg7dvxYMIDynMk60Jcx9CEHHmDZeweNhzHqsWPH3TEvBzIgU+yxYhpn8jzs Xyp+6o8+HX6H0+LonAgUQmDqaKaQO+ZJHH54H3/8cffXf/3X4QGGkvPggw+6Rx55JNM5onPFRTkY gBJyFZN8aXMfgxMGEzagsPgMVPjL4g2HDX5QunMHSPaWF2Wiu7vbxIvyaWW1PJOZMIjkOkYIlJtY IVknSWMM+SXrhHgx5ejvZ2HBMVdXXzdjnRBnIQQGSNQ7oa5u+gKvDX6+Pf1oeHhomrGmnHxQphlY JRUAS5/+S/uwvmTnY3yidMAkrZ/YOfoK3kyxAvnn66/IQZ1l2V8r+Qzt87zpr/meoRh6Ywee04Tc ZxfnaLcMTIkTc80U8qCs1EeaHJyjv7KGirVX5Ct3oL/yhwEo7XeNc/ZcKXfeyfRm66/Wlwrtr/as IQ/Kx/255Uvmr+8iEIsAff3IkSP+t3ckvKhgUWiMj/v27fdeR1syG8fjOcvLOjwC272h9Yb/jUSu NWvW+t/Blb748V+e8MLomPc+5IUNHmis/3Pk6JHwsmDLls3RZeA58H/+z/8JMuxjqtsL+1zXki73 j//JPy5xPabSjVKx2pvSzZ5AzXpA/MEf/EGYA/lv/+2/dR/84AeDS+If/uEfuo985CPR35yVsxp5 k4NCSkBRLjRwDw9wQtrgmYEahg0GajGDWatzZSBPG0BmMXg2OSzPZJlNNgZcMQN1YrzT5LA6sfqO JctQWH14wuiRm8dUFrX/I0EfGfEDHEK9N8jkBqunW/1x37rQPpGFwb7Vgcli53gWxH6za/0kVwZk sXMWx+Qr9yccLI80JdLkyKK/VsMz9NatCeOClTvJ287FZkGeVieWZ1IO6yex35rP9gylvTBYjs3D WNA3cxV066/ImlV/zddPTNZkfc30nbq18iC7tf+Z4uu8CMQigLEBb7z163vCWlUYOZnu1ba4NSjC sfsW5cKIyFSrlStXBMW/wcuAAQA5Tp9mymR8wy/PshMnToapiEytgkN7R3swRuAFgXEii3D16rWQ 9yMPPzLNCy6L/JVH7RGoSQ+IAwcOuCeffDIYHlg4jB9U5vDiTvutb30rXPuZn/mZeVGbGAnsQWuD vEIEZ+DAQIxgA4rkfXaOQVIxho1kGoV8N+8Nyy/tntgKN3maHGn5m2z54qTdV+w56oS/meoxSzkm ZE8fPHONtjM2Nu6VzvjW/WI5ljM+5Rz1dTIRppfVEwqXhv0imTEDbQ9ZaAPWDtLy41mQpmykxS3l nCkc+WSwZ0u+OKXkbfeEOrnt9WXnkp+Wbxb9FVkIlmeaHLGfoeO+H5oMuXLYMXLiJRFz4dhC2kbs ZzkymBzJusj9HlsOa3vGPzd/jqmT2P3V5EjL384V01+TBgi7z9LRpwhkRYC2h9cBXontfqHsZOj0 6xFd8+sU4RGIQh4zYAShT3R1LZmSTVdnV5gOggHApuNNiVDGA4wwLNK9dOmyKaniadbgp4gyFWzZ MtY1mz5+mXLDHA54zj3yyMMhBQyaMccgcxBTt84zAtNf+c2zAqSJy6KHPLz+x//4H5OuvMyRvP/+ +4M3wV/91V+l3VaV5xhg2ICrmE5vA+fZCmVpzxav1Ov50rfBW744peabe18heRQSJzfdYo4nlITZ vQoKGVQWk29u3ELKiayFxMtNe+7HKFJzT6WYFArpK7FZFJp+ofGKKX8ybiHpE6cQZsl0i/leaNsr RNZi8s2NW2gZY8tRSPrIGrvf5JMjq2f5RDlnf0AUWne5dV7ocSEskCFfvELzyhevkPQLiWN5JA0Q ybGHXddnZQhgyGLdAZRR86KsjCTZ5EoZB/w0gyb/tj9XscYDgHDz9tTJcBDhH5jf8lNR6/xuYLkv jdgKmWceXgGxg03lmmDxYm7I5EXwU84GwjSVF6/omwjMDwI16QFhW4Rt2rRpytt9W8zN5rLOhyri IWjzoYtxK2UgkS/YAG22ePnSKOSapW/5Je+xcxYnea3c3wvJo5A4c5GL9G2gnpZOVjxeLOfMA3nk fDFemrRxzqFE8SY3y8CP+EzBq3XhUmwWhaZfaLyZyjPb+ULSz9eGZ0u/0OuFyFFInELzS4s3W/pZ 91fLL01W6iRfO067p9hz+erdZJuNWbF55safKGeeDnv7hnyy5qZZynG+choLZMgXr5R8c+8pJP1i WGCsYBoNYw7W0lCoLAHq49lnn3Vf+cpXwlpmjF0Zx+LB+1M/9VM16wqPAYJxAIp+bjBjwGDk6cOM vYf82k84FtTlPFztOTQ4ODF1spB+mFuOQo9Z74GQy8JkGBvzcg4Nh+kRhaapeCJQDQSiGSBwXeKN PX90FB6k/NnbXbaIItDJeajSgS0uP+AWF2snLk72I8p5rMDW+fi0+GzXxsNpx44d7mtf+9o0vnhG EO69995p16r1BOUxi28xDzniWnwbEKWV0dJOu1aOc/ZjkZaWyRVbBvImDwZUlmdSHjtXjIdJ8v5C vyfrJN89+Zjlu6/Qa5Ptwt9A2a1vcb+xwJW71qdfUF7KPskD60dOMB6x2yh1jhzkZ3nmiBIOY7dR a3tpMtg5kzVNvnKcg0O19Vcre1r5YreNOt82CBjDkGOm/po8nybnXM9Z26skC2t7jCPS5LBzJutc yzzT/fn6id1DfcSWw9K3clvefNo5ZC20bRCXNGnTVsZkmvqeLQGmDT/66KPujW98o3vggQfC7gf/ 7b/9N/fhD3/YveMd73C/9Eu/lK1AGeU2MjKx8LpNgUxma23Z1m5KXivn9+ABNOqntfnfI9+BpiU9 KYfXYWJNY6YPj4xOrFFl+ZkgJlKIE3mKqOWpTxEoJ4EoBgiMD295y1vCjx6dhj86if0gMm/rk5/8 ZFjF9vnnn3e/+Zu/GQwTpghQQIu/e/du94EPfGBynhUrsf7Wb/1WYJDskMR/3/ve517zmtek8uFB /vd///dh7/D5sv4DBaGMtvtFMTsjMIDI5W5gYIUhhwFG7MGzbV9nhieTgU9kINi2ZeEg0j/2A5Em h50rhm8pYsKaP+YnW57JdKxOTNbktXJ+nyxnmFuO0v3ij+uY/8ElsOJ08nw4WYP/8MxhHiXeRdYe k8WEB/1oklnyYhm/08fJBxnsOWnJW39FMYitFNizJpXF7f4au31SJ5YHBurcQN/Jok6q5Rm6+Pb2 sBNrQVSuv1qdpLUNq6fY86HpA8jBG9K0Z2hWbSPZT3L7K3z4I05W/TWNhdWT1VtuP0o7pl8hN3yL uS8tLZ2bO4HPf/7zYbywZcuWME7auXOne/e73x3Guo899lgY727atGnuGUVKgfbEX7Fh0gsy5VYz SoyOTRgpik270Pjhd9ePlRtSZEimYf0sea5c30nbizBjMLbjt3+bZ4wY40IeuWbPbhaosyegGDVA wJv2yh/4AXvve98bjBC8dWaRFNZkuOuuu9zb3/529573vGdyqy4erBaXOMTFI4K4//pf/+tg5U0O au688073L//lvwyLz/z/7J0JtF1Flfcrc3iZk5eBDC8v80QCBILS2p3YNqjgEiHggNqAwFotIL26 W0DadkHaxm7A7qXwAaLQGgdwoFFWf4CKA1FwfQQhCElIyPgyzyEhA0OGr351376pd9659557361z h7d3ct+Z6lTt+tdwau/atQtFB2FPPfVUc/nllxuUFXFER/K9733PmRb+7d/+rZk0aVJcsKq85w9u /PNCzDJ4RvjnHRkcyjtc07HhxEY6MHlW7iMegyHSjPIvfGG5EpqkDkmafnoInwg9Mqj0n5XznDKB D3CIOknzy4RwISmzdVxXq1nHv0jbr4ho29NQCuXKY1uOcoUqz33qf7/+fV1kOHpqS8edCSblkYYC gjQQJKJ1lGvqDG0pdHulftIWovUTXOQe/nRCEnmU9oogFCX4QDhKo71WQx+KEzYskjJ9RNvWkW2v fRqiMJX9WrbXlHrgJ0A5UW4Sxn9WznPqJmnEfddIB96oF2m0V/oF2mtUAJH2msb3lXYC7nHLM6Wc QrfXcpavxtUWgfV268UXXnjBPProo9kHkydPNqNGjXI7NPzpT3/K3q/GE9oHdZP+IdpO8vNbWEAN /S2U+O2nNy+JQiRvoBIfwoP9n5NkTC285gxYZQ9QHrG1efsxV5UxquwERSCIBQQf3nPPPdeZvLMb xauvvuoygUIBBYJPCJ9itfCDH/zA7N692203c9VVVxm0vVEaMmSI+fSnP20mTpxobrzxRvf78Ic/ HA3W5vrpp582Tz75pLnhhhvMOeec4z7YbQLU6cWwYcPcfsVsc+kLlLLtZWgPwsBKugySSJOBmQjX fIy4x4BSlBQhi4ElP2ynFN0mjg8jH0gGaWKtEZoP6jh8+IoXFGlQY2OjwyQkD8zKDRo00C1l4gPg z9IdYt2v/eKJv5SQfOSMu9AXP+eLpT0YNHCQ2bF9R3abQYmFdZUMoKifYuosz0Ic6ds2Wu/e0faK 8zGI56EJwY16ieMr2ijtE2KgI20nDT5or5s2sc3ZwTZZpr3yg8c02muhPpT2Gpq62Vn//v37WSwy fai0V8pE2uuQweHrxqBBg5ziR74fkm/aCOWE4C+KI3kW4ggfW7Zsce3Ejx++qLPUzzTa6/Dhw903 Ba/w8l2DH8Enje8a7ZVvLO2V+iCCiLRX+Cq2vfKu0gkEKE8woS/kRx3jmh91X76V/piGcFIWhCEs 7dZXmkr4jJCZkTIJx31RGl144YUufsatQn77lzTkWbUd6asZ73CUMU4SHrtbrCDwiNKx1nuCQ/R5 ua4z5Z1b+hfW4rbvLhcP0bqRK15ZppfreYj7lExc+SRJCweispReltskeU/D1BcCQRQQAhEfR7/D FWeK8jx6pCOns0Jzn88BEs+///3vOwuLQsoHlmzcdddd5mtf+5qZPXu2+yjQwdO51DuxVzFLTxC8 +aBRHgjbXIM1A6g0CKegLLXZunWr27uYNNlCibJubm5uIwCH4gel2MiRI13eEegQYOg8wYIjCq00 PuYINJQJA2gGhpQJH+cNGza4wTvP0yCw2LdvvxV4NxmskMg7A+k9tkyG2SVSMgAKzcuqVaucd2/K RNr+M8886z5O9B0zps8w3XsE7aYc7gi8WGD1t9tr9e3bYAeWx5wA3LNnLzfbFBoH4m9qanI8UCdF 6UF7RSmB0JWKsGsHfmxZvHTpUrPezr6NGzfOZZ3BAmXENXU2NNFeR4wY4dqK315pJww86TfSaK/S h7a0tMT2oWm119GjR5sVK1Y6PMaMGePyjtBJex1u+4w0lDGUO3V03bp1rp6iwKbvpD+jD2PL69BC AfWOtkpbIF14QABnTABf8EjflgYxC02fsXbtWjNjhu2nrKKI9krfDl9ptVe+XXxfUdhRTyD6UtoN 7SSN9poG3pVIgzb21a9+1SnYKF9fAYFigX7qy1/+shvnvvjii4ZJNPol2gFH2gfjTcLSPq6++ups Nh5//HHz1FNPuTgJy4+wtOUvfOELbnwwb948w8+nNWvWuLJlDEec1UzkhfZI/1CMQq5Hz4wVKHhE SZZnhFYyUoakIQoknw/KlR/1AQVxKKJOoESkX4liITxgHZdGv9s+j+2VQ+3DxN9hW9OhQzNj3e4x jkbj39K79YZAuJZjkaJx+oIMnXkuQhDhww3RWSEM5aJnnnnGfRDOO++8XEHcfT4IP/vZz8yXvvSl rPIBR5SPPPKI+eIXv5j33Xp4SOeFQ04GSCtWrHAfNhQ7CHbNdmCSlhIGYZ9BEgoIZspIl7rAQDqt wSLlSXp02tQBBg4MWvm4YGnjW4iELHvKBBNKBstYBpEu9Z2PDIqa0B9VyRt1YKJNj0ErAxrSZcsp Bs0opuAzDXrppZfMypWvuQ/opEkTbZJd3DKsX/96mysj6m9oBQT5bLJ1Y731ut3Sst6VyRHqhh38 TBg/LrW10GBOXaS9Llu2zLVXZt9QPiD4pdVe6bOpowj7y5cvd3WD2SsUEwjkaRF9FO31tddecwpD BmFcY0WXZnvN14em1V4ZyI+zeGzZstm217W2TLq5mXbaa5p9KEoh+kzqBksgKQ/6L+oL/XxahNJ0 vVWQ8V2jL4cHBugoAtISuFkGRHp8T+CDa75vKJZRBKTVXhFE+b4yfqKdUifhg/aaZt1Iq+zTTAcc 6fPAlsksGcPyraZfAHepb1j/UO60jT/84Q+OTQRDnJ7Tf9NGfJL2wjIKsaKh/yduf9zsv8P5b37z G6d8mz9/vuMh+rwermlLXbt2s2O09j6AGLdBfezStJDU0ypB4OOQtTxzvrG8lbH0eyhC+vbpF3ys RN9PvcOC2CcmSSwb7lsQ3aLTDxfs3KZtUSgpevBTUgSCKiCAl9kKoVwWEHQo3/jGN5xASFgGmmLG KO/KkfuExQOwaPvlmX9kxuqBBx4wl156qev88S1BpWdgn9bAwOenUud8FNGSM0Bj4MjgBGE3LQFT 8s3AlVkh+IDgIa3Bu/BAuSPUUG/gg2t4SJsPPij4OKGeUyYMUvjQpV0m/axZ95SpU9yHDT4yZYLp YxeBLPjxkksuCZ5GkgQwYRw/brzDgv6IulGJMom2V2knadcNhCgUH2CBYAcf1NM0+aAMGIwjSEl7 zQxM07Veo0xmzpyZba+V6kMH2mVT/Qf0r2h7pfzpPxHKKBOuK9GHIvQhAMIDA/NK9aFSN2gnJ/rQ zM5fSfqdcoVBaSzfV8Y50l7LFX9njQdccfrIuJOJLFEs4Aid+ucTygN+WMWwawXKCxRld955Z3aZ hh+ebTT/8i//0nz72982P/rRj8w///M/m7/5m7/xg7Q7RxnLlpwf+chHzHXXXdfueb3coD33t76Z Xn99n/sGUZ+FKAue9+sb1hcR4yCsNhD+33r7LdOr9wnLP7HQHtIYfukbPOy0yl7yLUorsMh8m486 hWd3Dx/BKfTR6RBUjxAa5rqOP7gCwje7yqWAeNr6aFi5cqU544wzzPPPP+8GFDS2OKLzxewVHxO5 BsOYC2PChtLhP//zP7PRMJBmhuCaa67J3usMJ+AkWvpK5pePBsJ3pakSA+a4PFdDmYigHcdfZ7xX DXWjWtordaMa6mg1lAltoRqwqJb2Wi19OQopfpWkamkn8OEvea0kJvWWNrj6VldiCRGXT/orrHJQ QGDZgGIqF23fvt387ne/MzfffHNB5QNL81CCfOITn3AKDp+fXPFX+n7GUqB4KZVv4MiRONo86BQ6 YsnDpA3jewTxfv36Bs8eivjt1j/U1i1bXZnSxsgTy78aTmpwSvrQTFCXBtr8othiuR2TJfDAMiuU JGlZJe7YsdMujT9kdltlCEcUEK+8/IqV196x4/oGZz2bSyYLjZHGX5sIBFdA+Bq7uE4bXwAPP/yw wekkpscoIGhccWEx4cf3A2vy8g0GWVeHiTvER8AnGgjmcEqKgCKgCCgCioAioAgoAopAPgQQPBEE heLGp/KMpUEIyRAz5fkUEDhpZwmNOGKXOKLHPXv2mK9//evmQx/6kGEnN5QcpIGCA+vSeiQULKNH j3HLzt62Fgg9etglTtYaAcVDxg9OeEs4cB5vl2FuaNlgZYq1TtCmTJEjmsel478MZS9LRFvs0p7V drlsH5acvfWmrVtvWp9MzakpYH/1q1+6JZkoPYYPz9S53/z2N2bRH37v/IZdd+21drnsCUuV/HWy eKVU/vj0aS0iEFwBgRmvUJwXXCwaIEzKWDIhFLWWwNyIfY/PPPPMnNttyruYdVeLabfwpEdFQBFQ BBQBRUARUAQUgdpDwPfLEB2fSm4Y4zJJxtIgCEteLG/jCOuHX/7yl24pBUJmLmK2/b//+7/dko/L L7/cLQ9kku5//ud/3LIbxs71So12iQM7AKHwQZHT3zq0RBGU5kw76U2bNtVaT7/htuVGMcLEapo8 9LRLzvA3Qv2ibjVYS+Kmpv5uKUpaZc/ug0qKQDkRCK6A8JdgYI1AxykNF4uGX/ziF25JBKb5NHQ0 zXTY0Q4eywgcomGCVimCd3guhfh4+R+wUuIQ3Ep5V99RBBQBRUARUAQUAUVAESgeAX/8lssC4oUX XnDOpdkhBQfPWChEnQeSsigQ8BFx9tln52SGd/EPwaz7u9/9bjcG5l3SZ+eTz3zmMznfrZcHLLHy fclVIl9drFwyYGB6Tnbj8sj437cojwuj9xSBWkIguALCbzB0omjvZPnEd7/7XefQas6cOQ4zFBBo glFA+B08677Ysuhd73qXc+pTKYDRWN92220FkxdFgX9kx44PfvCDBd/VAIqAIqAIKAKKgCKgCFQC AX/cIueV4KPa0mSphFCcNS+TZvfff79zkP7ss886BUTcZBpxsGsKDi0XLFggUcYeH330UfPjH//Y ORx88skn24TB6ShLEZQUAUVAEahFBFJRQPARQ2uLBYQoINi6im2N/uM//iOrkBALCID0LSDYdoh3 L7jggopiTIefRAEhTMrHm2OpjrJ4V0z4sA5RUgQUAUVAEVAEFAFFoNwIyFiNMYeMX8qdRq3G51tA +ONTyQ/KAnbNwJ8DW8ALiT8IuWY58Q9/+EPz/ve/30yaNEluxx737duXc+IKhQhjUiVFQBFQBGoR geAKCKwdEL5la02ObKmzcOFCM2/evDZbGbGllayFEwsI1tA98sgjzkcEnXslCd78JSVp8CLWIChw sCBRUgQUAUVAEVAEFAFFoNwIYPLP2AsBO+rAu9xp1Vp8TJAJgRFjMlHSsEMF/sy++MUvuuUC/g4V UQXEc889Z9ZYZ4Js7ynvS7zR49VXXx291aFryreUcSRjXz9PHWJCX1YEFAFFwCIQXAFBx4V/BxQP ssIjiiAAAEAASURBVARj8eLFhrVyONERhQOlgUZXrsXEjfVvbIWDsiIX0aFjJeFv14alBUseTjvt tOxr7KGMlpoPB+nQGZ9//vnm1FNPzYapthP4FNM/WbpSbTwqP4qAIqAIKAKKgCJQ2wgwhmKmnzES AidHpQwCvgUEE2OMH5lMg7BomDVrlpk5c6a79pces9ObED4hHnzwQTN//nxnLSH30zoy5mZ5SFJC QcIP5ctXvvKVggqTaLzyPvcLKVui7+p1PSPQpZ4zp3lLiEAqCgg+ZGiBUUCgKWY3CzyqopjwiU6O DyBEB0/HjcKAdXL5LA/Yhuj00093VhUs7UDp8LGPfazddpsI8nwwfvKTn7gwKCjGjRvns5D3nH14 +XiU8lH+i7/4CzN37ty88cc9LCWtuHj0niKgCCgCioAioAgoArkQ8Mcb/nmu8J3pvm8BwfgUZQLj yY0bN5pf//rXbptMmSTyFRAsoxBi202W0uITrBL00Y9+1HzgAx9IlLSvMGBc7l8nisAG8t/RJcRJ Uav/cF27ZhQQ1A+tF/Vf3rlymJoCAgYw6WPXCzpuOsIo0cGLBQTr5Nh6CM+/hSwUxo4da/jhFRgF BOvqzj333Gj0br/k6dOnm3POOcd8+ctfzvqeaBcwxw004KU6koS/UsgfBPjnpcSl7ygCioAioAgo AoqAIpALAR1nxCPTu3dvN2bEmpexrCwnvuOOO8yll17aZjmxv/38nj17XIQs2/jWt75l/uu//qvk 5QwoPhgb+2PleG7j76IgESVJfIjy3oVXJhKxSG63pKeL+jQrL9q1ExuT0fv373PKB7ZXVeqECNj2 n4oCQiwd0AT//ve/N1/4whdiO2AsFEQbtmvXLvPYY4+5LYjEKqJQEQ0bNswF4d04YhcLPA9fccUV JXXCfID8JR1xaVTjPT6UfAQ5UhZsaSSmg2nyy7IaLGHQelLW/ixBWnzQ2YEFH0N8kzBQkPqZFg+k wwcZPlDGVbJMSJ8yOW6djvW2dWOg3WqqS+oDg+Nm3779bu2xW3LUr79p6NOQZnG0SYsB08svv2ze 8573ZC2y2gQIfEE7pQ9j8Ib1WGNjY7ZfDJx0m+hpr/TZCCTSXv0ZrTaBA13QXnfv3u3aK/0vlnCV aK/V0ocePnTYloltr7ZM+vTpa/r175t6e8UvEW0EgYpvM2WC/6a0SYQbhDLqBn15msKV5Je6gXUk JvngwJJRGcdImDSOtFWZbadMmDBJu72mkc9KpcG3iXpGefPdpP4xnmVr9osuuig7eQZ/vsUu/Sh1 gwm1M888s6DjyWj+1q5da/74xz+6yTXSZOxA22NXuI985CMV36Iyyq9/Tf9Au4RnsPOpS/eTzPEu +9mT1L+t550Agd69T7JjCjvWtHmtRF/ZCSCu+ix26dEQXgFBByTOaxi4nHXWWc5LcBw6hPMHEB/+ 8IfNyJEj44LG3hMnlTt27Ih9zjq9CRMmFP0BiI2sRm6Cxbp169x6Qz4ECL0tLS0OBwZKaRCD5ZUr Vzohgm2jUH6sX7/eDaJnzJiRmjKEgSrbX5E++3TzMf/zn//sPuBYzaQ1WEOgwlIHD9aUAYNGfKKw HEjqcPhyOW42b9psdlpBd/To0U4ZwyB6y5YtbqvbtIQ8BmarrUOuI+8caW3rx81aW1979uxhJk6c mMrHCR5QTrJvO/XhmWeedXUC66ukys9ylRd80DZOPvlkN3CjrtBeqZ/+oLZc6cXFg9BPn0FfgeUW ijrMjLk3bdq0dgPJuDjKcY/2umzZMjfbxzI7Bv6UD22kubk5tfZaDX2oHaWbDRs2mr179ppRo0e5 MoGvjZs2unbifzfLgX2uOBAkXn31Vdd304fSZ1JGCN7UjbQGkyhN8elEm2DiAYUyOw/wfaE/S4vo M2gbTU1Nrr3Sh9J+J0+enJqCnfbK9xWhlHZBn4XQynf3lFNOSe37mhbmlUqHuk0953vNDC6KB8aU LCdGQeuTf01dpe9kG8277747cb/Fdwl/DT/96U9dvb744ovdN5J+kDb38MMPuzr/r//6r4m/Db/8 5S/N0qVLfVYTnaM8uOaaaxLznihSO9HRtaHRHH/TKlSPvpPoFQ1ULwhklE78pZ9S6jwIODmrp504 6dkvvAJCtMbAy8z7hRdemPeDKOvs+KDTsRdDCHR8JBgERInBCQOFSy65JPqobq/5gDFIY6DIwATC 6gDBe9WqVe5jGtVKu0Bl/sMgjRldBkMiRCHgypIZBq6hCeUXnqc5sgxHlvowWEMpwYABISc0USYM FsFh/PjxLjnKhBkVBinU/zRmE/fufd1ss8Iu7YwZdogZsxWvrnAC70SrqOtuFTWhCYXHwQMHrfOu U5xQRXoIU6tWrXbttdSlS8XwzWDyG9+4y3Tt1tWMGjnKWYMU8365wlIHaJe0V8oFkrpBHcXPTRqC Jv3ntm3b2rRXlMMI/yjOaMehiXZKO+FjhTAn7ZV0EfBoI2LxFpIX6UNJK64Ppf9Mo73u3r3Hfdea xzabIY0ZxTFpr1m9xqxft95MmTrZwsB8UjhioEgfRXvx6yJ1EqGG72sa7VXqBt8Q6gZEf8p3bcOG Da48ULaHJnaKoC6i8KDNQrRXsJB2kkZ7pQ9FUTl79uxsXaS9YsVFf8L3TqnjCDBWEMU8yp6nnnrK fTPjlvvKOIdUGfuw9ILtOVGqJyHq8ve+9z23s8YnPvEJg/LBV4ajoKZPZNcNHEt+9rOfTaQcQEFX isBHXxNkgqZrD9OlYWjgnisJ4homTQT2H20wm/ZmlE5v9xhsuvZLPtGcJp+aVlgEgi/BEK0x2Uiy 4wQfcGaoL7jggqKsH4ifgSDvM3NHBy7LDBhE0pljJpfm7Ag8VZIQ/PloRPeKRtBmphUNPrPuIYmP HY5HEfD9WQE+aMz+M3hCox96oMYsBKaQDBR9YQYemOEFK4SM0DN4YAEmUeEJgZMyQfDDSic0bdu6 zbaPnlYpeGKgTl0ZZK/BYr8dXKMwDEmUO211kF32QRkI0Y779evrloZQV0PXDRQvCxbcanrYASZK mT/96U/CSqpHBDjqX7S9YgWGgMFAVgSdUIzRb9ImGWj7g2jKAEUV9Zd25LflELzQXhnkM2D22yvt BpxoK2ksTZE+NKqclD40rfZKX00ZDBw0IAu3W0LW2l5ZwsS3LyRRHrRX+ky/TdJ+UJzCI/XTF5RC 8IPlB5YYoqSTNLimvCgT6m4QgUkSs0faAu0VYdAn6ijKs7TaK7hTBr4ijO8r7QMs0mivfv7r9Zx+ SKx5sYLA8eRtt92Wvefn2/92YsFGfb355pv9IDnPsWj58Y9/bO677z5z7bXXukkzvw+UF1ECotD4 wQ9+YLAWTmI9iTKqFhVSjOEZN8mYXjBI84jiEz7gIXTfkitfYEBdgofQY9V8PDBOoJ+vFA+5eNP7 tYVAKgoIPsj4Tihk/QB0CF80LpQVxRKDIAZhrE9loCQdMmZnDJpYf1epjqPYvHQ0PB8xZjLpIKKd NoNHcACj0AoIyoLOigFRtLPiHh0qwlUxS21KwQaBhg+IP3CWeOCDQRpmvAzkQhFlAh/gIDMpkhY8 UCY85yMXchCP4H/w0EE3cIqWSc+evRxLDLD8QZTwWc7j6zYN8orviSj16tXTtuN3XLlEBfJo2HJc R8ujHHEWEwd1g7ZCuUfbK3UDoj2HVkDQBvjFzSCLkoi6kYYCgjxHy4X6Ch/MPqfRXqUPjfYb0ofS XkMTFgdYxwywStxc7ZW+PLQCgj7StdfW+ujnGzzAgjCh+w2+F+Ag7UL4kLoBH3xzpL7K83Ieaa8o GEgjKhxKXaE9p9Fe6c99ZaHkE97gM432KmnW85G+WRQQjCVYTozVSRwxjiA87YXv+tVXX524XWC5 cv/99ztLs49//OPt6pekR/miCEXZheJNxrvyvBqOjO/4lUosw8OKh3pMfkUpzXgxrbE8fQlLmujb 6HPoj+njkFXS4oF6JIp3FI0oIfg2ooAClzSIckShieKVNMGF/LM8NOS4OY28aRqVQSC4AoJs/d3f /Z2rqNEPdVyWr7/+ehc2OtCKCxu9x8dBBmE0EjpkBpCs0+uI5+FoOrVwTYclH784LLlHR0qnErIT 5YMBxZW93JMwIXHlQ5aLDwYKDCjAI2RHSnnw4QBvybufZykTOnZ4CkUIM1DG2WRbs+1utl5AEsZd BPpzyCpBoFz1k2dp8EE6lSbyKYOsKB7UF+7RTkK3V2knUR7AR+osgn9Ioi0KH3HtAD7AKo32Wg19 qODd1eY7uswizfYq/XRc3ZByQjEUUgGRpA9FIOcXcmBOe6UtUhejeHDNL43va6F2QjuV+hOyzXaG uClrUXpRx5lQk3ofl38mxFCGnXHGGebss8+OC9LuHtY9CxcudHUHi11ReLQLaG/QT1LPIZRh9Ua0 MZYzoVwTf1D0QdyjH0hjuRdtfPny5Q5rJlGZHABzeODnL+cNhT/ljPIB4R+rF+og+WdJMcoq+MpX D8vFF+mzBA/FC7IVfCFngcOsWbPaWGCVK02Np74RSGUfHBqHDF4LwRn3QS/0jjynw5eZOzTCCHvf /va33dKLqAmtvFOvR4RYBulQVMHANT86V/mAhcKBMojjwb/HQC00CR/RwaLPh4QJxQtlwofDT9NP S8pEwvjPynkOH5CtAu7nx92ldX9meDh+/Jj/qOznR7J8tFWCuIRad+J45+3O4ZyKupekvUrZlb0w WiOU+CvZThjYSFusJB9gkaRM0uxDabM++e0VR5UhKV/dkHIKrTCkXxI+ot818i58pFUmcXjL95W6 I7zGhSvHPYlf8u3HKfhIW/Kf6XnxCICxLHMpZjkxW3RigZuEECqfe+45lw7bxecj6hfL5SBRjOQL X2vPEG5pxyzDk/rNGB8sybco30LmC8UOlg/wIJaJWDix3AvlElZOoQmlC8uskGGknJGpsAKhv5U6 EJIP6hrlgeWjLB+mTESuWm994SBPKCkCxSCQigKiGIY6EpaOQUwRUUA8//zzrvM477zzOhJt1bwr A4okDBXqDCSu0MJurgG8nwcJ498r93m+NAQLBJ+QRJlIGpKmn57cy8erH77Uc+EhM5vaVqIRHjK8 hv2gHD2awVvS9PMjXB091jn2iJYyj8PCx0XC+ffKeS51Ix8fhJFw5Uxb4qLu5cun8BaSB3iplj5U 8pnJt7SMDFqCRTrtNXdbFD5Cf0/AolC5gIxglkGp/H+lfkq+/RT8exLOf17O8yTxJ8Usypefj+iz zniNwFXscuIPfvCDbulvUrxw9ItSiZn16PKzaBwIpQiFCMahl7BG0w59Tb1mqRV5i+KAlSr9DM9D Ev0MFtRMioriSdITS9k0LE9YXgceYt0tPIALigisZpL0ifJeKUeUMEwUogDy+wXKBysdlnmJMrSU +PWdzolAODvvCuEp2jl2f2Dt2Gc+85m8ZmwVYjNxsn5j988LRRC6QyqUfvR5HD+Sn7hn0ffTuA7N B/FLGnKMy1e+Z3Hhi72XJH7LaugJ1SzbFpXsebsTx0i7uyncyMNTgNQLlYm0lQBJlxRlIX5LirTI l0LzEDr+pNlNwkfFmkkkE0l4jbxS9GWSNJKEKTph74V88fOM9povjBdVh05DpOH3Nf55hxitk5c/ 97nPubKVGfl82fr3f/93N3OfFEOE6pdeeslFOWXKlHxRu2dYSiCYYv5ejf4fCmYgTwBm9rHciSof eEWWG4ReWoRAjdBNWUfLmzLll4YFBMvaILHAcBf2j/CFlQg/sY6Q5+U8ogSB4izZKQ/qIRYpIZe9 lTM/Gld1IFBXFhBAKgqIxYsXu3VjdM61THTAaFvRPOZbDxjNY7TDjD6XgUuhcNH3ir3OF7/wENep FZtOofCShqQZFz4fr3Hhi71H/PnSEN7yhSk2zbjwJ+KP0TK0SjKYdnft1na2NS6ujtzL8pFH1u/S lXXv6VPaAp1gIXUgmmO5L+Giz8t1LfFLenHxMvCScHHPO3qP+KulvebLi2AUEgvSl/gz6UUai9de LWxBScokLhHBQoSDuDDluAcW1A9I0oyLVzCLe1aOexJ/Lh64H7qdkI98ZSL5FLzkOt8RAQIza8Yb zPoW826+eOvlGXhL2RfKE2GLxQ+hl/hlKXGuNJiZf/rpp91jtpYP7RQ4Fx+h7iNQ04bi6rfcC73c i7KQ5UvRMpe2jdIopNUXGIAFFO1bhScsnITPUOUhWEuafjrCl4Txn+m5IpAPgbq1gGDN1qc//el2 jTYfGNX4DC/a7PWMhhFvs0mJTlo66lyDJOISj91J4y02nHROcTzIvdA8wDMDKzTmkqafDzHXDa29 lTKBhzg+hKeoplvul+vol4llxQ6STsTMNdTNCv4ZJ5WZ6xB/u1vzPSgOi+Oty2G6d6+MAiJEfvPF SZlTP3LVDcEodB2Vuifp+TzLPepPsQNrP55C58RdTe0VfiXvcbyH7r/YLld4qGR7lboh/aWPheAT un7SRuCD72EcCW/Ca1yYctyT+Mm35F3ilXtSj+V+iKPgLfn20xC+immvKB3+/u//3s38stZexhB+ vHoeBgHKiR2fpNzypfLb3/7W7X5xyimnmPe97335glb0GW2glG+FtO+4d+VeSMEf0GhT/GgDkmYc mPAhY6q45x29J227EA8dTSff+yF5yJevfDzps9pHoO4UEAjsfDjZczm6ZqoWi2v16tXmlltucR0h WzL90z/9U6JsMChGWymdqP8SHzg6eAZRoQcYsnYu7mMhHxkJ4/NY7nMZqEmafvzwRicY0oSN9CgT MEejHceHlIkMbn0ey3neu1dmW8djx9A28DuhgRCfC2yDGZoEb/m4+ekdbVVA9Oqd2RbUf5baeasy Jo30sHTK1V7BhzZL/Qn9sRZhOlf9BAspt1C40CdJe83Vb6TVXnOVSZp9aIOtG9AxJ3RXrr3mqxuy /jd0X07fSN3ANDqujtJW0qgbYhoubdNvC3Ivzfaaq53AVzHtFcd6t912m/OdNXXqVOd1X9qin0c9 D4MAuz38+te/dttq5kqBLSF/9rOfud1mGA+GHi/k4qPY+3Gz57nioA0VoiSKmkJx5HtO/EnSSBIm Xzr5niXBgfdD8kD8SfgohQf66tBjGvhXqk4E6m4JBlvzPPTQQ+bUU0+tTsRL4IrBBR2AmGIliYLO HgUM70UHJ2LeJg47k8RXahi2q4IXTMSiHZQMWguZG5aatv+eKKOipmrwBB5osGVQ6b9XznO/TKJl 6ZdJSG06+TmpobcdtLBHPPWqrZQtZZKGSefA/gMcvG+9ldkW1McaPsCrf79MGP9ZGufRuho6TQaR mD3TVnO118bGxtBsOB7gJdpOSFjupdFvSHuNmnVSLtQNlqOl2V5zlUkaWPTt18f2T7ZMbL5ztlfb 14cm6ROkHvjpUU60V3HO5j8r97l816J9KHUD3ngeWigjfnGGF+0rqJ/cS6Vu2D4D3ON2kgILBvfF 8IFShx/jhiRCR7nLtrPH94EPfMBMmzbN/OEPf3BKoCgeOJ38yle+4socRdHMmTOjQarqmrrELg78 ZGyRhEHqdCEKLbjCQ5I0koQplJdcz30cov2M/05IHkgnSfw+rz5vced8T6kTuayS497Re/WHQOFW XoN5xgIiSYOphazRqKVhRwfBhfjHMzKDCHEgI+FxnMMsI9YioQlhGr8c0tlIeuQFL8YIEgheoQlF CLNz5N0fWMEXg2e2E0pjpgcTS9JnpsknPBlTzjiTCl93u9j8DncDAnFwJLzse32fG7wXM2iVd4s9 9j6pt1u7um/f/jZCN+Vx6OAh96yhITP7W2zcpYbnI5/95XOOWWoCed6jvTJgi7ZX6gbtVfzb5Imi w49oA6SDQOM7+YIvPF0j9NOWQhNpMHNLe/UHXrQbBE/6+DTaq/ShrLn2SfrQdJy/dTFDhw11+fa3 nqMfkfY6JIUyQbBHWUw98L9F8EQ/goIstFKIMiAd+kr48OsG7QSiLw9t2Uc6o0ePjm2vUjfSUBjS BvimyHcMviDaCPjwbS2mvdLOBVMwTAPHDMf6FwRYOnzllVe68rvxxhuzWywivD/55JPm+uuvd2FQ QtTCJBvjP8Z4/IqpSzIJ44/VpIZI/ZQwcr/cR3/sLWnGpRGSD8aCgluUB7kmTEgeyLPEL2n6OMg9 4dN/luuc+KgTfOPl/Vxh9X79IlB3SzDqrahoqDRsOmJ/0Jckn8zQnHzyye4jhvDN4JHB0bZt29z9 NGar4BOrFAZDLS0tzo8FeVpv9w1mkMQaxvACd8aD8Lhx48yKFSvMhg0b3D7OzBCx7zaDZgaTaRCY M8hg72ZmFBmkIlQxs4EwIzO/oXkh3b179ppNGze5jwDmwtt3bDcHDh5w5ZWGcMcHvslisdLuWEOZ jB3bZOv5cYvNZsMSjKYxTanUDdrWAw88mG0bDOb5/eu/fsW1mQkTxpuLLroodJG4mUrqAHWDekBd QUmH8EudSUNRR1ukLdBeWf7FbByzvdJeMctOu72ypTLKUvoLzI/BJg3lKQXu96G012gfmlZ7HWHr xX5bJhs3bDQTJk5wypet27a69kq/Vszgr9SKTBrNzc1m2bJlZt26dYZ0GTxSJvTpXKdRN+iv4YM6 SZmgcEBZRh+CYioNizowRLAnPfjg+8oPJQjtlW9eWu0VLNgmj12/2D2Bsti4cWNJ31f6QhEIiIc+ WildBObNm+f63YULF5oFCxY4JRfCGsrQf/u3fzNp9cHlyDV1yRfkk8Ypy4bixrxyT8IkjbPYcIyB +Pah/IkqQsgX92gj/EIS+UTJK/mWtKStgm9oHkSxjIIySsKXhIk+j7vmOyF9SxrfjDge9F7lEQjb ciqfv5rngEbKwI9OsBgTNjJOw54wYYIbRLNnNAN5OtXJkyenNkiDD9JkNxIEq5UrV7oBDgP70047 LZUZM3iAGJii8ACLpUuXug6Q2SOEvtAdeIaDTJnI4HT79u0OEz5yOBhlMJsWUacmT5lsdtjBMgP3 Y/aD2rtXLzeAZbCTFvW2wsTUKVOd8mP16jWuzjJwHzVqtOmZgh8K8kk7mX366a59dbHtzV46YnHK UWupE3qgk0ktwwf1ACEGAYKPPYqhGTNmpKaYghfaK6a99BfSXikT2nCadQMlGbN89BvSXhmEY6GR hsANFtKHomioZB/q2uvkSVa43eUUubRXfEMgcKZZJtQD6UOXL1/u8MFaim8K9SYtoh6Qb8oEgZ/+ u6mpydWNtAa0fJtR0KHQR7leyfZK20SJLe2V72sp7RXBShQQfJfSwjKtelMr6TAm+Yd/+AdnnclE CWWBgMexMxD55NtH3qOEIhqSJWHR5+W6pk+hHe3atcu1bT9ehG7aSTHWRf77xZzT52IVKfmWd+EB JQT9bjHCv7xfzJGxM+OBOBkEvsCqmO+Q9DHw4J8Xw5OGrX0EVAFR5WVI50JnjFm6aBqLYZkBBB8z fjT0Sg0oyEeznamBKskHHxR+leSBMkC44ldJPhhAM7PKr5J84GgS4SHDAzWkVQPAaQpEeZw558wU UiqcRLW0VwYU1dJeEbIrWT8pExSV/CrJR7du3d1sPzP+leSDmX6c5cEDBD5pE2kyKOZXSSzgAytD fpXkg+/ruFaLlI6UCUoUme1FAFSqHAKM+zqLwiGKMt8f+luEXqyb6HMg2hjLvajvaSwThQcmBLAw 8gVsrmn7aSyLZGIKJTyWsvAjhExAe2XsFLoPZsxM/ikL+gexXkD5AB/w2FnrqpSHHotHQO3riscs 1TfoaPlBcY6mimEmdCeVlJdq4KMaeAAv5eNErclgkb4wc4KD6jrTunGiPBSLE1hwVg14wEO18NEW ncpcVQsWHeFDZnZBMPSsamVKSVNNEwEUBvxKIZYdYn0oy4mIA0sAlhGzTNBXCJQSf5J3UHAyUYSV qIy/OXKNMiCNJXgoX1A6Y+mFEgI8sQzB6oo2ivIzNGGBR3mQdyy+4IG+Yr1dfoayCKteJUWgWATU AqJYxFIO7ysgWAfmax9TZkWTUwQUAUVAEVAEFIE6RYAZTbG0TEO4qlMYNVtlQADBlmVEr1kfUUuW LHGCLvewvvItAcqQVM4oUOaxvAyfNywDhLiHQgA/RGIJkDOCMjwgPaybsDAAC6750T6nT5+eCg9k A8xJF+ULFhkQChCWSaa1TNUlqn/qBgFVQFR5UdLAxRRSBgdiEVHlrCt7ioAioAgoAoqAIlAjCIhZ N+ymsb69RmBRNiuEAGNf2WqUWXcE4LSJ2X+UHlAlJwCx+uBXKRzAHiUEP3iAKlEeLmH9UxcIqAKi yosRBYSYmqGAYM2XkiKgCCgCioAioAgoAuVEQMzMiTNNp8jlzIPGVZ8IVIOwm4bFQ6HSqwYcqoGH Qjjp8+pHQH1AVHkZoQHGCy7Euq84L7RVngVlTxFQBBQBRUARUASqHAGczAmpBYQgoUdFQBFQBBSB ciOgCohyI1rm+NA04ggHYnAg6zPLnIxGpwgoAoqAIqAIKAKdGAF2GIAYd6gPiE5cETTrioAioAgE RkAVEIEBLkf0Ygp54MABtYAoB6AahyKgCCgCioAioAi0QYBdBiD8TKkCog00eqEIKAIBENDlHAFA rZEoVQFRAwU1fPhwxyX+H3bv3l0DHCuLioAioAgoAoqAIlBLCIgCAt9TAwcOrCXWldcqRADhUgRM OVYhm8pShRDQOlEh4KskWVVAVElB5GNDFBCE2b59e76g+kwRUAQUAUVAEVAEFIGiEGB5p6+AUAuI ouDTwDEIsGsEvsv4yc4JMcH0VidDQOoFPu1UCdHJCt/Lru6C4YFRradjxozJsrZ58+bsuZ4oAoqA IqAIKAKKgCLQUQQQEmWC4+STTzY9evToaJT6fidHAKUDii1+6kC9k1cGL/tYc1MnUD6gjFDqnAio AqIGyp3BgND69evlNPGR7Tv37t3rtNDsqoF360oMLvbv32/4Qf369avIGlM6PrBgu7Hu3bs7M1PM TdMmBnvwQdlUskzY950yOXbsqOWjt8VjgP0opG0Yddzy8IY5eOig6WrT7te3n2no05B2kbgP4t7X XzdvWDzY+hYT5N69e6fOB3WCpVZ8oEm/sbHRVGL7LxzSZerGsWx7TXu2gvbKrCztlT6LMqlEe62W PpTKyKB+yYtLzOAhg01zc3Pq9ZMB4+u2neAUuVu3bq4f79OnT+p8INDQh9KH4bMAZ830pWmT316p m/hsqkR73bdvnxEnkv3793dttpj2Shvbtm2bg2/y5Mlpw6jp1SEC9A98SxlrVWLMWYeQ1kWW6K+p F1AxfVRdZF4zkUUgbUkjm7CeJEeAwYQoIVpaWpK/aEPu2rXLvPTSS06QQJhBoOCaAWRaxIB59erV 5rXXXnOdDR+jtWvXmldeeSVVrTgD1WXLlpmNGze6ASsD2D//+c+OlzTNAxk0UwYcGbDiXPTFF19M 2b/HcTfYXLFihTly9IgdHPS0dWWnWbFipRP20qobCNmrV60xGzZsNN26dnPJrlm7xpVJmppx6gZ5 f33vHidQkfarr67IzgimhceOHTtc3UAYoL3STl944QWDcJEWIfTTzyxfvtzNTiDUrbeKT9oOwlZa RJnQPrdu3eoES9rryy+/7Npvmu21GvrQw4cOuzbx7B//aBYsWGBu++pXXX+aVllIOihOX331VbNu 3TqnfKAcqBf07Wm2VwRt6sbOnTtdO6FeLlmyJCtAC7+hj1gM0HeTPn25X1dCpy3x015XrVpl+68V 7haCHteUC/1rUuJ7xA+aNWtW0tc0nCKgCCgCRSHgf7/986Ii0cA1j4BaQNRIEc6cOdMNxLds2ZKY YwYfDAxHjBhhxo0b54T/YcOGuXsrV640p59+uhPEE0dYYkAECGZWTjnlFDeDSTQoVZYuXeoG1VOm TCkx5uSvMThG6cEA+tRTT81q49HCgkXfvn0N2IQmyoT0sAAh32h/hw4damQQCT9pzPDue32/2bx5 ixk9epQRHyNYxry6/FWrDNhgJk6cYAWM8N3D1m1bzT6rFJsxY3rW2qBPQx+zyiqsqOujR48OXSRO cFq7dp2tE93NuOZxpnur6TGzmGAEP3379Q3OB0IMijra6/jx41161EnaCcLF7Nmzs/U2JDNYX6Ck mzZtmpvNJS3Wg6M0W7NmjZk+fXrI5F3ctFfaCUfSQ2kJIVyhHKGdyO5A7kGgP9KHktakSZPa9aEI amm01//7+P81zzz7rGm0fGB5UIlBG2mCPcI/fajkm76MOkqZpNVeaQ/MolFHxdoA/ujj4YPvS2gC B9oDkwN8XyH6chTrtGO+2WnM+vJtRRFCmVAWEPlHWQd/Sb+v/thC+h8Xmf5RBCqMAN8BfvIdqAQ7 9C+M07DqqOSsPd+kSvIgONDvSt9bifLQNGsfAbWAqJEylBkJZnyYCUxCmzZtcgNVBkV+h4nAiSDu DziSxFdqGIQZBkT+oJBBIqbl5CeNWVVmkvkhSPiDQnhgIAuPdKyhCR8efMQoA79M8PNBuaZVJgj+ 4NDYOCSbZfgZNGSQEzBYEhGa3n7bLjXYtcct+/CXOiDs9+/fz1mEJK3rHeEVq6BDdvkHChhRPhDf MNtujh0/ZjZv2ZxK3aAOQrRXn0aNGuXaK20lNIE3/Qbl4Qv4WEHAF8oJLHZCE22VcqF9+oNO2g2D L9pJGu1V+lCETL+9pt2HXnjhheaO2283//Iv/2ImTJgQGv7Y+Cl3yp92IsoHAtKvswQDrBgchyas hLCOQTnnD4DpQ+lbUXinUTekrx45cmSbLMMXSiKsIUIT7ZX8okAX5QNpUj7FtlfKDwJT3++Uu6l/ FIESEKAddqQt0p6xvhMlPFZ5aVoDkmUUH7R1eEApjmVRGt/iKNz0vYIDPDBegLc0iW8Ak5qUAz8U vmn0+WnmUdNKDwFVQKSHdYdSYqYHosNhdqUQ0ekzAGIwgYDtE8IFg2nxeO0/K/c5ggTKDtL0B4uk wz34FLPPcqftxwcffMxIM0rcY0DLoDEkkVeEKnDwB/CkyTVlwsc1dIfOoJUPSffu3dqVSa+emTXU 8BmaUHIcOfJOOyxIt2evnk4hs29/+KUHUv96WiHbp65W0O1l287Bg4fM0cCCFXWTOoqwHW2vUlfS EGhY+kE7iGsnwhd8hiZJQ/Iu6Ul/xuxzGu1V+tCobwHwkfYqvIU8Ui/gwVeChEwvLm76BPqOKBaE pZx4RrmEJuoG9SBaR1FMUUdpz6H7UNordYNyIV2fhC/pV/xn5T6nvfJDoR8lsJB+Jfos7hrhBpo4 cWJsGce9o/cUgVAI0J9gTUS/M3XqVGdBi1KNe2kpABizYUWEcg6roBkzZjhrJ+6xzInnaRBKRqy+ sDDDkhhrPDAAi7SUEPRnpEdfj3UXMglje5a+UVZKikCxCKgColjEKhSeGThmOSBZ65mPFToGOgUG rFHBn/e4d+jQoeAdqAiy0UEaPMg9CcO9UEReIUnTT4d7dOISxn9WznPKBEUHZRLHh5RJ6MEzA1Yo 42yyS5ssSl0JjQWJvnEgI6xImj4j3Wz9hN489KZ/u+znDCAED0nTTyRTN46aNwP7PqBe0F5JL4qH tGHaSegBj1g3RHkAE6mzEsbHqZzn5FHqn2/9IGnAG4KVhJH75T5WSx9a7nyVGp+Uu9QDPx4pp9AK CMqd9pqrD4U32lFoqzraK9+MuPZK/eSHgiyt9ir4+2Ui5STl5j+LO8enBsRSDiVFoNIIsLSI/gSL L1F+o4DAETFWEaHbOPnnmwsfTU1NzsqLe1h7NTc3u2VPofs70qO/w+cO1oDkHUIJDA+kLzvXuAeB /tCPgTl5xzpK+t+xY8e6cuBZ6L4uUNY02goioAqICoJfTNKYQ8s6U9ZiFyIGYQzWIDoLn7jmR4fB IDskSfxRHkhT7okAmAYfcYKV3AOzkET8olyQvPvpSZmE5uNEmVAGPgcZxRR34DP0B+XIOxlz7Tgs TOtOHMJrWy7Ld4UQkaudkIrUDQSOkEQ+c/EBPlI3pP6E4kXqnuTbT0fKKXSZgIOkEceH3BNefR7L eU78ScpEeC1n2tUYl+At+Ps8yr3Q7QQehA+pj3F8hBZO8pW5tFfqTq20V6w5WNoC/dVf/ZUPqZ4r AqkjQNuhPqJEi1r3sNsN/QxLEkITygd48JcPkybLnRgfSZsJyQf5BA8UED7BA4pHeAw9VkPRgUKV SVC/38UagvKh/wjd1/l51/P6QEAVEDVSjnSCc+fOddwmMbsq1CFJJyKDuVAw5BrA++lJGP9euc/z pSFY5AtTDn4oEzGXkzT9eOVeaD6EB6t+sMm31UAIDz6vPo/lPJd8Spp+3MLVUbs9aEhqm09J9USK VvR3F0fsACAk5cPCTzf0R17qRlyZCB+EkXByr5xHykTwiItXeMsXJu69Yu/BRz4SPkL3ofl4SPNZ PrzTwoJ6V6hcwCQfr+XATOKXfPtx+vdCt1fhw08/ep4EMxxWgqs4W43GodeKQJoIoEBkYipu2Zks AQttfUC/jtBNexZrIsEAwZ/7aShBZDmi77sMPuAJxS/KmHwKUeG5I0fySf8Q5YE4KQ/6uTQmEjuS B323+hBQBUT1lUlOjt797ne7da9oG9mpIB8lGaTle7/cz+L4kYFa3LNyp58kvtB8EL+kIcc4vvI9 iwtf7L0k8TvZK7/8VWyypYV3jJT2atK3EiWRKFDSFNuHK1Qm0lbav1mZO4X4TYOr0DyEjr/DGFVD +4xkIg3MkqSRJEyE9aIu88XPM9prvjBFJZYncLnSEKtKzN1957N5ktZHikAwBFBAULfFsspPSJQB oa2cUEDwoy1H+eAePwT/JEpAn/9iz0W5EF1mRfoQ6UuYYuNOGl4s26I48L6Uh4RJGqeGUwRUAVFD dYDBAWvgWPvMXuz5KK6j8MPLwKVQOP+dUs6lk4x7V3iQDiwuTLnuJclnkjAd4Yf4k+CRBh+ZfCDB RKSY1ku+bV27tbcI6Ej+o+92bV1mIfUg+pzrLl3bOniLC9PRe63f8dhorMrI3e9iyy4kSZnnwkLu S7hQvEj8kl5cOtRhCRf3vKP3ksYfkgfyUCh+wahQuI7iket9qZu5npf7fr58Chah+3LqRr4+VPKc j1cJ05GjxC/5jsbF/aT1OPpuMddJ8C6EF7OX4ldqzpw5ifAthkcNqwgUi4BYDkk789+X+hxa6Eaw F+WCpBnHR0gLOPoRwSLKg1z7YXz+ynmeiwfSED5Cl0c586NxVQcCYUfV1ZHHuuEC86fzzz/f5eeF F17Ido5xGWRgIp13rkES74k5W1wc5bgnJltxPMg9cTBUjvRyxSFpxJmOyz0JkyuOjt6nTESLLXmP i1Mwi3tWjns+D/b71oaEr27dMDEM2z1079HdpS1p+owct6bWEDt1hKSuVqCRQfxxu+VmlI4fywDE bhghiTKnvYJFLB6tBZVm3YjmV/gCLxl0RMOU45q4pS1K2/TjlXsSxn9WzvNq6UNz5UnKI9fzct+X uif4+/ELL6G/J/Rdfv/l88C58Ca8Rp+X61riJ9+Sd4lb7lGPJZw8K/dR4pd8+/ELX4XaK7tqsZ0e 67vf//73+1HouSLQIQRoA/KtkGOSCOPqc/S9JGGi7xRzTfvJl4bkJ1+YYtKLC+vHLekVChf3vKP3 fD6icQlf0t9Enxe6lvcLhdPn9YdAWAmj/vCqeI7e9773OR4wmZS1YXFMMThn4BHXiXIPzS6DF8KE JHEgJJpkPy25h2fd0CQD47iOFD7oBCVMKF4oEzAX/KPpSJnIoDL6vFzXks9jTrhuq4EQnwu97DaY oUm2q4stk1YFhPAaihcsG3r0yOQ1lo/WunFSzPat5eQJLPK1V+oMWIhSsZxp+3FJmUjb9J/JvejW mH6YcpyTRyl3mXnx402zveYrE2mvoftQP+/+ua0SqZKUidQDP3Epp9B9OcoH+MjVh9KG0+jLpZ3A Bz+f4CHt9ir4+3xIORVqr8uWLXNjCbY6xMO9kiJQLgR84bKYb5f/Xi5ekoTJ9W6S+8SfLw1p98Xk K0m6fhg/fUnPfy7nfji5V85jvviFr3xhorxIWI5yHg2j1/WPgCogaqyMWYbBQGG93fYGZ5S5iEEx nnsZDEUHJ5hK0WnIlj654ijH/cGDB7sORtL04xTTNbwah6YBAwa4JKLrBsGBtWsMbEVZEooXv0yi 5mqy5pEykRm+UHwwIO3evYdVQh2x9aPt4FnKpG+/zJavoXgg3gEDMltKRcuEZ/DBhx1PzyGJj9+g 1q2tJO+SHnXjLdtWEDa6W8VRSEI5xQwkPETbq9SNqBfsEPzAA/UvWj9JS+5JWwqRPnFSJpJGdF0p ZQJG4n07FA/E67fXaJlIf5ZGH5o7j23bbu5w5XkiZRJtJ8ROOaXVXuW7JvVRckfd4B7PQ1vH+O2V dH0CH+4JXv6zcp/TXsE92k5IByz8tpQr7d/97nfu0SWXXKLCQC6Q9H5JCNAW2Mpy3759sXU0V6Qy BoqbFJD2JmFyxdHR+7QrfpCkGRdnSAU06YuAHuXBvw7JA3mW+P00BQu5J2Hkfr4jDiupE9SNuDLO 964+qx8EVAFRg2U5f/581yn95je/ycv9ySef7Bo3Dd0nrunUeB6amM1HcMJjsS9o0uns3bvXzWaF FjLJI0oOhJZoh4c/DQZq8Bh60AofpEPeox6cxZsyz+WDQ/gQRPzDRwx3ghz59+mN/W+4j82ggeGV Qied1Nv0tdYvYOELeJTHoUOHnfIh9IwqeR84CKVPD3PgwIE2A43M9THnlK2Yj6uPZzHnw4cPd5ZJ 0fa6c+dOVyfw/xKaaAPUQQYIvlDDIINy4nkaCkPSIC3aqwxwyDvX1A95HhoP6UPpq3ySPhSs0ibB I+2BG4I9v2h7pZ7gLR5lTKHZ9nJghZNE+rBoe6WMwGbYsGHZAXM50ssVx6hRo2Lba5p1Q9ojWPjf V/pTyon+M197xYoSa0owk122cuVX7ysCxSKAkoA+A2VcMZadYm0V18fJPep+SCJ++KdPkT5X0uMa PlAQhFaESD4l3z4P8AEPxWAr7xdzFIuvKA/EIfekzJLEy3eCOsHYP/R4Nwk/GqYyCKgCojK4dyjV c8891wkJKCAYeOQiGjgDi02bNjltI+EYwG/evNmMGDEilRka0hw/frzrJLHaYGBEh8UuHgg5U6ZM SaUDohMfN26cw4v8wwOCzLp169wHpLm5GVaDE2UycuRIVyYi1DBQBA/KCouRNGiYFWZ79+ptNm7Y 6HDgQ4YSZP8b+01TU1PWBD4kL3w4wZ2yyJQJTp+OmK1bt9p68o4ZM3pManXj5JNH2L2sd7v9rPEF 8aatmy0tLc4qIS0hE8GKH+0VgQ6ivW7fvt0g7KShqGMwMHbsWGf1wdpwzLipGxs3bnQ8TZo0KRXh jvZK3UCYk33OEbBorwhV1NE0SPpQ2qcohvw+NK32+ordJvG++75pvv71r5vnnlvssv7znz/mrr/5 zftdHQmNBwNtyoRyAA/qBv057YQ6MnHixFTaq5Q//QQ7QkG0F/wYIGyn1V5Jh/pBnRRlHX36jh07 HE5ptVdwp2zWrFmTba/Sh/Dtzac8/fnPf+54v+CCC4ILUqHrp8ZfPwjQ/yNUM0aLkrS10O2L9LGK pW+T5UzCi3wXUbqGFqAln76CET6EB/gsRviXPBRzpJ+D/IkieR++GMsVY0EMpkL+udzTY+dAIOMF rnPktW5yyQDsvPPOMwsXLjRPPfWUufDCC2PzRsc4efJkJ9AwOKPzYKDC4D0N6wdhio/JrFmz3EB1 yZIl7jamo6ecckoqQpXwgXDH8hUGZwg14EPHyr3QWmzhgTQZFNJZo5BByKPzZu0tQmZaxKB0ytQp Tth/7bXX3EeWj9jECRNNv/5hlz34eextNeHTpk4zW7dtMcuWLXdlwozJtGnTgn9UfT6wPkDLj/BA 3aBMhjYONcOGDws+wBA+qBvkGx5WrlzpBhiUyfTp01NZLiV80F5PPfVUJ2TSXhkgUCannXZaUYMM ia/UI2XCTAnKDxRU4IOSDgVePqGq1PTi3iNN+lAEXQRNTIrBh6VwaQm68IWVzsyZp7h8n3322aaL vccQDuUdA1GZoSJsSKK/pD5SJi+++KIrEwbIs2fPTrW90l9SN7Zs2eL6UcoJCyG+bZynQfQRfMPA Aj8KlAV1AyVdmnVDvq/UT2mvjBH45vKdzUUIcr/61a8cbn/913+dK5jeVwRSR4DxGMI91n/UU79/ Y+KK/j+fZU+5GEbBjFKRNH3rLpkgYEwZmuhL6OeYqPLTQ/BnTJ/G95DvP/2M5FvyTJ+HFS3fAL+M 5LkeFYF8CKgCIh86VfyMGQsGD48//rjBIoIBRxwxGGOwxg9BIq3BWZQXOm8EfaiSfPDR4ldJHigD FED8KskHH/lqqBu9evcyzc3jWrGghqQjQJCSTwhX/CpZJtQNBG9+leSDuoGijF8l+WDgM2PGjIry QJkg3PKrFBZjxjTZtpqO1YffJuLO/TLhOfikTdVQJuQZJQQWQ/wqVTfgg8E/ykt4gJKUyTPPPOMU 4J/4xCec9Yp7Uf8oAlWAAPWXNoXwj/CN9Sr3UASglGDslE+5Vq4soPAmfSy+GBug+EDo5pp+kOeh CeGebw8TI6NHj3aTZbRz+IIfxnChCeUDig6UrXv27Mla6sID5dLc3ByaBY2/DhFQBUSNFiozPR/9 6EfNN7/5TfP888+befPmFcxJkkFJwUjKEKAa+KgGHoBS+ThRoRSLE1ho3VAs2iJQfVfaXtuWSTXg kZQHZk5/9KMfOWHmU5/6VNV8h9oiqle1jgCCsijFis0Lk1ZYW2HZ87JdgsY1M/AoH9KydELAhwcs VZcuXep4QAmC8gMFCQrI0ESbZpkVPOB4nslGrEJIe+bMmcH9P0j+UH7AC+WBEgjLO5bIoPhEGaOk CBSLgCogikWsisJ/5jOfMf/7v/9rvv/97xvMckOvA6uirCsrioAioAgoAoqAIlACAiwZ4Xf11Vc7 a6sSotBXFIHgCGB1gAUcQj8CL7P9/lKI4AzYBEgPHlB+oLjDGoAltEmVfeXgEUUISwHhgWWAWCii iEhDASL8k1+WCbMMRBQg8AAvSopAKQiEV9+VwpW+kwgBOsIrr7zSLF++3Pz+979P9I4GUgQUAUVA EVAEFIHOi8A999zjfFV8/OMf77wgaM5rAgGcLDLDztLZtJUPAhCCPksh4AGhO03lg/BAmlhewAO8 pKl88HmgDOAB5ZAqHwQZPZaCgCogSkGtit7BGSUOpB566CHnpKaKWFNWFAFFQBFQBBQBRaCKEMBv FObk1113XU7fUVXErrKiCCgCioAiUIcIqAKixgsV06xrrrnGOev57W9/W+O5UfYVAUVAEVAEFAFF IAQCeNL/1re+ZebPn2/OOOOMEElonIqAIqAIKAKKQEEEVAFREKLqD8C6rM997nNuVoO1ckqKgCKg CCgCVYjAsXeqh6kjb1UPL8pJcATw3v/www+77Q2vvfZa50E/eKKagCKgCCgCioAiEIOAKiBiQKnF W+ecc47bqucf//EfS/Y6XIv5Vp4VAUVAEagJBKxH+GNvvl41rB4/+rYxR1RhXTUFEpgRnE4+8cQT ZsGCBW7bzsDJafSKgPOVUAl/CQp97SCg9aN2yqrcnKr70nIjWqH4cEhz2WWXOccweMnFQaWSIqAI KAI1hwBbtx05ZMzxYzXHek6GydM7B22ejucMUokHxw7vNV26WyVEtx6VSL78aXbpavPT2+5v3K38 cdd4jIwRbrjhBtPc3FzjOVH2axGBSjhNrEWcOgPPonTgKOedId+ax7YIqAKiLR41fcU2nOrVuqaL UJlXBDo3AnaJwrFDu6pOUK/nQjl+5E1rCWF/dULHu7xhujY0GtNVhzd+kbKVoJIikCYCbJ2J35G3 3nrL/dJMW9OqXgSoD/v373c7eRyvMqV89aJWf5zpF7r+ylRzpAgoAopATSJw7M29qnyoyZKrIqat 5cxxu9SlC0oIJUVAEagYAmzTyLaVbKXJT0kRAAEstNlSFFIFhIOhU/5RBUSnLHbNtCKgCCgCVYjA 0SNVyJSyVGsI4N+iS60xrfwqAnWGAMIlSy90+UWdFWwHs8OyC6kTugSjg2DW8OvqhLKGC09ZVwQU AUVAEVAEFAFFQBFQBBQBRaAWEPCtHvzzWuBdeSwfAmoBUT4sNSZFQBFQBBQBRUARUAQUAUVAESgD AkeOHHH+I/AncdJJJ1VkKQdp47cAB+/4WuOX9sw9gvqbb75p3n77bbeEAR7EiqAMMCeOgvThg+U1 vXv3rggPiZnVgFWNgCogqrp4lDlFQBFQBBQBRUARUAQUAUWgthBAaO7IDPe+ffvMqlWrnMCN8oHr IUOGmKamJtOtWzo73aB4WLlypVM+9O/f3xw6dMgpQSZMmOAUEWmUCEqY1atXu/wPHjzYKUOOHTvm drSBpzSIcty4caPZtm2b220PnlBGUBaUiZIiUCwCqoAoFjENrwgoAoqAIqAIKAKKgCKgCCgCQRBA 0H/11VfNiBEjnJDLbD8z7y+99JJTaowfPz5Iun6kCPlLly51CofTTjvNKT2whnjllVfM8uXLzaxZ s4IrQuBh7dq15vXXXzezZ892yhiUASgk4EHu+XyHON+0aZPhN23aNDNo0CCXxJYtW8yKFSscDv36 9QuRrMZZxwioD4g6LlzNmiKgCCgCioAioAgoAoqAIlBLCGzYsMFZHYwePTpr5o/J/7Bhw8zWrVvd No6h88NsP4qQMWPGZBUNWF6MHTvWHDhwwOzaZbeMDkwoHnbs2GFGjRrllA8kx/IPrlGGYJUQmkgH zFE8iPKBNFEOoRhCQYKiREkRKAYBVUAUg5aGVQQUAUVAEVAEEiJw0A5ed+zabTiWQh19v5Q09R1F QBFQBCqJAL4WWG7Bdo34GvCpoaHBCbt799otmwMSVgZ79uxxigeWf/gk17t37/ZvBzlHAQEvsm2l JCLYwGNo4R8eWIoi+RYeUD6wzSrKGJZjKCkCxSCgCohi0NKwioAioAgoAlWFwE4r4D/9zHPtfstW rCpZ8C9XBtdv2Gyuu2GBeeJXi0qK8vkXX3HvcxR6/sWXbXxPy2WqRzAF7yRUTNgk8WkYRUAR6BwI IOwi0Pbo0aNdhuXewYMH2z0r5w1xtoiQzc8nLBD4oQRBORCSEO4hybekBU/wIHzK/RBHUfZElUGk xT0UIIcPHw6RtMZZxwi0VS3WcUY1a4qAIqAIKAL1h8CyFavNvQ/+MGfGLvnoh8wlF3wo5/Nae/CT x540LVaxMWPqJDO2aVRw9rHCWPTsYoMSBKXCNVd+ysx7b7zTMcI+8dQis9yWSaGwwRnXBBQBRaAm EUABAUUFf/9e6Bl3nCxiiYGQH+VD7rE0gXBR5YBjvgx/UG5IPqNON4UnhH/4DEn43oDId5SELymz 6HO9VgRyIaAKiFzI6H1FQBFQBBSBmkHg8k9eZM6cPdPxe+jQYbv0YY9Z+PCj5qc/f9L0sSa05507 r2byko/Ra6/8tFm+clUqyoc7737AKh5eduwMbRycjy2z4Pa7ndKBQIXC5o1IHyoCikCnRgDBHooT eOVeaKEbwZ6fCNi5CiS0AgIecmEhPAlecl3uYxIewKEUkvIs5V19p7YRUAVEbZefcq8IKAKKgCJg EWhoOMkMazwxM9/cNNqMs78bb7nd/NRaDdSLAqLZWj3wS4PwX3H+OfOcYmenVejkszRZv2GTszSZ PnWiXaaRP2wavGsaioAiUFkEEC5FwJQZ+yQchV7WkJSHfHxIvpLEVWqYfOn7cSYN579TzHmS+JOE iaaZBobRNPW6ehBQBUT1lIVyoggoAoqAIlBGBJiJb7Q/liz4JEsFFj2z2DmJRFkxx1pPRJdqRMP1 sUoOCTfUU3YQN74R7n3wIYMgDhFuxpRJ7jzJH5YsPPLYL5wVgaS5r5cBAABAAElEQVQTTYN4sOrA t8QtN33eRXvQWnt8zVoqzDndWn9YC1n8TaA4IO9nWR4u++R8F6cfN/m95spLbZgTCps4Hu9ccFP2 Nn428tF37rk9+7hQ2GxAPVEEFIG6RYCZc5YQ8JNZ9CSZTaKsCC28wkO+NETgzhcmSV7zhfFxkPTi wvvh4p539F6SPBbDg9SLQhh3lG99v7oRUAVEdZePcqcIKAKKgCLQAQRYjhEllgsgxM9777vM3CFn mfUbN7ulGigRrrFLHIT8cBdP+aDzbYBwjd+JO6xwjqIA4r1bbZzM/M+ZPcs0jxlll0msNt998VGJ Ku+R90mL+LA4wJrj6Wefs/EtbvcefKOsECJ/XKP44L2zTp/lHi1e8rJ53Coj1tnw+GSALxQs5JVl FfB7z523SjR6VAQUAUWgrAggNLNUQn5JIxefCnFKC7knYZLGWWw4ll7I8ot8wn9IPhD8c/EgPPlh is1j0vCSR8Hef0/uSRj/Wa5zloxQJ+A99PKRXDzo/cojoAqIypeBcqAIKAKKgCIQAAGWXqAUQNEg 9LjdQQIhHp8R/rKMex/4odtJY9573mWmWwePi62QTrjzz51nrQgucq8TD5YFxLvIKiLkfUknGud3 H3rUOmV8WpLOecRyArrlpuuzyytwnnnn3d92zh9zvug9QPngK0XgjeUnKB+ifN15l/XtYBUUWIak 4cjSY1NPFQFFoJMggPDMNo0Ip8UIqL1793bCaZxfAblHmJAEv2x1iXNFEbIlPYR/7vFcFATyrNxH 8nnIOveVfEv8pA8fWBHAR0iS7TejOJCm8NWrV6/ELMAv9QIqxnIicQIasCYQUAVETRSTMqkIKAKK gCKQDwEsBrA6EELxgGUACgN/aQUKAe6J8kDCz3vvWc7qAKUDCoizrMXAT75zlzzOHvFxYB4zdovP E5YV7BARFyfLMAopILB+gE92tYj6dphjrRn8LTizTMSc8L5YZPAYflh+ctDmJ5pX0kEBsc5aTagC IgZMvaUIKAIVQwDhH4GWpRsI2f4SANkVon///kH5Y3vJvn37uu0lo7P0CN3wNXhwfse85WCwX79+ Zs+ePYadKPw8wwMKAXAqRvgvhachQ4aYjRs3xu62QXlQXg0NDYmjBjsh/1zu6bFzIKAKiM5RzppL RUARUATqGgFm83ft2uvyiO8GFAT4YPjC9VdlBXPuoZhAUI86VBSFAksUfEI5wDaUvAdJOAnDfe5N L8Lfg7zLkd06oOlTrGKjzNS3iEFhmZPW6BQBRUARKAkBrApGjBhhWlpazL59+8zAgQNdPAire/fu NczIDxo0qKS4i3lp2LBhZvv27U4B4AvYu3fvdkoReAxNjY2NZuvWrY4H+BE6ePCgW77Q1NTURkEj z8t5JO8oP/bv3+8sHlDOQIcPH3bWGSNHjizKwqWcvGlctYuAKiBqt+yUc0VAEVAEFIFWBFgmIUst UAhcd8Ot1t/Bpjb4+P4gWJoQJXbRYMtOIdla0lknWL8OLHPYaQef4mhSwnHs0+fEe/59PVcEFAFF oDMigMKg1Bnuk08+2SDob9iwwbAMgVl+lAEoJCZOnOjuhcZ0wIABThGyadMmgyUC1wjhW7Zscfe5 F5pQtowePdopY3bt2mWwRsAaAlxIHwVFaEIhNGbMGLN69WpnCTF27FhnDbFu3TpXDjxTUgSKRUAV EMUipuEVAUWgrAggzD3x1CLnxR9HeUqKQEcRwMLhPOvMEd8MT1ifD/hTgFAgQFgr3GAtI/JRxtnk qnb+E7CIeNruniEkcR48eGJJhjxLcpRlE75yJMl7GkYRUAQUgXpFgFn2mTNnGoTcZcuWOUUGvgNm zZrVZilCyPyz9ANlBxYAa9euzc7+jxs3zgwfPjxk0tm44WHUqFHOzwOKEPAAGxQPKCb85SnZlwKc sNxk6tSpTgHxwgsvuHRRyMyePTu4H4wA2dEoqwABVUBUQSEoC4pAaARwqPe9h3/WLhnWiTc3jXTC WqEt+dq9XKYbzFYj7A0dMth56s8V7Z12q8HlVvj7P9ZzvwhtucImvc/6e9k6ET6IN9c2i0nj1HDV gQB+D/ALgQ8Gzilbfvg8WL5ylVs2ka8eiYXEnDPyK8WIg3aUJM44ZGh3xBG11ogLq/cUAUVAEegs CDDzjgIAwpIiLWHbx1cUACgBKsUD/AwdOtT9KskDyzBmzJiRtWqpRHn4ZaPntY1A19pmX7lXBBSB JAgwu7rDCtuNjYOsg72J7ocghtDNVn2ZLQR3J4mqYmHENL5cM8WydSIWGOx8cM2Vn7JWGLPc7DZ4 gI1S7SKAUP8xu+0k5bjw4f/JZuR9tqy5d++DP7B+HU7UeSwbrrXLNlByQeIQ8olf/S77LmGiviN4 SP0hzkceeyIbFj8UCx9+NHud6wQ+qXds7YnFhhC8+ddyv9aPgiFHJUVAEVAEkiBQDcKu8pApKXCo BiyS1BsNU70IqAVE9ZaNcqYIlB0BBCVZJy+R//TnTzpBB8GKbQCrla656lOWNX7lIdk68YbPX5W1 vMi1zWJ5UtRY0kZgrt02EysIlkzI9ppYQ6AsoPyxcmhuGu2uUUShdBhrryHefdxaT6Cg4/0+1gwX JR7+IHA8uXP3nmx2WOLBrhISFqsGFAhJrYouu/Qis8xaZdAWWTJyIq2JWeeX2cRq/ATssHgK4XSz xqFR9hUBRUARUAQUgU6BgCogOkUxayYVgdwIIDwhpDEDGyVmcdkGEKEB4YxfnFDlh8MZ35zTZ8aG I35mPltadxogXFJCQERwRAAUIq5h1vwdnjhHoCyUvrxrjndxioeo3wmUEAin0d0Qsu/pSVUhQJ1k m81x9hhHOKekDlN3hKjzWAL9yd7fYRUJLKG45IIPZhVRhMMy4VarkMMiYvGSV9z1Zadf6MKgKOC5 T3csuMkssoK1hGVbz7lW4YdCYQZbd+Yh4uL9P9mlUvL+595zqavbKD/8vM19z1lthHd8UJB/cIhS NKw8z/CTGzMJ5x8L4Zw0rFgwgbmSIqAIKAKKgCKgCHQ+BFQB0fnKXHOsCLRDAO//KBl8Qimx8KFH 2whuCA03WosBmSUmvB+OeJgl/q59DyEPwcgnBHuENyHCRRUA8ix6XGh9WKBk+Ml37nKP4JddChCm EC7X220YhUjj1ps+34ZPeSbHjEWFXJ04wj8kSz5OPNGzakQACwZ+uSjXcxRZvjIr7n3qO5YQ/Hyi bsdRMWGj76OESPJ+1IKJ93LxEw0raU63eedXDOXCMS6OfGGft8occC2EfVy8ek8RUAQUAUVAEVAE ah8B9QFR+2WoOVAEOoQApuII9v6MJPfufeCH7t491ukjQv8dC2506fj+ESQcs7CE+z933uKOzJai BJD19LyI2TX3eEZcxHmLVRJg2dARwnJjht3V4Dv33O7ivNzOeKOQ+G6C9ffRdGXdPkIdpvpKioAi UD4E6A/oaz6WQ4FTvpQ0JkVAEag0Ar6vAPUZUOnSqM70tV5UZ7mkwZUqINJAWdNQBKoEgUOHDzkL BWb5+SEMoFCAfKFAhPcbPn91VjHBrOb5dmtDhPvnrak4xDlWDlhFiAKD48Wtlg/+sg7W00PESVwQ s6CYyHeEXBx2DT1KAwjFATz4FhH54icPWFLceMsd5robFrg84QtD8pPvXX2mCCgCyRHA6gJFIUtT lBQBRaC+EfCFy65dVdyo79JOnjupFxzlPPnbGrJeENAlGPVSkpoPRSABAt996Gd2eUTb7TgR3LEa 8IUC1sxjqXDc/pMlCUTPzhlQS+tyB9/UmnCyjGOX56CP8Aj5vIOyICrYi+KAcKVQND7iYCkIypWk RBwNlkcsIMjDomf/n83//KSvazhFQBFIiEBH23vCZDSYIqAIVBiBd955x+zfv9+89dZb5s0336ww N5p8tSBw+PBhs2/fPoNS6tixY9XClvKRMgKqgEgZcE1OEagkAqwJF+/zKAl+0rokwl9uIEoELAiw CIijg/YDIoSg/8hjv8gr8Fez4zkEIrbgFMIvxePWceDQIUN0GYaAokdFQBFQBBQBRaAIBLp37276 9+9v3n77bdOjR48i3tSg9YzASSedZAYMGOCyqBYQ9VzS+fOmCoj8+OhTRaCuEED54DumQ3nAMgnW Zked0uFXYa715B9HWBhA+IBg+QLWEvhzwCICgZ54uV+LhEO/J+xyEbZV9BUztZgX5VkRUAQUAUVA EVAEFIFqQeD48eNZVvzz7E096RQIqAKiUxSzZlIRiEcA/w3LVtzttp28pdUrPssRUCIctP4ifGVF XAyP/2qRu33tlZ/OLs+ICydxdtThZFzcpd5jh4+GPr2tD4vz2kRB3l3+7ZIMJUVAEVAEFAFFQBFQ BBQBRUARKB8C6hWmfFhqTIpAzSGA1cM86xAOKwhxLEkmplvrB5Zg+LtYSObYdjNK+E7wafnK1f6l O8dKIi7Olo0nts9s91LAG+s3bjJPWAUKVhw+oSTBZ4U4yvSf6bki0JkQQEl374M/7ExZ1rwqAopA FSGAr4C1a9ea1atXm61bt5pKzJjjv2L9+vVm1apVZuPGjQbfFmnT0aNHXdrwsGHDBnMoMuZKgx/8 NWzfvt3hAB6UjZIiUCoCqoAoFTl9TxGoEwRYcsCMPztfIHhDl9tdJbBaYIeMhfY+SypQULCsgu05 RVlx1uyZLvw9Vkhh2QJbbRKG7TYh8f3Auex2IWGJ46ePPWmdYj7K49TpvNYdPcgjShXySB7uvPsB hwfWIUqdE4HFtm5+7Irrs/W4c6JgzGK7DIk2raQIKAKKQLEIoCwoVWHAe+vWrTOvvfaaaWhoMEOs TyYE3pdeesm88cYbxbJScngE7j//+c/OYeKwYcMMioAlS5aY3bt3l5y3YpnBkecLL7zgfGk0NjYa fGssW7bMbNq0qdioSg6PI1Gw37lzpysL/DisWbPGKSPUkWTJsHbqF1UB0amLXzOvCBinaEAYx/nk E9b5IuSUD3YrSnatwCEjSgUEcxQUN9gtN+fMnuXCYUEhgjqKBGZLCXOrfRfCykAIiwLehQhLfKTn O4CUsGkcycO1rc4nUaqQR/jCv4Vuw5lGCVRvGn1bl+FEOaSOoJCLEoornu2M7P4SDRfqWnyuiGIw VDoaryKgCCgCaSCAgI+APXbsWDNixAgzaNAgM3XqVCf0o5hIQ+hF6MbiYOjQoaapqck5TmxubjaD Bw82K1ascLt7hMYCawusP3r27GnGjx/vcBg5cqQZPXq0s8pAORGaUAbBA8qXSZMmufwPHz7cTJw4 0Wzbts39QvOg8dcfAuoDov7KVHOkCLRDAF8O+fw5YAXBzyeUEDiWRKHAMoUGK5SJ80k/nLzL0gU/ zE++c5cfzJ0j9POLhs3Hm0QCLz7BX1wahImG9d/zz+daXPixhSjWGj7/fjg971wIoFj7zj23t8s0 gn4crW/Z7Cxo4p6lcQ/lIbzNfU+809g0eNA0FAFFQBEoBwIIvJs3b3ZWBwj/PnGN+T8KiugzP1w5 zuGBXRqwfPAJhQjLQbCOQEESknbt2mUOHjxopkyZ4niRtLAIAYeWlhYzc2bGElWelfuIIub11193 iqBevXplo2eHEywhUBShkOjWrVv2mZ4oAoUQUAVEIYT0uSLQyRFwThmt9UIhKsZnQjFhC6Vbjudx ipVyxKtxhEUAxdgOK3xTn6inQnIfCx6f5L6ER2jH2qHRKrMWPbvYDvQOO0UcPk3wVzLM3h9qLWLE Lwhx8UzeQ2FF+jvtYBha37LJ7Ni521kOuRutf3hH4scXilgQSRifL2OO5w0r73AUPltaLY1EESF8 S1g//aFDB5sZdjcc8lUMkefl1ldMnz4nmTmnz2zzPs+gKN7CHxiPTdCHFMOPhlUEFIH6RIBZf5ZZ IOxGt2kUARiBOKQC4siRIwbrgq5du7bbQpQtReELJUhoBYRYOEi+pcRZhgFvKCfAK+Q2p+QTi5O4 NHr37m327t1r8JPRp08fYU+PikBBBFQBURAiDaAIKAKKgCJQjQjgPJVlPyzj8a1o8Gfy/Iuv2KVA n2+zvey9Dz7klAli3cCyCSxpIIR3CIselA88Y3kR1ywXkufyjPSGDhns/Ji4F+0fwkG+ZQ5+TsQn into/2SWOH0+K8Q//cxiF8/5585zPhfEFwvhZ0yd6JYEybv+UXiRe6RlHjOteGQUDCwPifpZQVlD 3pJuM3vn3d92eEo6xOcsn1r9pDxi06UswNVXBC2y+aIsrr3qU6qAEPD0qAgoAnkRQJjFCgIhO0py j1n5kIRQ//bbbztFA4K+Tygf+OEIEsE8+twP29FzcTYZFf6FB/gEi+jzjqbrv3/gwAF3GZdPsXo4 fPiwKiB80PS8IALtW3fBVzSAIqAIKAKKgCJQeQScJYFVQLDriq+AYKYeQgnBcgoIoZ6Z+jmnZ/yX uJv2D4oF4sFJqi88y3OO99x5q7vEMSWz/P4SHwRxfIjgyPSer93qlBIusP2D8I/y4XzrY+ViG474 8XuCUI7iBF8jPuH88gvWTwppwBeCP4I9u9FIPvzwhEPZgaPIOEUM91EWYHVxw+evdooP4r319rsc D9yPi9dPg3PeuWPBjc7SBGsQFC3kC0sK3gdT+Fxk0/OVGs8vecVFxa46SoqAIqAIJEEAwR+KE3jl XmgFBP4OsIIgPYT9KHEP5QMKgKh1QjRsqdcoYYgfivIg136YUtMp9J7wINj74eWelJn/TM8VgXwI tFXr5QupzxQBRUARUAQUgSpCAIEeCwHf+SJKBpQNzWNGu10chF0EZ+isM9qul8UaAeeoZ1klRHQJ gbxb6hEHrsR/md1VRpQbCOiy9W3UaeXHrJJCeOC9i1stDBDuSyEsIkhXlA/EkbG+yCg+nMVEgoh5 X5ZNcRTnrSh4IPyoQM/bXTOEWFYiCh/SVFIEFIHOhQBCsgjKxeQ8iYPJJGGKSTMalviTpJEkTDTu pNcoF4Ty4RiSB9JHGZOLhK98YXK9y315P18YfVafCKgCoj7LVXOlCCgCikCnQIDZdecotXUHikXW EgGB9/JLL3Qz91gPQCIsR2fjQ/n/QPjGcgAFCU5O/V/z2FGOp2WvZnhzF/ZPo13S4ZMoLfx7Sc9R AJA++Y0qALiGLxQb/nKPXHFH38fqAd7YphPiPBqfKE2iCp9caeh9RUARqC8ERIjnqIJm8WXrKyCK fzudN0rlkTrBu1ov0imnakxFl2BUY6koT4qAIqAIKAKJEEDw/an1e/D8Cy87838E37PskgCcHiIY c43A3LJxs1uKEBWmEyVSQiAR7PHvwC9twjkmxDKLOBpqvagbs9ocYu2uxalYwgGnT3Pf8642yzDE GuLMyJIX/x09VwQUgfpFALN8dpLAhB9HhUlJzPrjhFu5J2GSxllsOHwbJEkjSZhi05bwxC0CuuRb nvlH8cPg3yvnucRfLh727NnjdhEhb6GX0pQTB42rvAioAqK8eGpsioAiUAEE8LaPwIcg1aehoQIc aJKVQkBm41E0jLXCNrP+Z86e6eoCwvfylSzJmOeWA+B4MS1iW1cI3xTTra+EOJLlFnHPOnpPlArC RzS+g4cPRW916Bo/Gn1w/mmtIliSgcUJeRc+OhS5vqwIKAI1hwA7JIwaNco5cxwwYEBi/nv27OnC xi0tkHuh/C4Ikzi75Ed6kqY8QxDnhwAtvMqzch6JH+eSOOWM44G0UFKIgqCcaftxSR7jFBCy9ELC +O/lOh88eLA5+eST3eOQzjNzpa/3qwMBVUBURzkoF4qAItABBBB2ok748AuAMOo7xetAEvpqFSPA 7LssvcgsL8g4PeQ+9UJ8REy31hJpkVge9LH7pPsOMtNKX7bZXL9xU2ySWEigHGAnj2LJLXmx7/sK FOLCGSXOOAVvtutUUgQUgc6LAAKyP5OfBAmUC7wT59hQHCKG3vIRgRo+ZKcLn28EcRQCffv2zVoo +M/LeX6S/X6wJSlY+HlG8IcHcAqtjCGf27dvd045o3mjPFCUNBQ58QPfSp0bAa0Bnbv8NfeKQN0i 8BPrgI/dBlrslopK9Y0AigWEYpQQvlDsZuWtYLzQ1gNfMZEGGjhrJE0Ecvwx+MQ198tNvrUDCgHx y9DS6oBT0mO3DdqF20VEbuY5ih8NCQLOUNSyY957z3L3Be+k8buX9I8ioAgoAhYBhH+EbXahiM78 Yw2AwFuMRUUpoJIGM/UI+vDhkywbGOKWsflPyn8+cOBAFynbXPqE4A82/fv3D7oFJ2mCA3jEKYTg CyUJ1i5KikAxCKgCohi0NKwioAjUDALXXvlp64jwImeWXzNMK6MlISBKB5QQ/qw7QjiWCNyXMCUl 0PoS8bGbBttr+kK5+JX46c+fcFtikh50ud3ak/Nbb7/bKRxwTMm7XLONpYRrjb7kg1hboNRg601R eFxj2wA832G3zeQZ6bPzxX3WKgSeky5JuceGh28sG3ifrT15P2rZIcthyoV3yYDoi4qAIlBxBOJM 9pMwhbA7btw4F3THjh3ZVxC6d+3aZRDKQysgSHT48OHOwmDDhg1ZHsjTli1bnMDN89CEkqNfv37O l4aPJ34UUECAE3iFJCwsyCt+PLAIEaIsKBOW2ahFg6Cix6QI6BKMpEhpOEVAEagpBBDKRDCrKcaV 2aIRkNn+9TGz+iwLwD+Er5goOoHWFxDYsapBAEehcYv9QQjiGeE/43DynmmZXSKwAGCLT96594Ef tsZi3LvXXPmpsvlHwNrivHPmOSXBiaVIQ5yS4I4FN5k7rQIiLn1RnGQZizmhDYEhW4qypAniPX9r T/81lr2grJj7now1hP9MzxUBRaBzIeALzcXknJn9pqYm09LSYphlxyJi586dzi/DpEmTgvs9gFcs MaZNm2ZWrFhh1qxZ4xQBLIc4cOCAmTBhQvClD/CAHwryCw/Lly83jY2NDg8UM+PHjy966QNxFkso OCgLLCBWrlxpRowY4ZQfKGK4n4YiplieNXz1I9DFdg4nNpqtfn6VQ0VAEciBwLE3tuR4YlpnL18x 11x5qV0T/5CbxZVZSu7h0f6JpxaZJ6yQwX0n0Jw718yzwkSUmAFdZL36s60ggt8c6/APwUzWnBMe YYww550z18XLzCthLvloxgkgM7RRPgibxGSbGeiFD//MzeZK+qTNjDJCnczKYgaOQHrNVZ9y69zJ 19esIOYEUTth8MSvFrk8IEydZfNw2SfnuzgfeewX2bjBAXz8vJE/BCx5Hx6Y+b38kxe2C0dYn7r2 G+lf6nkEgXx1OBK0YpfUI+ovdYKy94k2AcVt7ckzlkjEvefH0ZFz2gZtOS79fHwnTTNf/BLHvQ/+ wCl87rnzVrlVkaO2tYrArokqAlkEEJqvuOIKN0s+f/58c/PNN2efJTlBPNm/f7/7sRQCJcSgQYOc UJ7k/XKFYdkHFgcI4CglsEoI7XchyjvLPuCBI44bsQLxfUJEw4e4ZimKWEFg8YAVCtYZxVpg3Hff febBBx90LN5///3mjDPOCMGuxlnlCKgFRJUXkLKnCJQDgfUtm51QfeMtdzhhmdnS9XZbQkyqMQeH EKZE4bDYerJnxnSYdVCHcC104y23O6Ge2V9mOJkRxeSbGeZbb/p8VgDfuXOPSw/FA/ESfpgV9CGE N/iASA+BCYUFs7TE4afnAnl/eHdBK7/n2zzIuzt3td/mEOUD6Qsh/HEtQhRbNULk9XGrjFhnwy9n ptzOWqMs8fHxhSlms1HUtMHA8o9J/h0LbsxiIOnqsb4QoD73sYqpOIoT/CVcvmcSpqNHFGa5KB/f ud6J3s8XP2Fpn2w5KkrA6Pt6rQgoAopAUgQQbBFy01hukY8n/BuMHFnZyQMUHrJzRD5eQz7DGmPo 0KEhk9C4OxECqoDoRIWtWVUEEKz9XSFQMiD8YwVwy03XZ2d0sWrIrFtfnFUIoGhAqPctGUAUx3OE xfSbOHxCmI+amhMWxcGtNqyYgCOwoNzAukLM2v145ByrCWZyMSuX5RVYVdz7wA9sPtorIeQ9/0ja vI9ABoEHaaN8YM2+j8+ddz3gthXEYR9bPEKifLjFKkuEBAOsIi6zfieUFIHOhABKveUrVztlJPmm j1BSBBQBRUARUAQUAUUgDgF1QhmHit5TBOoUgTlnZGb9JXuyLSHWBCKQ80ysEGTNN/cet8sOoOjs JmHF2/7O3Zk14i6g/cNyBz9eLBCIE+sBUT4QlnPWjmNJEY1D4kLxwPu8K8oHeSb8ynW+I+/7PJF2 Y6t1hq98IA7yBa2zAhYEDxlqu3JtrJ15RiERfb81sB4UgbpGgG1w8YtB20bh6Lftus64Zk4RUAQU AUVAEVAEikZALSCKhkxfUATqDwGsAgoRVgBRxYG8I47+1rdscv4W5H70KFtish4eiwmfmEWFduy0 6+vt0o8oyfPo1n/RcKVc982xh3UUFxQXzO5iqXHtDbc6BQV5R5EDNkqKQGdEACukudYSqo9tR75y rzNioXlWBBQBRUARUAQUgfwIqAIiPz76VBFQBBIgIIL6CQuB+Jfk+a5de+12WnvbBUpjnXy7RIu8 gbCFP4vFS16xPjRecWveiYL7anpeJJgavG4QqIW2Wzdga0YUgRpAAB8OxToorIFsKYtlREDrRxnB rLGoVAFRYwWm7CoClUKAmc2D3h7QPh+yVKOQ6bXMjl58wQfbLeXw44s7l3dxJllpmmt9VvCDWBbC rhjswmHsygwUEUqKgCKgCCgCikBnRsAXLtk1QUkRAAGpFxzlXJHpfAhoj9D5ylxzrAiUhAB+F3BC iaf7KOHIEirkJV98NyxfeWJ3imhcua5l+8L1GzNLNXKFC3kfZQNLRzgKsfTimis/7S5xxKekCNQK AiyFyqVUrGQe4IufkiKgCNQuAmwTOWnSJDNx4kQzZsyY2s2Icl5WBNjNgzoxefJkc9JJhZf/ljVx jaxqEFAFRNUUhTKiCFQ3ArK8gO0yfSUE/hCwgIg6sozLjTisZKs+dpMQQghie0t2o8hFWECIrwnS FIIX/1ruhzh2sZGyG8hCy6uPwZ/sdqaQbwGCkkIUM+6h/uk0COCvBIsYv45UW+Zps9fdsMDW5Z9V G2uOL9lut+qYU4YUAUUgEQIoIC6//HJz1VVXmbPPPjvROxqo/hE49dRTXZ347Gc/a5qbm+s/w5rD WAR0CUYsLHpTEVAEogigPBAHjDfecoezdkCIYaYSK4CLEy49wFrg1tvvcgoHdtZg7TgCG/4hRMkR TVuu2eJymbWeQLhDgYFVBO9iWbFTAgU8xmEA3/CA8sHnH4UKTjfHjRmd3cIzIGsadRUhwHassr3t vPcOqSLOlBVFQBFQBNJBYMeOHeZLX/qSefvtt83FF19svvjFL6aTsKZS1Qj84he/MA888IBbfnH/ /feb2bNnVzW/ylwYBFQBEQZXjVURqCoE5sye6QTkhoi5G4I7QvM4e4wS94cNbbsbBf4N2IYT4f/g 4cMuzstOv9DMmd12e8/M9pXx8SKo33PnrWaRtSRYZpcsIMDPs1twnml5LLSTBFYQdyy4yWBxgBNI ri+x/iTIB1YVfj7mvucsw44ZkmccZZInWQbi51fC+vc4j8PHx2CH3XaU/MyzOwCwjaj4qeDdj9m0 ltttRcfGYMtzpfpFgDrSPHZUu3ZRvznWnCkCioAi0BaB48dPbFftn7cNpVedFQGtE5215DP57mIr wIkeonNjoblXBGoagWNvbKlp/jsD8137jewM2Sw5j1qHS4au6BexXmIrWRSK11z5qaLfD/nCx664 3in2UFSWStrWSkVO31MEyoPAypUrzRVXXOEsIObPn29uvvnmkiI+duyY4de9e+XmTBGVjh49arp1 61ZRx4lHjhypKA+CA05FS3Uset9995kHH3zQ1QUsIM4444yS6oW+VNsIVK411zZuyr0ioAgoAopA FSDADiRsh3rZJy90/gxkOU/GOeilVpA9sQQC3x04UiXsvQ8+5JbO3GktasRK5XlrWfPEU4uck1Hn c8Ra5WA1I3Hg/6RPQ++s01HJPmniS2HO6TPNeefOc8svFlmLnMs/eVE2bsJm/JX8wvL7srP8wWkr 1jPnnTNPonJHHJ2iILjlps+3uc+yjrh48YHCfZZDxfHdJpKYC/yaLHp2cTbf8HTZJ+e3CxkNh2WV j4+84IfjHmVx3jlz21mE4PsFvJ1FlbWEAg8smpQUAUVAEUDgX7t2rdm3b58TdhH+8RkwYMCA1MBB 8bFp0ybDchLSRwAfPXq0GTZsWGo8kNCuXbvMhg0bnPIDHhobGx0fpSoBSmF+//79pqWlxSmUeL9/ //5m3LhxFVUMlZIPfac6EFAnlNVRDsqFIqAIKAKKQAkIrG/Z7ARncVqIMM+SIJyA3nr73U7Ql2hR PnAfRQLbuSIYH2fvVEsI8dyHEKpxeIpiw4+jj13CxFIflg35hBKEeGUJ0c6de9z1AW/bWpQPxIXy gSVHpIGy4LsPPWqVIT/wo3PvEl+U4uLF1wg+UcQHCXzDI35akjjBhB8UM81jMsuxWKr0uPVh4baV 9RgAHxQjkOBDOj4+PPPDXW59thAWxQjYLo/kiTIjnYziwS5bsTxIGRCXkiKgCNQuAgjK/Eoh/Ea8 9NJL5q233jI4LcRPwNixY80rr7zilAGlxFnsO/D+2muvufRmzJhhTj/9dMMRpciKFStKzluxfGze vNlgTcKuEfBwyimnmNdff93hg4IkDdq5c6dZtmyZU/5QFvABLVmyxLzzzjtpsKBp1BkCagFRZwWq 2VEEFAFFoDMigKCL9YEQgj2KAWbX8cngk1MAePecZYIVhFEg+FYHy1ec5QRsiQPLgIwVwnNt0lpm fX2gABBLCj8tOcfiAqsGfJj4fkiETxycwlcxhCIE3qJ8wyeKAZxh4rg1H6FwuPWmzJIHwoHhdXZp BvkU3LDwQFEQXa5BPlCACD7wIwoFH0cUQuxwg2KDfEJYSaAQisYpvLtA+kcRUAQ6JQJbt261SuJD ZurUqaZHjx4Og4EDB5rBgwebdevWOUG4V69eQbFByEfw9reLJE2sMFavXm3YTjK0NcZh62sLqwMs Lvr27evy27NnT9PU1GSWLl1qwGnUqPY+vMoJDJYoWIGwqwnbqXbp0sX9OMcyhPJgu1XuKykCSRFQ C4ikSGk4RUARUAQUgapFYM4ZbR2hIjxjYYAgHSURrOU+gjF0ccT8H2EZxYLEwTVxPr8ks+0q7yCc o1hgG9pchIIDiwasE3zlA+GFT5yylk5tZxnH2qUMKAB8hUyuuFFekEch8tdor8mTEIocKJpH4if8 crszDcT5T75zl1Wy3Oiu5Y/k+ZAdTAsJhiiOfBIFhX9PzxUBRaDzIIDAy5IDfD40NDS0yTgCP9YR PA9JWD/Isot+/fq1SYprnqOcCE2kAR4oXnxCGQA+8FiqlYkfX77zgwcPGn4oQHwlQ+/evV357N69 W60g8gGoz2IRUAuIWFj0piKgCCgCikAtIyCCNFuhFqKWjZkw+CPAF0KUfGGc3U4QyJntJw2WIUDR nWD8ONa18pDZHcZ/khHaEfixBiiWSB8BnmUPOJQkfpQc0+1RloMUGyfh+0YG/Tt37XXRsG1uHO3w lBU8R9nyJ6uk2WHfY6lLHPEO/PvKj7hwek8RUAQ6FwIsu2Dm/yS75M0XeEFBrB4OHDgQFJT/396Z xNZ1nXf8yJYoipM4SKQoiaQGVpPloXHtAHYGZ2fUixhF2iCpkQIZFk3SLoKkWbrpri6QVYsM6CJZ dJFFgKyMwFk0TRAEsetalmRatqnZmihRIkWK1GBJ5f/Y38Ph9X3v3cf37n2P9/4O8PTucO453/md J9nf/37nO0r2KKdb/Sv3Q1jk+Ou6HO/x8fHwVsOPlXdBxaJArAPZpPwPN2/e9IKMcbH7jfzWOCVy RG1QH4rG0FzIDh1TIJCUAAJEUlLUgwAEIACBVUUg6khXMz5OrFgKNnWDQSLLJ5cSL0qAUMSCIgD0 Jj8aRRDtx5xwLXeIK7Lz9ELtAoTaUgTFoI/SeNXnrDBBRNej0QVxfSe5phwOKtpWNlo6l2wPRQSJ IbYM4yFtgzvy4ZijOS3EpByPaB+cQwACxSEgAUIOb1yCRRMD5PCmWZTXQB8JDVE7dE0fRWIoOsFs SsMesVCJ7gCi/lUklMiONAUIYx3loP5t7KqjpJQUCCQlgACRlBT1IAABCEBgVREIk0BWMrxjaWcL lX9dWjqgt/KVipYIKNJAwoMiDRQd8TdBPom4Z83RNiEiWkd2Vus3+kx4/tmlrTT1UZGjL4FEIoDy a0aXm4TPJT02+//9316s+Ijl0ojmddBDEibCojavRCInwvscQwACxSQgp1olzuE1xzvtxIcSFvSR DdZnOBt2TXaYEx7eb8SxRBjZoGL9WbvhufGye43+tvbDPq0PuyYRhAKBWgiQA6IWWtSFAAQgAIFV QUDOsCIakixF2DGy3Y8pLg+DtoqMlgN7/2zJ0Z/0EQe6p/NKxWywvAdhXbMzbMMc/rBe3LHEBu1M EUYXqK9vfu0FX30lyzri+nnoo/FFd7GI1rWlJgeWIh+qFbWpZSzV2qzWDvchAIF8EUiys0OSOvVQ kfNfqQ9zvCvVqad/PRu2bf3FtRnWi7tf77VK7Ztd4rWSYs+v5FmeWd0EECBW9/xhPQQgAAEILBF4 +ZX/LnGQaKDdGVS0q0K1YskU9ZZegoAVHWurSO3yEBZL7qj6yrkQLkEI69mxohu0PahEizASQHba tpN//fyzVt2Zw/8/QQJNiQyWDNMqKghXu0loG83Q7v9d2lpTpbOzcjSHr5Tgj+eWlpqo/MeS2BH2 o2PlnjA+XR9FjygCw+ppjLZ9ZyjmaCmLiubJ6uo85KNzFT0XtvnhVf6EAARamYCcS3Mw7TuJvUnq JqmTpK9yddR+pT7M4Y6L0ijXZq3Xw/6tv7g2wnpx9+u9Vql9s6tSnXL9r+SZcm1xffURYAnG6psz LIYABCAAgQgB5T549f+O+HwN2plCb9efWHJylTSyWpFA8L1/+LoXA771vR/4qAk9b45xdGcG1dcy DAkKSQQO9a+lENotQksjJBoor4TZqVwNO0Y/jMJQXbUp8eE//vO//NaVWrqhPAzaTSLMuy67LAnl P734km9DdqtdiSKNygGhtr71tb/1AkTIx/qxBJyy55mntVXpq0vbbr60tKykw9ut58UsTOYZ2q42 NX7xVvSH6oZFnLVd6VtvT7rv/ePXw1scQwACLUpAzqnC9y2EP6mZlu8g7s27ObxWJ2mbtdaTsGDi gvUZ10Zayy/UV7j8I2pDeJ6mDbLD2g/71HUVu2Z1Prxa/U/9JsLxVX+CGnkjgACRtxllPBCAAAQK SOClf/m+m3j7PffqG0f8jhRy1hV1EJbPLjnH5ZYHyCF+6Qff97kd5PDKaX7izx/2iSajDrHalHN/ YO9k7O4XH+52oeSQA6Xu1Yba1zKPt96Z9ALJM0viyF8siSS2RMMqyxlXXUUWTE1/uFvE3z/9Ze+Y v/aRyGJ1JWwo54LVld0SMCS8xNltz8nR/1D4+Pge8nGclGNCjKr1882vv+DHo3lQ/385+llvi8Yt cUQfs0u2K4GmeKgY7+hSGPHRFqASlCgQgMDqICABQckJlUjRkikmsdwSKsYJEHbN6iRpbyV1tKOD RA45yuZkWzs6lx1yoNMWQjROMbRcEKENskOOf9zuFFavEd/ablPF2Idt2jWrE94rd6x8EZbYMjqu cs9wPX8EECDyN6eMCAIQgEAhCYTJGOMAyFGvVOS8S7SIChdxz8gZ1yeuVLpXzUZrT7b83Zf/yk5L 32GkhF0sV9fux31LBCiXoLIcp6T9xI3Rlq1EbUlSV7bGsYi2xTkEINA6BOQcd3V1+e0Za9miUXXl VMeJFua4dnd3pzpQ9d/Z2elmZmY+5vzLaZbz39/fX3GZRiMM1M4Ss7OzfqeLsD0TRiSApC3G9PX1 ufPnz8dGsmiONM/aMjVpkb36XahYlEnSZ6mXHwLkgMjPXDISCEAAAhCAAAQgAAEIrFoCcqrl3MvR N8HBBrOwlA9G93t7e+1Sat8DAwP+rf+NGzeW9WHnmzZtWnY9jRMTOSRChEWOv0SIwcHB0hKJ8H4j jyWCSBSycVvb6l/zofu1RECEESXhsbXLdzEIIEAUY54ZJQQgAAEIQAACEIAABFqagJITjo6O+jf7 Z8+eLYX+z83Nuenpabd169bSG/Q0ByKBQZEWZ86cKUVjyPHXuaICshAg1L9EhqmpKe/sa7xy/M+d O+f5bNv28SV0jWYiwWdkZMRdv359aevkK755iUOKilD0g+aKAoFaCbAEo1Zi1IcABCAAgZYhoLwA WhrQUUMIaMsYjyEQgAAEckpAb7dX+oZbb9T37t3rTp486Y4cOeJD/PUGfsuWLd4ZzmIHBTnXe/bs cZOTk+7YsWPehsXFRS9+jI2NZbJ8QOPcvXu35/jOO+/4ZSGKCtHShf3796ee/8F+msPDw14IOn36 tBeBJEAol4PmSBEQFAjUSgABolZi1IcABCAAgZYhoB0YbBeGljEKQyAAAQhAoC4CGzdudI8++qh3 dOXw6k18Lbkk6ur8o4eV2+Dhhx/2ERBmg/JDZCGAmP0mhCjy4c6dO6XEk1nmT9B4t2/f7iRESHjQ uTjINgoEVkIAAWIl1HgGAhCAAAQgAAEIQAACEEiNgBzdtJMsJjG+2TaYwy+nv5lFgkMtCSebaSt9 tzYBckC09vxgHQQgAAEIQAACEIAABCAAAQhAIBcEECByMY0MAgIQgAAEIAABCEAAAhCAAAQg0NoE ECBae36wDgIQgAAEIAABCEAAAhCAAAQgkAsCCBC5mEYGAQEIQAACEIAABCAAAQhAAAIQaG0CCBCt PT9YB4HkBJaSNVFamMAa/rmtOjv8hqsiokICAvxdSwCJKhBIl4ASJ+pDgUA5Avw+ypHJ/3X+jzj/ c8wIC0JgTVtXQUa6Ooe5Zl3H6jQ8Q6v5DWcIO8dd8Xctx5PL0FYNgdC5ZLvGVTNtqRtqvwt923Hq ndJByxFgG86WmxIMgsDKCKxp63bu/j3n7iy4+/fvr6wRnmo4Af8f2CVxyM9Pw1vPV4P8hvM1n1mP hr9rWROnPwiUJ/DBBx+469evu1u3brmbN2+Wr8idQhFYXFx0s7Oz7oEHHnD37i39PyulkAQQIAo5 7Qw6rwTWrN/o3NKHoMe8znD+x8VvOP9zzAghAIH8E1DUQ09Pj7t9+7Zra2vL/4AZYSICGzZscBs3 Lv1/KhEQiXjltRJLMPI6s4wLAhCAAAQgAAEIQAACTSZAVGaTJ6AFu+c30YKTkqFJREBkCJuuIAAB CEAAAhCAAAQgAIHKBOSgXr582V29etWH6nd0dLitW7dmHk0xMzPjLl265LSkRJEcw8PDrqsr25xb CwsL7ty5c+7OnTtu7dq1bvPmza6vr68ywAbfVSTLhQsX3Pz8vLdB/cuOWvM4hMJDeNxgc2muxQkQ AdHiE4R5EIAABCAAAQhAAAIQKAoB5QaYnJz0Trec3LGxMe/oHj582EkQyKrI6X/33Xf9koEdO3b4 74mJCXf+/PnMcm1JgDl06JBbv3695yAex48f93yy4iAB5MiRI07fmouhoSEvyrz11lvkcchqEnLW DwJEziaU4UAAAhCAAAQgAAEIQKCZBPR2e6VvuBX5cPHiRe/sDgwMuM7OTn+sCISTJ0/6aIS0x6Y3 /epLzvaWLVu8DYODg/6tv67LGU+7KIGn+lIujZGREW+DIg9GR0c9H4kTaRfN4YkTJ7wAND4+7qM/ ent73a5du7wYJJGGAoFaCSBA1EqM+hCAAAQgAAEIQAACEIBAwwnI4VWEgXZJiC4z0LmEgSwcby27 0PKCTZs2LRujRAhFaEgkSbtMT097oUMiSLjUQQKAknzK+V+pyJPUdtu1Qokj161bV3pMolB7e7tf lqHlKRQI1EIAAaIWWtSFAAQgAAEIQAACEIAABFIhoFwDN27c8PkWQqdbnWkZgoq2cUyzyKGem5vz IohyLoRF57JL4kDaRduYqti4rT/ZIIFGYkzazr/l4AjFB7NDAoSiNPShQKAWAggQtdCiLgQgAAEI QAACEIAABCCQCgG9cddb/ajjr87MCVadNIuSPUoIkdAgRz8suqaPlmAoEiLNYss8oizMhrt377qb N2+maYIXYtSBIi6ixeySYESBQC0Elv+tquVJ6kIAAhCAAAQgAAEIQAACEGgQAXujH3X81bwcbxUJ BGkWOfZmh/UZ9mfX0rRDIkw5G6z/sE5oXyOPxULF+gzbtmtmZ3iPYwhUIoAAUYkO9yAAAQhAAAIQ gAAEIACBTAiYw1ups7QjD9R+pT7M8U5ia6VxVLoX9m/9xdUP68Xdr/dapTGaXZXq1Ns/z+eTAAJE PueVUUEAAhCAAAQgAAEIQKApBOScmoNaiwEreaaW9pPWrWSHJX6sVCdpP6u5nnFY6RiKzm+l3PLw HAJEHmaRMUAAAhCAAAQgAAEIQKAFCcQtpyhnpuUaiHNu7ZrVKddGvdfVfhLnOE07xMxssHHHjcvy MMTda8Q1G2OjbdDYbHyNsJM2VheB5aldV5ftWAsBCEAAAhCAAAQgAAEItBgBheVrlwYlc6wlV4Lt +BAX1m/LDdra2lIdrZJdyrFXf9andShHXNckEFhSTLvXyG855xqnkkxGWZgYIBvSFiBsPqIcNFaz y+okGb/yReh3IdttHEmeo06+CBABka/5ZDQQgAAEIAABCEAAAhBoKgE5mHJM9dFx0iKnW2/d40QL iRkqXV1dSZtbUT0JCxs2bCgrQMhxlg1pv8GXDSpRFnL8JQiIU9piTHd3t7fBxAZ/8tEfsksMzM7w Xrlj1ddvQnYjQJSjlP/ryf9FyD8LRggBCEAAAhCAAAQgAAEI1ElAjqYceX1qcdRVf+PGjd7pjjre 2pZSTndPT0+d1lV+XPb29fV5J//WrVvLKtvWmIODg8uup3EyMDDg2SliICyySQJEf39/qlEY6lMc xDy69anEA22/2dnZ6QWF0L5Kx4rY0Bzru5bfRaU2ubf6CCBArL45w2IIQAACEIAABCAAAQjkjoCc 0pGREe+gXrp0qTQ+iRFXrlzxDrGc4rSLBAY5yWfPni0tNZDT/f7777v29naXhQAhIaa3t9ddvHix ZIPGffnyZS9AjI2NpY3BRyoMDw+769ev+491ODU15W0aHR2tKcIljHoIj61dvotBgBwQxZhnRgkB CEAAAhCAAAQgAIGWJ6AIh127drnjx487RRwoZH96etove9i9e3cmb861RODAgQPu7bffdseOHXNa iiAnXELI/v37fVRA2iAlgIyPj3sbjh496sUX8ZAdskFc0i4mCGn5izhIeNHxtWvXnOZCURgUCNRK AAGiVmLUhwAEIAABCEAAAhCAAATKEtDb7XrecA8NDXmHe3Z21r9pl7MrYSLLsH3198QTT3hnW8LD 1q1bvU1Z2qBoi8cee8zNzMw4Lb3QkgeJEhInsirqa8+ePX7JhZaDdHR0OEU+yDYKBFZCILtf70qs 4xkIQAACEIAABCAAAQhAoHAEFIWwefPmpo5bCTSVi6GZRYJHFstOKo1RNijxZtoJQCvZwL38ECAH RH7mkpFAAAIQgAAEIAABCEAAAhCAAARalgACRMtODYZBAAIQgAAEIAABCEAAAhCAAATyQwABIj9z yUggAAEIQAACEIAABCAAAQhAAAItSwABomWnBsMgAAEIQAACEIAABCAAAQhAAAL5IYAAkZ+5ZCQQ gAAEIAABCEAAAhCAAARankCWu4m0PIyCGYgAUbAJZ7gQgAAEIAABCEAAAhBIk4CcS3Mw7TvN/mh7 dRHgN7G65qvR1rINZ6OJ0h4EIAABCEAAAhCAAAQKTOD27dvuypUr7s6dO25+fr7AJBh6SODGjRtu enraaXvTu3fvhrc4LhABBIgCTTZDhQAEIAABCEAAAhCAQNoEHnzwQbd+/XofBbF2Le5G2rxXS/v6 XbS1tZWiY1aL3djZWAL8i9BYnrQGAQhAAAIQgAAEIACBQhOQo9nd3e0UCSEhggIBEdiwYYP/XehY vxFKMQmQA6KY886oIQABCEAAAhCAAAQg0NIEPvjgA7+Mo5lG3rt3zwsp9+/fb5oZ6vvWrVtOtjSr yAYJSvXYEDIMj5s1JvptDgEiIJrDnV4hAAEIQAACEIAABCAAgRgCi4uL7vjx497p1hIOOb6jo6Nu cHAws/B9iR+nT592V69e9VEcEgCGhobcyMhIZjYoT8KFCxfcuXPnvA067+rqcjt37vRLGWLQNfyS hALl8zh79mwpd4OWUezatct1dnY2vD8azD8BBIj8zzEjhAAEIAABCEAAAhCAQGYE5LSu9A23xIYj R464np4ed+DAAe/0Xr9+3U1MTPhoiO3bt6c+Dtn+9ttve+Hjsccec+vWrfNiyOHDh93c3Jzbt29f JksI5PRLfHj88cdde3u7H//k5KTn8+ijj7os8mtMTU259957z42Pj7stW7b4CIgzZ864o0ePuocf fth1dHSkPh90kC8CLMHI13wyGghAAAIQgAAEIAABCKxaAnK6FW0wNjbmxQcNRGJEf3+/fwuv6Ii0 i3ZqmJmZ8TZIfFBRLgvZpIiI2dnZtE3wQsf58+fd8PCwFx/UoWyRACMGFy9eTN0GRVxoPnp7e330 hzrUDhbbtm3zu1icOnVqxUJT6sbTQcsSQIBo2anBMAhAAAIQgAAEIAABCBSHgBxeOf9ytJWwMCwS IbQsQssB0iyKftBbfyVJjC4xkA0qup92EQfxkPMfFnFR5IMEiJVGmYTtVTpW5InEDi37WLNmTamq 5kdsJNJoq1UKBGohgABRCy3qQgACEIAABCAAAQhAAAKpEFDkg5ZgWNRB2IktN1hYWAgvN/xYIoec bjncetsfFl3T59q1a+HlVI7n5+d9u1EWskkfcRKvNIuNM27HCs2HBJIsIlLSHCNtZ09g+d+q7Pun RwhAAAIQgAAEIAABCEAAAu7mzZv+rX6cw2vX0nZ49UZfzn05AULOv0QKfdIqimwQCxUTXqwvE0Xk /MvONIuxtj7DvswuszO8xzEEKhFAgKhEh3sQgAAEIAABCEAAAhCAQCYE5FSryPmPFruWpuOvPmWD bTVpfUZt0XnadpgNcX3btaxsqMSBJRg2G3wnJYAAkZQU9SAAAQhAAAIQgAAEIACB1AgkcbqT1KnH QEUfJMmtkKYdYduVnP8kdtbDIrQj2o7ZtVIb7Plou5znnwACRP7nmBFCAAIQgAAEIAABCEAgMwJy Ls3BjAvfL2dIkrpJ6pRrP8l1tW+2V6qfph1h25Uc/CR2VhpDtXuV2je7QlurtWf3K7VrdfjOLwEE iPzOLSODAAQgAAEIQAACEIBA5gTkYJqTabkbkhhheQXi3rybw1tLe0n6jNaRQ21OtfUZraNzszXu Xr3XxM7GGbUhPLc69fZX7nkbY9in1bVrtdhgXMPxWXt8F4cAAkRx5pqRQgACEIAABCAAAQhAIHUC ck6VIFGfWvIUrF+/3ttmuSBCQ62d6PacYZ1GHGvXCX00hqgQYtfkmJtz3og+49oox0I2yQ45/tEd MuLaqedae3u7fzzKQRdtPqxO0n7sd6ExUIpJAAGimPPOqCEAAQhAAAIQgAAEIJAaATnHbW1tPqlj 0k70jBzvuO0lbceH7u7upM2tqJ5s6Orq8k5+VAiR0y3Hua+vb0Vt1/KQbFCJ7jJhSTJlZ9pijI3T 2If265pEkM7OzvByxWPx029C4k2cqFHxYW7mhgACRG6mkoFAAAIQgAAEIAABCECg+QQUam9LMKJO fCXr5FT39/d753RhYWFZ1bm5Of/G35ziZTcbfLJp0yYvnNy4cWNZy/Pz8/58cHBw2fU0TmSDGM7O zi5rXltjiumWLVtKjJdVaOBJ4y34fgAADtVJREFUT0+PU4SDOIQRC9r5Qnb09vbWFIVhvwX9PmpZ utHAIdFUCxBAgGiBScAECEAAAhCAAAQgAAEI5IWAnMtQgAid12pj3LFjh3d6z5w5U4qEmJ6edlev XnUjIyOuo6OjWhN13x8YGHASAE6ePOkdbTUoh/vUqVNO4oMc77SLIgu2bdvmpqam3MzMjBcAFHVw +vRpH/mge2kXzePY2Jgf+4ULF7wNimI4ceKEj2LYuXNnTSYgQNSEK7eV1+Z2ZAwMAhCAAAQgAAEI QAACEMicgN5wW8LBWkPtFZ5/4MAB7/xPTEz4N+USMCRMDA8PZzIWiSe7d+92EkGOHTvmnW299d+8 ebMbHR0tjS1NY2SD+pIIYA6/HHgJMPv27cvEBo1PY9YcSoCQECQbNEcHDx6seQlIKEAo2oVSTAII EMWcd0YNAQhAAAIQgAAEIACBVAhonb8JEJY3wSIiknSot/9ycCU8yPltRri+xjA+Pu7NlQ02niT2 N6qOxi0RQp9m2aB503IPfWSDzmuZy5CFfgsqYpl2Es+wX45biwBLMFprPrAGAhCAAAQgAAEIQAAC q5qA3m6bg6nIAXvzXeug5Og2Q3yI2tkM8aFVbVip+KDx6Legojbs9+Ev8EehCCBAFGq6GSwEIAAB CEAAAhCAAATSJaCdLEw4UALDlQoQ6VpJ61kTsMSiEqhsm9GsbaC/5hNAgGj+HGABBCAAAQhAAAIQ gAAEckNAOyeEAoSF3udmgAxkRQRsVxFFP6S9heiKDOShTAggQGSCmU4gAAEIQAACEIAABCBQDALK n2AOpt56I0AUY96rjRIBohqhYtxHgCjGPDNKCEAAAhCAAAQgAAEIZEagv7/f9zU/P88SjMyot3ZH tgRD4pREKkoxCSBAFHPeGTUEIAABCEAAAhCAAARSIzA0NOTb1ltvSz6YWmc0vCoIzM7Oejs3bty4 KuzFyHQIIECkw5VWIQABCEAAAhCAAAQgUFgC27Zt82NfXFx0ioKgQGBmZsZDGBgYAEaBCSBAFHjy GToEIAABCEAAAhCAAATSILB9+/ZSs9PT06VjDopJ4Nq1a+727dt+8Fu3bi0mBEbtCSBA8EOAAAQg AAEIQAACEIAABBpKYHBwsNSevfkuXeCgcASmpqZKYx4eHi4dc1A8AggQxZtzRgwBCEAAAhCAAAQg AIFUCezcubPUfuh8li5yUCgC77//fmm8O3bsKB1zUDwCa4s3ZEYMAQhAAAIQgAAEIAABCKRJYNOm Ta69vd3dvHnTnTt3ruaubt265a5fv+7u3bvnOjo6XFdXl1uzZk3N7dTzgLYPlQ1aOqCx9PT0uAce yPb97f37970NyqWxfv1619nZ2ZQdJJTHQ59169Z5G8SjlnLy5ElfXXO4d+/eWh6lbs4IZPs3KGfw GA4EIAABCEAAAhCAAAQg8HECctRtrf+FCxc+XqHCFdU/fPiwm5ubc3LAjx8/7iYmJjLdTUPCw6FD h9zly5e9pRcvXvQ2aVePrIpEGHE4c+aM7/Lq1avuzTffLNmUhR0SgN555x337rvvejFIc3LkyBEX RjQkscNEKAlTfX19SR6hTk4JIEDkdGIZFgQgAAEIQAACEIAABJpJYN++fb57OasSEpIU5Ys4ceKE Fy927drltmzZ4h555BG3sLDgr8shTrso8kGCR3d3t9uzZ4+3QWNRBMLRo0czEULM8dcWpgcPHvQ2 7N692ym3hsSArIQQiT9KILl//34/J1o+oeU1p0+fdleuXEk8FSZYjI+PJ36GivkkgACRz3llVBCA AAQgAAEIQAACEGgqAQkHKnprrk+1IpHi/Pnz/k27hAcriqbQm3M5vLOzs3Y5tW9FYEiEkA3hsg9F dGg5Rha7esjpVxSGEjaGNkiAEA+LKEgNwlLDEj801v7+frdhw4ZSVzpfu3atO3v2rLt7927perkD RXIogkRFIgql2AQQIIo9/4weAhCAAAQgAAEIQAACqRD4xCc+4Z1l5Q5IsgxDDq8iIBRp8OCDDy6z SbkPFBWQ9o4aEkHUh/qP5jmw8yySamq5hWxR7ouwtLW1eaYSKCSSpFkkPmhONB9hkQCivByKwrCt NcP70eNLly55MUVCikXFROtwXhwCCBDFmWtGCgEIQAACEIAABCAAgcwI6E350NCQd1KPHTtWtV8l rNQbdb1djxYTJLQUI80ih1p2yFmWox0tuqbIhKRLSqLPJz1X0kkVG7c9J7v0kfigyII0i8apErVB 1zRHYmB26lq5omgNiRW9vb1+SUu5elwvBoGP/60qxrgZJQQgAAEIQAACEIAABCCQIgE5nLbm//XX X6/ak962q5Rz/HUvyRt31VtpkQ0WWSBHPyzm/Mvxtjrh/UYdq30TF6LOv7FRNEiaNmgsxtr6DMdn 18zO8F70WIk0VZTTY9u2bdHbnBeMAAJEwSac4UIAAhCAAAQgAAEIQCArAp/85Ce9oKAIiGpvyy2f QNTxl612LW2nW85/kkSXadohG/QJx+1PIn+kaYO6svmIdOtPk86HWL722mv+GSWyjC7niGuba/km gACR7/lldBCAAAQgAAEIQAACEGgaAe3goN0kFIavvAaVijndleqkfS90/uP6Msc77l6jrrUCh2pj SWqjkoZqxwwViVEUCCBA8BuAAAQgAAEIQAACEIAABFIh8NBDD/llGArVP3ToUMU+bLlBnHNr1yz0 v2JDddyUDZX6yMKOsH/rL25IYb24+/Ves/mo1E41G7RlqBJmKh/Ik08+Wakp7hWEAAJEQSaaYUIA AhCAAAQgAAEIQCBrAooYePrpp323f/rTnyp2rx0eVOKcbru2bt26im3Ue1NOtzne1mfYpl0zW8N7 jToWMxtndDmI9a86cck6G2WD2ilng+6ZXVZH1+LKH//4R3/52WefLXGNq8e14hBAgCjOXDNSCEAA AhCAAAQgAAEIZE7gqaee8s7y73//e78bQjkD5NTLsbbkh2E9S1Cp7R/TLLJBeQrk6JuTbf3pXNdl g+xMs2zYsME3b+O2vpSXQTZIJElTBFF/2vpUJS4XhOwSA7PTV4z8oRwVv/nNb/zcP/PMM5G7nBaV AAJEUWeecUMAAhCAAAQgAAEIQCADAtu3b3cHDhxwc3NzpYSEcd3qbbqcezm3UefftsbcuHFj3KMN u6YlBT09Pb7/qPMvYUTOv5YTpF20g4iKxh0WOfViI3EgbQGir6/PiwxxO13ILgk1lQQIJR69dOmS X4Kj+adAQAQQIPgdQAACEIAABCAAAQhAAAKpEWhvb3cvvPCCf2uvKIhyRW/1bZvGmZmZUjU53Moj oGSWEgfSLoODg97Jv3z58rKurly54scwNDS07HoaJxI55OBHbVBSR4kQEnXSjsKQGCQhxPq0cUpI kighDpWWgfz617/2Nn7+8593+g1QICACCBD8DiAAAQhAAAIQgAAEIACBVAl8+tOf9k6zBIgLFy6U 7UvOvz6Tk5Nuenrab9158uRJ7/Du3r07k20cu7q6/Ft77dxx/vx5H4Wg74sXL7odO3b4KI2yA2jQ DUU37Nmzx0eNnDp1ytswNTXld5SQSJNFFIYEjvHx8dI2qgsLC07C0PHjx93AwEBJLIobsnhp+03Z +bnPfS6uCtcKSuDBf14qBR07w4YABCAAAQhAAAIQgAAEMiCg6AY5sL/73e98FMPjjz8e26ucXjmt Wo6h7Rv10bF200g7/0NokEQIvf0/e/askwCgPAj79u1zmzZtCquleqyoAdmgKAgJMvPz825sbMyN jIyk2m/YuCIcJDYo4kE7WmgrVQlEEoMq7YChef7Vr37lvvjFLzryP4REOV6ztI7pPhggAAEIQAAC EIAABCAAAQikSWBxcdF96Utf8mH7v/jFL6ruiiA3RZ9Kjm6a9lrbWgIiYSTtJQ/WX9y3ODSzf9mU lINs/cY3vuHFG82z5bOIGxfXikeAJRjFm3NGDAEIQAACEIAABCAAgcwJKGHh888/786cOeN++9vf Vu1fDnezxQcZKRua7fw3u/9aOLz++uvuzTffdF/4whcQH6r+yotXAQGieHPOiCEAAQhAAAIQgAAE INAUAgrJ11KGn//857HbOzbFKDptGAFFSfzsZz9ze/fudc8991zD2qWh/BBAgMjPXDISCEAAAhCA AAQgAAEItDQBRUF897vfde+995575ZVXWtpWjKudwB/+8Af3xhtv+CUYW7durb0Bnsg9AQSI3E8x A4QABCAAAQhAAAIQgEDrEHjkkUf8Uowf/ehHPqlh61iGJfUQUJLRn/70p+5Tn/qU+8xnPlNPUzyb YwIIEDmeXIYGAQhAAAIQgAAEIACBViTw1a9+1W+p+ctf/tInmmxFG7GpNgIvv/yy3+nk29/+dtNz ZtRmObWzJIAAkSVt+oIABCAAAQhAAAIQgAAE3ObNm92LL77oJEBMTExAZJUTOHfunPvxj3/svvOd 72S6Tegqx1ZI8xEgCjntDBoCEIAABCAAAQhAAALNJXDw4EH3la98xf3kJz9xd+7caa4x9L5iApq7 H/7wh35ZzVNPPbXidniwGATWLO3Ter8YQ2WUEIAABCAAAQhAAAIQgECrEZibm3NdXV2E7bfaxNRg z+LiolOCUQoEqhFAgKhGiPsQgAAEIAABCEAAAhCAAAQgAAEI1E2AJRh1I6QBCEAAAhCAAAQgAAEI QAACEIAABKoRQICoRoj7EIAABCAAAQhAAAIQgAAEIAABCNRNAAGiboQ0AAEIQAACEIAABCAAAQhA AAIQgEA1AggQ1QhxHwIQgAAEIAABCEAAAhCAAAQgAIG6CSBA1I2QBiAAAQhAAAIQgAAEIAABCEAA AhCoRgABohoh7kMAAhCAAAQgAAEIQAACEIAABCBQNwEEiLoR0gAEIAABCEAAAhCAAAQgAAEIQAAC 1QggQFQjxH0IQAACEIAABCAAAQhAAAIQgAAE6iaAAFE3QhqAAAQgAAEIQAACEIAABCAAAQhAoBoB BIhqhLgPAQhAAAIQgAAEIAABCEAAAhCAQN0E1i618IO6W6EBCEAAAhCAAAQgAAEIQAACEIAABCBQ gcD/A/txw1fqTqFLAAAAAElFTkSuQmCC ) (1)(c) Expand out W\_{\text{net}}^1 = W\_{out}^1 \cdot W\_V^1 and W\_{\text{net}}^2 = W\_{out}^2 \cdot W\_V^2. What do these matrices tell you? ![](data:image/PNG;base64,iVBORw0KGgoAAAANSUhEUgAABCEAAASCCAYAAACM+va+AAAMbWlDQ1BJQ0MgUHJvZmlsZQAASImV VwdYU8kWnluSkJDQAghICb0J0gkgJYQWQHoRRCUkgYQSY0JQsaOLCq5dRLGiqyKKbaXZsSuLYu+L BRVlXdTFhsqbkICu+8r3zvfNvX/OnPlPuTO59wCg+YErkeSjWgAUiAulCeHBjDFp6QxSJ1AHOKAC OjDi8mQSVlxcNIAyeP+7vLsBEMX9qpOC65/z/1V0+AIZDwAkA+IsvoxXAPFxAPB1PIm0EACiQm85 uVCiwLMh1pXCACFeqcA5SrxDgbOU+PCATVICG+LLAKhRuVxpDgAa96CeUcTLgTwanyF2EfNFYgA0 R0AcwBNy+RArYh9RUDBRgSshtoP2EohhPICZ9R1nzt/4s4b4udycIazMa0DUQkQyST536v9Zmv8t BfnyQR82cFCF0ogERf6whrfyJkYpMBXibnFWTKyi1hB/EPGVdQcApQjlEclKe9SYJ2PD+gF9iF34 3JAoiI0hDhPnx0Sr9FnZojAOxHC3oFNEhZwkiA0gXiCQhSaqbDZJJyaofKH12VI2S6U/x5UO+FX4 eiDPS2ap+N8IBRwVP6ZRLExKhZgCsVWRKCUGYg2InWV5iVEqm1HFQnbMoI1UnqCI3wriBIE4PFjJ jxVlS8MSVPZlBbLBfLFNQhEnRoX3FwqTIpT1wU7xuAPxw1ywywIxK3mQRyAbEz2YC18QEqrMHXsu ECcnqng+SAqDE5RrcYokP05lj1sI8sMVeguIPWRFiaq1eEoh3JxKfjxbUhiXpIwTL87lRsYp48GX gmjABiGAAeRwZIGJIBeI2robuuEv5UwY4AIpyAEC4KTSDK5IHZgRw2siKAZ/QCQAsqF1wQOzAlAE 9V+GtMqrE8gemC0aWJEHnkJcAKJAPvwtH1glHvKWAp5Ajegf3rlw8GC8+XAo5v+9flD7TcOCmmiV Rj7okaE5aEkMJYYQI4hhRHvcCA/A/fBoeA2Cww1n4j6DeXyzJzwltBMeEa4TOgi3J4hKpD9EORp0 QP4wVS2yvq8FbgM5PfFg3B+yQ2ZcHzcCTrgH9MPCA6FnT6hlq+JWVIXxA/ffMvjuaajsyC5klDyM HES2+3GlhoOG5xCLotbf10cZa9ZQvdlDMz/6Z39XfT68R/1oiS3ADmBnsRPYeeww1gAY2DGsEWvF jijw0O56MrC7Br0lDMSTB3lE//DHVflUVFLmUuvS5fJZOVcomFKoOHjsiZKpUlGOsJDBgm8HAYMj 5jmPYLi5uLkCoHjXKP++3sYPvEMQ/dZvurm/A+B/rL+//9A3XeQxAPZ5w+Pf9E1nxwRAWx2Ac008 ubRIqcMVFwL8l9CEJ80QmAJLYAfzcQNewA8EgVAQCWJBEkgD42GVhXCfS8FkMB3MAaWgHCwFq8Ba sBFsATvAbrAfNIDD4AQ4Ay6Cy+A6uAt3Tyd4CXrAO9CHIAgJoSF0xBAxQ6wRR8QNYSIBSCgSjSQg aUgmkoOIETkyHZmLlCPLkbXIZqQG2Yc0ISeQ80g7cht5iHQhb5BPKIZSUV3UBLVBR6JMlIVGoUno ODQHnYQWo/PQxWglWo3uQuvRE+hF9Dragb5EezGAqWP6mDnmhDExNhaLpWPZmBSbiZVhFVg1Voc1 w+d8FevAurGPOBGn4wzcCe7gCDwZ5+GT8Jn4InwtvgOvx0/hV/GHeA/+lUAjGBMcCb4EDmEMIYcw mVBKqCBsIxwknIZnqZPwjkgk6hNtid7wLKYRc4nTiIuI64l7iMeJ7cTHxF4SiWRIciT5k2JJXFIh qZS0hrSLdIx0hdRJ+qCmrmam5qYWppauJlYrUatQ26l2VO2K2jO1PrIW2ZrsS44l88lTyUvIW8nN 5EvkTnIfRZtiS/GnJFFyKXMolZQ6ymnKPcpbdXV1C3Uf9Xh1kfps9Ur1vern1B+qf6TqUB2obGoG VU5dTN1OPU69TX1Lo9FsaEG0dFohbTGthnaS9oD2QYOu4azB0eBrzNKo0qjXuKLxSpOsaa3J0hyv WaxZoXlA85JmtxZZy0aLrcXVmqlVpdWkdVOrV5uu7aodq12gvUh7p/Z57ec6JB0bnVAdvs48nS06 J3Ue0zG6JZ1N59Hn0rfST9M7dYm6troc3Vzdct3dum26PXo6eh56KXpT9Kr0juh16GP6Nvoc/Xz9 Jfr79W/ofxpmMow1TDBs4bC6YVeGvTcYbhBkIDAoM9hjcN3gkyHDMNQwz3CZYYPhfSPcyMEo3miy 0Qaj00bdw3WH+w3nDS8bvn/4HWPU2ME4wXia8RbjVuNeE1OTcBOJyRqTkybdpvqmQaa5pitNj5p2 mdHNAsxEZivNjpm9YOgxWIx8RiXjFKPH3Ng8wlxuvtm8zbzPwtYi2aLEYo/FfUuKJdMy23KlZYtl j5WZ1Wir6Va1VnesydZMa6H1auuz1u9tbG1SbebbNNg8tzWw5dgW29ba3rOj2QXaTbKrtrtmT7Rn 2ufZr7e/7IA6eDoIHaocLjmijl6OIsf1ju0jCCN8RohHVI+46UR1YjkVOdU6PXTWd452LnFucH41 0mpk+shlI8+O/Ori6ZLvstXlrquOa6RriWuz6xs3BzeeW5XbNXeae5j7LPdG99cejh4Cjw0etzzp nqM953u2eH7x8vaSetV5dXlbeWd6r/O+ydRlxjEXMc/5EHyCfWb5HPb56OvlW+i73/dPPye/PL+d fs9H2Y4SjNo66rG/hT/Xf7N/RwAjIDNgU0BHoHkgN7A68FGQZRA/aFvQM5Y9K5e1i/Uq2CVYGnww +D3blz2DfTwECwkPKQtpC9UJTQ5dG/ogzCIsJ6w2rCfcM3xa+PEIQkRUxLKImxwTDo9Tw+mJ9I6c EXkqihqVGLU26lG0Q7Q0unk0Ojpy9IrR92KsY8QxDbEglhO7IvZ+nG3cpLhD8cT4uPiq+KcJrgnT E84m0hMnJO5MfJcUnLQk6W6yXbI8uSVFMyUjpSblfWpI6vLUjjEjx8wYczHNKE2U1phOSk9J35be OzZ07KqxnRmeGaUZN8bZjpsy7vx4o/H5449M0JzAnXAgk5CZmrkz8zM3llvN7c3iZK3L6uGxeat5 L/lB/JX8LoG/YLngWbZ/9vLs5zn+OStyuoSBwgpht4gtWit6nRuRuzH3fV5s3va8/vzU/D0FagWZ BU1iHXGe+NRE04lTJrZLHCWlko5JvpNWTeqRRkm3yRDZOFljoS78qG+V28l/kj8sCiiqKvowOWXy gSnaU8RTWqc6TF049VlxWPEv0/BpvGkt082nz5n+cAZrxuaZyMysmS2zLGfNm9U5O3z2jjmUOXlz fitxKVle8tfc1LnN80zmzZ73+Kfwn2pLNUqlpTfn+83fuABfIFrQttB94ZqFX8v4ZRfKXcoryj8v 4i268LPrz5U/9y/OXty2xGvJhqXEpeKlN5YFLtuxXHt58fLHK0avqF/JWFm28q9VE1adr/Co2Lia slq+uqMyurJxjdWapWs+rxWuvV4VXLVnnfG6hever+evv7IhaEPdRpON5Rs/bRJturU5fHN9tU11 xRbilqItT7embD37C/OXmm1G28q3fdku3t6xI2HHqRrvmpqdxjuX1KK18tquXRm7Lu8O2d1Y51S3 eY/+nvK9YK9874t9mftu7I/a33KAeaDuV+tf1x2kHyyrR+qn1vc0CBs6GtMa25sim1qa/ZoPHnI+ tP2w+eGqI3pHlhylHJ13tP9Y8bHe45Lj3SdyTjxumdBy9+SYk9dOxZ9qOx11+tyZsDMnz7LOHjvn f+7wed/zTReYFxouel2sb/VsPfib528H27za6i95X2q87HO5uX1U+9ErgVdOXA25euYa59rF6zHX 228k37h1M+Nmxy3+ree382+/vlN0p+/u7HuEe2X3te5XPDB+UP27/e97Orw6jjwMedj6KPHR3ce8 xy+fyJ587pz3lPa04pnZs5rnbs8Pd4V1XX4x9kXnS8nLvu7SP7T/WPfK7tWvfwb92dozpqfztfR1 /5tFbw3fbv/L46+W3rjeB+8K3vW9L/tg+GHHR+bHs59SPz3rm/yZ9Lnyi/2X5q9RX+/1F/T3S7hS 7sCnAAYHmp0NwJvtANDSAKDDvo0yVtkLDgii7F8HEPhPWNkvDogXAHXw+z2+G37d3ARg71bYfkF+ TdirxtEASPIBqLv70FCJLNvdTclFhX0K4UF//1vYs5FWAPBlaX9/X3V//5ctMFjYOx4XK3tQhRBh z7Ap7ktWQRb4N6LsT7/L8cc7UETgAX68/wsztZC2JFMiHQAAADhlWElmTU0AKgAAAAgAAYdpAAQA AAABAAAAGgAAAAAAAqACAAQAAAABAAAEIaADAAQAAAABAAAEggAAAACSNtJsAABAAElEQVR4Aey9 yW8lyVr+Hy6P5SqXq8rloea5er6Xe6FhBwJ2LL4LQCzYgIRa6g2r+w/8WCA2SA0SG4RgAUgIcRGo rxCIQUgIMalBfYeurnm2XXaVXeWhBs+/+MTx6wpn5Tk+tjPjHNtPdLtOnsg8+T7xvG9Evu8bkZkt K744FTEgBsSAGBADYkAMiAExIAbEgBgQA2JADJTIQIsv+0o8v04tBsSAGBADYkAMiAExIAbEgBgQ A2JADIiBNQaUhFijQhtiQAyIATEgBsSAGBADYkAMiAExIAbEQJkMKAlRJrs6txgQA2JADIgBMSAG xIAYEANiQAyIATGwxoCSEGtUaEMMiAExIAbEgBgQA2JADIgBMSAGxIAYKJMBJSHKZFfnFgNiQAyI ATEgBsSAGBADYkAMiAExIAbWGFASYo0KbYgBMSAGxIAYEANiQAyIATEgBsSAGBADZTLQVubJdW4x IAbEQNMy4N9OvLL40rmV5aaFKGCbZ6BlX7tzbV2b/6F+IQbEgBgQA2JADIgBMZCEASUhktAsIWJA DDQVA8sLbvnlU5+AWGkqWAKzfQbQaEv7ftfSdWT7J9MZxIAYEANiQAyIATEgBgpnQLdjFE6pTigG xECzM7D8+pkSEM2upG3gW1l45dzS/DbOoJ+KATEgBsSAGBADYkAMlMWAkhBlMavzigEx0LwMLC02 LzYhK4SBlaW5Qs6jk4gBMSAGxIAYEANiQAwUy4CSEMXyqbOJATEgBsSAGBADYkAMiAExIAbEgBgQ A1UYUBKiCjGqFgNiQAyIATEgBsSAGBADYkAMiAExIAaKZUBJiGL51NnEgBgQA2JADIgBMSAGxIAY EANiQAyIgSoMKAlRhRhViwExsPsYWF5eds+fP/cvxdBbMXafdtUiMSAGxIAYEAPpGZibm3P8qYgB MVA/A0pC1M+VjhQDYmAHM0Di4Q/+4A/cb//2b7vFRT2YcgerUtDFgBgQA2JADDQNA//yL//iPvnk E/f0qX/1t4oYEAN1MaAkRF006SAxIAZ2OgP/8z//4773ve+53/iN33Dt7e07vTnCLwbEgBgQA2JA DDQBAz/7sz/rBgYG3B/+4R86VlyqiAExsDEDSkJszJGOEANiYIcz8PDhQ/c7v/M77ld+5Vfcu+++ u8NbI/hiQAyIATEgBsRAszCwf/9+95u/+Zvuv/7rv9xf/uVfNgss4RADTc2AkhBNrR6BEwNioAgG SEB0dHS4X//1Xy/idDqHGBADYkAMiAExIAbWGDh79qz7xV/8RfdHf/RH7ubNm2v12hADYiCfgbb8 atWKATEgBnYHA//xH//hvvzyy7ASotptGPbAyrn5Obe/a787dOiQ27evMTna+fn58PDM5ZVld6D7 gOvp6WmQIlbczMyse/HihWtra3OHeg65js6OhmBZ8Vw8e/Y8PPirq6vL9aKf1taGYEE/3PeLzaCb 3t7ehuBA6OzsbLCVVs/F4cOHHbNxjSg8b2ViYsK9evXKoZ8jR44Em2kEFp73gn4WFhbcwYMHg34a 1Zdfvnzp7faZa2lpCWMKeBpR0M/U1JTvzzOus7MzcMJnIwr6wVboR/CB3cJPIwr2Ojk5GUTTjxup n+npaccfyXI4aZR+GNfoPzxksbu7O/TlRvUfMGArzTDW0nd4qDXXwlpj7f/7f//P/f3f/7373d/9 3XBrRiPsWjLFwE5hoPX/82WngBVOMSAGxMBmGBgfH3ff+c533EcffeR+7dd+be1ZECvzM2unwTm/ feu2W1xaDI7ftHc2hodHfDDVGb6nc5BXgtwHDx4EuQSW42Pj7tnzZ8EZrJZAWWtIgRsECLdv3/FO 11SQveC/Pxp+5JaWltyBA90+aEiXoEE/t27d8bIXQ4A7i35GRoMuawXdLW2drqW12EAL3TDDRaCN M4p9jY6OhqAutX7AMTIyEgInnPR79+45gt7UQR3O+Q9/+MO1oAUMt27dCvykDOoItMfGxtxXX30V AjmCuSdPnvg+NRxsGJ2lKugDDrAXAjnGELYJYkhwYjupyuvXrwMnBLjoA/3cv38/PJwXLOnGNxf6 y9WrVx1jG0E2wS76OXDgQPieihNs5fbt24EHZMPBo0eP1vQDvlSFQBubJVmFftDXnTt3whuc+J5S PyRk0A+JIsZW7BVboe/wlxILculD4EAfjHXYC5ykHGvh4tq1a+7x48eh75LcZKxFTyShswka+jtc /dVf/ZXr7+/X7Z+pOpLk7DgGfsuXdJ7kjqNHgMWAGNjJDOBo/sM//EMITH7pl34pd5aYYPuudyhI OFy8eMENDg668+fOeUfngLt7526Y2U3FweTk8+DoDA0NuVOnTgUsl69ccnOv54Ij6Fyq14quuPv3 7vu2v17j5KTHc/rUaTfiA+7JyWepKAnOMA5fZ0e7O3/+Qnjw17nzFf3gpIIxVSFIIJBEN+fPn3cn T550V65cCTO6169fT/raVwImZgjff//9gOf06dPuwoULwdbBmKqQlCJQwBF/77333IkTJwI3x48f D/VwlqoQZN+9ezfYCFyA5Z133gkBAfohmEhVCKwJWi5evOhYoo3NoCsSauiHsSlFQT8E24xzcAEn cHPmzBnHc3JIoqUqrNohcdbX1xd4AQvP5yEhc+PGjaSvNySgJWFlXKAf+CEIh69U+rFkCLaJfMYU xhZshnEPnKkKNoIeCKLRC/phfCNRRf8heZWqkGzAPhnXzvnrMbx88MEHIfBPfZsDYwr9Fk7AgW7g xWwoj5Of+7mfC33sT/7kT8Lqo7xjVCcGxIBzSkLICsSAGNiVDODs/+mf/mlwGH76p386p42VmdMl 7wAe9w6Xcy3hGGZ7CKKWlpeCo5rzw8KrmDkdHR0JiRKcdCutrW3u2EB/WHb/9Gll2bDtK+tzenrG TU1PuWPH+tbNTh7qPeRvVekKziF4UxQcPRz0gcGhaBauJQSaOPAEeymKBQvMrpOossKMHAEdqwGY dU9RmIGDl2PHjoWAwWRyCwS2w8qMVO+rJ4gluMQxtxlB+g8BDMElyZJUAR2ysEuCFpuxBQN9GT5S BXQEc9gly/vjvszsKDqDM4KaFIUkEH/YbLy8n6f4Y8sk8khUlF2wAWShH/qL6Qeb4Tu3RWC3KQr9 B1msgEAfVphxx24JgOnPKQorDUh8oI94VRf6IhmAHaXqP8jCFgj8TT988p36VPrBRrAV2g8vVqwv k2xMldzkVkSSiUePHg0rMAyL3YYHZ/T3bMG2ft0/f4r9f/M3f5Pdre9iQAysMqAkhExBDIiBXcnA 3/7t34blpL/6q7+a2z4cKxya1n2twSGPD8LhIQGAs5Mi4MYx5o+gFtlx6ersDF+fPUuThDAHj9su 4kLAQCBjvMX7ytjmORAEa+ihs3P9syjgqb2tPTyvIsUMN7OABEoEbtml2gQPOOvGWxlcxOckiCUw wUnPFrDAB8FNimIz6QTYccGG0RHBHI582QWbhP9q+kE+K0c4ruyCzdrzKLKy0A/jSSpboc3YCkFR tlBHcga7Lrtgk5W+/PZYa9hScYJN0uY46Lf2W59KhcX6ch4W6ghwSfKVXdAPbbYxPpZHX+aP/Smu hfFY+9a1cHWcSZXwpf9QuAUkLoz32Ar6qWYrP//zPx+SXN/97neTJUBjjNoWAzuBASUhdoKWhFEM iIFNMcCsKA+HYhY0fxWE80H/nJufm3ct+1rWZnFNCE4GDhlOF8mBsguBGsFCNsBFrtWBt+xZMc5v QQnBf7aYU/j6dfmBy/zcQgio0QP6iAt1+1r3+f0LuTNR8bFFbNvMKLrIYrE6dJgiIWJY4CBbzFZS JCGwFcNidhHjoY6gP8UybpKJG/Uf7DpFEsKSLqaLLCd8t2PifWVsmxwSQtliOjMdZvcX+d24N5nx uamjT4G17PENuRbU5+nH6oy3GGfR21xbTI7JjWVYnR0T7yt625LgYVzNjCt2LSSxlqIvox/soJqt 0Hbr70XzkD2f2UoeFvoUOKvph4Q9t2WwAuuf//mfs6fWdzEgBjwDb3sxokUMiAExsMMZ+Ou//usw 8/lTP/VT4Z7WvObgVHHLBbdhZANLjt+3GvimWNpuiY48HCRJeBAkARS3jpRZOP/ioufEtz0PiwW+ cz55U3bhQaE46uYEx/Ko4w8nED2WXcwGrP2xPOMJHCmSEGYrFqTEWAwfx5Qd0MEJMqz9MQ62DYtx l91f5HcLjkxmfG6rC/0nwUqIWlhMZ6bDGGcZ228Sim8/aDElllo4aLf15RS2YtybXcS8W50dE+8r epuxzcYukxvLsDrjLt5X9Ha4Fq72jWx/5jtYUibkaZ+1P24rWPhjnC17rI2vLdZXYiyGr5at/PIv /3J4Js3f/d3fJbsFK8aobTHQ7AwoCdHsGhI+MSAGNsUAsxe8lpNlvr/wC79Q9bd2L6f3abxjs/6w 4Iit1qWYPTUZQe56KKtPqvCPpfS3Jywvl/tAu2UfVCInQ8caIsNneNd2lLCB01sJct8+eYzPHPm3 jyquplZ74YQ/jql1XFFo6pHBMWUnIdBPPaXsYAEMxonZZzVcjcZi+FLggINaOjIsxl01zoqoNxwm M3tOq280FsORQj9wYrxk+eC7YUkxvll7kWlys5gYT1Lop5YMwweWWtxlsW/le6yfPE6szrjLk3H+ /Hn34Ycfhred/Nu//VveIaoTA3uaASUh9rT61XgxsPsY+M///M/whHMelvetb32ragPrdWLKDuYA uCGW1ai7bCyc32SYk5VHYNnJkMCJx1K1BD4qpBjeqscWsMP0U4sTcNhxBYjc0ikMXwpOLFgwmdUA p8BSS0aMr9Zx1fBvtr4eG6jnmM3KzR4ftzXmIHtcaixZ+fH3ZsIS8xdjLHLbZDSTfvKwWF0K/ZgM k5nHN7zZcXn7i6jj/KafvPMZPo6pdhzH/MzP/EzY//nnn1c9Lu/8qhMDe4EBJSH2gpbVRjGwhxjg VgzKT/zET1Sd1WG/Ladku9bbL83ZCMeV9M86LHkyLB6PlwDkHbfNOtrKf5ViQt8+6T5/i0jZxW6H yZUTQWsK/XiQ6HBDPeY2ZnOVtdprznDQo9dlmcXaajKryaqFt9pviqyP8aXAUo8M467IdmbPFeOI OcgelxpLVn78PQUWk1eLE46J+bPfFP1pMmphScGJ4aB9jcZST3vBW89x29HXRuc3nsAS85eV+fHH HzveXPT111+H1/Nm9+u7GNjLDCgJsZe1r7aLgV3GAE9g/+EPfxicgp/8yZ+s2TqepE9hwj2KaVfr 3lS2+genlV3swVfm2MTywEfB0WnNPDSssqe4f3G8eAYFIk1ufHbDl3ePbHxcEdsBi38WRi6OCsIg xrgrQma1c5iMvNk3OOEPTmo5o9XOvdl6w2K6iH9vdSn0s5EMw2J4Y5xFb7/py9mevF7SRpjXH721 b7WwmP3kPShya9Jq/8q4N13ER1udHRPvK3rb2msys+c3XlJgMRvIw2J1KXAwvlmga3JjXqzOuIv3 Fb1t7TWZ2fNTz9hmx2X3F/m9Vv8Bh2Ex7oqUHZ+L85utmH3G+40rOybeF29fuHDBXbx4MTzg+p/+ 6Z/iXdoWA3ueASUh9rwJiAAxsHsY+O///u/w2jneivHOO+/UbBhPr644EN6xyXnWgq8Nv+9cTVbU PNk2d9orDnNxBMdrOWAtOyECHx08Sd/LzHO8bAl+e8fbT9vfJgVv/by93b8m1TuC4Mhi4XYQc0bb 29e/vvOtExVQga1QsjisDiw46OZAh4NL+sdkmC5iMYYPeyo7IQInyKDt5pBXwxLXl7Ftrzi09scy jCdsO0VAZ33Z5OZhMR3G+8rYNrvNu2/d6lJgqYXD7AdbSoHF9FPLVuyYMnRi5yTINXvMw2J1Ztv2 uzI+wWHBtMk1OeiHOvCm4MXam8UBHurAA9ayEyLYo9ltXl+2/mPHGF/ZT3iz20L/8R//Mbtb38XA nmZASYg9rX41XgzsLgZ4CjXlm9/8puvt7a3ZuM7OjuD04thU3pLx5vBQt8SbGfb598l3vdlR0hYP 0cRZ4Y0Q2WLOzv4EgSWOF1goeQ9Es7ru/ZVjsliL/E5A0trW6h1PHvi4/iGI6GfZ1+GIdiZIiNh7 4tFFNuDGQaUO57lsxxh+e3p6As2mi5hzqzt8+HBcXco2tnLo0KFwbrPRWBBY4MOCinhf0dvgAI+1 Pz6/YUulH7OVPCxWZzqMcZaxXU9fNrxlyLdzErwSOJourJ5P46S7u7v0xBnyjJM8LFZnx3B8WYXx 3rg3ubEsq0uBhb6BjirXPd4Y9aZYHeNxir4MJ3BjdvEGiVuzH+vv8b4ytq2f5mGhjjHHdFhL/gcf fBB237171z19+rTWodonBvYUA0pC7Cl1q7FiYPcyMD4+7q5evRoayBOpcRBqlX37WsO9mjhZ2VdO 4mAsLi64Y/19/jzlD5M4gPzxirpFLzsu4RVgvi19fcfi6tK2ewlgvTx7R7oJIhkAL+1t7T7BUwmE bV85ny1eP4fDE9nn5l6vE8GbTeYX5t0hH5C3sXKj5EIgwF/QT+Y1qcbTsWNp9DM4OBicdHsVpDWd RAi2QuCfIgmB3KGhoSDeODAs4CA5g4OeIogiaDl69KgLduFtIy6GDf1sNCbEv9vqNgESOpienn4r YQWWlPoxm5yZmVnXHMY86gj8+Su70Gbui0du1m7hidLf3182jHB+cDB7nffqyxcvXgQb6evrS4KF /oNNZl/zSF/GVgj6LRAuExAJIvoPfTb7mlT6FAkROKGflV2wR9qNfrLBv+mMlY4pCpygH7NRk4l+ sGMSM9bHbF/e55UrVwK//I4HZ6uIATFQYaD8EUVMiwExIAYSMPDgwQP3/PnzMON26dKluiQe84F9 a2ubGx5+FGbcKz9acaMjI36pbIcbHBis6zzbPQhH58TxE27OO3xjPplihYTE2Ni46/HB3GEfkKco BI1HfCJicnJynaM+OfksOMsnT530jlmaS8fg4FDQAwmmN0/uWHF8xyE+cfJkCkqCDO7txUkfGxtb k4mTPDo6GlYE1OOMrv1wGxsEUAQv2HrsHKOvZ8+euZOek42WCG9D/Lqf4qRjLw8fPgzcsBNHe8T3 H7g5c+ZMcOLX/aikL7SbYPfevXtrEsDw+PHjkOBLFbgQmNBuAqZ41pOgEv1gJxut0lprwDY3CLgJ 7p88eRISNHY6bJaEwOnTpwNnVl/WJ+Mbbyti2T+2gY1Q6E/Dw8MhWTUwMFCW+HXnBYPppzKuVHYT VGIr8FXP7Pa6k27xCwkGdIQ+SIBYAYf1H6sr+5MxxfoPeqGgJ66rNuaUjYHzYyvoh2QItmEFPvhO 30mRmEEuyZC8sZaxlySejTmGsdonmN97772w+1//9V+rHaZ6MbDnGGjxg0ztJzntOUrUYDEgBnYi A3/2Z3/mfv/3fz8EhH/+53/uTpw4UbUZyzMja/sI5HC0unxwt9/PwhA8zM8vuFOnTq4tN187uMQN hmKc4ifjT9zBnoN+hr/NzU7POp6/QLBAcJOqMPtFYMkMXe/h3nDrAzwRQA0ODiRLQtDe6ekZN+KT RPv8bB3BAcECK1fQT61grqWzx7V0FLtigwAKR5iVBuiDB6GSDCHpxUqWVIWZSZb2ohMSAdgOjjGz 8CRLUsxYWlvpLzdv3gzBA/L5jo5w0AksCSpSFRIxt2/fDkEK+iDwJ9iGk1SBJW1FH/fv3w/BP/pB H+iKWd5z586tPQsgBS8Ec+iHT2az6dvwQnDFX0pbISlz586doAuSVwRyBLzoJ1VgCefYBPohoQgn BN/0H2yEZEnqsfbGjRuh35AAIdhmXLH+Y89qSGErcIB+SDrABYkR7Ob8+fPJVlfRTvoP4yzJGMZ4 Ekdwgl6wlVRJVrAw1sIJtoqtgG1iYiJcC0mW1Nt/fu/3fs/hl3AN/e53v5t0PKIdKmKg2RjwvkGL khDNphXhEQNiYNMM4Mh+5zvfcf/+7/8enJS/+Iu/CCsiqp0oTkJwDLcaTE3NeAdwPjg8BFMpnb8Y J0ECAcvKyrIPbPcHZyVlIPcGy4p3vGaDc4yjRZCQ0vl7g6MSNOCEznsHnQdn1qOfMpIQYMIpx1nH 5gikwNIY/ZCgmQ6BAvLRD3gaUXDMSQDADQEDnDTKVug/2ArBHDOZ9dhKWZwRxBG8wA+6QUeNsBXk wwnJIQJucMBNI4oF2egJGyGh16ixloQZvMAPCaJaSc0yuUI+doK9MNbCSaP6Dwka+jL6oS+TRGuU fsAAFjCZfhrRf9A9+iF5h3zsZLP953vf+577rd/6rZCs/uM//uMNH5xdpr3p3GKgGRjwfaml/HfP NUNLhUEMiIFdzQBB4a1bt0Ibuf9ys05T5fkQaW532EgRzPYwW9L4UglsCVgaXXDMWbbcDIXggOcy NEMhwOav0QXHnFnCZij0n1TPF9iovSQeGpUYirGhHwJb/hpdCGybY3yrLLffbDBZBn/op1n6MmNt s+iHvsxqnWYoXAe3cy3kNZ3YPqsLr1+/riREMyhVGBrOQJobexveTAEQA2JgNzPAPc8s3aRs9GrO 3cyD2iYGxIAYEANiQAw0FwPcmmYJL5IQKmJADDinJISsQAyIgR3PAPfHW+GeURUxIAbEgBgQA2JA DDQDA6xCsufS3PMPz+VWIBUxsNcZUBJir1uA2i8GdgEDPDjKCg/0UhEDYkAMiAExIAbEQDMwwLNY 7GHZPAg7+1rWZsAoDGIgNQNKQqRmXPLEgBgonAEu6hTurbULfeFCdEIxIAbEgBgQA2JADGyBAZ4L QeGtLDzkUkUM7HUGlITY6xag9ouBXcAAF3UKyx15mJWKGBADYkAMiAExIAaahYFTp04FKLzt49Gj R80CSzjEQMMYUBKiYdRLsBgQA0UwwLJG3j9PaZanehfRLp1DDIgBMSAGxIAY2B0M8HBKKw8fPrRN fYqBPcuAkhB7VvVquBjYHQzwZoyXL1+GxjTLaxx3B7NqhRgQA2JADIgBMVAEA/ErcicmJoo4pc4h BnY0A0pC7Gj1CbwYEAMjIyNufn4+EKGVELIHMSAGxIAYEANioNkYiP2T8fHxZoMnPGIgOQNKQiSn XALFgBgokoGZmZm1JER/f3+Rp9a5xIAYEANiQAyIATGwbQZ6enrWzjE5Obm2rQ0xsFcZUBJir2pe 7RYDu4SBqakpx4OeKPFyx13SPDVDDIgBMSAGxIAY2OEMxEmI6enpHd4awRcD22egbfun0BnEgBgQ A41jIL6Y83aMzZalxUU3PDriXsy+8K/49K/5dPvCWzYGBwdcW3v7Zk+3reN5tsXo6Kibm5tz+/aR I25xAwP97siRwx5b2pzx+Ni4m5is3LcKlnbPxdDQkOvu7t5WGzf740WvHziZnpp2rW2tzq0419N7 yPUfO5b8TSgvXrxw9+/fD+94573vFDjp6+tzra0eW6KysrLiuKeYJ6zzWlr0g3xeT5s6EWdPegcP OPjbv3+/O336tOvs7EzESEUMt2XdvXs3PCMGHHCDbtBRSv2AhnEJW1laWgqy0RlPx+e5NeBKVZBL /+HZOWzDC28QAstWxsvt4GZcgxP6EfoAz9GjR93x48ed9aftnH8zvyV5Tf/BZsCCTrATlsyn1A+Y h4eHw8OV4QPZ6OfMmTPuwIEDm2nSto9lrOV113ADDmyFwBlbYfxPWXiFJQ9vfPXq1Vv6AVfKwq0T 9CHrP9gL4xvcbMZW4JC/hYWFwHHKNkiWGGhGBpSEaEatCJMYEAN1M8DtGFY261TjVNzxQQvO8blz 50KAzds2CGRmX8y6S5cuegcozTAJhlu3bvsArstdvnw5OF7T3hm8e+9+CKrs9V7W1jI/cbqGR4aD 80kQRyCFo3zz5s2ALV0iYiU4xTikvGOd4Bae7ty+42a93q9cubIpJ3A7nCH366+/Dhg+/PDDoB+c 9WvXroWkBEFDqkLAf+PGjeAIE8BRCKjA8u677yZLRNB/CCp5Lsv777/vent7g36wk6+++sp99NFH yYIX9EP7sVU46OrqComA69evh6CXPrWZgGE7uiQBQfu5PQy7IGgiCQCWCxcuuMHBwe2cflO/pd8S zJ0/f94nNAdCAMR38H3zm98MPG3qhFs8mMCL9mMz9Fv6MskI6z+XLl1Kph/GE/oy+gGL6efWrVvB fkhGpCr0HQJ/9AMe7Jfrz9WrV90777zjDh06lAQKekE/BP3IJQFCUpw6rrH071SJIq7BtB8MjLXI 5fYFxhVsBq5SFcZa5NKPSfJa0pX+Y2PeZrBw7eS6wXjFX+pE7Waw6lgxUDYDadOJZbdG5xcDYmDP MRAnITbrsOFg4BAcHzoeZgVxRnESTp06HRyvp0/TPMEaB5DAYGVl2Z08eTIEbmA57GdNjx497MbG xoJzmEK5OEZgYZaHWUFwMHsDLg/TPXwATr+RoMD/5LNnXj9DwSEFC8HLCY+FQILZqVTl3r17YdaU IJKZSmbDmMUloCQQB0+KghN8586dwAezcTjo/LENN7dv306mH4JtglwcdGb4TT8ElAQz2FGqElbL eDwkG+jDof8cPhx4Ian2zNtRimL6Qb7ZCvohWUSSFFuhj6UoBGwkp9ANgTWYCHrQF7PeBFep+jL6 Yaw9e/bsWl9mjIGXx48fhyRNCk4I8tEBXICFsY2+DD9cP0gApNIPfQQsjCPIBwdjC/0H/bAvVeEa Q6CPbaAX+MFesWH0lupBitgj4xufTAzAB1i4FvHHeJNKP9ZHSIZw/UM/2Asc8YmtbLb/2OoW7JD+ qSIG9jIDSkLsZe2r7WJgFzDArImVzayEwHkgCYFjcdA7XXHp7t7v69u84/XEV5cfcDNLSEDH7C2B ZFx6eg4FRyeVE/jMO6JwczjcAtKyBqW9vc0Hd/vdzOyMnyF7tVZf3obXj09CtHoH9EDmNpsDB/YH 5xSnebNO4Fbwoh9mstEPf3FhpQgz7E+fPo2rS9tGDk44qw7igh0TPBDYEDSkKAQu8J/FQhKA4MGw psBCIEtgAAdxIbAEoy2njveVsU0yij9LysQywILuUtkKNovtgiUu6AYs2Ek8fsbHFLlNYgb90E+y tmIPEwZrioJuSEhhJySH4kJfJvB8/vx5XF3aNm1GXvYWKvqy6SdFwE3/AAtys/ohaIanVNcf2gv/ XAcZR+ICT2Bl3ElR8A+s/5AIsWL6YaXIZpObdv0gCcFYrSIG9jIDb3rVXmZBbRcDYmDHMmCv56QB NstQT2NwdnDAcS5aW9cPhVY3N/fazc8t1HO6bR3DjAiOOs5edsm4OcqpZk1mVmf0O9o7Mm1qCUEe lS9flj+Dg/P36nXlfmCcvri07msNepufX/D6qbyeNd5f9DbBGs4vOKrpx44pWnb2fOb0ml3E+wnC wWnHxPvK2MZJpxDUZgv46GM46mUX+gb2QvuzxeoIPnH8yy4kE9FBHie29JpjUhRLRuVhoQ6cKcYV gi3GaXSR7T+GLRUnpp/smII+rE8Zb2XqKO6nJjeWRx3HYLdlF/RDP61c99aPtVbHtTJFQgR7pJ/m 2Yr1ZRt3yubF7MBsNJZHHTjjlZjx/mrbZnf8NgWf1XCoXgw0AwPrPe9mQCQMYkAMiIFNMBAnIcxJ qefnBC3MQlGyjjF1+1YfBPnKJyLKLjYjgsOXLdSBD7zLy+UGUfCBHEoeFkvWpHCeFhYWvZO3HNqe 1U+L5wT9wMf8QvlJCNOPOZCxjkw/2GFsi/ExRW7bzHU1LMhKEbhY/0E3Wf2AwfClSEJYe/NtthJU oZsUSQhrr7UfLqxYnenQ6sv6NLvNC3INi+EtCwPnBQcBdR4Os5947CkbC+e39seyrM54i/cVvU3S 2cZRkxvLsLoUSSL6Bv3ZdBHjsDr6Tgpbsb5s7Y+xxGNt2X0Ze7V+mme3VmfHxDhrbdvvYv3XOl77 xMBuZuBtj3c3t1ZtEwNiYNcxYEEzDctzXKo1GAcQR8OnIHxA++a2g7XjV6sWF8tfCWFtyEHhHUMQ tQSsS4uVV5GuYSx4A8eIwH9V6Ftnt2DT8L51QIEVJBh4RoY5wW+depWsFFhqJavABUac4rIdY2RZ e00X1FmxOvBWbNv2FP9pOKqdOcZS7Zii6i35k5eEMBzI2ghzEXgMS965DEsKHMivZbfGVQos1i+s /VlurD4FllqcGK5aOrRjtvtZGWsrSWVrf3xOq0uBJR4vTG4WC3hT6Kee9oLF9BjjLHIbGfxRqnHC vnrwcpwV81EYn8tug8nUpxhoVgaUhGhWzQiXGBADdTFgTgCOgl3g6/nhG8fYH52J/mOnY5mgvOSy tOrsVAv+wYfTsuyD8jILMkLg74XEHLyRWSHKnLM39cVvLS9XfxaHJWaQanosHsGbM5qMPE6sLnZa 3/yy+K1a3BuWih6r81cEKsNhMqud07irtr+IesOSd64YX63j8n67lTprbyzXzmN1dozVl/EZ24DJ zZOTAstGMgzfRsfl4d9sndmAyYx/b3Uxd/H+IrdjGSY37/ypOcnDYnXGXR7OouqsvSYzPi91/MXc xfuL3EaGtbcaFuTZMfXKjldCpEjq1ItLx4mBRjCgJEQjWJdMMSAGCmEg6yjkOQvVBHkfo2mKd6ua BouAvM3AZuzq7V+nr6FfNEvZadw1C2+pcEg/qZjeGXJqjR219hXVut1ujzZREvsuRXGn84iBncaA khA7TWPCKwbEwBoDOCz82YzCZhwY3vZAIV7Lxmyxs2XLlteElrCxL/PgxbdEeIzW1rf2FVixbx+3 puwLKZGYAxPh10mEzSScrD4Lw2THnxV9VbCYUxfvL3rb2pvLyarxcIwdV7T8+Hy1ZBg+jtlMX4jP X++24TCZ1X5nx1XbX0R9LRkxvlrHFYGDc9hMZyzXzm11KWw2Hi9MruHg0+pScLJRew3LRsfF+Le6 be01mfF5rI5jyu4/G+nHcKXgxGRY+022fVq9HWf1ZXyaDJMZy6COv5i7eH+R27ENVMOCPLOnemVz PP4KKyg2+9t6Zeg4MbBTGFASYqdoSjjFgBh4iwFzDuxJ2vb9rQNzKnhSfUt4+GTFsckeYufKezJ2 9tjtfu/s6AynMJnx+ZZDkFtxvCzAifcXuc1rSYMTuOrsZc+9snrbSLt/cnvZhbbibK742zKyvPA9 0OJBbOZhpFvFbDZQbekteMBrDvRW5dTzu1pYDB/HlB1EGY6sbqwNMRarK+vTXmtrMmM5hg8+UtiK vYLP5MZYDJ9xF+8rY9vkmNxYhtXZMfG+ordNhi21z57fuLLjsvuL/G42YDLjc1tdChwEoIbFdBFj sTqzp3hf0dvgsPHCOIhlUAdee7tLvK/obWuvtT8+v2GLuYv3F7mNDLODPCxWZ8dsRjZ8c347x2Z+ q2PFwG5iQEmI3aRNtUUM7GEGzEGplwKcB1ZD8DsLru231JmDYE6R7Svjk/eh4wTmOek8kwI84MVx KbNw/q79XUHE4uLbb+Kwui7/DveyC44aQT3PwSARERd0w4Mr2f/2q0TjI4vZtiAX/WTtzOrQz1Yc 0s0iNHvMe6iZ1W3mVbWblW/Hk3SxoMTk2j4+4QWbNu7ifUVvHzx4MJwyD4fVoZuyk3iAsPaa3Lit VmfHxPvK2GZcodhKsViG1aXAggxswdof47D+Q3/fS/phrLX+AwfZYlyZDrP7i/wODvpHfN2z81fG 2uUw1qawFcauqtdCP+6DkTGw7Gsh7Tfura8YJ3xaXQpOYrnaFgO7iYFyPdrdxJTaIgbEQFMykA0K 6wXZ0dHucHhwshYzTmCo80F4d/eBtdmqes+7leO6vZPOKgQcm2xCxJydnkM9Wzn1pn9jAd1c5tWk 8MxDQHEQew6Wj4UEA04giY+FzBtKcNBx3HGeO7s6Nt3Gzf6gt7c3OL3oAtuIi71lhWPgpuxy7Nix IMJe7xfLM/0cPXo0ri5t27DYw2FNUKX/LIaA3Bx521fGJ0EJcsCRHQ+Mp56enhBIlSE/Pid2QMl7 zSOv88NG7Jj4d2VsHzp0KJw2Dwt19DF4KbuQYKj05cW3EhGG7ciRI2XDCOc/fPhw6MtmF7FQq+OY sgt20NfXF8Rk+w+V1BFomw7LxMM4ih3Qb+16Y/KsL3NdSJFkRQ7JKNqfN9aCq7+/3+CV+mncm43G wujL9J/t2Ep2rIrPr20xsBcYUBJiL2hZbRQDYiCHgRZHEIWjMzExsW7/s2fP/PcVd/z40Lr6sr60 eSf9yJHDPnB57Z5PT60T8+zZZHDKBgcG1tWX9QUc7e0dgZN4hg4HHWeMALezq3L7SFkY7LxDQ0Ne CytucnLSqsLn9PS0T0wsuqGh4/57+YE/zubJkycdHNh77AGC7YyNjQVnFKwpCsEaAcPz58/XrZx5 +fKlm5qaCkFLisCSth4/fjwESuPj4+uaTv+BqwFvsykCF4TDP/rIYuE7+jt16tQ6jGV9YWaU4BL9 EKhYIXEGLwRZqZJEg4ODYdb46dOn65IzL168CH0Z/dhsvOEs49P6D4E3WOLy+PHjyvjmsaYoJJ5p N/0l1o/1ZZIlqfQDDuyF6w/yraCfmZkZh/5SrA5BLv2H8T7bf8x2Uo1vJKy4LqOfynXYWHHBduAD 3lIUxlrsBf3EyRnGNq5B2Mlmx1olHlJoTjJ2CgNKQuwUTQmnGBADhTPATAeO3tjYuHvinWNm/3F8 RkcfB0doO7McmwVLkLTf3wox8mjEzc7MujkfwIyMjHhn7LU7dfJUSAxs9pxbOb6trd2dPXsmzEQN Dw8HR/3li5fu/r37wWEmGE9VenoOOpIvExOTwQF97Z0/9PNk/Inr8w4gCZNUhXYzg33r1q0Q7JOQ efToUXCWL168mCSYo60EcufPnw/NBgtJEYKWe/fuhYTApUuXkqzIAAAB1Llz54KTblyQMAILgfiJ EycCzhT/kBChL9+/f989efIk2C2YsJfTp0+H4D8FDmavL1y4ELgxW0E/d+/eDasA0F2qxAwJBmwT W0Unlqi6efNmsOUzZ86koCTIQDfMYD98+HBNPyQgCHLBkWp1CP0HeyDYvXHjRggm4QV+SASgO5Im KQoBNf3H7AMcJB+wG65N4ExVuNYxxnHNQS/YDME3+iIBYas2UuChj5AMQickVkkWPXjwIPDEvlSJ Gfoy4ymJA/ovemK8vX37drAf9IM9qYgBMbA1Blp851p/s+3WzqNfiQExIAaSM8Dw9cknn7gvv/wy BGCff/55cJg2ArI8M7J2COcgcOKPmSAcHGY4mAVJ7WAwW8pMFI4ouAgimPVhNiZ1wSHGGWVZLDzg FPf7GSpWbaQs8EAgOTE54RYX0E9r0M/Ro+ineh69pbPHtXQUu9ScIIXAFseYbYJw9JMyWWXcEySQ JMIxRj8EcQR69swIOy7FZyVxNxpmCwng6D8ELinu247bh04IomyVCFwQ+Kaa2Y6xMFuKraAfCn2Y pAw2k7pgr6Ojo2F1CjphbCNpkyrYtvbSl7FZ04+Nb43SD1gIKsHFChU4IfhNXbL6YTzBVlLrh/5D UohrENdCEjX0H1YmNOJaiH5YcWBjLWOK3SKRUkdxwtmuhdjKVlYRffrpp+6LL74I4/Rnn33mPv74 45RNkSwx0DQM+L7UoiRE06hDQMSAGNgsAziP201CxDI5X2Vio/GzGxUszYEDjnC+Gl02w0kZSYi4 /ZvBEv+u6G1wUHaaformIXs+6SfLyJvXcspW3nDTbP0HZM2in2bCsls4URLiTd/T1t5mwPfplvLf s7a3OVbrxYAY2EEMNIOjY3Q1C5ZmwQEvwmLW8eZTnLzhIt5qFl6aBQfcCEtsIZVtcfI2J9SIl7d5 aSZO3kanGjGw8xiovpZ157VFiMWAGBADYkAMiAExIAbEgBgQA2JADIiBJmZASYgmVo6giQExIAbE gBgQA2JADIgBMSAGxIAY2E0MKAmxm7SptogBMSAGxIAYEANiQAyIATEgBsSAGGhiBpSEaGLlCJoY EAP1M6D7NevnSkeKATEgBsSAGBADjWNAPkvjuJfk5mBASYjm0INQiAExsEUG7ELOp21v8VT6mRgQ A2JADIgBMSAGSmHAfBQ+U7/CuJQG6aRiYBsM6O0Y2yBPPxUDYqDxDPCO9bGxsfBO88XFxcYDEgIx IAbEgBgQA2JADGQYmJycDP7K/v373YsXLzJ79VUM7C0GlITYW/pWa8XArmOgp6fHHTt2zLW2toa/ XddANUgMiAExIAbEgBjY8QwcPnw4+CtdXV2OPxUxsJcZ0O0Ye1n7arsY2AUMsKSRBISWNu4CZaoJ YkAMiAExIAZ2KQPmr+Cz2K0Zu7SpapYY2JABJSE2pEgHiAEx0MwMrKysNDM8YRMDYkAMiAExIAbE wDoG5Luso0Nf9iADuh1jDypdTRYDYmA9AzgDPE+CT5upaNQsxfLysltaWgoADct6tKm+wcmSAw9c tLUxc9OYvDV6WfL6WfJY7LabRukHLPPz80EJYGlra9xlFN0sLCys6qetoauB6D/YLTbb3t6eykjf ktNM+oEPe05NR0dHQ2c+sRPsBf1gs+o/LujGxlpsFm4aUcL45m3F+k+j9YPNYiuNHmttfEMnjR5r bXyrXAu3PtaiaxUxIAYqDDTOe5IGxIAYEANNwMCCDygfDT9yr169dp2dHSHwbm9rd6fPnE4eTE1P TbvRx6OBlTYf4M55bNxDevz4UNIEAM7f2NhjNzExuXbfKkHMqVOnHM/gSFlev37tRkcfu1cvX7rO rk63uLDo2nzAcOrUSa+vzpRQ3PPnz92DBw+CTIJKsKGf06dPByc5FZiKfsbcyMiI4wFnOMZzc3Nu aGgo/KXCgRzs4v79+25mZibYigXdFy5ccAcOHEgJxb30NnLr1q0gk6AS/Rw8eNCdP38+ebLoyZMn 7tGjR2EMIYB69epV0M3x48eTJgCwFWyWB+JxDzqBLoHQ2bNnXW9vb3L93LlzJwT/YIGTQ4cOBSyp k3kTExPu4cOHa8EttjI4OOjH2rT6QQHoB3vp7u4OwT86O3fuXOAmpYIYQ+7duxcemMi4Qt+mH128 eNEx3qUsz549C7wwtllfPnr0aBhrUyaL0MXjx4/9NWh0bazFVhjz+/r6kvbllPxLlhhIwYCSEClY lgwxIAaakgECptu377jllWV38YJ3tEISYtHd8XU3btxwV65cSZaIePnipbtz9647cqTXB9inw6wc QdXNm7dCgHn+/DnPYYv/K7usBIfryZOnPnjDEe7xQYsLjtjNmzeDQ5oqeMEBJMBd8isyLl6qOMLo 7Nat214/N917772bLLgkyMYmcIQJEAgsp6en3dWrV4OzfunSpWQOKckHAiiCA3OEcZIJ8HDaCaZS FILZ27dvh+TMRx99FIKoin5uua+++spRRzCTohAYoAvkXb58OfRb+s/169fD3/vvv59MP0+fPvX9 9mawk4GBgWArBL3UwdnJkydTUBICWgJcgqh33303JB2QT5/6+uuv3QcffJAsqUiAi0z0gy4ILKlD Z/Srd955J1kij4TMtWvXQvKDpANBLW9Zwmax3zNnziTRD7og6B8fHw82e+TIkSCXOrCgM6srGxDt hhPGXPRDgpckBHU/+tGP3IcffpgsETE7O7s21pJAZKzlTRLgwGa4Lqcq9B36C+M7Yy2FOnjBZvv7 +1NBkRwxsOsYaMzas11HoxokBsTATmSAwGD2xWyYoWSW3WZdCBIIaggmUhScUVZj4AwPDgwFpwss zCQfPXrETXineXY2zeu85l7P+dUYj33AcijMxHELBrgsmBoeHg6BVApe4H/GO6SDQ4PBKTb9sDKE WyJGR0dSwAgyCPpx0JkBs+XSJGMIYnBKCWJSFIIFsDDDbzpBP+Cgjn22xLxsPMxWoiNWyGCrph8C BzCQLMG2UxTskpn1cz5BZLc+gOnEiRMOnPT1FAUbIfAnyEYnZiu8wYdZf/QDzhSFYI7kFMETK3bQ D7aCDaMXklbgTVHQD0khZJt+WA2BftANgXiKQv9hhQpBNiuHCHDhBX7gCf0w9qco8IF+SDSQ3AQH f6xSARfBb6rCa65JqqIf9AIO9HTO9ydwsj9FwS7v+mQ88rkOW/9hbCPgx05SvdoSWyEhhGz6L32H P5K8JNOwo1RjbQruJUMMpGZASYjUjEueGBADTcEAzjczYm2tbe7ggYPrMHXt7wpO4Jh3eFIEUczu EDB0+UQIyZC4hNsfvGPGct0UJQRrXh4BNo6gFZxBlgvjkLJqo/yyEgJcbkvZv797nbgDB7qDg/zs +VSSIIqgBFshcMkuSyaAgKdUCSuccJxjAtq44BxjKwS4qRIiBCb0DwK4uBDEEIRjs9h22QUM8IJu sreAwBP7UyVECOToIxb0x20HC7PLqfoyNomtZFcu0ZexFbCmSIjQZrBgowR0cSH4pqTihDbzh53A Q1zQmV0X4vqytrFZgths/4En8HFNSJEQoc3wb3YRt5dgm4QI+ktxLcQeWXWGXP7iAk/W1+P6srbh BP1kr4XwQV8mGcIteipiQAxsjQElIbbGm34lBsTADmcA544AqWWfnx1sXT8UElTiCC565zmFE4gz gyOIExgH/lBMkoTyIgT+5c4q4+Dh+FKyDnpcN/uyckw4sKR/5ucW3PzcvNdDq3eC1+sn1Hm98XwI e0hkSTDCaQnq4cZmTWNZxhOOMzosu5jTa3JjeVZHwqTsAh8mx+TGMqkj+DR7ivcVvQ33BNvVcCCP PsYxZRcCXLjJJquQa3UpOEGe2YrJpc4KSSJKCiw21iIzO74ZNnDAW9nF9FPLVlIk8Wir6ScPC1xx DHjLLgT+/KEbxri4WB3XyhQJK+yAwB9OsrZiPLGyKYWtmB3kPXsIuwUDY4+KGBADW2NgvWe3tXPo V2JADIiBHccAAZItpdyX89YHq0sR5HILBCXrdIU6H2xzS8TSUuXtA+HAkv6BjwX//AUKSZhssWSA 4c3uL/L7Iu31QT2LMbJv5YAn6paXl5IkIWw2P4+TChafEPEBLjZVdjF7zMNidczEl+2k01aSLtb+ bLsNS6okHvJNZozF6tBPiiSEBWomNw+L2VO8r4xtk5MNLJFldSn0UwsHWLChMPYk6D+GJU8/xokd A7ayStzePCxWl0I/jCngyevLVsf+FFgYuyimi5h/wxJfu+P9RW4zftYaaw1fCk6KbJfOJQaaiYG3 vcxmQicsYkAMiIGSGMCRwdGoODbrhYSbEFbvRFhYTBBYLlZe+QiWbLEqAr6lpXJn2pFBYJ+HA1yW DOB1mWUXnN4V/8DQPP0E2aafBIGLBa95vFTwVZIQdlyZ3FiioxoWZKfAsZEMw2d4U3BigVssy3Ck 4sUCl1iu4TF8dozVl/VJH6LUwpJSP3k4YnwpsJjd5mGxuhQ4uPbU0o9hSRHkVsZauxbmXYNaQsIx BS/19I3KtbBi28HAS/gnlmH9NhZj+qkHb/y7eNvOEddpWwzsJQaUhNhL2lZbxcAuZmCzF3RzAAMl Wb/LIn+/c2W5/CXCyxslF1bxlY0Fx5g/Si0+lxJwYjgCmMw/FfVUSMFZLLuYjFqccEwtzEVgjPWT d74YX9lY6uEEjGXj2EhGSk7Asm5coSKnGHc5uwqrim0l5iArIBWWrNz4u+Grh7v4d1vZtvaazLxz xNzl7S+iLpZRC4vhLUJmtXPEMvKwWF2KvmxYTGaMmTr+Yu7i/UVubyTD8G10XBaT/Y76eDt7nL6L gb3AgJIQe0HLaqMY2MUM2IWc2QrbLrS52QRFoSevnKxu3CVj8S5eXa3b528RKbvUzUmdmLeDtx4s 9RyzHQz8NoWM7WLM/r6ZMKfAkjdr2ghO6m1rvcdl21DG93q4267cZmpvPW1JwUmMI0WiIZaX3a5H P/Uckz1v0d9jnjaDx47lM7Vui+ZA5xMD22VASYjtMqjfiwEx0DAGuJDjDLAk0pbZ1gvGHnLFxP/q 5P/aT2MHwx4MubazhA27vzTv1AGbx5jCaQkP6fTPWmAtRMyB4bK6FM4TMsxhM/n2WdFXZcVGW9v6 B6nZMUV+mn6s/fG5qeMPvKl4QX41LOyrxR37iyhv+k9FD9lzGj7jLru/yO+GJe+choN9KfRjWGK5 hsvqUnASt9dmlw0Hnymx1OIkNRbjfiNOqo09MYfb2Y7Hc9NF3vkMb96+oupMRjUcVm96LEpu3nlM hsmMj6HO6svuy5zfZJjMGIttG3f2faNP7G4r/spG59V+MbATGVASYidqTZjFgBgIDOAc4MzxpGoc hlrOQpayyrvQGQLfODbxMXYue4p8vK/obXv6dt7tFh5dwMjzGDbr8GwWJ+dvJaiPnL34HCurtz6k 4AQs6HTZ3/phujAsfPf/B92nwGL6qRW4gMMcaMNZxqe9TSAPi9VxTNlBlPFe0cXbiQjTmXFXBhd2 Tvoyxdpv9XEdfKTQj2Gx9udhMe7ifWVsm600GovhyNMP7TZ8dlwZXNg5a3GfEgdjm9ljHi9WZ/Zk +Mv4hBMbL4wDk8N3/hiPU+jHXstp7TccfBq2mLt4f5HbyDBbycNidZvlxPwV+Exx+1GRnOhcYqBo BpSEKJpRnU8MiIGGMGAOSr3CcR46V1+zZQ6F/ZZz8RBIHIYUTiDvhEcWb4TIliX/tgrwdHV1rs3M ZI8p6juO1/7VgI7Xk2bLwuoDKc1RzO4v8jv6wVFDN1lnje88QLPVv75zs07gVjB2d3eHn7HaJmtn YKGOYDsFloMHD65hybbFVgPxDvuyC7oxXrL6QTZYsKcU/Yf2hv6T88BUe5geOCzoK5Mb+jIl74F1 Vmc6LBMH5zYseW97SIkF7tGPyYzbjX6s/2BTZRfjxOwilmd1dky8r+jtuP9Yv41lWF0KW2E8R0fo IduXrY6+Y/09xln0Nu2t1pdtrEU/jC1lF7ODPLu1PpVCP2W3U+cXA41ioPxe3KiWSa4YEAN7ggGc pK0UZjkO9hwMQe7iwvrg3wLfnp6eJIELTmB7e4d30t+8NtTaZA5Qb2+vVZX62bMawL56/XqdHHgG CysyDnleyi44mTi9OJ4WHJhM6nDSuw90+8C/3apL+8TRxAlHZtZJ5+n1cJNKP8eOHQvtNCc4bjRY cOCPHDkSV5e2PTAwkIvFdIb+zJEvDYQ/MX2Zvop9ZscD6z+HDx8ufSURbSQhgu3aqwbjdqMz9AOW FMVs0l4bGssEXxgDV5Na8b6it00O46rpw2QYT6lsFv3Ql/PeOmF9yngzjGV90perJWfQGXaUIqGI fmgz+smOtdQx7hlvZXFh52XMIJkLDkvE2D7TmY07Vl/Wp9mB2WgsBywkkrZjt9mxKj6/tsXAXmBA SYi9oGW1UQyIgVwG+vv7g+P15OmTdfsnJp767ytuaGhoXX1ZX3Bm+vqOulevX7npmel1YiYnJoJT BtYUhQCpq7PLTXi5ccCNI4bjNTDY79q805qinDhxwotZCVhiec+ePQvYBgcHfXX5D8lEP6dPnw7t n55+ox8c9MePH4dgroIlRlnONsE2zvHk5OQ6J312dtZNTU0FpzjFjCWtwyYJ6EZGRtY19unTpyGI oP+wP0WBf5z6LJaxsbGAoWJL5SMhoUhwiZ3EwQsBFXZL/0qVhCBYwxboy9iqFeyEgBv9EICWXeg/ p06dCgH3kyfrx1r6DzaSaqwlKYbdoptYP4x1o6OjjoRjKv1gJ6xAgJM44DbbOX78eOCsbP1wfmTB AfqIy/j4ePiaSj/YAnbLePb8+fMYSuCJFWd9fX3r6sv6gh2QfKH/xMkzEkQzMzOhn6O/zRQlHjbD lo7d7Qyk8Q52O4tqnxgQAzuSARxSgksCF5wfHFCcn7GxcXfy5Mkwu5qqYQRJODcP7j90vLKT4CAE c/52jHNnzyWZxaWtBAznL5x3d+7ccXfv3g0OIbNSOOg4ZSe8s5qqEEChBxxjdIR+Xvjg4cn4k1Bv M1Up8OCkYxu3bt1y586dC4khHHSClkXOQQAAQABJREFUuffffz98T4GDmdNLly6569evu6tXr4bg Dsf24cOHjgD48uXLyQIXHHDkwQm2wqwggd3w8HAIalIFLvBOEoIk2aNHj8IMMlyQqCGQuXjxYpKl 5KZ/5IEFHTG+MKNN/+GTfakSM4wh6OfatWsBCzbMGAMWAnESA6kKATd9GTul0LfRDX8XLlxIOtbS f9HPV199FWTTp+AEvq5cuZJsrMUekIed3LhxI/QZgt0HDx4k1w9jK7Z57969YJ+MrSRD6Mtnz55N ttIL20Ae4wjXIJJn6IVkItxgz1yjUhTsAnlff/116EP0FxI1jDEkg8+fP59srE3RXskQA6kZaPHO y9bWMqdGKnliQAyIgQwDDF+ffPKJ+/LLL4OD//nnn9c1o7Y882bmlnMQXDJLSbCNw0MwhVOGE5Ky MBs2NTXtnb+p4Hx17e9yx/qOhecNpMSBrLnXc25iciI46/DAbRpHPS84zinLG/1Mev0sBgcZ/eAE 1tJPS6ff31HsbSM4xMxaMpOMM0qga7PNKTlBFkEUSSpslwIfYMF+UxeCFQJ+AlwChKNHj4ZZwtQ4 0A9BLVjoy8yaEmzDTeqCfPTDjCm4CLpJAjRCP+gFuyWwo/8yu4utNKIvoxsba9EPyYkUtxxk9Y9+ mOHGdk0/JLLAlLq8ePEi2Cyf6IQEAPqpNb6VgREesFfsloAfW6X/oJ/UWLgWWhIRXIy16GezKw+K 4Imxlv6DfsK1cHWs3Uoy8dNPP3VffPFFaMdnn33mPv744yIg6hxiYMcx4PtSi1ZC7Di1CbAYEANF MmBORSMClWw7cGq4LYO/RpdO/yDMVEvYa7W1mfRDgIAjzF+jC854ypnsWu0lSGlEIJnFhH5IgPDX 6EIAR9KBv0YXArgzZ840GkYI4FhKn2o5fa0Gox9W6qRcrVMNDyvyUjw7pZp8q7cESMoVZiY7+8m1 kEQMf40ujLWsaFIRA2KgWAbSTmkVi11nEwNiQAyIATEgBsSAGBADYkAMiAExIAZ2EANKQuwgZQmq GBADYkAMiAExIAbEgBgQA2JADIiBncyAkhA7WXvCLgbEgBgQA2JADIgBMSAGxIAYEANiYAcxoCTE DlKWoIoBMSAGxIAYEANiQAyIATEgBsSAGNjJDCgJsZO1J+xiQAysMcADDFXEgBgQA2JADIgBMdCM DMR+Cg8CVREDe5kB9YC9rH21XQzsAgbsos4F3bZ3QbPUBDEgBsSAGBADYmAXMoCvIn9lFypWTdoU A3pF56bo0sFiQAw0EwNcxJ8/f+7GxsbCO815t7iKGBADYkAMiAExIAaajYGJiYngr/Da3FevXjUb POERA0kZUBIiKd0SJgbEQJEMrKysuMOHD7vBwUHHSojW1tYiT69ziQExIAbEgBgQA2KgEAb6+vqC v9LV1eU6OzsLOadOIgZ2KgO6HWOnak64xYAYEANiQAyIATEgBsSAGBADYkAM7DAGlITYYQoTXDEg BtYzwGoIFTEgBsSAGBADYkAMNDMDsb8SbzczZmETA2UxoNsxymJW5xUDYmDHMLC8vOxmX7xwC/ML folkhztw4EDDHhrFcy1mZmYcuZWurk7X3d3dIB5X3MuXr9zr16/DbS5w0tbWmEsGzhqcLCx4/XR0 ugMHu71+GpNDB8PU1JTXz0rQDbw0qnBPMbxwG1JPT4/r6OhoFJTAydzcXMBw6NChcHtUI8DQf9DP 0tJS6Mf0n0Y9AA79vPDjCgX9NGr5NbY6Ozsb7kHHRrDZ9vb2RqjHoZ/p6enwiW4aOdYytsEL9gEW 7tNvREE/2MnLly9D/2mkfrgWMqbMz88He8VuG9V/sBWe+URBN40ea7EVbvts9FjbCBuVTDFQBgON 8SjLaInOKQbEgBjYAgM4fw8ePPSB3D63v3u/d3qeeQd5yZ08edIdPJgywFxxT5488Q+tGg8OcXtH uxsfH/POV5c7fvxE0gAT5++h5+SVd9IP9hx0K94xHX407PoH+l1//zHPcrrXoeKYg6VlX0twRNHP 8uMVNzQ0FJzBLah8yz8ZHx93jx49WgviRkdHA6azZ88m1Q+JkJGREcdDznCIcYwfPHjguN/49OnT SYMGArnbt2+vBf1gu3v3rjtz5kzAs2WyN/lDAjkCFmRzvzV/6IfE2fnz55Mm8wjkhoeHQ38mcALD /fv33dGjR4N+UibzCCbv3LnjSBAdPHgwBP8EmdgJz9JJWZ49exbslPYT9PNAYWz34sWLQV+psGAr 9OOnT5+GvkwSj/7D84Xoy2BKVRhrb968GfRDX8Z20NepU6fcwMBAKhhBDnZBX4YPbIXrEck8OCGx mLLQd/mj/5A4oz9hM4wrKZN56IMxhbGlt7c3jK30ZXSDj9CoBE1KXUiWGCiLASUhymJW5xUDYqDp GcABxOHDqbnkHeF93vnCQaXu1q2b7v33308WXE5Pz3hH+JE7fmLInfBJB8rCwry7du26n5W67y5f vuRrUgT/K+7hw4duys9WvvvuO2vBAQ77Pe98EUAcOXIk4Cv7HxxgnOKO9o7Q/srqhxV3+9YdX3/H vffeu8kcUpzQW7duBccTp5xCAP79738/fH744YfJHFISEOjoG9/4xlpwQIKEYAanmAAzRaGv3Lhx IwRQ3/rWt4JtWP+5du2a++ijj9bwlY2HWUraTyB5+fLlwAP9+6uvvnLXr18PXKV6cO3jx49D0uHK lStrgSTJzh/84AchwCWQShG8WECL7WIrBHEUAjpsmdUQJEZSFJKJ6IGx49KlSyHQBd+XX37px7hr 7oMPPki2OoPkx71799w777yzTj9gYcwhKZJCP/QV9IDtvvfeeyHwRxf0b/oV+1Mliugr6IHkHdc9 +gryqbt69ar79re/nexaiL0y7tNP+KOQTGOsJZnGWJuqkJyiP9N/SBJRJicnAyckq06cqFyrU+GR HDGwmxhIl+7dTaypLWJADOwKBnBGcW6YVScBQcH5PH58KDijBHYpCs7e8PCIdwA7Xd/RvjWR7T74 Ptp3NCQEnj+fWqsvc+Pli1duwjtZBCc4pFb43ulnpFg1At4U5fHYYzeHfrw+3tx+0eIGhwb9jOGS e+xnylIVZr9IwMROJ/wwY4nTjGOaorDSgCASfcSzk8zMEYCzj4AiRWElBkvrCRTghkL/IUlDEAOW VLbCrCkByrlz59YCSDAxW0mQl6ovm36Ywe3v719TA98JwJmBt1s01naWtGF2iW1YAgJRx48fD/oh 2UrQXXbBBkiawQ0JMksGEcRhO8zAE3inKIz3yEIfx46xqqtS+M51ADsiYZKi0HdI7mInrDywgn5I jGMrqQqy4CbWD32Z74wn8JKiYCskiLDXOAHDagi+Y9PYS4rCeAIvjLWWgEAu/Rh7wY5SjbUp2isZ YiA1A0pCpGZc8sSAGGgKBpaWKvcmt+5rXRdsA67DP3egtbXNjfvlqCsry6Xj5d7xV69ehpnA7FLT 7v3+lhDvmOGspigTkxNBXk/P+ltRCBi4NYTVGSmcQHh/NvnctXk9ZJ91AEesjng+NR2WL5fNC0Ej AQM4LNg2mTinOOsE5CkKs3IEjnHQYnJxjAn0WPaeooCFoCGLBY6YZQcHtl12YUadpeMkhbK2YtgM a9lYsBNWyJAgwi7iYsvtk/Vlb5PYSpysAg99mbrKuFO+fgjUsAWSD3EyBCwkziipkniMXfRncMBD XMxWUvVlbJb+Q7+NC3YDFpIhKRJW6IfgvjLGr38uBmMt/RmbBWvZhfbSbsaP7LUQnsBAX05RTD9x AgK56If+Qz9PNdamaK9kiIHUDKwfgVNLlzwxIAbEQIMYmJubD04EzxrIOqM4GdQtewf+1avXpSMk GCCQyga4CG5rq6zQ4PkMZTuBnN9mAUnCZIvhe+kTJmWXef+QUBIe6CGrn1Dnn+FBIonZqrILM+kU 5GYDSwIr6jgmxayyJYBsNjluu+knhWOMrRC4UPKwUAcfKYIoAn/6T9ZOwGZ1BAwkaMouZiumi1ie 1aXgBLlmKyY3D4vhjfcVvc34Bve1cDDulD2+0S5shZKHxew4hX6wV9OPyQ3AVv+xuhT6oW+gI/qK yTUsdi1klYRdG2xfGZ9wz7iR9+BU0xk6TGErpp9qWMCQQj9l8KxzioFmYEBJiGbQgjCIATGQnAGc KhxBP6+xFqjEIPb5wJLCcWUXHEBKNsC1Om5FWFpcKj3IRcai//NA8jnxq0Yo86/LD/yZnVteXgmc vLkVI4hfrWsJ+kuhH0t0WEBbQfEGC1sEWSmW5taDBXsq20mHd2Rgs3l2a1wZ3pizoret/2QDKORY HbpJkSQyLNb+uK2GhaAvRTHuTW4s0+pSYKmFA0xwxVicsi/n2WxKTmivJcXybMXqzJ5i3RW9DQ7r G1le+A4W9qewFUt0WPvjtoKFvxhvvL/IbcY2s8c8LCltpch26VxioJkYUBKimbQhLGJADCRjAEem EkSRhlhfwvfVykU/2152seA16wAi1/tcoXB7QiVpUvlexr/LQcbSW3yYLMNnDqvVl/FJW2kz7TcO 8uQsLJY/u11bPxXHGE7suDycRdWZDNNFfF6rS6GfjWQYFgu0YpxFb5sMk1nt/MZdtf1F1NfCYvhS 4KAtJsfkxu2zOjsm3lf0tskwmdnzW70dl91f5Hez27zAMiUOrj2GxeTG7bQ6C4TjfUVvg6NWQhF5 7E+hH+s/tdoYc1fruO3s4/pj19tatlIP3mo4TMfV9qteDOx2BpSE2O0aVvvEwB5hYLMXdHMAAz2r gf4aVVHUu+Jn48su5uxUlWOJiJKx4NzxR6nF51LJOJBvONjOlop6KqSk1E8tTmLusniL+r6RDMO3 0XFF4LH+YzKrnbOWHqv9ZrP1tWTE+Godt1mZ1Y43Xqrtp76eY2r9vp59cVtjDrK/3XDsyf5gC99j LLV+noKXetoL3nox12pPrX2xjEbrJ+YkD4vVxcfVatt29pkMkxmfizr+Yu7i/UVug6OWDRi+zWKx 34E13i4Su84lBnYKA2/f9LtTkAunGBADYsAzgONqsxGbuajb7RYbkphNUGz4g80f0CxY1vNXPfmy zz9Ho+yyHksNaVHCqMZR29pVD5a82bJtCc35cT04+Fm9x+WIqLuq3vamwFILdBxIpMBSDy/1HFOr TfXsi9sac5D9bWosWfnxd1viHtcVvW281OIEmXZc0fLzzlcLS2r9gKVa26vV57Vpq3X1tLcZcJjO wLIZPOav2PMttsqTficGdgMDSkLsBi2qDWJgjzLAxZ/7VHmAFBd1m0Wph452/6YDChP//MWxbHAw VmNw3s5Qdmltry4DbA58OQ/QLBoXfO7zz58IIlfbH8swx6seRzH+3Va2kRGeBZGLgzNWdrT5ByCW XcxhtPbH8qjjjwAqBS8WqOVhMfuvcFduoqgWDvgxfMZdzFnR2ybDZFY7fwr92EPs8rBYnXFXDWdR 9cgh6DG58XmtLgUWk2EyYxxsW70dl91f5HeTYTLjc1udHRPvK3o7jLWrz1rIO7dhMdvOO6aoOmuv ycyel3rwpsBiMvKwUGdYyu7Lph+4yMNiHBl39n2jTx68ib/CxIk9K2Wj32i/GNitDOh2jN2qWbVL DOwBBnAOeJUZ7/HmVW84DvUWXv9VcWTyl9762nAqXtdZdunq7AoiLIiM5VUcr+UQ5NqbMuL9RW7j ALZ3+IRIcPbefjWp4cu+BrFIDHYusLDigudUZJ3AldWlsug7BRZ7VVzecnHjJHDnXytXduE1lBST G8uzOo7ZTF+Iz1HvNpwgo2Kfb2eKDItxV+95t3KcvfbRZMbnMJ0RLFiAE+8venv//sorDmthMR0W LTt7PpNjHMT7rS5F/6mFw+wHW0phKyYjO6bAjenM8MZ8Fb2NPRr3potYhmExe4r3Fb0NDgvqTa7J MP2w37izfWV81urLhgXuLNlXBgbOGbc3Tz/2fAzTYb04eLUn/gp/m/1tvTJ0nBjYKQwoCbFTNCWc YkAM5DKQ50zmHpip7PSOF04Vv886GZU6Hoq4z79PvhL0ZX5e6Fd7Z30WB0Ksbn8ILMsdsgkEuvd3 h7YtLLz9QE677aW7e/177QslY/VkHR3ta6tbwhs7IiFLPDRsySdm/Ns60GPZhffTU9BF1t5w2vkj WCjbMQaDYTFdUGfFHOPe3l6rKu0TWyEBSDG5sTDqCBYsqIj3Fb2NYw+eajiQR19P4fQbJ3kPFLQ6 O6ZoHrLnM1sxufF+q7Nj4n1Fb8M9tlDNZulTBP7osOxi9piHxepScEKQa1hq2W0KW0E/jF/VroXg Y2xLwQvthZuNOElhK8a99ZXYNq3Ojon31budvZbU+zsdJwZ2CwPlerS7hSW1QwyIgV3HQKufaT/U cygEluZ8WiNxgJb8WzGOHGF1RfnDJM5oe3uHX545/5bzZa9F6/UrPVKUQ6sBrL0qzWQSaMMTDmJP TyX4tH1lfMJ7T0+PW/SB/2LmDRjg4O/gwQOurcatLEXhwtHEUcfxtKSQndt4OnLkiFWV+jkwMBDO b3YRC+N1fsz2syooRenv78/Fgm7oQ+gvxUwuAS6Jlzz9GE99fX1JglwSIuiAZdfZAhaCp1S2YnLM RmM81GHT2wmi4vPV2ib5gy0whmSXoM/OzoafMjOcotA3wGN2Ecu0OuMt3lfGNn0ZezC5sQzqwJlC PyQY6D/0WfpuXPjOmMd+xv6yC4kO7BI7ySYi7HWlNgaWjcXG0Wz/IXkAFnjbrN0q8VC21nT+ncRA +SPKTmJDWMWAGNhTDAwMDgTHanR01DvIS6ttX3GP/fd9fpb9+PHjSfjAET1x4rh7PffaPX36dE0m zt+TJ0/CDNSxY8fW6svcIIg65AMGcMQO6dTzqeB4nTx5MsxqlonBzg3/7T6gGx8f91W25H8lcMLt Mic8Fv+0DDu8tE/0c+HChcBHBUtFFE7yyMhI0E8qx5iEFcH/xMTEukD3+fPnbmpqyoEjReAPA+Ag YHjw4MG65Az9CW5OnTqVzFZOnDgR+vLdu3fX7ICgF/2AMVVfJnBEFsH1s2fP1rAQyKAzkiEE5CkK YwYBdbYvP378OAR5p0+fTnKLCv3n7NmzQRa2YYEYn9gO9jo0NJSCkrDiAv2gj1g/BL3YCpyl0g+6 QdbY2Ni65Az6At+ZM2eSJM4g3voP+qDfUNDPo0ePQh+mL6co2Ap2SfvhxQrXQvQDX5sN/O0cm/1E FuMp+rAECOdgnOW5DtgRY4uKGBADW2Og+tPQtnY+/UoMiAExsGMYwIHAOb5//757+PCR6z7Q7V6/ eu2mpmd8AHUyWTAHYThWzJ6OjY0HJ5BZFgLLtrZ274yeTuaM4gSeO3/e3bx50925cyfgwgEk+O7v P+adssrsdwolox8c0gcPHrq7d+65nkM9wTmd9vqh3pYzp8BCwEAC5uHDh2HGncCJBBGzgxcvXkwW bNPW814/6OTrr78OjjA6w0Fn5g57TlUIuN9991137do1d/369RDwEswRPBBA2UxiCjz0HxJF9+7d C32FGVWCBYKHK1euJA0WCNiYyb59+3YIYuAJTsCU0lawzXPnzgXdXL16NWAhsUgSAlseHBxMoZog g+QmfZb+Awa+ox/sBU5S9mUCbsbaGzduhOCblSuMb+iHvpVixh9S6LeXL1926IY+RMBLvx4eHg79 OlViEyyMZ/QTOGHcZwWGJWrgJKV+aDf6gQeSmdyqw1hLP7p06VKyayG80Hb6Mvqx/sJYSzIxVWIG HCpiYDcy0OIznTa9tBvbpzaJATGwixlg+Prkk0/cl19+GRzHzz//vK4ZteWZkXWs4OhMTE64eX87 BI4ODgaOaSMKQRMzdDijOMWpbgnJthVuwfHy5Qvv9O0LyYhUM+xZLEteP0/Rj78dosPftlKPflo6 e1xLR/EzzjjmzIyZflgNQDDRiDI5Oemmp6eDaIJwArtGFGZOCbJx1uk3BBGNmiEkqCWgpE8TODGz ze0ajSgE2fQh+hIJGf4aYSvIJ4gjsIML+g9jSyMKNkL/oS8znmArjdIPfQf9YL/0HXhpREE+fZnV M9gHY0rKoD9uM/2Gvkw/og+jHxLijSiMtdgtYy2rEujLjeg/tN3GWvoSdrLVsfbTTz91X3zxRUis fPbZZ+7jjz9uBLWSKQYazoDvyy2N8bIb3nQBEANiQAy8YYDAaXAg3azgG8lvb+GYNyrYj9Hg7BHY 8tfowvM7mkU/BAfM8jdDaRb9MHOc6naHjXi31TMbHZdiP/fR89foQl8mmGyGwqx2s8wgE0huNZgs kkv6DwE2f40uXAtZJdMMhbE25cquWm1ulrG2FkbtEwM7jQE9E2KnaUx4xYAYEANiQAyIATEgBsSA GBADYkAM7FAGlITYoYoTbDEgBsSAGBADYkAMiAExIAbEgBgQAzuNASUhdprGhFcMiAExIAbEgBgQ A2JADIgBMSAGxMAOZUBJiB2qOMEWA2JADIgBMSAGxIAYEANiYOcxwLNaVMTAXmZASYi9rH21XQzs AgbsQs7DvWx7FzRLTRADYkAMiAExIAZ2EQOxjxJv76ImqilioG4GlISomyodKAbEQLMxwEXcLuSp 3u3ebBwIjxgQA2JADIgBMdD8DNjraPFXGvUa8OZnSQj3CgN6Rede0bTaKQZ2KQO8Y513zvP+bv5U xIAYEANiQAyIATHQbAwsLCwEf4UExOLiYrPBEx4xkJQBJSGS0i1hYkAMFMkASQdWQnR0dDithCiS WZ1LDIgBMSAGxIAYKJIB/BT8FZIQmjQpklmdaycyoNsxdqLWhFkMiIE1BnQhX6NCG2JADIgBMSAG xMAOYEC+yw5QkiCWyoCSEKXSq5OLATEgBsSAGBADYkAMiAExsNcZUOJhr1uA2h8zoNsxYja0LQbE wJ5kYG5uzs3MzIR7NTs7O13PwR7X0dmRnIuVlWX36tXrgIVnXXTt3+8O9fQ4e5hVSkALC4vuxYtZ j+dVkH/gwAHHX/qy4ubm5t309LTjflr0c+jQIdfe3p4cCg7kixcv3NTUlEM/8HHkyJG1h6OmBLS0 tBQ4AQ9LfA8ePBh4SYnBZKGXycnJ0H9Yaox+9nvbbUR5+fJl0A/3W4Oht7e3YbaCzc7OzgYawNIo W4ELbBZuWAbe48cU7KURhfEELNaX4aRRfRndoCP6NX0ZW2nEbX3ox2wFLtAPeOyhyyn19Pr1a/f8 +fO1a+Hhw4fDmJsSA7IYX60vs93d3e3A0ohroeknHmuxFRUxIAa2x4BWQmyPP/1aDIiBHc4ADvHN GzeDY4yDM+Wd0us3boTvaWctVtzIyKi7efNmcNBxhsdGH7s7d+6EREBKmnnQ5927dwIecCz47+AY GRnxzuFSSiheD9Oek1s+MTMdAqgZ9HP9hnv27FlSHAh7+PCh++qrr4KDDi+jo6PuBz/4QUP0g53c vn07cEBCgu+3bt0KzntKYkjeff/733dPnz4NAQKOOpw8fvw4JYwQSCITLAQv9OWxsTH3wx/+MPTl lGAImtDF9evXw8PnGEfu3bvnrl275hNqcymhOILKH/3oR254eDhwQhIAHA8ePEh+T/r4+HiwDYJ/ 9IPNgI0APGVBH/Sdr7/+2tF3KPfv33c3/LjP2JeyYA+MKegDTtDX1atXg76wo5SFRCL6YGwlWUUy gv5Dfdpr4ZuxlmQVY+2jR4+CvujbKQsJCOyC6x84+M5Yy3e2VcSAGNg6A1oJsXXu9EsxIAZ2OAM4 gLe9M3HYz2pcuHDet6YlOFs4hAQR7733XpiBSdHMp08nQ+B29uxZd+zYsSByYKDfffWjqz4hcNe9 //57AV/5WFa8vHshWHrvvXfXZikP+tUhN2/dDM7pwMBA+TC8hEXvgIKl5+ABd/78hdWZwRV3/96D wElnZ5fXT5oZ94mJiRAoXLp0yQ0NDYX28/l///d/IYj48R//8WQzlwRM4EFmV1dXwMLqAwIIHOUL Fy4k0Q8BAg45wRN9BdkUVkPQf1i1wkx3ikLgRD8ZHBz0tnI+6OL48eMhcCHohisCqxSF8YOEyIcf frjWfnD97//+b0hGXLlyJYmtEGATQBEsfeMb3wh6of3M4hLooiez5bJ5IdkLlpMnT7pz586F9p84 cSLYLMmAH/uxH0s2404ykQTV+++/v6YfxrQvvvgicPXBBx8k0Q+BPZyQbAAL/YXCSgg4Yf/p06fL Vk04P9dCZHLtMfu0ZA228q1vfSvZSjh0g44uX74c+jMA6cuMtfTlb3/720k4QQjJBmwX+7TVXUeP Hg3JTsaTM2fOJMMiQWJgtzGglRC7TaNqjxgQA3UysBIc0RXvABIgkICgsAR2aIjvLtlsLsHC48ej rnt/ZclpEO7/aWnZ5475RASzP0+fTlh1qZ/Pn0+FVQf93hmNl0n3HOoJTtijR8Nrs4elAvEnf+Rn b+FmwOvnzdLkFv+9P3wfGX5UNoRwfpxxnFECfksQsYPgm6QRs8vM8qYoyMJJJ2iyBARyCSz7+vrC vlSz7U+ePAm3p8CBJSDAQsCAg04gAXcpCjP9BHMW4CITmyHQZXab/SkKskhAoI84AUNf6u/vd3DG DHOKQmKGVQbYCgkHK9gJtkMyy1YC2L4yPrEBZrL5PHXqVNSXXQiyU+rH+iq3o8T6gR+wGWdl8JA9 J7IIcBlTLAHBMXwnEYHNpuo/9FUKSSIba/nkO4VVcCkK9ggWbkfBTq0wvjCusNKKBGyKwiovxlpw WAICudgOt4awEi7VWJuivZIhBlIzoCREasYlTwyIgaZggNlBnAwCydhBB1xra1v4I1hIsSSWJbg4 M23tbW/N1natzo7hsKYoFiDt767MsJtMnEAcZW7HgLeyi8lBP3EyBLltbe2u3f+98MmZFEticXzR EYE1eOJCMIeznko/JDsITOIEhOHBUYaPlFiQHQdQfDedsewe7soutJlgjn4cJ0OQazwRuKQIuMHB CpG8Zy6ABd2l0g9tRh4BXbaAj+A/xfJ2+IAX+k+2LxtPNu5kcRb9nbGLRITZRXx+q0uFxfpyHOAa HvoUvNkzRay+jE/kkKyi72SvheiMP2w2xbUQe2SsxU6QGxfjidt4UhRuQ6Fk+w/jPc+ogLdUfTlF eyVDDKRmQEmI1IxLnhgQA03BAA87nPd/OBTZwMXqcLp4UGTZBceLYCEb4CLX6l6/ngvHlIkFDDiA FbnrHUDqzCl8/fpVOKbMfxbmF0NA3dq6b21mzuShr32+nuCTQKrsYkkXdIFtxMXq0GGKhIgFJWYX WSx8J+Aru9A3DIvZRSyTOoL+FEkI9LNR/yHwTJGEsPbm6cd4ShH4o4ta+jF8Ztux7oreNu5NZnx+ 6uhTcJIiyK2lH8NnvMU4i96mrWYHJjeWYXWGN95X9DZjPn9hXPVja1zsWsjYZnjj/UVvwz192fpK fH7jxPp7vK+MbeM+Dwt14EzBSRlt0znFQDMwsH60aQZEwiAGxIAYSMAAsxhL4SGLLW8Flojftxps zs+X/yA5C/yzAS44WvaBb18IoJa8I1hm4fyLi/5Bbb7teVgsWUMCp+yyuLS4GpTkJ4nAhxOIHssu luiw9sfyjCdwpEhCmK3kYbE6joGbMgucIMPan5VlWIy77P4iv1sgYDLjc1sdQV+KJEQtLBZEmQ5j nGVsmxyTG8uwuhTLyWvhAJP15RS2YljMLmJOrC4FJ9ijjV0mNw8LCZyyC+OW9Y1sf+Y7+MBr3JWJ p1b/AQt/4C17rI2vLdZX4nabzlJwEsvVthjYTQwoCbGbtKm2iAExUDcD5vB6n8Y7Nut/Fhyx1Tpz ztYfUew3kxHkZk5t0FZWlr0jWG5gueyDSuSYzAyUtYDT8Gb3F/kdp7cS5L591hifOfJvH1VcTa32 mmPMMbWOKwpNPTI4puwkBPqpp5QdLIDBOMnrPzHGRmMxfClw0O5aOjIsxl3MU9HbhsNkZs9v9Y3G YjhS6AdOjJcsH3w3LCnGN2svMk1uFhPjSQr91JJh+MBSi7ss9q18j/WTx4nVGXdbkWHn2Mpv9Rsx sBsYUBJiN2hRbRADYuCtWyo2oqReJ6bsYA6cG2JZjbrLxsL5TUYtB6nsZEjgxGOpWgIfFVIMb9Vj C9hh+qnFCTjsuAJEbukUhi8FJxYsmMxqgFNgqSUjxlfruGr4N1tfjw3Uc8xm5WaPj9sac5A9LjWW rPz4ezNhifmLMRa5bTKaST95WKwuhX5MhsnM4xve7Li8/UXUcX7TT975DB/H1Dou+1v7HfXxdvY4 fRcDe4GBt2/63QutVhvFgBjYNQywfJOHavEgq804JracMhBBvBtPr0fspHAU1mGJZK9tGr4qGNeO 2+YGbeW/SqmeBNjnbxEpu9jtMLlyImhNoR8PEh1uqMfcxmyuslZ7zRkOeswu79mcmA2PtraazGo/ qIW32m+KrI/xpcBSjwzjrsh2Zs8V44g5yB6XGktWfvw9BRaTV4sTjon5s98U/WkyamFJwYnhoH1g ib/HbU6BpR4Z4KvnuBj7Zrc3Or/pDCzV+MqTyTMv8FdYiWmJ3LzjVCcG9gIDWgmxF7SsNoqBXcwA 92viBNRynvKab08B9z6Xi2LacGhwMFYrWzNP6M4713br7MFX5tjE5wMfhTa2Zh4aVtlT3L84XjyD ApEmNz674cu7RzY+rojtgMU/CyMXRwVhEGPcFSGz2jlMRl6SC074Mzusdo6i6g2L6SI+r9Wl0M9G MgyL4Y1xFr39pi9ne/J6SRthXn/01r7VwmL2k31DxNYkbfwr4950Ef/C6uyYeF/R29Zek5k9v/GS AovZQB4Wq0uBg/HNAl2TG/NidcZdvK/obWuvycyen3quP3Zcdn+R32v1H3AYFuOuSNnxuTi/2YrZ Z7zfuLJj4n21tuHR/vLOW+u32icGdhsDSkLsNo2qPWJgjzHAa9V4pzqveuPiXm/hFWgVB8I7NjnP WvC14VSd/rV/ZRd7NVwujuB4LQesZSdE4KPDrygh8s9zkGzmpr3DH1NyafevKyXpAo4sFm4HMWe0 vb18/dgrKLM4oIA6sOCgmwNdJjUmw3QRyzJ82NNm+kJ8jnq34QQZtN0c8vi3MZa4voxte3WfyYxl GE/YdoqAzvqyyc3DYjqM95WxbXabd9+61aXAUguH2Q+2lAKL6aeWrdgxZejEzkmQa/aYh8XqzLbt d2V8gsOCaZNrctAPdeBNwYu1N4sDPNSBB6xlJ0SwR7PbvL5s/ceOMb42+uR1n/gr/JXdho2waL8Y aDQDSkI0WgOSLwbEwLYYyAuA6jlhZ2dHcHpxbCpvyXjzq1C35B/Q6Gfi9+/verOjpC0cE5w83giR Lebs7E8QWOJ42TvR8x6IZnXd+w9kYRb+nYCkta3VO5488HH9QxDRz7Kvw4nrTJAQIcFFQRdZe8NB pQ7nOYVTifNKMV2EL6v/WN3hw4fj6lK2sZVDhw6Fc5uNxoLAAh8WVMT7it4GB3is/fH5DVsq/Zit 5GGxOtNhjLOM7Xr6suEtQ76dk+CVwNF0YfV8Gifd3d2lJ86QZ5zkYbE6O4bjyyqM98a9yY1lWV0K LPQNdFS57vm3I0XF6hiPU/RlOIEbs4sIypr9WH+P95Wxbf00Dwt1jDmmw63Iz15LtnIO/UYM7GQG lITYydoTdjEgBrbMwL59re7IkSPB8cq+chIHY3FxwR3r7wuJiC0LqfOHOID88Wq4RS87LuEVYN7Z 6es7FleXtt1LAOvlce9qXEgGwEt7W7vr7a0EwvH+4rdbvH4Oh/tm5+Zerzs999POL8y7Q8wmsXKj 5EIgwF/Qj09ExMV4OnYsjX4GBweDk26vsjMsOLTYCoF/iiQEcoeGhoJ448CwgIPkDA56iiCKoOXo 0aPhPmt7641hMWzoh6Ch7EKAhA647zsbZIAlpX7MJmdmZtY1m8CSOgJ//soutNnG2qzdwhOlv7+/ bBjh/OBg9jrv1ZcvXrwINtLX15cEC/0Hm8y+5hG7wVYI+i0QLhMQCSL6D302+3pS+hMJETihn5Vd sEfajX6ywb/p7Pjx42XDCOeHE/RjNmpC0Q92TGLG+pjt2+gzOyZsdLz2i4HdzED5I8puZk9tEwNi YEczcMwH9q2tbW54+FGYca80ZsWNjoz4pbIdbnBgMEn7cHROHD/h5rzDNzY+viaThMTY2Ljr8cHc YR+QpygEjUd8ImJycnKdoz45+Sw4yydPnfSOWZpLx+DgUNDDeODE7vdfcXzHIT5x8mQKSoKMCxcu BCd9bGxsTSZO8ujoaFgRsFlndO0km9wggCJ4ef78+TrnGH09e/bMnfScbHaJ8CYhrB2Ok469PHz4 MHDDDpzsEd9/4ObMmTPBiV/7QYkbtJtg9969e2tSwPD48eOQ4EsVuBCY0G4CpqdPn65hIahEP9hJ b2/vWn2ZGwTcBPdPnjwJCRqThc2SiDh9+nTgzOrL+mR8O3v2bLj9ANuwQIygd3h4OCSrBgYGyhK/ 7rzcemD6qYwrld0EldgKfG1ndnudsA2+kGBAR+iDBIgVcFj/sbqyPxlTrP+gFwp6evDgQRhPLOFY Ng5sBf2QDME2rMAH3+k7KRIzyCUZkjfWMvaSxLMxxzDqUwyIgc0x0OIHGfPsNvdLHS0GxIAYaDAD DF+ffPKJ+/LLL0NQ+vnnn6/NztaCtjwzsrabWQ4crS4f3O33szAED/PzC+7UqZNry83XDi5xg7bg FD8Zf+IO9hz0M/xtbnZ61vH8BYIFgptUhdkvAktm6HoP94ZbH+CJAGpwcCBZEoL2Tk/PuBGfJNrn Z+sIDggWWLmCfmoFcy2dPa6lo9gVGwRQOMKsNEAfU1NTwe4uXbqU5H5p0z8zk3fv3g1JCBIB2A6O MbPwJEtSzFgaFvrLzZs3Q7IB+XxHRzjoBJYEFakKiZjbt2+HIIWVRQT+BNtwkiqwpK3o4/79+yH4 Rz/og/7DLO+5c+fWngWQgheCOfTDJ7PZ9G14IbjiL6WtkJS5c+dO0AXJKwI5Al70kyqwhHNsAv2Q UIQTgm/6DzZCsiT1WHvjxo3Qb0iAEGwzrlj/sWc1pLAVOEA/JDHhgsQIdnP+/Plkq6toJ/2HcZZk DGM8iSM4QS/YSqokK1gYa+EEW8VWwDYxMRGuhSRLNtt/Pv30U/fFF1+E68Vnn33mPv74Y8SoiIE9 x4D3DVqUhNhzaleDxcDuYQCHYLtJCNjgVoOpqRnvAM4Hh4dgKqXzF2uEIIGAZWVl2Tsq+4MzmDKQ e4NlxTtes8E5xtEiSEjp/L3BUQkacELnvYPOgzPr0U8ZSQgw4ZTjrBM8EUiBpTH6IUEzHQIF5KMf 8DSi0A9JAMANAQOcNMpW6D/YCsEcM5n12EpZnBHEEbzAD7pBR42wFeTDCckhAm5wwE0jigXZ6Akb IaHXqLGWhBm8wA8JolpJzTK5Qj52gr0w1sJJo/oPCRr6MvqhL5NEa5R+wAAWMJl+GtF/0D36IXmH fOxkq/1HSYgye5LOvZMY8H2ppW0nARZWMSAGxEAZDFSeD5HmdoeN8DPbk2ppf20slcCWgKXRBcec ZcvNUAgOeC5DMxQCbP4aXXDMmSVshkL/SfV8gY3aS+KhUYmhGBv6IbDlr9GFwLY5xrfKcvutBpNF 8oh+mqUvM9Y2i37oy6luA9lIn1wHm+FauBFO7RcDO4mBNDf27iRGhFUMiAExIAbEgBgQA2JADIgB MSAGxIAYKIUBJSFKoVUnFQNiQAyIATEgBsSAGBADYkAMiAExIAayDCgJkWVE38WAGBADYkAMiAEx IAbEgBgQA2JADIiBUhhQEqIUWnVSMSAGUjPAfbUqYkAMiAExIAbEgBhodgbkszS7hoSvbAaUhCib YZ1fDIiBUhmwCzmftl2qQJ1cDIgBMSAGxIAYEAObZMB8FD55CKiKGNjLDOjtGHtZ+2q7GNjhDHAh 5xVrvH+ep67zKi8VMSAGxIAYEANiQAw0GwO83hl/hbeyvH79utngCY8YSMqAkhBJ6ZYwMSAGimSA 96vzCjxeb8a7zG2WoUgZOpcYEANiQAyIATEgBrbLgPkrXV1dYeJku+fT78XATmZASYidrD1hFwNi wLW1tTneJ66ljTIGMSAGxIAYEANioFkZYMUm/gp/mjRpVi0JVyoGdENSKqYlRwyIgVIYYDWEihgQ A2JADIgBMSAGdgoD8l12iqaEsywGlIQoi1mdVwyIATEgBsSAGBADYkAMiAEx4BlQ4kFmIAbeMKDb Md5woS0xIAb2IANLi4tueHTEvZh94ZdHOuefWe0OHjzoBgcHXJtfOpmyvHz50o2Ojrq5ubnV20ta 3MBAvzty5LDHljZnPD427iYmJ0LzudWFZaRDQ0Ouu7s7JSVu0esHTqanpl1rW6v34pzr6T3k+o8d C0taU4J58eKFu3//fnigGLcBUeCkr68vPJMkFRYc2YmJCffo0aOwpBf98EyUEydOuMOHD6eCEeTw MFhwgAcc/PHQtdOnT7vOzs6kWObn593du3cd/QgcLHdGN+gIflKW6enpYCtLS0tBNjo7deqU78tH Aq5UWJBL/3ny5EkIgOCFpeBgYZxLWRjX6D/0I/QBtqNHj7rjx4+H2+pSYuGBxtgtNmPPE8JOjvlx JfUy+eHh4fCwQvhANvo5c+ZMeN5RSk4Yax88eBAe9gwObKWnpyfYCuN/yjI7O+sePnzoXr169ZZ+ wJWyjI+Phz6EfpCNvTC+wU1qW0nZbskSA2UzoCRE2Qzr/GJADDQtAzgVd3zQgnN87ty5EGDzxGoC mdkXs+7SpYve4UgzTILh1q3bPoDrcpcvXw6OzrR3lO/eux+CKoKGVAWna3hkODifBHEEUjjKN2/e DNjSJSJWglOMQ3rx4sUQ3MLTndt33OzMjLty5UoyJxC5X3/9dcDw4YcfBv0QyFy7di0kJQgaUhUC /hs3bgRHmACOQkAFlnfffTdZIoL+Q1A5MjLi3n//fdfb2xv6Enby1VdfuY8++ijZw9fQD+3HVuGA B7+RCLh+/XoIeulTqQIG5NL+/v7+EEwSuJAEAMuFCxd8gnMwlamEfkswd/78eZ/QHHALCwshuAPf N7/5zcBTCjDIpf3YDP2WRBXJCOs/ly5dSqYfxhP6MvoBi+nn1q1bwX5IRqQq9B0Cf/QDHuyX68/V q1fdO++8Ex66nAILekE/BP3I5QGKJPOom/FjLf3bEq9l4+EaTPvBwFiL3MnJyXD9wWbgKlVhrGU8 Y3wnyWtJ1/+fvTcPsi2p6v3z1jzfqlt1h7rz1PNEt6AyxQ+DUANFJNAHIooaz8YOn/5EBX/+NEBe +Ic+nyG+n6IoP33oU5+KED9bBOMF2DxBRWRouunuO491h5rnefrlJ/dZp/Ls2nXqVNXeeU5VrRW3 7tkn9z65Vn5XZu61Vk60H+nzQsmifBSBnYZA2HDiTkNPy6MIKALbGgEMDBzJ7kPdblQQYxQH++jR Y87wGhiIZgJkXUgMQByF5eUlc+TIEee4IUu7HTXdt6/d9Pb2OuMwaznIH2cOWRjlYVQQORgFQy4r prl5AznD7MMB/kPDw1Y/h5xBiiw4L4etLDgSjPCGomvXrrlRU5xIRioZDWMUF4cSRxx5QhBG8JUr VxwejMZhoPPHNdhcvnw5mH5wtglOYaAzwi/6waHEmaEehSI3W8bKQ7CBNuzaj50VAi4E1YZtPQpB oh/4S11BPwSLmHlAXaGNhSAcNoJT6AbHGpmYnYK+GPXGuQrVltEPfe2JEyfybZk+Blzu3r3rgjQh MMHJRwdggSz0bbRl8OGUJQIAofRDG0EW+hH4Iwd9C+0H/XAvFPGOwdGnbqAX8KG+UofRG20oBFEf 6d/4ZGAAPJCFdxF/9Deh9CNthGAI7z/0Q30BIz6pK6HaTwjslYciEBoBDUKERlz5KQKKQEUggPFA EALDosUaXT41NTXa9BprePXb5OwdbkYJcegYvcWR9Km1tc0ZOqGMwGFriIJNu1sCsicvSm1tjXXu Gs34xLgdIZvOp2d3YfVjgxDV1gBtjk0bb25udMYpRnMIIxD9MJKNfvjziZkijLBz9nsIgg9GOLMO fKIe4zzg2OA0hCAcF/CPy0IQAOdBZA0hC44sjgEY+IRjiYw4wSHqCsEo/iQoE5cF3YWqK9RZ6i6y +IRuwIV6wqhz1kRgBv3QTuJ1hdF/CFlDELohIEU9ITjkE20Zx3NkZMRPzuyaMsMvvoSKtiz6CeFw 0y6QBb5x/eCAg1Oo9w/lBX/eg/EZd+CErPQ7IQj7QNoPgRAh0Q8zRUIFN4W3fioCOwmBlVa1k0ql ZVEEFAFFYB0EMHYwwDEuqqsLu0JJm52dMXOz8+vktPXbjFhiqGPsxaeMi6HMMyFoPDeiX1dbF2O3 xzl5JE5NZS8Lxt/0TLQeGKPPp+qqaqe3ubl5q585/1Ym1zhrGL/IsZZ+5JlMBPAyFaNX6oV3y+kH OeUZ/14W1xjpEE5tnJCPNoahnjXRNqgvBCHiJGk4n4yCZ00EE9FBEiayRwbPhCAJRiXJQhpyhuhX CIyx9wK6iLcfkS0UJqKfeJ+CPqRNCW5Z6shvp8LX50caz1Bvsyb0I/uoxHGJ3oXV7l0ZIiBCfaSd JtUVacvS72SNi9QDqaM+P9KQk6UqSoqAIrA5BAot783lob9SBBQBRWDbIYDTwigUFDeMSavKbQQ5 bQMRWRNGIOSPtghP0pAPeZeWsnWiwAM+a8kiwZoQxuj8/II18pZc2eP62WMxQT/gMTeffRBC9BM3 0AUn5MPJ4i9rkpHrtWSBfwjHRdoPZY/rBxlEvhBBCClvUvsROdBNiCCElFf4goWQpIkOJT2rT6m3 SU6uyCLyZiUD+SIHDnWSHFJ//L4na1nIX8rv85I0wc2/l/Y1QWfpR4Wvz0PSQgSJaBu0Z9GFL4ek 0XZC1BVpy1J+XxZ5F4Zoy9RXaadJ9VbS5BlfTr1WBBSB0hDQIERpOOlTioAisMMQwADE0MDIqrJ/ PrlvuaSFhexnQsznHOlCKSKJItH2uJkSiwtLvpipX2MYi+OflLmc0CGBiqRn0kojwMAeGWIEr8o3 B1YIx79YsAq5kJFnQji5gj084yRpyELdzpJEjrV4iCzrPbfW7zeSLnUgKQghcpBfSFmS5BdZRN6k Z9JMK1ZvBasQmEi7kPLHyyjpIWQphonIFUI/UV8bBZWl/MKfT0mTQIV/L+1r6S/gKXx9HqQhbwj9 lII9sogefTnTvIYHf9BamHCvFHl5TkkRUARWI6BBiNWYaIoioAhsQwSSDIVixRDD2D0T9+c8B2/J jsZnTYtLOYfR41vAMyffknXKsyQcVxx/qBieYpxlKcuSYJLAJIIpAiWELFJXkjCRNN9oTRA5lST0 U6y8Ikukx2yDECKH8FyrgPLcWvfTSC/Gw5ev2HNpyEEe4hz5fCVvSQshh18HhK/I4X9K3fbT0r5e j4fIJ9ilzd/PT2QRnv49SfOx8++nee3zEL5J+Yu8SffSSvPrY5IskuY/lxbveD5SXuHp3yeNPx87 /36a1/CQ8q4lC/zkmVJ5+3lJILDU3+pzisBOQ0CDEDtNo1oeRWCXISAvdZmqmXrx4wGK1BlYhz+D PDeTpTXxNvOzTH5TUbKsFRzySi710EtK/TIEj7SF3o4ybwWD7VbeSpI3hCwheGyl/sR/G1penO9y UujylquslHO3lLVcGCvfykdAgxCVryOVUBFQBNZAgJc4RhNTIjc6iiZrOrG54naXb4ixCWLWJHst JPFxJqH9L4TRsqeKpSlV7jwQHwORy86TcJchRnCqqtc20iJ9RbIkrR0WedP6FB6JmOQqD5gEwcXy WYtEvswCch7jYpjwmC+L97NMLkWWpMxFDu4Vey7pt5tJW+lXovrp5yGyyDP+vbSv6S+kPgpfn4ek hcBEeAhPXw6uJT0ELsVkETl4JmsH0e/PhW8cF76LvEn30koTHmvJIenyXFp8k/KROiA8/WdI48/H zr+f5rXfh64lC/ykjZXKm7zEXsm6jpUqkz6nCJQLgbWtmXJJpHwVAUVAESgRATFI2KkaYyDJWFgr K3aqj/Y4iAyb+HOSl+zIHb+f5ve6unqXnfD0847SIsNLDDT/fprXGJnO0MwZe/G8l3NrZGvtzu1Z E0ekYqQt22UZcVz4bv85CqEf4bHW1FvkQTchjHTZqT1JFklD3qwNXMEkrhupF5Iu8kp6Fp9ybKqU 3+chcoBH1u0HviKL8PVlEfkEO/9eFtfCR/j6PCRNnvHvpX0tdUB4xvMXrELIInVAePqySFoIOXhf CZ8kXCRN6pMvZ9rXyCH9hWDg8yANeUWP/r20r+UEGSm/n7/I5mPn30/z2i9vkiyStlFMwJnf8K6Q pSdpyq15KQLbCQENQmwnbamsioAisCYCYqCs+UDsRn19nTUCo2PQxLmWR8hLjIwQRiDnoWOcJM3m YDNK5MFwwTDKksi/obHBsUiShRMroAZ7hnvWVFdXGxlqNvBBIMIndMPGlTU2ULH6KFH/yXSu5bx6 jMZ4PQMn0c9GDdLNSNeYwz5pkzjRWUtLy2ay3tBvcOakbQhfPwPSqNOCnX8v7evW1laX5VpycBPn RhzQtPn7+Yl+iskSAhNkEj5JdUXS5Bm/DGlfg8ma/VuuTeEIh9CPlLeYfkSHaePg54cTKg53kjMq 8om8/m/TvqYd03f57z3hEfW10fHRIXCh76KuJGEi/S9yZP0upPyCvbQVwYRPSZNn/Ht6rQgoAqUh kK1FW5oM+pQioAgoAptGIO4UlppRbW2daW5udsGGBWsI+4ThtbCw6O7LaJV/P+1r5GDkf25u3jrc hZtPyukcbW1tabNNzE8cuviu7OCM4YWBGEKWKrsMBlwwPOdjJ5RgoJOO8VzfUJdYjjQTKS9GL+WX 4JTkL6es7N2712Ej6Vl9dnV1uayTdmVHFvTT0dGRFfuCfPfv3+++x+sKuuEPAz2E44IzR10Bk3h/ IEfoocMQM1Xa29sdJknHGXL0I/qhroQgaadJspBG3xYiYAUf6gLtVpxrKb/IFqrOoh/astQLkYNP qcf79u3zkzO7lvazVltGzhB1hQAEdYW+TZxrKXT0Llxw9SREkJX6SDAKTOJ9reAkuImMWX1KW046 spX6Q3+ylXob76uyKofmqwhUKgIahKhUzahcioAikDkCOHQYOoODgwW8hoeH7fdlc+jQoYL0rL5E xky7M4JHxkYL2AwPDzmj7MCByOEruJnBl/b2vdY5qXOY4EgKYaBjjGGghzBG4Xvw4EGrhWUzNDQk YrjPsbExG5hYMAfs/RDbeqKfI0eOOP3IOfYIQt3p6+tzxmiouoJhTKBoZGSkYLQQZw5ccCYkkOTA yvA/9IOj1N/fX8AF2agvOAsh64rowxcG2dDf4cOH/eTMrgmMdXZ2mtHR0QJHF+cbXHCytuK4bERw 9IM89G++wzM5Oelko/+T0fiN5LvRZ6X9EIAZGBgo+Hlvb6/r35A1BBEMOXDggOvL/EAEdQdZuC/O Z9bygD9BOvQDfyH0Qz8DJmAXguBFf09/5hP6ou6E0g8BCHChP4vewyvSIAv3QwUhCAAR3EQ/fnCG vo2+lnch9zdCfjvcyO/0WUVgJyKgQYidqFUtkyKgCJSEAA4bxlVvb5/ptwbO7OyMM3zu3LnrDKFQ xijCHj161BqkDeZ2z20zMT5hZu1Iy+3bt60xNmOOHjnqAgMlFWqLD9XU1JoTJ467kahbt245Z2Vq cspcv3bdGcw446GotbXFHLQOw+DgkHNeZqzxh2Ha39dvOq0B2NERjTqHkIdyY5ReunTJOZgEZHp6 epyxfObMmSDOHOXEkTt16pQrMrLgrOC0XLt2zQUEzp49G2RGBgLgQJ08edIZ6YIFASNkwREP5fgj S3d3t2vL169fd0ERHExkor4cO3YsyIg/chCUOX36tMNG6gr6uXr1qpsFgO5CBWYIMFA3qavoBMeO 4AmmqAUAAEAASURBVMjFixddXT5+/DgiByH6WZzHmzdv5vVz9+5d166RI8SIPwWl/VAfmJ1x4cIF 50yCC/gQCEB3oRx/HGraj9QP5BgfH3d9DO8m5AxFvOvo43jnoBfqDM43+iLASnsORbQRgkHoRIJ5 N27ccDhxD9xCEG2Z/pTAAe1XgkOXL1929Qf9UJ+UFAFFYHMI7LGNq3Cx7eby0V8pAoqAIhAcAbqv J5980jz77LPO+H/66adLmr2wNH47Lyt54Djxx0gQBg4jHIxWhjYwGC1lJApDFLlwIhi12+hoS75w W7jAIMYYZQosOGAU77cjVDXWeA9J4IAjOTg0aBbm0U+108++fehn7Tj6nvpWs6cu2isgLXlxUnBs MYy5xglHPyGDVVIWnASCRBjG6AcnDkePke/QhH7u3LnjRgtx4Gg/OC4Y8SEJneBEySwRsMDxRZ7Q xGgpdQX9QLRhgjIhlqfEy0p9RT/IhE7o2wjahHK2RR7aMnVW9CP9W7n0gywE8ZCLGSpggvMbmuL6 oT+hroTWD+2H2Qa8g3gXEqih/TAzoRzvQvTDjAPpa+lTeA+FJj/gLO9C6spmZhE99dRT5stf/rLr pz/4wQ+aV7ziFaGLo/wUgYpAwLalPRqEqAhVqBCKgCKwGQQwHrcahPD5kl80sFH+0Y1IlsqQA4ww vspNG8EkiyCEX/6NyOL/Lu1r5IC2m37SxiGen+onjsjKUZhaV1awqbT2g2SVop9KkmWnYKJBiJW2 p1e7GwHbpveEmdO0u3HW0isCisA2QaASDB2BqlJkqRQ5wEVlkdqx8qmYrGDhX1UKLpUiB9ioLH4N ia4Vk9WYkKK4rMalkjBZLZ2mKALbD4GwcyW3Hz4qsSKgCCgCioAioAgoAoqAIqAIKAKKgCKgCKSE gAYhUgJSs1EEFAFFQBFQBBQBRUARUAQUAUVAEVAEFIHiCGgQojg+elcRUAQUAUVAEVAEFAFFQBFQ BBQBRUARUARSQkCDECkBqdkoAopAeRHQ9ZrlxV+5KwKKgCKgCCgCikBpCKjNUhpO+tTORUCDEDtX t1oyRWBXICAvcj7lelcUXAupCCgCioAioAgoAtsGAbFR+Ax9hPG2AUkF3TUI6OkYu0bVWlBFYGci wBnrvb297kzzhYWFnVlILZUioAgoAoqAIqAIbGsEhoaGnL3S2NhoJicnt3VZVHhFYKsIaBBiqwjq 7xUBRaCsCOzdu9ccPHjQjSpUV1eXVRZlrggoAoqAIqAIKAKKQBIC+/btc/ZKQ0ODIRChpAjsZgR0 OcZu1r6WXRFQBBQBRUARUAQUAUVAEVAEFAFFQBEIiIAGIQKCrawUAUUgfQSWl5fTz1RzVAQUAUVA EVAEFAFFICME1HbJCFjNdtsgoMsxto2qVFBFQBHICgGMgZnZWbM4v2BqamtMfX192Ta5XFpaMtPT 0waZamtrnSxZlbt4vstmbm7e/s1ZLKpMQ0O9KddyF6efmRnDnh+1NRaThjonU3H5s7m7uLhopqam nH7q6uosLg3ZMCoh1/n5eVdX2OQMOagv5SLqLPJQR5qamsrWfqgrrLWmHaGfcrZl6qu0ZaZel1M/ s7Z/4w/9gElNTXnMP/RD+wEb5KgU/Ygs5Wg/YEI/O2P7OPRSbv1QZ9EPslBvZTPF0NjQhmnL4FNO /VBu0Q9YgMlm2w9lUVIEFIEIgfK8hRR9RUARUAQqBIF5a/z13OqxzsKMNXTqrPG16BzdY8ePBXca xkbHzJ27dxwyNdZZmLWytbe3m+7uQ0Gdboy/3t67ZnBwKO9k42AePXrUtLa2BtUchvmdO3fNtHVc 6m0gZMEFimqtLEecYRpSmJGREXPjxg3HEgcX2dDPsWPHggZoIv30mtu3b+edBBzMQ4cOub+QmFAv rl+/bsbHx11dwXmBTp8+bZqbm0OK4pzbS5cuOZ44/OinpaXFnDp1atNOw2YL0N/fb3p6elwfguOP Y4d+uru7gzp11BXqLBviEagiiIYjdOLECcN+OiGJ4MOVK1ecg4ssYNLW1uZk2axTt1n5BwcHzc2b N127hTd1hb2FQusH+dEP9YXgHfri7+TJkw6bzZZvM7+jD7l27Zpz/HG0adu0ozNnzriA3mby3Oxv hoeHHS44/dKW2U+BvjbkqRLo4u7du/YddCff11JXkKOzszNoW94slvo7RaBSEdAgRKVqRuVSBBSB zBHAYbpw4aLZU7XH3Hfffc4gxUC/cOGCeemll8wDDzwQLBAxNTllLl2+ZPbv3+8MHAqPUfjSS+ec gXz69CmbsidzTIxZtkGZW2ZwYMDce++9eUfy7t1eh8vZs2eDOS8YgFcuX3Flvv+B+6zxWe0cqAvn L5jz58+bBx98MJhziZP94osvOkcSBxvCiXr22Wedfqg/oUYMb1n94PhTPzGEob6+PocJ9RdHKgRJ WwGbJ554wjkqpF28eNF8/etfN48//rgz3EPIgmPwwgsvuPoKLuiC0ctvfOMbLv3RRx8Npp8B23ao nzhvoouxsTHz/PPPOwf8+PHjISBxzixOJacHPfLIIy4gA2OcXnB5+OGHg7Vl+jJ4EhRCFgIz9L/P Pfec0w+ykBaCCEDQlmnHR44ccSzRD7IgJ0GrEERbuXz5sg32Drr3DwFNiLYNVvS/XV1dIURxuoAn Dj5thcAM/S9tirb82GOPBQtETExM5Pta2hBE+6avJZBF+w5FBHqlryUIAtGesA/o8w8cOBBKFOWj COw4BHRPiB2nUi2QIqAIlIoAzsL0zLQzRMUAxnk5duyoGwXifgjCGGU2Ro1dauAbNUxB7ezaZ4bs qND4+EQIUczszKxzaNvbO/IBCBh3dXU6I/TmzR4XCAghDKODU1Y/3Ye7XQACnujnkJ0ZggNz2zrj oQhDFAOd2SBCjBbixCAnI3chiHIzgssotgQg4Iuz0tHR4e7xTAhidJ1y41QzMwRCP3wHK2YCULdD ELxwVHAgkQFCJvRDkIQgTQjCcbt69aqbdcDMByFG/HEycWpCHc2HM8coLn0Kzr/Q4cOHnX6Qk5kR IYg6i4PPDAzpa3F00Q8BAEaaQxBtgyAMbVcCRPBFP+CEfgguhiAcavRDO5YABHwZZWf0nzodipAD edAPeoFow7Qn9IbjHYLoL6iXlF8CRPBl5gwzVQjY0J5DEHWFfp++VgIQ8KWvpT1Rp0P1tSHKqzwU gdAIaBAiNOLKTxFQBCoCAZwFHKiaarvutaHwqKy6umj/g17ruIRwojDycBjYd0GcOQGppdk6D9Yw CxUQwciD3969bSKC+8QwxRCcsUEBZm1kT8vO4GRZSn194b4LTU2sr68zI9Z5QY9ZE84tSzEICmEc +4SBitOLQx6CcKZxHHGafMJhYPkDdQmnLgQhC+0jLgv1hPpCXUKerAkZCATRdnAufWL5EPdxokK0 ZbCnzFIvfFnAiSnuro35NzK6ps+grsSXUKEb0nDmqNtZE2WmfVBHWXLgE4EzKFT7QT8EgWgryOMT +pH3gp+e1fVa7Udw4p0QQj+UmbpCvYgvoaLPI2jE/RDthwAQ9ZJ2DG+f0A8yhHoX0qeATbx/Aw+C EMg6Ojrqi6jXioAisAEECnvgDfxQH1UEFAFFYDsjgHGHs8BSjKrqwq4QpxJDcMEazyGMQNlIDyNQ RnEFW4Ik0KRz/LMdVcbAw/CFkCVOkjYxlf2sjLlZuynm7JzVQ7U1ggv149Ks3tgfgin3WROGJthg fK7STw4nDOcQARGCIZDowi+7pIVw6MBD+AjfuCw4n1Kf/HtpX4M9I5JryQE/2liIUUucXLCJBxOR QdJCYAI/qSvClzQhCaaFkEX6WnjG24/IhhzgljWJforVlRCOJWUV/STJAlY8g7xZE840f+hGZqkI T0njXckzWRP1gMAZmMTriuDE4EGIuiL1IB4MAQPqLTLQ9ygpAorA5hAotOw2l4f+ShFQBBSBbYcA DpJMRa6ypz/ESdJCOLksgYDiRpdLs842p1MsLi7k5XUPZ/AfeMzbjTmh+CghaRIMEHlJy4oWKK8d hWJmPeX3CZxIW1paDBKEkNH8JEwiWWxAxDrB1KmsSepjkiySxrTqrI10ykrQRcofL7fIEiqIB3/h 6csiaegnRBBCHDXhmySL1Cf/XhbXwifuWMJL0kLop5gcyEIdcn1PgPYjsiTpRzCRZ5AtK/LLmySL pIXQD30K8iS1ZUnjfghZ6Lsg0YWPv8jiv7v9+2le038W62tFvhCYpFkuzUsRqCQECi27SpJMZVEE FAFFIEMEMGQwNCLDppCRW1EeLSu3TnkAx3IhGs1HljhJEg7f4mK2Sw/ggWOfJAdySTBg0Tp0WRNG 7/KyOLkJ3EQ/ARwXcV6TcInqTxSEkOcSpE0tSQIda8kCoxByrMdD5BN5UwMgISPhIY6b/4jIQdp6 Mvu/2+y1OC4+X8lL5JNnJD2rT9oQVEwWwS4rGchXeCTJwX1Jl+dIy4qkDghPn4+khZCDd08x/Ygs IZzcqK+Vd2HSO2iPCziGwKWUthG9C7Pdy8TnIe02qa6UIq//O/9adOyn6bUisJsQ0CDEbtK2llUR 2MEIbPSFLgaggyRud4nnb28uL2U/RXhpveBCTr6sZcEw5g8qhudiAExEDidM7L9IPREoGItZk/Ao hgnPFJM5DRl9/STl58uXtSylYIKMWcuxHo+QmCBLQb9CQgIJdgm3Ukvy64qPQZxBKFnifP3vIl8p 2Pm/28y1lFd4JuXhY5d0P400n0cxWUTeNHiulYfPI0kWSQvRlkUW4enLTBp/Pnb+/TSv1+Mh8q33 XFwm+R3p/nX8Of2uCOwGBDQIsRu0rGVUBHYwAvIiZ7RCrlMtbjxAkWrmUWYly52xLNbEK6l0VXaJ SNZUMiYlyrwVeUuRpZRntiIDvw3BY6syxn9fSTKHkCVp1LQcmJRa1lKfi5chi++lYLdVvpVU3lLK EgITX44QgQafX/y6FP2U8kw837S/+zhtRB55ls/Quk0bA81PEdgqAhqE2CqC+ntFQBEoGwK8yDEG mBIp02xLFUY2uWLgPzf4n/+pb2DIxpD5mxlcyPrSpKydbFbGEEaL26TT7rXAXAgfA5FL0kIYT/AQ g034y2ekr2jGRk1NtSRn9in6kfL7jEjjD3lD4QL/tWThXjHsuJ8GrbSfSA/xPEU+wS5+P83vIktS niIH90LoR2Tx+YpckhYCE7+8MroscvAZUpZimISWRbBfD5O1+h4fw61c+/256CIpP5E36V5aacJj LTkkXfSYFt+kfISH8PSfIU3Ss27L5C88hKcvi1wLdvJ9vU/q3WbslfXy1fuKwHZEQIMQ21FrKrMi oAg4BDAOMObYqRqDoZixEIeMYwSjPQ5WDBv/GclLdpH376V9LbtvJy23sNJZdpSzKnGzrjRlwaCq xqn3jD0//+Xc0ocQmCALOl2ySz9EFyIL3+0/p/sQsoh+ijkuyCEGtMiZxaecJpAki6TxTNZOlOAe 6WJ1IEJ0JthlgYXkSVuGpPyS7qeBRwj9iCxS/iRZBDv/XhbXUlfKLYvIkaQfyi3yyXNZYCF5FsM+ pBz0bVIfk3CRNKlPIn8Wn2Ai/YVgIHz4zh/9cQj9yBG7Un6Rg0+RzcfOv5/mNTykriTJImkbxUTs FfAMsfwoTUw0L0UgbQQ0CJE2opqfIqAIlAUBMVBKZY7xUJ87ZksMCvktebEJJAZDCCOQs9nhxYkQ cVq0p1UgT0NDfX5kJv5MWt8xvBpzDh3Hk8ZpPrchpRiK8ftpfkc/GGroJm6s8Z0NNKvt8Z0bNQI3 I2NTU5P7GbNt4vUMWUjD2Q4hC+fTQ0kzfyQtfq69+0HK/6EbwSWuH5GP+hSi/VBe134SNkyVzfSQ Q5y+lKEoyI62DCVtWCdposOCH2bwRWRJOu0hpCxgj36Ep19U9CPthzqVNQkmUi98fpImz/j30r72 24+0W5+HpIWoK/Tn6Ag9xNuypNF2pL37cqZ9TXnXasvS16If+pasSepBUr2VNhVCP1mXU/NXBMqF QPatuFwlU76KgCKwKxDASNoMMcqB84KTuzBf6Pxj7PC3t21vEMcFIxAHFmMnbgTO2jSoo6NjM8Xc 8G/2tre730zFzoRnFgTyMSOjrbV1w/lu9AcYmRiBHE0qzoHkgYHOX1Nzk8WtVpIz+8TQpL4gR1w/ HMlIHWzP4ZaZELmMDxw44K7ECPb5IQvOTai6IrLEd/BHN2AFbmLI+3KmfY1uWm2dBJN4QFFwApMQ Ti71gLo7OTm5qpjghIMVSj/CR44N9QVCPvqcEE4UfNAPuok7dILTvn37fPEyu0Y/ONTxOgtDSQvZ lqkPUkf9QqMz5AwRUKT9UGb6tnhfSxrtGTlCBfEI6FJP4OuT6Ef6Hf9eFtdSD6SOCg/6e2QBD2lj cm8jn5u1XTbCQ59VBCoZAQ1CVLJ2VDZFQBHIFIGu/V3WgTTmbu9dy0eCGctmYGDAOhJ7zMFDBzPl L5njHHV1dTrDZmRkRJKtbEtmaGDQjVJ1dnbm07O8wNhsamwyQ0NDNjizMhtifGLCyQcmNdZoDUHd 3d3OoUMfvn6QDYfm4MFDNj37TTLRz/Hjx52zMDw8nC86Bnpvb6+bBRHKMMap37t3r9OP77xMWP2M jY0ZnLkQI5aAsH//fhecuXPnjgvECDDoC2fm0KFDQRwX+FJXoJ6eHvfJf9SRvr4+52zL/fzNjC5w oKgL4+PjBYEIdEXbxrFBfyGoq6vLBRkGBwcLgjO0H5w89IMDmjURlDl27JgLwKAPIZww6g5BioMH w/S1BHwp99TUlKHNCIHH3bt3nW5C6Ye2Snvu7+8vcP7pYwhCICdBihAEL/Rx69atPDu+gwn6434I oq+lLqAf6q0QslB30F+odyH1gPZKe5EACPIQlKB90843OtOLcigpAopAhECNAqEIKAKKwG5FAGft 6NEj5vbt2/bvjjPYMUz7+vqdUxNilFCwP3So2xo3U9aJioxAnAOcuQXr6J6wDnCIUVxkgc+JkyfM lStXzLXr152hhUOJs4BB1h0oMIMs6AfnEUNY9DOJcWoDM4cPH7YOQxuPBSEMY5z8q1evOmMdxwmj GOfy/vvvd45UCEFwSs6ePWvOnz/v/o4ePerkuXnzpjPQT58+HcxxwQBHlkuXLplr1665UUGcBxwZ 9BYqMAPuBETgTT2l7eCs4Dzg+J86dSpYYAZZ4IcsFy5ccM43Thxy8XnmzJlggRlwgN+5c+dcXUEn OLfIAl5HjhxB3CCEww0/gkTUYdo2uqFNnTx50s2UCCKIZUJABEfypZdeMtJepN6AV6i+VuoDbfni xYuuzRAMuXHjhg1KdwXVD+86sKAdE0jDAUc34AJeoQIz1AH48R6+bt8/1BXqMcFe3kP0N6H0A2/q A/WEtkxfS+CZOszMHgLTPKOkCCgCm0Ngj43KaVhuc9jprxQBRaDMCNB9Pfnkk+bZZ591Bv7TTz9d 0ojN0vjtAslnpmdM/0C/Gx3EuWSkBScmtIFBeTD8ZKSf5QZdVpba2roCeUN8wdhihA5jHWOZaafM kuA6NDEK1d/fZ/Uz7wxS9IMTU0w/e+pbzZ669JeNoBuCQ+CD4Y6jjdEempiqjH5GR0cdazBBRyGm TMfLin4IFOHg4jDg4IZ0WkQe2g8jlASHcFioIwSPNjpaKflt5ZNZGKIfrsGDuhLKgfJlx7HFiaMt w1/qSrH24/8+rWvp32g/yIR+wIS+NjShE0bamXXANX0bspSj/VBX0Q+ON/0rAQjacmj9oAMJUhFc lRkqzNYILQt1Bd1IX4vTT79Sjr6W+kH/xrsZHAiobbavfeqpp8yXv/xl1yd98IMfNK94xStCV33l pwhUBAK2Le3RIERFqEKFUAQUgc0ggKGSRhBiM7z1N5WNQFZBiMoutUqnCCgCioAiUKkIaBCiUjWj coVGgBhE+CGt0KVUfoqAIqAIKAKKgCKgCCgCioAioAgoAoqAIlARCGgQoiLUoEIoAoqAIqAIKAKK gCKgCCgCioAioAgoAjsfAQ1C7HwdawkVAUVAEVAEFAFFQBFQBBQBRUARUAQUgYpAQIMQFaEGFUIR UAQUAUVAEVAEFAFFQBFQBBQBRUAR2PkIaBBi5+tYS6gI7AoE7B43u6KcWkhFQBFQBBQBRUAR2H4I +HZKOU6a2n6IqcQ7GQENQuxk7WrZFIFdgIC81Hmhy/UuKLYWURFQBBQBRUARUAS2IQLYKmqvbEPF qcipIlCTam6amSKgCCgCARHgJT4yMuLOWK+trTULCwsBuSsrRUARUAQUAUVAEVAESkNgcHDQ2SuN jY1menq6tB/pU4rADkVAgxA7VLFaLEVgNyCwvLxs2tvbzcGDBw0zIaqrq3dDsbWMioAioAgoAoqA IrDNEOjs7HT2SkNDg6mvr99m0qu4ikC6COhyjHTx1NwUAUVAEVAEFAFFQBFQBBQBRUARUAQUAUVg DQQ0CLEGMJqsCCgC2wMBZkMoKQKKgCKgCCgCioAiUMkI+PaKf13JMqtsikBWCOhyjKyQ1XwVAUVg 2yCwtLRkJiYnzfzcvJ0iWWeam5vLtmkU+1qMj48bYisNDfWmqampTDgum6mpaTMzM+OWuYBJTU15 XhkYa2AyP2/1U1dvmluarH7KE0NHhtHRUaufZacbcCkXsaYYXFiG1Nraaurq6solisNkdnbWydDW 1uaWR5VDGNoP+llcXHTtmPZTrg3g0M+k7Vcg9FOu6dfU1YmJCbcGnTpCnWUPnXIQ+hkbG3P796Cb cva19G3gQv1AFtbpl4PQD/VkamrKtZ9y6od3IX3K3Nycq6/U23K1H+oKez5B6KbcfS11hWWf5e5r y1FHlacikAUC5bEosyiJ5qkIKAKKwCYQwPi7ceOmdeSqTGNTozV6hq2BvGiOHDliWlpCOpjLpr+/ 325a1ecM4tq6WtPX12uNrwbT3X04qIOJ8XfTYjJtjfSW1hazbA3TWz23zP4D+83+/V0W5XDHoWKY I8ueqj3OEEU/S3eXzaFDh5wxuAmVb/onfX19pqenJ+/E3blzx8l04sSJoPohEHL79m3DJmcYxBjG N27cMKw3PnbsWFCnAUfu8uXLeacf2a5evWqOHz/u5Nk02Bv8IY4cDgu8WW/NH/ohcHbq1KmgwTwc uVu3brn2jOOEDNevXzf79u1z+gkZzMOZvHLliiFA1NLS4px/nEzqCXvphKTh4WFXTyk/Tn9vb6+r u2fOnHH6CiULdYV2PDAw4NoyQTzaD/sL0ZZpT6GIvvbixYtOP7Rl6g76Onr0qDlw4EAoMRwf6gVt GTyoK7yPCOaBCYHFkETb5Y/2Q+CM9kSdoV8JGcxDH/Qp9C179+51fSttGd1gI5QrQBNSF8pLEcgK AQ1CZIWs5qsIKAIVjwAGIAYfRs1ZawhXWeMLA5W0S5cumgcffDCYczk2Nm4N4R7TffiQOWyDDtD8 /Jw5d+68HZW6bu6556xNCeH8L5ubN2+aUTtaef/99+WdAwz2a9b4woHo6Ohw8mX9HwYwRnFdbZ0r fzT7YdlcvnTFpl8xDzxwfzCDFCP00qVLzvDEKIdwwL/+9a+7z4cffjiYQUoAAh09+uijeeeAAAnO DEYxDmYIoq1cuHDBOVCPP/64qxvSfs6dO2ceeeSRvHxZy8MoJeXHkbznnnscDrTvF154wZw/f95h FWrj2rt377qgw7333pt3JAl2Pvfcc87BxZEK4byIQ0vdpa7gxEE4dNRlZkMQGAlBBBPRA33H2bNn naOLfM8++6zt486Zhx56KNjsDIIf165dM/fdd1+BfpCFPoegSAj90FbQA3X3gQcecI4/uqB90664 HypQRFtBDwTveO/RVuBP2osvvmieeOKJYO9C6iv9Pu2EP4hgGn0twTT62lBEcIr2TPshSAQNDQ05 TAhWHT4cvatDyaN8FIGdhEC4cO9OQk3LoggoAjsCAYxRjJvDh7tdAIJCYXx2d3c7YxTjIwRh7OEY sPxif9f+PMta63x3dnW6gMDw8Gg+PcuLiYlJM2iNrC7LF4NUCGelwS6FuH79hjVOlyQ5009Gwmat fg5Zfawsv9hjDh46aEcMF52xnqkAXuaMhhGA8Y1O8MHhx2hmVkIIwginrjDrwR+dZGQOB5x71OkQ xEgpU+sJyoANRPvhO04Mo83U7RAk5WbWgziQyMSIMk5eqLbMTBDKzUjy/v0rbZkRXdoQ95AnBOEs US9xZCUAAV/6N7Ah2IrTnTVRB3DmwAanUoJBOHEnT5500/9xvEMQbUP009XFrK6I0A+40OfIEhq5 l9UnbYfgLvWE+iKEHCw/INAYisAEbHz90I7o3whQ0L5CEHWFvpb6ymw3IWZDUI/pa8EtBLGcClzo ayUAAV8CaXwHE7BRUgQUgc0hoEGIzeGmv1IEFIFtjsDiYrQ2ubqq2u4zsOJsUywMnurqGjNgDfgQ DjfGDvsvMDIZX6vd1GhHL61hNjjQHwRxHBf4YZT7hMPQ0Fhvja55awSO+7cyuQZ3DM4aq4c6uzTF J2auMDsCOUI4UTglOI7UC3G2RR6cB4z1UEEIZjxQ5rh+kIc0HD1wC0E49jgNcVnAiHqMHNTtrAk8 wJ96EW8/Ihu4MfKeNbEfBYEiAkQSDBGe1BVkwPEMQbRl+PkOFHxpy6ShmxD6wVFjKQbBh/i+CxJI c/1OAFBwYCkzTi44+CSBgFBtmSA47Uf4iizUG+RDzhABK9GP6+O9wDPy0KZoz+gHWbMm+lrZG4P+ 1ifaMjKAWwiincIv3n7QDzqjnYeqtyHKqzwUgdAIFPbAobkrP0VAEVAEyoTA7Oycm0bPXgP8+YSR gUG2ZJ2b6ekZ/1Ym1xibON1xBxdmMmo4bQ2erI1A8henRPj6BRb5pmeydyzn7CahONToIe4suDS7 hweBpBCj/uIIwDfuWIITaRjPIQIirNv260WSfkIEIagrMiKZVFdIA48Qo8pggrMdrydgI2ksnQkx ainlTcJE2g9OVgiSeit8fZ6SJs/499K+pk8Be+Hp5y9pYJJ1/wbfYu1H6koI/VBfpa4IXx8XqT/y jH8v7WvaBn/IIXyFh7wL6WdD4EJ5wUbqhcjBp6TR74SoK9I2hG9cFmQIoR+fr14rAjsJAQ1C7CRt alkUAUWgZAQwqiKnMQo4xH9YZR1LaHZuNn4r9e/i+McdXBiRxlKExYXFzJ1ceMzP2+mluSBMvKBV dtYINDeTPSY4LUtLy678cSM9wmSPM1ZDBCEY8YLiBjppojMCJvxlTTgLUBwT0kSWKKiV7aglmGCE iy7g75PIJ9j599K+FucoST+SRn0KESQSWaT8fllFFtGhfy+La+EjfH0ekibP+PfSvhYewjOeP1jh eIZoyyJLMf1IfxyXM83vfnmTcBH5QshCvyUBOulDpKx8RxbajmAn97L4FKdeyu/zkL4GebNuy/Rt Uh+T9CNpITDxMdBrRWAnIaBBiJ2kTS3LrkKA9dh/9md/Zj7/+c/vqnKnVVhxGIk18OeT+5pLy9rY ge9aBiD3RDZmSiwtZjudfMnygE8MDsRwJAbqgjVIsyaMdCfLWsLkBJi3y0OyJtFPEh8xjKknIeqK 8BBd+DJJGs9kPVKIfoqRyCLtrNizW70n+hGea+Unz611P4104ZEki6TJM2nwK5ZHKXUlhCzF5ED+ kLhIvRWePn6SJvL699K+pn2KLEl5iyziCCc9k1aalBeewjeeN/KGqCvr8UA+ZBGZ43Km9R3diH6S MJG0rfRvkkdaMms+isB2Q0BPx9huGlN5dzUCbJr1b//2b25N5Fe+8hW3s/jP/uzP7mpMpPAbfaEX NWLE87eZL9vR+KxJjJ01+eQc8awdS/IXHsXwZIZC1lSMR6SeCJSQ+imGiY9dVtisx0PkEx1mJQf5 SvsRnmvxCiFLMR6+fOu2s7UKsYF0waXYT0p5ptjvS7nnl9XHIP5b/7n4vbS+F9OPzyM0Lj5v/xp5 +SuGm//8Zq6FB78txieEfnweSbJImv/cZspcym+Eh/D0fyNpPnb+/TSvkQM+a1FcFvm+1vOS7j/n X8t9/VQEdhMCOhNiN2lby7rtEWCq7/PPP+8292I3fHlhb/uCbaEAjESAC9O+N4JHlV3ikKe1bQ1r IeafyuyiKrYnxSpGIl/GshQaRcJ0lTR2em7GgliWpfIolHm1rGmkJE0NjudbyjPx32z0+3plFaOZ 59Z7dqO848+XWt6s5YjLFf8umJBeqszxPDbyvRQepTyzEZ5Jz/o8fAziz/rPxe+l9b3UOiBT3NPi m5SPyFIME34nzyXlkVaa8CgmS2j9FJNF5E2r/En5lFJe5MhalvXkEJw2Kgt2CvYKyzhCBN2SMNY0 RaBSENCZEJWiCZVDESgBAc4vf9/73uee/OM//uMSfrGzH8EAYLoqL3SmcW7EMKnNnbjAYAfutu9W OwMj54NzOkPWVG13H1+LZDCGsq1nGK2VR6npjocNzlB04ev/VgyvrOWAJzzcsZw5PRTKwbfoRgjH RXhI+QtliUZNeSYELuvJgmzllgMZBKukTd24nyYJD+G5Vt4hcJHTOZJkkTTR4VpyppUOHxwd4evn K2khZBEewtOXg2tJl+fi99P8LjyEp5+3pMkz/r20r+lri72vRBap22nz9/OT8gpP/x7XpCNrCFmE R5IspIksxbCLy7+Z7+Qv/UWSLJKnYCff1/skCCH7SKy39GS9vPS+IrDdEfCGArd7UVR+RUAR2G0I YBxwVNa+fftMe3t7UaMujg1Hj0VGxsoSBP8Zm+q+1tXV+8mZXDfUR0eEJs3kiAyvJbcpYk1NtDFk JkLYTDEAa+tsQMQZe6vX/It88aPTspAHWZgNEe1TURiJWM5NlcVQDCELdQVKGrkSTBx29ljKrKkh d4Se8PX5SRrPZG2kgwk8ovpZqB9kElkEO1/OtK85zhASnn7+ojOcBXFw/PtpX8sRlMVkER2mzTue n/ARDPz7khai/RSTQ+oPdSlEXREe8I2T6Ezkjd9P8zv1UbAXXfj5iyxSn/x7aV8jhzjcwld4iH64 L9jJvSw+i7VlkQXsJNiXhQzk6Zc3ST8SQBAdlioHR9Jir/C30d+WykOfUwS2CwJrD79tlxKonIpA IAQ4m/r27dvOkBUDX16KvLg5w/qee+5x0nB29PXr153jyMvMN9Z5FsPi5MmT+ZcQSwouXbrklhTw rJ8/v3/kkUcClXL7sUkyJkspRb01vDCqwB4jwx/RIM9FuwkkI/FNTVGAoJQ8N/uMnFmfZOxIWqNz LLONG1Pvmhqb7DF2E9EpGbECySZcTU3NsTvpf62zM1VwGin/gj21wz8yftG2ITbprLandaDHrIm2 DSELdUPaJ2m0Z2nTWRvG8EMW+hfRBWlCYhjv3btXkjL7BAMCgBx5KHx9ZqTRpsSp8O+lfY1hjzxr yQE/2noIox9MoKQNBSVNnnEPZvgfdYXTBoSvz0rSpG7799K+Bnvqwlp1ljbFO9FvV2nLIPlJfUyS RdJCYMJ7HVnQT7F6G6KuoB/wp07Qx/nBOnSDfPRtIXChvGCzHiYh6gqyDAwMFG0/W9EP2CopArsZ AQ1C7Gbta9lLRoDpc7//+79v/umf/qlgtA3ng5chBtb3fM/3mHe/+93u+zPPPGM+9KEPuRc6THhG nBde8Pfff7/59V//9bxR3NfXZ97//vebW7duuWflBcvnkSNHzF/+5V+WLKs+WBoCLIFoa20zff19 zjj2HRQMoMXFBdPR0WH1ka3jj7QYo7W1dTYINeeML98IlKmbe+1MjxDUZh3YXlsfWbfKaI0QdR0j HQOxtTVytOReFp/g3traagYGBy0mnIDRmGeDHPwhR01t9q8xDE0M9SQjHZwg6koIYi+Ymzdv5qf0 +jw5zo+6w6ygELR//34XhKCO+sY4uqENob8QI7n0vwReRkdHXZ/LdyFpP52dna5vlfSsPgmIoAM5 atDngyz06aHqCnx4t0gd9WUhjTrt682/n+Y1fSt1YWRkxAXa4Ss0MTHhLv2+Ru5l8UnbQB6pFz4P SQulH9oyTq7wjcuCnCH0Q4CB9sPG17RdXz98JzBB+6Hvz5oIdMCfZQv0If67UI4rBbcQJP1ovP1g yyELuG203mrgIYTmlMd2QSD7HmW7IKFyKgJFEOBl8wM/8APmAx/4gHnNa17jDEyMJwyrd73rXS6A 8Pa3vz1v5L72ta81v/qrv2re+c53upc6z2KUvvWtbzW/+Iu/6IIVvnGBMU8A401vepN74WOUPPbY Y+app55yeRcRzQU3it3Xe2sjcODgAWdYYXwtLcmxk8vmrv1eZUfZu7u71/5xindwTA4f7jYzszPO KJWsMf44ihXDrKurS5Iz/cSJasP5t8YxBqjQ6MioM7wIivlOntzP4hP8a61DhyNl14jkWCw7TFgu c9jKUribRxZSREHE06dPOzwiWSI+GMnMjkI/oQxjAlb0F4M2OOM7ujh4OOHIEcLxBwHkwGG4ceOG 67cEfdoT2Bw9ejRYXTl8+LBry1evXhUxXMAY/SBjqLaM4wgv+vzh4eG8LDgy6AxnjvdGCKLPwKGO t+W7d+86J+/YsWMFTl5WMtG/nThxwvGibogjxid1h/p66NChrNgX5MtSC/SDPnz94PRSV8AslH7Q DbyYZQl/IfSFfMePH8/bFHIvq09pP+iDgDOEfnp6elwbpi2HIOoK9ZLyg4sQ70L0A14bdfwlj41+ wksCRRIAIQ/6WWaAUY/8gM1G89fnFYHdjkD2Q0i7HWEt/45AAKeL2Qv88YL+7Gc/6xwSXszveMc7 VpWRFxd/GF6f+MQn8i9TAhZ+ZF9+iOH66le/2pw6dcrlTUCDAEbSs/Ib+RSDTr7rZ+kIYECgI5bO 3LzZY5qam8zM9IwZHRu3DtSRYM4cEmNY4VT29va5OkbgC8eypqbWGqPHghmjGIEnbT28ePGiuXLl ipMLAxDne//+Lluv95cO8BafRD8YpDdu3DRXr1wzrW2tzjgds/ohXaZWb5FNST/HYSAAwywEZkTg OBEgYnSQDWNDBWYQln4Cnbz00kvOEEZnGOiM3FGfQxH9Fn3iuXPnzPnz553DizOF84ADJSOJIeSh /RAounbtmmsrBIZwFnAe7r333qDOAu8FAsmXL1927wFwAhNkCllXqJsn7bI/dPPiiy86WQgsEoSg Lh88eDCEahwPgpu0WdoPMvAd/VBfwCRkW8bhpq+9cOGCDf4edu9Z+jf0Q9sKMeIPKLRblnCiG9oQ NgPtmhmROLh8D0X0Z7QTMKHfZ5BEAjVgElI/lBv9gAPBTAJH9LW0o7NnzwZ7F4I9Zactox9pL/S1 BBNDBWZC1QHlowiERkCDEKERV37bHgFexjgcGFIylXStQuFI8ic0Nja2ZhSf4ManPvUpZ7yXGoCQ fPVz8wjgvGAQDw4NugAEhs5DDz1YUgBo81xX/xKDFMeN0WVG6DB8MHQ6OthwM+ykNTB48MEHnRxT U5OOP8ZyqBF2Hx30s9fqZ8DqZ3pm2jmTDz74QHD90OYxSDFEZaQSYxl9obuQRHDmoYcecntD0KdA OBDU49AEz5e//OXOyabOEjh92cteFtTpp8zogNF0gkU4lAQfCILgtIQMECELGBCcwcmmLSOLBGVC 1xWcySeeeMI5cTh2YEHdweEOTThtzDSg/SALI80hgzJSXvTzwAMPGNqO9LXIRn8bmrAn0A/7vGBP UD/YAyqk0y9lRje0GQJmBCDoZ5DNt2Hk2Sw/wYCAIu2Z4AN1hX4X+UK3H96FzEpFP9QXBn3K1ddm ibnmrQiUAwENQpQDdeW5rRHAEcOIgXg58lJa68XICBQGlxBGKU5VEvHcJz/5SfPzP//z+fyTnoun 6UyIOCIb/44+Dx4INypYTELqVzmc/bhM1Gnq6lr1Nf58lt/Zv6NS9INzgENZCVQp+mHkmJHbSiCZ PVMJsrDOPsQmoeuVlbYcclS9mDyMalfKCDIBtHIE7uL40H5wsPkrN/EuZJZMJRB9bciZXcXKXCl9 bTEZ9Z4isN0QCDu8tt3QUXkVgQQEGFmSoAOzIfy1gv7jTKX/0z/904JNp2TU0n+Oa/LhWUafWZax EZL1mxv5jT6rCCgCioAioAgoAoqAIqAIKAKKQDkQ0JkQ5UBdeW5rBPyZEKzfZIOipKmTn/nMZ9wU PkYI2YwL8jfC8kFgM7UvfOELbjPLUOtRff678npxziwvzu/Kou+KQtvRXyVFQBFQBBQBRaBSEGiq XTJtDfZo6YYqU700bZbnxitFNJUjIwT2VNkl2TXZH/WekfiZZqtBiEzh1cx3IgIEHGQ5BpsmEYSQ DYukvAQbONLzPe95j9vnQYIQzI6IE4GMj3zkI4YTNVgLWozI94UXXnAzJ5599ln36Je+9CU3vZV1 m9/0Td+UGBApluduvUcAYnl+crcWf3eU2y6VUlIEFAFFQBFQBCoBgaYaG4RorLL7beSCELMahKgE vWQpA1bIntpGs6chzDHeWZYl7bw1CJE2oprfjkcgPhMivsSCzdn+6I/+yDz66KPm9a9/vfn85z+f x4TNjeL0uc99zu1W/lM/9VPxW6u+E/T48pe/7E5PYKdxjg1lT4ivfvWrLo0Nk5JmZazKKIOEf/iH fzAf//jH18xZlrDEHyD98ccfNz/xEz+RX+YSf6aU72vlX8pv9RlFQBFQBBQBRUARUASyRWBlht4e s3KdLU/NvdwILM9P20CE3Qy4uq7colQUfw1CVJQ6VJjtgAAzDthcC2I/hvgJGRxtyBGev/u7v+uO lPJ3IY/PhGBjyw9/+MPmLW95S0mb3bET/7vf/e6KhIlZGKVsIuUHC+SaHe3leqOFk9/xKdcbzUOf VwQUAUVAEVAEFAFFIEsECm0UDUJkiXWl5b28OGv2aBCiQC0ahCiAQ78oAqUhwOaUkCzHkF8x0+HP //zPzate9Sp39Bjp8izXLN0QYgbDpz/9abc79xve8IZt70ATIOEvJEnggWBQ4cs9pBTKSxFQBBQB RUARUAQUgeIIYKcsW3vFmOXgRwcXl0zvKgLhEdDTMcJjrhx3AAISWJCNKaVIX/va18yFCxfMj/3Y j0lSQRCCmQ/8BmKfiD/5kz8xP/7jP+7O5s7/QC82hAABCIJBc3N2o0ndA2BD2OnDioAioAgoAoqA IhAGgaWlRbNgbcCFhcW8LRiGs3JRBCoPAZ0JUXk6UYm2AQJytrg/E4KlFn/zN39jvvd7v7fgHPTW 1tZ8iTjOk98QiPirv/orwx4Or3zlK/P3t/NFb2+v29tiM2VgBsXhw4c3/FOCDows1NXVGT1VZMPw 6Q8UAUVAEVAEFAFFIBACe/ZUGZb0Rpub68bJgWBXNhWKgAYhKlQxKlZlI+AHFmRjyo997GNu1sMb 3/jGAuElYEHi1NSUC0B85StfMcyaeN/73lfw7Hb+8uKLLxo2p9wMnTlzxjz55JObWlKhsx82g7j+ RhFQBBQBRUARUARCIrBsl2EoKQKKQISABiG0JigCm0BAlmPwUzamHBwcNH//939vfvqnf9p0dnYW 5OhvTMlMCJ7/u7/7O/OKV7zCnD17tuDZcn1BLjbb3Mq+Ct/2bd9m+FNSBBQBRUARUAQUAUVAEYgh 4MUgNCARw0a/7joENAix61SuBU4DgfhMiL/+6782L3vZy8zLX/7yVdkTsKiurnYzIHD2/+Vf/sUF Ir7v+75vS07/KkabTGBvit/6rd8yb3vb2xLl7+npMYcOHcpNH9wkk23wM2ZU9PX3mzu3bztdlkNk NqyamZ01k3bGDNf19fVudk05lpqwbIiZO7NWHvg3NTWaxsam8LBYvczNz7s2g0wsvWlpbjY1dkpr aFpeXjLT0zNmfGLcLC0uWTwazd69bbYdh99eib1Q2OiWPmVP1R6LSYtpbkY/4Xdcn7f6YUYY+7Iw 1Zg+T04QCqujZTNj9TNmA72Ltq4gQ6uVpVx1ZWJi0rUhMKCutLa2lKWuyN5F1JWa2hrTZNtxeeqK 1c/MrG0/E2bB1hn6tza7XLE8+ll2upHNotFPW1trefRj9weYmJxw8mAr0JYbbX+7lUGBzbY7+nsw oS2jn9aWVlNXH/5YwXxfa2Whr2tAP7augE9o4r0zOTllpqanTJVd/snAUkuLPW5xU32tF4UIXRDl pwhUGALhLacKA0DFUQQ2g4A/E+Ly5cuGpRjf//3fbx2Svauy84/0ZN+IP/zDPzSve93rNrUHwqrM U0j43Oc+Z5555hkzPDycmNtOXu6AcdHb22cu2GNVn376afPffvu3zef/6fOJOGSeaJ3t/oEBc+P6 dedA4dgO2O83b9ywhvtM5ux9BvPWAO252WP6+/qcIYyTefPGTdNnsVrKbazqP5/lNQ7LdYsJm7pi gDKT6OrVq2Z0dDRLtol537l711y8eMnMz827wEy/DVqdO3fBBWoSf5BRIvq4du2a0wksFheWzLWr 18z1azeswR5tfJsR61XZ4mifO3fecDIQ+pm2hvr58+ddQI8d4EMR/VR//4A5Z3nPWGcbWWg/5y9c tE7VRCgxHB+cphu2vaAjglVLVrabN2+aK1euOucupDCzs3N2s+SLbr8egokEIpDj9u07diNfdukP Rct2xuCQqxtTubZMnaE9xY+5zloiyn3D9qu8u9EVdQc8rto2RHAkJOHsX7x00fK/7fraWfsdudhj CdlCEn3qRVtX+KT9jNrA4nm70TbfwSgcRfrguHP6Oupt7527tt5ecfU3nBzRCWi8f3ps+yW8S0Dv 2rXrtu/tcft7hZRFeSkCOw0BnQmx0zSq5QmCAJFwXtK8kJgp8KY3vck88MADibwZjeN5nChGGI4d O2Ze//rXJz4bOhEn/F//9V8d20cffTSRPfLuVHrppZfM//gff+ZGKJuamuxo7rjpPtRdluKOWEMP p5YNOjs6OpwMnfs6zKVLl1wdO2v3zbBWavayWWPz1q3bZnZu1pw5fTo/Sgk+GO7VtdWmc1/hkqOs hKJ+3rLtq8m2n6NHj0Yjgzn5bt26Zerr6u0IWUNW7AvyHR4esbNk7prjJ46Z/V3RUbRdXV2GvVBw pB566AErX5i4/q3btwwBzYcefMjUN9Q7ORnFPX/+gpsVcfz48QLZs/qysDBvA0TX7CypancksczY qa2tc8EzRlL3trVlxb4gX2ZiUCdYDid1hQ1vL1267IJWDz74QLDZXMwuoy2z8bDsCdRl5XrxxZec jCdPnozqckEJ0v8SOUxXnbN0//33uVkqcBluGXaObp1tP11dYdoygSCct0OHDro+jlH+AwcOuNOk cC7vv/9+N8spfRRW53jHOrQsoTxz5qybycQTtOXnn/+Gw+qee84Gacs49gRU0RNLM2kvEDO9XIDE 3j/cHeZ9xAyIy1YP7XYg5fTpU1YKe5Sk5U+fzzsI+4Z3QAgaGBhyQbMTJ044vcDzwIH95oVvvJhv y5ubhbBx6QkeYhc88MD9+dldtGkCrbwLD3dvfEPtjUuhv1AEdiYCYSymnYmdlmoXI8DUTaaFQyzN +Mmf/EkXlEiChOd4HiJw8Y53vMPs27cv6dHENEYC4oRxgINWCjGakpQHv8UQw3DHaMcg3G308MMP m//yX37dbRBazsAQswv6+/uckcOUU6E9dgSow9YVZkIMWaczBI3ZQBlTgwmE+NOkZYp9791wI3R3 7cyDRVt/Ozv3rTht1nnp3G8dJxuP6e29GwISZ4wzEkZb7miPAkQwpj1jhM7MTNuZAMkzidIWkLow MDDo+hAJQMAD/XTYoBWjy4yuhqBBy4uR9W7rKEkAAr44/9XVNeaudfZCjaDKyPHRo0fydQVH9+DB gxaPWTeyHAIT+lpmZLS2tuUDEPBlRlxHR7vTjywByFqesdExt/Sh0wYa4C9E266vb3DBzcXF0t4j 8tvNfS47p5IZCAfte8ZfZkD7ob6ivxDk+lI7IwOHmqVUQuBDgIRg1sT4pCRn+skMA2aBYA9IAAKG 6Ke5qdnOPOsN1H6WHf4sAaS9iIOPnsAECqUfAjJ3795xS4ba29sdb/4jwNtlAxEsERwZCTMLjoEj bCTarb+8jEEl7L7+vv7gs+DygOiFIrADENCZEDtAiVqE8AgQVMBowKD5wR/8wXy0PkkSty4551i+ 5jWvMd/+7d9eYIQl/YY0Xsaf/vSnzRe/+EU3LZNAB04QywZ4MTLa9tBDD7llIDJyLnlh+F+zU4H/ 8R//0RmaGBOM9LzlLW9x+zt8/OMfN1//+tfdzAzywQh6//vf75yJH/qhHyowniXPnfjpG8PlLB/7 QMzZKf4tLfWmuqawWxbjdNwax/tyMySylJURS8g3uviOk1lnDXXqPIazjPByLwtiWYFM266xDq1P NVXVpram1kxbWWgnBAOyJBxtZoY027XacV7MxMBAHhkdWbUpbRYyDQ4N2lUOywVOi/BpsGu3h2xw EueGQEDWNGSdOUjqqPADo1q79wBGPMszmqxDlSURkKXe0j/GZ6PU1UXONzNH4sGSLGQiwMAMkaam lWCV8CFoRN9MXcm6/cATPtSVxoYoCC5y8Nlk6+2wxWRm2tbrlsL25T+XxjWBGfoM6oUf2CRv9j6A 5JQp9yXD/6iTM7MzprN59UAAgRlobHzMtNqZRVkTAUPqQ7yvhS/7MExORXuK4PRmSVH7GXf6oQ35 RDCRP9oXAxp+sNF/Lq1r3i/MyqB9RMdYruTckJspAm5+gGLliXSv6EchGUSS3LEbSKOtU29D9LXC Wz8VgZ2EgM6E2Ena1LIEQwCjgJc1023ZC6IY8bLieT7f9a53lfwSv2DXYn7pS18yP/IjP2K+8IUv mF/7tV8zH/3oR90oCYEPNsH8yEc+4ja69PljUHzyk580v/zLv+zWRL/5zW8273znO02fXd//B3/w B+5Rlljwe4wKnudUi2/+5m82jz322CqHgqmYodem+uXZDdcYXhijVQnT+cXom7OGGc9kSeQ/Zw10 KO5sR2mRs8I686xpfn7BLNgAAxuBMSPEJzDBECQAsdYsH//5rV7juEBV1VYWy9cn0c/MlA2I2LaU NU3ZDdIg4evzq7LBGSjESDt9AqOSFhBbVwr1gwzVVhYCSWzkmTWxaRyj7Ely1OQCVDg21JesyWFi meC4xUmCaWyeGYLABWK5TJwk2DlhHd2saXZmLhcsXI0J/QxtyvWBtk5lTVM2oAhVJ2GSq8fT1OuM ifZDmSFptz5L6X+lPvn30r6mP5+zf+gh3q9IGjNmQrRlyst7SMrvl1XS3MbNGb8L2c9GsI8HQ5CJ 9oOcBKiVFAFFYHMIrH4jbC4f/ZUisKsQ4LSIt771re4UhfgshDgQBB/e8IY3mO/8zu809913X/z2 mt8///nPm1e96lV2feZp97J7/vnnDbMUXvnKV7rf8PnbdiPFr33ta+a7v/u78/lw+sZv/MZvuH0n 3vOe9+Rf5hwJ+pnPfMY9R8AB+spXvuKmcTND4tQp1oGuJtaXxw2T1U9pylYQYDMyqMqechAnjED+ WJbAsg1xHuLPpfEdJw3HH7IsV1FVzkgP4fgjiwu62MAMgYgCymHCfWRJGkkseH6LX2R5Q1KQKNKP 3cDMGukLdpf7LPVDMfKyxAIz3JNZADjcYINsWRGbc67wWM1HAkcib1ZykC/LYaCk8oocEnD1lyW4 H6X8nziWSbJIPxoCE4rFMhQoyckVWdiENmtiFhGUFCSKeLP/wJKbDeYvMYrupfu/BFmlrfi5S9qs rdtZE/VR+tHEfj/3LpjNBSqylIeBiEUbMKzKBYTivOh/56281KfoVJX4E+l9L9Z+OAUIHS3afpaA b3xWTXpSMIFoZcmr9CF+/tJ+QrVln7deKwI7BQENQuwUTWo5giLA5mc/+qM/WhJPouhvfOMbS3rW f4jfPfLII+aFF15w0fYf/uEfNt/6rd+af6Snp8dd+3s58AL/8Ido1REFAABAAElEQVQ/7GZdcAQo I7icekGggmNE3/e+9+V/TzobgjGVsNhGdvHpmfkM9CI1BDCqHCU4jSTxh1G0tGRHiFLjujoj1gQX cyzlFyHWkSMHf0nkMHF7lUe7lyc9k2YaDgOU5FgKH57BkM+aRJYkPlJ9qCeM5FmJkx5LJW1FjmQe IkuI2QfCI0k/fhonVWRNwsPnKzwljWBVCIrqQdR/xPmJLCFOU5G6Yl3IuBjue9S/2SWIQdpPrk9J EGWlzjKjKdv2U9i/rRZGsJoPMLtqpf2sriuunljxkFeeS1RiSonCQ+qnn62gRMBK6rZ/P81r8l+p t6tzFllE3tVPrJ8iOl7/SX1CEdiZCGgQYmfqVUu1AxBgGQbR9k996lOuNK997WsLnKCvfvWr7v63 fMu35EvL8g13goEd0fi93/s9t2SE9drMwPjN3/zN3KZT0eMcO8USDfaokGmO+Yy24YWMTGxV9JyJ utVsNvR7jKr1KDJa139uvXyK3bduvyEQAYmRlfT8WsGBpGc3myaYJMthU3M3QspSrCzES9aImRT7 2QbveYGZJGDyaTy32qHYILOij3PsJITjluQwyI/D6Ke0VrtUQjsTuTf7KY5Lsd9L3S72zFbvRTzW xyVrZ45y5OtAvn7GSxfdyL79rMiS5AD6achcrF7HS7CZ71JXJPjh5yFpyy6g6N9J/zqvn3WyLvW5 dbIpelswWfOhXB3KWhbyFx5J9UDqijyX9ExSGWS2TXRvzQaR9FNNUwR2HAIahNhxKtUC7RQECAww TfK5555z082ZFSHEi4+jNZnBcOrUKUl2Z40zxfM7vuM7zM/8zM+4jZ2Yqp4UZOCIKTZeYm+I7Uys 22RzKKZZr2vAlFBQMTxKeDS1RwoNk+RsnZEjlmnyI1tOdYZVjkcx90UMsC0zLJKBYJIsx0pqWsGn IqJYZ2T1ngfx59FPxuqxLOEhVnhcAvs9B0sIWfJiWJ7F2kzSVOYEybeUlMdknVyC1JWEJVVxsUqp T/HfbPS744GS1vHsS8Vuo/wLns9XloJU70tUceUx70bql8Ijqc5atzPiZx8KgYvwSFJRTpLEJXpp gyJyuHxhvIZvXPBc2kLk8lu3jYp8a8iYlliUVcpbrK74z5XCm4162aQVOy3EzJ9SZNJnFIFyIbC+ ZVUuyZSvIqAIuA3mOC/8wQcfLDhijQDCFbuUgj0j2HOCjSg5z1qIDTP32vO+2RBTAhC8SP0N6559 9ll3KgbHc0Ich/jMM8+4wIfksx0+WbZCGTdqDFRS2fIbxyVYoyS5ZGsUrWugbbFQrAmutrNvqCt5 g9zLU4wxnsuaWD8eGYFijq9wjDCJ0qV+r9xN/0o2sUsaSRdMkJfNGLMmKe9qVFaCAdEeANla6bJZ W1I9AQPBRTaGzBIX2edBePq88mm2/Qh2/v20r4vJIkHSfHtPm3ksP8E+j4F3X2SRuu3dSv1SNuRM aj8wc/2b/UzazDNtYYRHEiaSFrXjbNsPfZv058LXL6ukVdvgetYk7cL1qzFmTg7b2SBv1vvdwDrf r0il8OSRJCdLwp443qNbvkQ3RfWTC1gJdqUylHzd/ihSoFJ/rM8pAjsMAQ1C7DCFanF2FgK3bt2y Z873u70h/JJ99rOfdV+/67u+yx3XSRCipaXFnD171p1/7j8r11evXnWnY2B8Ylhw+gZHc3JsHfQX f/EXbg8JMQLkd5X+yXITgi2c+x45rZUu8Wr55IjDpKnRTKtGXwQHNmrwrOZUPMXxsEEdKGkasDgu tbljD4vntrW7OAIYbJTdGcJedm7vCluP0TdHQWZNdbXRsXXLiwmyOPkiB6q6NvsghDi5SWv5RT/1 9ni/rNsCcrjR9gT9oA+py/WxI/+y0JVsTCrl93lIGnVbnHL/ftrXpcgSap+d+rp6V7ykdetSf6Ru p42Dnx9Hx0KLCXtyRO07ast1ddm3Zem7kpbESF1psEepZk30bfm2nLBMSGRpsoMMWRP1MXq3rOyD 4PNER8grR2T699K+brRHx0JSfj9/qSvImnVAhP5T+q4kWWQfp422Zd712CqNjU2Zv8997PRaEahE BDQIUYlaUZkUgRwCX/ziF50D9sQTT+Qx4UXMdD4MmMOHD7tZEK9//evd0Z0sreDUjL/92781d+7c cTvp375923ziE58wv/M7v+NOzMCY4AUrp3qwmeXHP/5xN0uC/SG2G4FHupR2futLh+HFDulJGz6K 4d5ol9Vk7VhaBm5mDRKzFChOspt7Y0P2hjH1G6eRKaucDOITRiF7EmCMhnCimq3RCEXTZwvrBw4e Dg06DOHkEmx0stgjTOMkOmtra4vfSv07dbG1NZJF+K4wiXaWZwS83tbbrIkgJAGR1XKs1OP6erss LeFYxrRla25qdllKW/HzlzTkDUGNTVE7Fb4+T47AhULIUmudXILbCwvRiSqFckQnUdCnhFim0tKU q7PzqzcHlfrD7MKsifcwzigkfH2eCzn9NFlnNWsiaIlz7PrVhL6Wfo/+OERwpsm2H7BJwkTqcVtr W/bvQgt6Y66dJp2A4WSxfaDosFQdFb49Sv2VPqcI7EwENAixM/WqpdohCGC4PfbYY+bxxx/Plwjj n00rSfulX/olt+zCP33j537u59zRoR/4wAfMe9/7XvOhD33IvdB/5Vd+xaVLRk899ZT7Lem9vb3m F37hF5whIvd362fqMY0SgGywDhKjlrPzc/lj2+RnM9PRWfJ729slKdNPcWAnJ6Pz2oUZx4NieNVZ Y7SlOXvDmIBIm11SxGkDcmybyDJn5cChaW1tdcfKSXpWn43WWcBZ4+jLuHFMQBDq6Nhn/892Cjd8 uuzJPARfpmNH9xEI4cz6mppahwvPZk37u/Y7FuzL4tPc7JzTT3NLswvO+PeyuMZpaW/f6+onOvKJ E4Kgzs59QZzcJts2ODqQelE42r5swInlACyVC0Htts/AsfeX4cEXuWjf0Yhs9g437zH6FQLGU1PR capSfiebbev7rH5CUGtbq9tjaWZ2JqYfYzGZcHhJgD5rebq6upzDPWPbrU/oB5yYVSOBJP9+2tcs 36KuEISI97XS50ldSpt3PD/KzB+OP/2IT/RvvBe69nf5yZldt9t2SvuhD/HbMvUYWZgZAi4bIs/A WGsp24by04cVgW2MQPZz37YxOCq6IlBuBN7+9rebt73tbfnRaZHn0KFD5oMf/KAzumVkVO5h2BB8 iIzg5chRs0Z6nJg18eijjzrHijwyH2WPC1AB3zn3/M7tO2bBOroXL11yErFZ5wvfeMG02FFecOE4 1szJGlYclXqj56ZbXoN+Ic5CHx4ecg4wDncIYhSQkaaxsVHT1bXPBqaikexRu/knQQhm34TYbJCy gv2IPUqW42QxCDFArTVohoeGXX0Fs1B05MgRc/HiRaufISP6AY+BgQHXxnCCQxCjyp1dnWagf8C1 cWn/1FvaPLJtdIrwZuXGoaO+sJ8MS7sIjmCg99klZMxeYalXiNFt5D948KDboPaOnfl1MrdZ70JO P9RhghAhiBHjbqsDjlAeHh5xuMCXAMTo6Jjp3NcRZPYBPNusfjo62s3IyKhz6qJ6YfXT1++cqoMH DzknmGezJN4t9Btj4+O2vfTb4Mdx136ZRdRvZWHJAfUnBBEQoa6wh9KQ7Uekf8f5jr53rHrfZiUX gU369SHbp3Ta8hPshPpt22ZW3NGjJ7JivSrfrs4uVy+YOdlig4fRvjLLpte27Vq7HA27IgS5utJ9 2Fy6ctn027pCvwvRlsGl1b6TNzr7YLNyEwzZbwMeAwODZmpy2hBUhdgIm/bMflrUJyVFQBHYHAJ7 rMGgs4M2h53+ShFQBMqMAN3Xk08+adhkk9HQp59+Ou8gFhNtafy2WZ6bNDevnjd//Mf/3RoS0f4D 8ht6RQzkhx9+yLz5zW+W5Mw/h6yz3W9npTRZ4xSHYcw6lkxnP2ydOUZXQxFlv33rtjW0Jg0zMBYW Fs2kdXC7rPO7zwYGQgaspq2xd9suLWIfMAJDU3YEFwO92zo1xaaS76m1U/Tr0p32Pjg46JxLHH/0 QoAEPZ0+fcrNQAilH9bz99j9YgjGyKjtiHV4O6yDe+zYsVBiOD6MWF69csXqZMm0tbeZmakZO0tj 2smx4VHCLUqOk3/DHj1cb9f0N9sZCTgLBEFO2aCE7LuyRRYl/nzZ3LLthwAVMwBYvz5mAwGtbS3u RKNQgRmEpS1fsZsbT9mRW9ruvNUXI7sEq0IG8ZAFfVy7ds3N+qItj4+Nu8lDJ0+edKPfPBOK2G+J YExHx143O4UgHs738eMnggRmpJy0ZTafnrYz3jpsIAJnm9khBG3obx1A8nDGn9QL9o5iGVWLDWBN Tky6vpZTuCTYmbEI+ewJOKAjAml1drnI6PCoabBL3o7b/o1AbChidggBKwKKBDL5PjIy4oIyhw9v PMj6/ve/33zDDnLQH/3SL//f5pGHV049C1Um5VMeBPbUt1qbJMxgUnlKuDGu1o7co0GIjWGmTysC ikAFIbDVIMTyfDRVu4KK5GYbzNkp5Ywks1a3zi7TCOn0r2CxbKfmzjqnhZkPjAqVa9RHpgmzTrrG bkTp9sdImN2zIrs13TMIQpA/sx+YissyEZxdcCmPfoybOi3LDxgdZBS+HEQ7xIGJ9FNtR5LLt+ka ASqmsrNxXKQf2s/qmWAhcEI3s3ba/5Ld0BQHirpSDkI/jPQjD7NVmL1SrrbMciaWmM3b5VQ4YshS rvZDAA1c6F+QI2ygaqUmrOhnzgZAon0GyqUfgiK0H5a7MQMCXAjwl4Pyfa2VidlM5e5reR9yAi8z Vjbb12oQohw1qTJ4ahCiUA+239+j84gKMdFvioAioAiUFQGMm80aOOkKvscZfeVynPyyYASHmoLr 8026rhz9mIrRD05k6JHSJN2Qxp4LoZYurSWDpOPUlsuxFRn4RD84k/yVm3CumQVRCcQspmiJSnml qST9sAyjUtqy9rXlrZfKXRHIGoHyhDezLpXmrwgoAoqAIqAIKAKKgCKgCCgCioAioAgoAhWHgAYh Kk4lKpAioAgoAoqAIqAIKAKKgCKgCCgCioAisDMR0CDEztSrlkoRUAQUAUVAEVAEFAFFQBFQBCoQ gT0BjnSuwGKrSIpAHgENQuSh0AtFQBHYzgiwrlZJEVAEFAFFQBFQBBSBykRA7ZTK1ItKVQ4ENAhR DtSVpyKgCKSGgAQf+JTr1DLXjBQBRUARUAQUAUVAEUgBAbFRCEXIdQrZahaKwLZEQE/H2JZqU6EV AUUABHiJc777wMCAO1GCo9aUFAFFQBFQBBQBRUARqDQExsfHzMjwkD2yt9EemTtXaeKpPIpAUAQ0 CBEUbmWmCCgCaSLA+erNzc2mra3NHs1XrSMLaYKreSkCioAioAgoAopAagg0NjaZ5pZWd3Rvba26 YKkBqxltSwS0BWxLtanQioAiIAhw7jxnvVdV6eoywUQ/FQFFQBFQBBQBRaCyEKiuqXazNmtra3XQ pLJUo9KUAQG12ssAurJUBBSB9BBgNoSSIqAIKAKKgCKgCCgC2wWBZaO2y3bRlcqZDQI6EyIbXDVX RUAR2EYIEMhYXFwwxDOqqvbYv/Iv7ZiYmHAjJvX19eVBEkzsHhvss8HeG9V2psmeMs02ifSzaPWz HMlSxqU3y8tLZmFh0epk2VTtqTaMbJWL0M3CwrzFpMrUWDn4LA8tO0wWFxddPalx04zLsws8dQRM lpaWLSY1bplWeTAxru0sLCw49ky9Lp9+jFmYn3ftmXZcbXEp16Z40n6ou+XWD/V1cWnRHpRo+zfb p5RvNp3Xfqwc1dW2ry1TW47aD+9C279RV8rY11JH0BEksrgvZfjP1RUri3sXbqWuWFyVFAFFIEJA gxBaExQBRWBXI7BonYS+/n4zOTlpnUprjlrDi3Wb+7s6nbEeGhyclgsXLpjPfvYfzeMve5l5zWtf E1oEx29waMiMjIy466qck3vgwAFT39AQVB7RD0EZDGKopaXF7OvoMDV2SmtImp6aMrdu3zZzdkMx gg84L522nnR0tLvAVShZcBCGh4dNb2+fY0m9RZ4DBw7a/VFaQ4nh+OAo3L171262NmKqrPOEs0Dg rLv7kF0mFTaANj83Z3pu3TIz0zPOWdhjA4rt7e1m//6uoPoBmPHxCXPnzm2zaANW4AIdOnTI7N3b Zq/CBWioK/22fxseGjZLNoCGc1tXV2tl6TZNTY1OrlD/zc/N2/bTY6ampl1bRjann66u4H3t+Pi4 q7fzNjgT9St7bPvZ7+QJG6BZNr19fU4/ogeWF1JXmpqaJCnIJ33tLVtnJycmbT2hlla5vvbgwQPB +9op29feuXPHbt446/oUpEE/9LWhAzSDg0O2DfWtBGbsIEX3YdpPk+tngihHmSgCOxABDULsQKVq kRQBRaA0BDCCb97sMRiiJ06ecDMPMMSuXbtmDbFxc/LkKefclZbb1p7653/5F/PVL3/FBUTGxsbc CNAjjzyytUw3+evBwUHr4Paao0ePmlbr8DN2w/fLly+bU6dP2yBNIOfF6qenxzqVMzOW70mrnzrr 1KGf62Z8bNycPnM6mBE4OztjLtnyN1vD89777rWOS5WZsMb6hQsXzdT0lDl29Ngm0d74z9DPzZs3 zeEjR2ywrMtlgMF+6dJFc/r0GetI7d14ppv4Be3nxo0bZsgGrO677z7n1C4uLrl6cv7cBfPAA/cH c17mbADi0qVLzmE5c/aMa8vT09Pm4sVLLsB46tSpYHWF9ossBw8edM4kgRkwunLlijl27Jjpyuls E5Bv8CfL5rZ1Kvt6+82pUydtAGSvWbZBIwI1BDrRT6iZVszEuGjrZ62dhXHmzBm3jw86O3/+gg1K TDn5QjmXBDSpFzjX3d3drl4QcHX6OX7cHNi/f4M4b/7xO3fuur71xIkTLgDCDKvrtk2By7333uM2 Xt587qX/krZ82dbPORsoOmvbD/WC9+LFixfN6Niobd/0eWFcBgIP6Ke1pdn1K7SfSauz87avnZjo MsetjkIR7fb69eu2fz/qAs7w5V147tw5c88997hNsUPJonwUgZ2GQLnmbu40HLU8ioAisA0RYDR5 YnLCHLDGKKNPjIAxut59+LCZto7vkD1KKxQdsI7Jq1/9KvOuJ5+0n68OxXYVHwxAHNrW1lb3x8wQ jMD91jDn+o6dCeDWraz6ZfoJQ1Y/45PjOf3U5/Vz8NABMz07bY3BaCZA+pxX50gwhFHtI9YYZRo5 DhMYUXd67/Za43hi9Y8ySGHmAbIwCnfQzkxBN/zhTJF2s+emVU+Yo2oZTR6wAZHDtr1wSg2YgM3J kyfNvF0SccfOkAhFfXY0ecoGHXBQpC2Dx2E7Ykk7HxkZDSIKTv4t6+TXWIcN3jKdvbOz09aXFqu7 Hje6G0KYyckp09834IJSzDigf6uys4mOHDnslopcvXrVje6GkKW3r9cFG+hbcXCRhc9D3QddgGZg YDCEGG7Z3W3bh1FPCRLRdpClw86sat/bbnpu9lhHfDaILATJcGgJDkUj/HZulW1Dx48dt3LtcYHG IIJYJsyWIXhGnW2ws93AhHZ0/PgJp7f+/oEgokhgE/7UFWk/Lbav7era54L0BKVDELMSCUA0NzeZ rv2d+b6WGYHM8qIth+prQ5RXeSgCoRHQIERoxJWfIqAIVAQCGA+jo6POyGmyyy98wjjG+GEUJJTD fc+995onvumbzLHjx5wB6MsT8hpMIBxsDEEhjPbGhkbn6M3YWQGZkx2ZY3QSPcRnXjRYOersrIix sZEgThTT/IetLHX1ddb4LFwC0m4dCHAaHhnOHBIYUCfZ84AlKT7hTLXYkcM5G0QaGwsTEGFGBu2D uuIT7Yfd33H+Ge3OnpYNTmytdfwJPPhEcATHRqZT+/eyuJ60o/r8tdllF/GR/ebmFqu7hWDBTfCn rnCEsU+MaFNXkJOgY9bEevohuxyEdoI+fMLxh2Tpl38vi2sCMyyVoZ7Qp/nUZmcQLdk9IkatMx6C ora8kNPPSl9LW0Y+lgiGaD8ENqkr9LXUUZ8aGngX1tj7LM3Lfj8Dyktws972tQRDfGpttfXYtuXB oTABKzCh7rbaIzX9tgxOBBQJhozZGXlKioAisDkENAixOdz0V4qAIrDNEcDYwWHDMGbtuE+kYQgy HXXO/u0asgYe+x5AcQPdT2NNd9aEs+bWA9uRQXThE9/Zp4INIllnnjVNWGcA45dRZN8YhS8GKeuV WUcdYlSM0coVvu4y/x/OAg73mJ0+nTXBx80usG0lqa4gC21M6lOW8lAfcbajDTELOYlsjDrLJneF T6T7zc2IsdgQhImT7GEyZR3hECSzcxJlsQ44szZYCpE10Y6pC+iCvtUnJ5tNQ1bqVNaEY087ZSPX ONW4tmzMRADHkrLibENRH1IoDVjxDPJmTegHh5p+lWVmPkkay9FmZ7IPKFIfCYok1ZWamqhNjY2G CRLl208s8Aw+1FvkDKEfXx96rQjsJAQKe5udVDItiyKgCCgCRRBwDqw9EQOKO7mYyWIshxgpdEJU wH/sFD9vnX8oHpghTTbYw6HImhbsCBTOAk6/6EJ48h35MAKZ9p81zeam/1ZbWeJE3cGvYi31/HyE XfyZNL8L9vE6Cw9Jw+HOmggScaKM31Z8npzCAM0EqCviSEv5C+WInE0CiiGCEIJ9MVlC9SkzM9Es h7WcXKefAFPb2cgVh5rAVJxcW7a1CN2w30vWJFP5k/QjaTNW3qzJ9V25ADenMcUpL8tMmLZMe4bi fS1pBHyRdybAMpVi7YclKshHH8iMlSyJ+ip9bVL7kbRQbdkv63/+L79jfuFXfsPOZFpdNz76Pz9h fuq9/9n028004/Sxv/20u3f9xq34rYLvn/vCv5m3/tj/aT71vz5XkK5fFIG0EVhtUaXNQfNTBBQB RaACEcDhZnZpZATHBMTJde6VCeK4xLiX7SvHGmJsUnQpvy+MGKhisPr30r7GKUEeCEfXJ/87zmXW REDEkc84xtQ5UfJc7F6aXyXQIbrw85Y09JP1rAz2xyhGEsRiQ8KsaaUOrFYQASKh9WSW57byuSCB KI+v5Cf6ER1KelafK0GXJGEiriHa8kIu2Ovrwi+zpM+vU6f832z2WjARXfj5SJrI699L+5p+dkWW 1bmLLLOz2befqL+Ijj/mpB2f3Df7H/Iy2yhrmp+PAkCFUkRcI9Gi4PPiQrb73jBLaMlusguJLiIp ov8lbaXv8e+Wdp30ji3llyeOHTHXbvTYDUx7Vj3+7199zvTZpWn//pXnVt178dwld4+9NYrRgYT7 BDwIfnzs//t0sZ9u+N7n/vnfXL7rBUY2nLH+YFsgoEGIbaEmFVIRUATWQ0CMgvWek/vO2ZYvSZ85 K2jd55J+u03TMLwYAVrPOAqBiTjRScaotQrzhqEEKrKEXMpbrI6BW9aywENwWRWZAYAcWDyXNS3a uhKRZRpzXHzegp2flva18EgUw0vkiMqsacWxXF1zRRSRN0tZqCdSV4Svz0/aeAhZhMfa7SfCKuvR bcpfTBbBJOoHs68r0k6TcBGdhcTE1Y94tRVBwC7nlLvnMvpvMRd4XrNPycmXdVteoj/P9ReJ+sl1 tlKfSobDwzPfYZf84+jBk8ePuItrsRkNBCb6B6IZEP/+tdVBiBfOXTT8tnmdo3kfvP8e89f//f8x 3/Udr8tLxpI3fp80wyL/0CYu+vuHXL4TAZaFbUI8/UnGCKyeG5cxQ81eEVAEFIE0ERADgemrcp1m /ps1FNKVIUxu2eC3Wdnj1nByPgU2XfIjW08tQZQQcsCjNB2VIPAWUSkob5GgR2nyblGYSvq5QJ99 HGidUosg6zwW4nZBZSnGMJzMNpy3SpDCtOxlKdJsVslWEQnZQyJx1LIX16+yEizyhVqpKxsDRQJd 9IssL9kMveKJR435oz+3zvulgkABMx2gh+67x82U8PMmgCD3JJ20FhuQYGbE//7nL7mZoQQe2LCW AAczIvZ3dUbBh1xwg3vyuxPHj0pW7pN0ZGhubjSvePwR99uCB7wvwqOfzY0tyUyIkzZPP0hCntdv 3rJ7b0y7AIoru5ePXm5vBDQIsb31p9IrArsaAV7kGAis3dyos8M6V0erbdGC5PhGXWEBX0O4rISw eDpMEwx0HxRZr5yVGOQbrf91dtFqNlbnYsCHkKU6t2Y7yRgV4bZiVEoe63+inyL1Nldd4lOq1893 40+s4L5GHRVZEta7b5xb8V9U5zYZlDrhP+3rTPTo30/7WtaK+3yFh6TJM5KexSf1kboSzYhYzSGk LLI/iPCMSyPpIfraPI+kapuvs1kFtFdKHvUX7LXA3jerhZGkEHVF2jI8+bNVJ0++bGHaz9oTtB1K yJfbGyIvZAYXtJ31bATYbjSQAJ4s4UjaeLPUYuCk77cBghfPR4EF+d2/f+15l/5/vOabzQt/REDg omFWAyRO/oP3n3XfmTHB8oqH7Pc+e8138iQIQQCCe//he99g/sOb3+Cu3Y/sf//+1efd3+te8y3m J//jO3J5DZr/+jv/r/udPPfRv/iE+y15JBH5/J4NpAixlwX0od/8QD4IQVp8Xwpk/IWf/nETD4BI Pvq5vRDQIMT20pdKqwgoAh4CvNAx5jjPHEPKN5i8xxIvcVwwZhjRwLjx7K7IEsMas5S04767EeK/ SIQQnBwPDF5naFq+SVjK9FQMqKwJOfJOVIxZJFtOPwFkkTqQhAlpVBU23atJ2HgvJvqWv9bW1tid 7NfRj30mH6zYMsfkDKQORFqI/vefFKyQN2uqr6t3LGT5QQG/nGj0ExKsKLif8heOFoRWI8JSgCg1 BCbIwHGybJyXhIukiR55PiuqzZ1qsPbU9RwuueeykoN8pbxSfp/XSp2NTmHw76V9TX3kNA72WRC+ Pg9J47jbrImTHqL+gr5sdc0ljfdr0ikractWl2/LyXLQshx2Gff7lFf6fXnv+WWVurxRTJCd3/Cu laVbfr6lXjPbgf0UWB6xv3Of26SSWQMEB2SmBI6+BCFeyAUs5LvwYTYFz//I29/iZj5Iuv/J0gyC FP/pvR9w+UvwQZ75gA1YsFzjV/6vn7ZBjXvcsx/9nx93+0c8dN/ZvAzyPJ/IyR97TLBhpvxWniGN AMR3f/vrzPfbQAiBF/a7IDABv9/9ryvBCvmNfm4/BNYOOW6/sqjEioAisIsRSDKeisFB4AKnkd8t xTYUZD0o64Kh+trIqSiWV1b3VpthWXGK8sXwqrfnwkNJBpJsYhfKMGYEFWdBDL5IymhdN2kYcrUZ G6PwbKiPzqsHk3g9i9ZsL5t66lPt6l3uRea0PhsaIlmSNs9bzG2K2NzcnBa7NfPBmXMOg20riwlr xZEPg7shV5/WzCiFG81NUXkT5cjt+k97DzGq3NDY5EqUdNIDp4lAjTkdui8Z/iftVNqtz0o20JT6 5N9L+5o+hbrAiURxkjZFfQoRJGpoaHQiJG1SKhvQhsDE9V25ox+L9bWNjZG8cdzS/E7gjMAYfZu8 9yR/937M9bUhcGlqasrVlaitiBx8shkl8iAH76qsqSmHfVL7kVOZNo5JOm91mdHwwkvRbAgCEBDL IHDYmeFw7WZPHiJmNyTtB8HMgvfamQXfbAMRLIXYKHGSBgGKH/3Bt7gABL8nz5/8jz/ksvr7//W/ N5qlzW/QBSfI50dsvrI8g2DJf7KzL9gkMz5DYsNM9AcVgUD2rbgiiqlCKAKKwE5FIO4UllpORuea rMPA71c22Yt+jYPL6RkYITV21KJctNmybUVecWDjR485nKxDhzPR2tKyFRYl/bbKBhgaLP44lnEj EGcBwx3HH+cya2ptbbFGb7U7gjOuk9nc8YPNLc0Wm+xfqe3t7W6+9Lw9EjROc3aaL/ppa2uL38rk +77ODpevHGUnTNANfzh84vTJvSw+66wThbOGHKv0kztWkHodIgghbUOOGvTLy/GQrv0E0k9Lrp1O JxzzOG2P76yxfWBLgIAVgUICIgRh4m15ZtpO67G0d+9eH6rMrltbW+0MuCpbV5LaT3Skaaj209nZ 6co5lzsRwi80x5rS59D3ZE21NtBO++C9J4EY4enSbPCo2QYHNjrqL3ls5DNqpzVOP/GAiJzOsW9f 8dMdNsKv2LMttq5AckSz/yz9PoGQtraN1Vs/BGFDPn6WG7pmxgH04vlL7vPLuY0oZc+EB+1MCWY5 4LDj1BMoSAoyHLB7PmyFZAPMLjsbg1M55I89Hwgi+IGQUvlctQETiFkQcWImB0EJZoEobX8Esp8n uf0x0hIoAorADkWgY1+7GR0bNSMjI+bgwYP5Uo6NjTlnpmt/Vz4t5IUcayhGV0jeGOk4DaOjowZj Txy3OTute8b+te9tDxaYYVOs8Ylxgz4kOAIWk5OTzpnp7u4uXMCcEVA4AwcPHjB37951vMEIwkAf tBtrgdH+/fsz4l6Y7d69bc4hABNm8BCsgXBwJyYmLE4tBVgV/jrdb12dXabvbp/DwHcix8fHDQGR IxaTEI4Lperq6jI3e3qcLFxHtGyGh4acfg4cWGnfuZuZfOBsEyhCP7PW0c/PLLIBPHChHodycsFh YGDAjAyPmEMHD7kACIWesg7CzOyMOYB+AgTxqKPdh7rNtevXzPDIsHVOVtrK4NCAWyIhDnkmSvEy JViFwzRkMSHQKrNFXFu2ThT39wYKEnV0tNs+pcHpB11IEHPaOo5T01OuTtP3hCDqyvDwsGs/R44c ybMkjSUQnfk2lb+VyQV9KbgM2mUGI/bd3NEeBTphNjw85OrKvn0raZkIkcuUfr7RDlSM2Hdhtw0U EayB5m2wc3Ji0gXO2IRxQ2QHPdIgHHyccZkBQcBBAhPkz0yIj/0tezg8Z5rscxCzJNKmqakoiMge EmnRgA2YQCJ3PF/SWf6htP0R0CDE9tehlkARUAQ2iQAOG8YXjiQGRqsdzZ62ztzAQL81hPbZUagw I8qI/8UvftE8++yzbrT95s2bznF65plnzIULF5zh9W3f9jpz3333b7Kkpf+MadHdhw///+y9aYxl WXLfd3Lf18rMyqyqrL26epnhJo/4gQJmvliwSEAyJFCALAO0IbVMSKI/WAsxpDUkP+gDKUO0LIn+ IsE0JEqAJMtiAxoaJAwPAVE2ycGwNdM93dW1Z1Vl5b7vq+N37ouXJ2/efLm9e97LzIjurHffvffd iPOPOOdGxNncawnoJiYmfCICB33s7VuZltDkBgb2AojjP/V0d7bKMHsCXRxhhr0SwNHDPDMz7QO9 WMEc0pOkwvEcGRlxw8PDfioCctCTm3zPf0RGgmKNu3Hjhnv54qV7+fKluzqYBJdv3rzxa5zcvnWr GGyeDvXj/wqd3Bi+4eA9Pj4mTnm3D7zfvH4jyapOsZWB4z/sjHeSBCK4fit2SiDDsO4ZSUAsLC65 waFBsZ1kmsQZ2Rz5c3pHb4p9PHn61I28GnGDoh/q1MTEuB/Vc/PmzWiJGUYJ3RJ7ePbsuXv95rW0 df2OBOdr0Q8jNkgMxKIrEjQtLC5IOzLmF/xrl7aXBOPMjCRIBKMYPf6UlZEog1Lu5SXZAeDFC0fA XSf6mZqc9EnF27dvFxN7eWPDOjLXrl2XNuWl/L3ybQwjiF7JMaNYfJI1byEKz6ctpY0bH59wjdLO d8oIDAK9t6Iv3pF+BFYkWWjfqMujr0ddQ12D1Jd6Ny11mYD3xvUbUr/jhC6+Lt8cds+ePnNvRBba Wpmv4kZHR6VO13ndpVaTioRQwuYrP/wDfkQAC1Ay0uHPy9oJSjpigJESOq0kaySE3n/Wz/Q6EWd5 3mHJh7M8035bnQjEqcnVWXaTyhAwBC45AjikOF7MW5+fY0TErA8YcFLp2eV6LLomzjBBA44Pf0oM ScUx7e0927BJfd5xPnFI79y545MzPrgVHHCKe2RkBAFVNCroh95JRqvMzsz6eeP9/QO+tzKmfij3 /Qf3xUkf94EuiwxiN7fv3JYAKhkZEQsX+KksJEVYVbWzvdMnJ04+R/lsUpNoYGHISUncEVD6USGS qCJwCe34bFyO/jW8SMCMT064SQkomcbT3NwoCYEbrospLBGJIO7+/ft+5Mzo6BsZVVXjkyAPH75T 7HmPJQ7t2P3799yE4ELQzTor3d1d0u4NiH7i9LInZa1xtyXAH5P6Q9KXBCejEO7cuR1tKoZiDl/q z5gkzkjOMCqepNWDBw/8SAi9L8YnvfoE2SS+0Q9tWpfoh3pFXYpF8CX41+TdtIygoc27KUF4T09P 1Hch5UYX2MgbX39kzR10JnZMIjom0dY+fPeh1OW3kvR94Vm3t3e4YUkmxpgKWKqsrPHgfi/ZsYL7 mIIREkkKRkow5YKRE/yVm1gUE/rgvQd+gcxyPJ+tQSHd0SN8ZjK9ZGbfqI/wuh2fLwQiepPnCxiT 1hAwBC4HAjhf3TLkkz/mk/u0g5yLTfSe8lctROCPU6rzcplHXQnyTrkEUgRTXj8V0I2Wm0CX3kn+ Ki0LyQZ6uYmgGOELTpUigib+EkyQojKyYKNMO+Cv0vohQGHUA3J4RCqoHwIp/iotCzY6JL3J/FWF foZNP944C/+gH6bF8Jfox1tueEu0YxIg12REHlRpW6GtvX37TsXrTxp8XZzy3//Ot/yik+kkA9dZ O4ERLbpWRPoZZ/3+lR/5sufxzd/+f2SHjT+373HwZncMplWehHQUB7/XnTH099+UskJfk21Ijc4/ ApXxKs8/blYCQ8AQuIAI+EBOHDGjPQQI7CqVgNiTIjny+kmfrND36pGlpqIJiBD+BJPqqD/Voh/k MFlCK0mODZNsTKoJl0olE9PIVBMm1SILGDG9QneOYMvONGnigdEDrBFxFiLBAS/WmGBHDKaAQPDg 2eyC8Wv/9J/7kRe6leav/ZPfcL/7H/6gJFuVn+QCz0VWiOkdHP/tX/hln+TQZ7KlJ7J89cd+tORz 7eL5QMBGQpwPPZmUhoAhYAgYAoaAIWAIGAKGgCFgCHgEGO3wh9/5nvvPZERCmgjwSRCwaOVtmZZ2 VvrJP/On3K//y38ryYbfkJEIP+oYsQD9zZ/50G+pmSQSkqQDvLmfkQyl6KvyHEZyUAb+fkESDCyw SXKD7Tj/1W/+liOZoQTfn/oLf1a/2uc5R6BGhjklYwXPeUFMfEPAELh8CNB8ffjhh35BR4bKf/TR R36Rs6OQ2Fkcdbsby253c/moW+36OUWgpkG2zGyMO3/4nEJlYhsChoAhYAhEQOAb3/iG++STT/0a Fz/38193X/7SweRBBDFOzSJZk2HaT7HQUQzhw16MvPa7Wpx0689Sv2PbT6aUMK0ji2fIv5qPa5pk e+DGuOtHVTUeMqzIRkJUs4ZMNkPAEDAEDAFDwBAwBAwBQ8AQuFAIyEStc1cekgBtMg3kMDrtDhyl fnfShMZhstn56kPA1oSoPp2YRIaAIXAKBKppruYpxLefGAKGgCFgCBgChsCFRuD8JR4utDqscBVF wJIQFYXfmBsChsBZEdDkA596fNZn2u8NAUPAEDAEDAFDwBAoJwLqo5CKiLmFcTnLYM8yBMqFgE3H KBeS9hxDwBCoCALz8/NuXPaeb2hocFtbWxWRwZgaAoaAIWAIGAKGgCFQCoH5+Tk3PT3l2PZzZWWl 1K12zRC48AhYEuLCq9gKaAhcbAS6urrc1atXfa9CXV3dxS6slc4QMAQMAUPAEDAEziUCXV3d7sqV Pr8wJYkII0PgMiNg0zEus/at7IaAIWAIGAKGgCFgCBgChoAhkDsCu842JMwdZGNwbhCwJMS5UZUJ aggYAlkI2C7DWajYOUPAEDAEDAFDwBCoVgTySEgsyxQPtrQ0MgTOAwI2HeM8aMlkNAQMgVwRIJGx sbEua0psu/r6OtfY2FSxRS53d3bc2vqa2xGZGmWdi4aGxlzLftTDd0SesbExd21oyAkoR92ey3Wv n3XRz7boR6bcNDY2uprayuTQt0WGtbV1t7u7I/ppdI1NldPP1uam2Mq6qKVWhvc2iO025IL/0Q/d detrG25jc0NkqJf5ztSfyugHvayurjn0hJ14W6mQ3bJGzbroB/ttbmpy9VKfK0XIsbGx4Ziy1iS4 1ImeKkH79OPrT0PFbEX1Aw60s42NldIP759NbyteP9Km1NVVSj+7vk3Z3twSe6330wZ0McXY9sK7 Z3V11dcf1nxqkjpUKdqUtpY6BBbIQTt3KpK2IE/63//F/+m+9Xu/7/7x//SLrv9Kr2f1zd/5lj/+ yo/8QJ6sT/xsEia/+3t/4JaXV/1v33/3vvvg3Qcnfg4/eDHy2v+u1Faf+uBPP3/s2Gr0OPfqb+wz HwROWYvyEcaeaggYAoZAbAQ2xTEfnxj3gSXOuSYihq4NRQ/qlhYX3eTUlGPEJk4OQV1HR4fr7+uL H3SLszQ+Mel+///7f93v/cf/6P7u3/27p3e8zqDUDXH8JiYnRT9r3vkjcACbqwNXoycAFhYW3Ojb t65W9nfHQd9YF/10drjBwUEf3J2hmCf6KYkq7GRC9INDjGOMk9zf3+/6+q6c6FlnvRl9vHnzRhzJ ZS8LwT9048YN19raetbHn+j3BCwjI698goighaC7taVVZLkePeiemZlxY2/HvJ0QWGK/6GdgoF/K FC+Zh61gs/Nz8z4hQ2BHD+y1a9d823IigM94M/p5/fq1X0AYuwWT9rZ2d/36tej6mZubc6Ojb329 raurdZuSBOjr75P6I21t1KTVrpdjdnbW1x/0Q/t/TTBpb28/I+In+znvwtdvXvskXpMkQngXNkhi c/jmsF/4+WRPO9vdC/ML7u3YW/8QEs/rIlt3d7cbGhqMmrQigTgp75+JiQlJJDYLb+dlGZKkPPLE tZWTYzo5NeN+/V/8W/Ehel01JSH+8Dvfdb/2T3/DLa+sugF5Z/nRG7/p3E/8ya+6n/oLf+5YBdUk xh9+53uOxMJP/pk/dWhi4ST3Hou53VQWBCwJURYY7SGGgCFwHhEgYHrx8qWrlV7bu3fuuFpxdnA6 Xjx/4Z49e+bu3LkbzflaXV1xI69GXE9PrzhaMupAiCDq2dOnvgdmWII67wHlDDSO6D/8R//Izc7M eiwIvBmVUQnCIR959UpCthp39+5dv/io6ufFixfu3r270YKX5eUV91R0QaAyfGPYw0EQ9fnnn8so gDV3R+SL5ZCOi0P85s2ouyvl7xFHGJqanha7fe4DcILdGESv9osXz93S0or74IP3fV3x+pE6BS7v v/++XwU+hizrMnroyZMnrqWlRezivuii1tcfzj2Wv4cPH0bTDwHlc2lDbkrwprpYlATjF48f+wCc BEAMov4QVE5LIPLw4TuFpJAEvZIceSyyPHjwIFoigrbs8eMnIkOL1wXbE5LAevTokejnqchyP1oi jwQEdkE9vjp41atiaWnJy0JvNwm0GERdGRkZcXOSILp7905RF6Ojo+6LLx77cwS6MQhdwLOmtsbr h8QZ8n3xxRfus88+c++99160d+GKtLVPnj7xdWd4OGlr0ctnn33uE1dgFSuRx85bJKt417AINjQt be1T/IPbd2SRyWS0gb9Qhf+QfPhrf+kvSnKteuQk8fDr//Lf+rbgb/7MX/ajH0iW/Ot/903373/7 d2XUxhX343/yayXR/Ne/+Vvum7/9LZ/EYFRDKfp7//CfuO9LkgK+4GFUPQhUZrxk9ZTfJDEEDIFL jADBAs7NwNUBn4AACgLJQelt2ZCeZa7HIJy98bFxPwS3T17ASgwl7+7pcWxDSiY/BpGI+Ymf+An3 V/67v+L+zt/5H2XUQeV2HKE3makpBP66pzr66R/ok6kZm7I160QMSDyPNxLMkQwZujpY5Mnq5gMD A25G7IRkTQwicfZWerbb29uKCQj49oqddHV1yrUxH9zFkIXgaU56LOmdZOQBhH6YulNbW+en8fhu 3QjCjEn9obeUZJ1OBaH+sHMOozRmpmciSCGllcD/zes3fsoDvepKjGjqklEz9KjG2pqPck9NTvlA aW9USo2MIur39YlRCTpyReXM63NMerWZ8nb9+vViXfYjmiQJsLQkI8CktzkGbW9v+frT3Nwi7che so5RB1d6pUdWRhetyXSeGMTIkCkZ0dTd3SUJiL1RD0l9qi/UnxiSSBJT5FhdW/X6IQEBUZeHh2/4 UVZcj0G8C0mcMbWMtlWJkTNXJICkrSUhHINIzJAQ6hDb0AQEfEkMMcKKdhh7qnb66p/40VNPc8ij bIxaIOnwtR/bk4vkwE/9V8kIiE8/f3IkW6Zf8Ptf+NmfcSQyShGjLPTevyoJGaPqQcBGQlSPLkwS Q8AQiIjAjvTiLkhwj8OV3iqLgKpegiiC4AF6lcUZy5PoJVwRh7RNhq+n543TswvNzc65tra2PMXw zwYPeo0rTuKMzkuPpdePOKAhNcmwWJzUpeVFH/TlvT4EiaoF6clGP3WppAzBJc46soaOaihvOY/p hSNwhG9IJGlwjOclKbAsvbpdEXpQqR/SXXpg2HgyZ7rOTwFgrYgmWSMiTyJwIWHIGipNqW3vCL65 Pjk16Xql1xJd5UlLEvivSbDdL0OM07zaZOpBkriZizJVhR5/RkPANyTWGwAXbJq2R9uY8J5yHhPM UW5sNN3WdnV0elbUH6Y15U2Li0uSBFp13dKrrYlN5dkmiYCp6Sk3vzDvmlvy3z6RuoxtJtMu9uyS JFqL8Kcu0/ZQn/IkbIT6Uy920SLJmZBYH4k2mOAf/aRtOry3HMeUl1EpYEISMSSm7ozvjvv3cox3 Ie0b2LQHCSLkAQ/4T0mbQh3q6e4JxYx2zFoP35SRAwTZBPFfk2RDFv3SL/9DmaZwXaY5/Fl/mSTA v/nN/0umL/wX7tNHT4ojClgjgXNM20g/+4//yJcPTJOYFL7c963/IOs6FEYZfCDrOjCdotToBOof lB6VoL/BFzqK/tbPfFi8hfKUor/3Sz9bvHzUvcUb7SAKApaEiAKzMTEEDIFqQ4D5/DjgOFVpZ5Rz BLbMs/dOYCqwKXdZ1qQHCmcHnmknj/mw0IpM12CusHTGXwoCe0ajoJt0kqFOznEe/fGXDjzLDdCi OMUE216W1IKLOKQoheATHaZtqeyyFEZcMI89TbWFBAmjE/JOQhA8EViKwfo1OtKygAvTVbDbvJMQ y0vLfvRHVjCd6Mcli1VKMJxO8qXlPut3pl1gK1mLhOpokRWZvhKDvCzCqEHWL0mTl0XkZLREFm7p +8/yfV0WcqWeElSm27cGFoMUGyKIwabS18/CN+u3lJdpRKzpkqYGCcIhgmBG0ORJlHVJEiIQI0LS xMKh/h6RJe8kBPWU9xxTMWpT7Qr6oE1jmh735W0r6Id2FEzStkCSBFpYkDoW4WWIHUBab/2Xwj/1 Dcl0FaaOVCIJwXSEf/3vfssH8j/xn3/NS/Wt//D7PsEWyslxOvBmFALn+GMhyB+X33OOBS2ZukAi 4cXIGz96gOlTnGeaBEnun/wv/1Tx8b8oyQ1+B3/u888QGfj8hZ/974v3pQ908cnfleeGiZM/kHUi oHRyIv17+35xEDjY8l2csllJDAFDwBA4FAF657Z22G2B1dn3R/Z813MEwnkHuaur617OrABWExP0 fu+IvEyXuAzEThg4o/Ta1qb0I8rx+uH6pugx335CWZtDnG8oSz+cQ7zNzS2ftMo7YMAeE1kO2kFt TXKOpFbeTjpJIoK5sK54wQr/MB0DYveOvGlVpuxAWfrRJATDprcJbnIWhkDtMFkUEwLyGMSOC5Bi EPLUc+syaiNvYoFdSMu/nx+TnGp820abnBXw7b//bN+YEgJp+x4+jSAciqEf2vPNrUQ/On0olEVt We0pvFbuY3BHHhJnrI+UJs5tyvQ3cMk7CaFTYQ7TD1jR9iAv74a8iATQZtFuD2JSrD+F+p6XHFnP ZQSCJiB+RXr5dQQB6yj84i//Lz6hl/W79DkSAOH0hK/IaAeSEEyH+Md/T3bXKKyfwHP/+t/6RT/q QZMQJAxINrAYpJ7T55O0YA2G9w/Z6YLn8jsSKb/2T/+5e//hAzcp0+VY40Gv6bPs82IjkF8Nvti4 WekMAUPgnCNAAEtnCgHk/hREUjB1ggiG86ZtcQIh5Rny8/KJgDhF29vS035JkhDohzJnUaKv5F8c 6LzJ24owIVg6jEgQ6X2H3VOO86XKi61ArGoPdln2lNxx9n+3hUcpUll2ItQfemmhrPKG58Au7yTR liSjDpOlthDkkjiLQWorIQbKV89p26Pn8/hUOQ4kEwvMsBWqOsFl3kkI6gakCYeCCMm5gtGSUMw7 iUf91PZCdbFPlkJbAyZ5EwF90l5Qh/Zz81+9fiQoLyRN9t9R3m8kO6BMTAqykfzkXZjnqxDd7Gwn 759MWVQ/Z6jLpd4npVDVNRP+vIxK0AQE9xPAfyAB/bemfr/Uz4vX3n94v3jMgW5ZyUgFTUBwHh5c C0dU6JQKP0KTmwpEQuKrf+KPu1tyfyn6cdkF4w//6Lt+KgfTOZT+lqzvoLxZB4sRUiGxk4bRxUHA khAXR5dWEkPgUiOQ5SiUAoQ1IY5DLDSXNx1HFpxE/qsYRWatuKd84qT44imrs6yOfJ64FPVzsEOs yJYgamcnX5C8DcAIygKmeE4sRW5TjJIflPffvR1T0EWR8QEmMfQDLsehvPWDDMXyHg7J3j3HEfqU 9xCoJYF06QfEwKSon0MxSS4U7yst8pmuKo+sAHDvHPUn3yQehSjKkoGLVqmiPZ2p1KV/vI9HWhYV BHlzbt+QckeSCyWpIF/esqAbbfez2je1Fe47ka0EeGY34iVL7y8yVQLqy2lnDk0CeCaH/MO6Ef0y koFpGi9GRmU0w30Z+XDfT+846vckF/72L/yKfzILS3I/SQ1GRjDFg0SEX5dCns25kLhfp3OE5+34 fCJgSYjzqTeT2hAwBAoI+GGthd6iLGfhMKDSvtah9+1zGg6762znjyP3ce45mxSlfx07AaI9lUeF lzFwKQ6XLiEMZpK3qSQ8jmO5JAZK6/OsV8PnazCV9cwY+slMyGQIE8qccbksp4rlLWkrOStHSuLl oMBHJGiK8pal9Ic85EjgFaz8cVEJs2y22MZ5eePJojJlfUbRTxbjrHMRIDl2eXOWJZSjpK1I4xPe mwVbeI4kCyODknVA1O7DO44+1lEIA/2VGxXA6AimbPyurAHxrd/7gyRZ8JvJaIz/RhbAJIlwGP3h d77np3L8tb/8F/clFP7qX/qv3V///BdlisZvuP9Nfu8THYUpIfosGwmhSFyMT0tCXAw9WikMgUuJ AC9/FrJiFWte6iQkjkvFxQ4P8QP0tM7NPe5zT3Ofzi/NChiIIfjDLz5sOPNpeFb7bwj8D3XuCr1P lCGGfpRHljOqOCJv1nxqvV6ez2ARVTXQ8MGFc4md5OulKyZinaEEe8cqS55jpgvcdMG6rHg71Fld YZ2KPSHLf6S7p4R8lYueY2HV/ElsRRoNRqwo35CnntvTY3i1vMeKu/JMP131povwpq+X87su6Ko8 w2erfOjn0LYn/MEZjnl+ktxMpkKkH6WDDorvhvQNZfyuPMCEP941SooJ37We6bU8PlWWrGd7nSEf C2jmXIfQj/LIshWVT+/R70d9Li4u+C23WevjtGuP6EiDFy9fu/6cRkMcVQ69zvaf/DFt4tuyTsS/ kpELrCvxizJi4bA1IZiGAaWvk9j4yg//gF8IkzUi2NGDP6OLi0CMN+HFRc9KZggYAhVFAAeJbRFZ ybxfttI8iUPgV98WZ4ber0K8VCyLd7wKnkfWKurFG8t0wDZo0N4Q970HqxNI2Sq5HoTKsSdZvkcE JDiCDCs/TD9cj6OfZKu4LAw4h6lgTxqA5omMzpnfN4S6wJB1KSB2HACbPEnlSHST1tDetAS2zcyb dJ2HZArCfm46bJv6E8NWmpoK+4IXpAAAQABJREFUtrJfDP9Npz74HSEyrpf7VGNDKbtNhr2rHsvN O3ye2kCWzSb3JfaTtYtH+JxyHO/tjpIx7L9gxjEw8fZYSNBltiuFaYB57ywDppQ3SYgcnrAiORAD l2bZfhnSeuu/FP5J3tTJLkWlkhXhb057jH60vDotI3yW2nIju7ucgLplO88rV674v/QWpMd9DLtX QDotI/zdMrtoRSC25mTEgq7ZQAKBZAQLTkJZsqlYbbLLxmFE8gFqLWxNfth9dv5iIGBJiIuhRyuF IXBpEchy4I4DRpME/g2yGji/Ty+e58+JE0gg11Jwio7zzNPe0yp7wsMraz4ssiFPk2xvh2NUKTot zqeVFweQHlScPQ2u9Vmc25auQvBoKgRaei2PT7Ymg3QBt5BHsujermw/2CBOa/6DC1tbE1l0sb/9 siRJiI6OzvB0LscEAR4Xsc2tzfQIJDknu1Ggv+act7elcO3t7b7+ZGGii+mRqIhRf9pa2z3eulhm CL6utq86DK/lcdzSmgR0WT2ueq6trS0P1vue2dTM1pyypa6fNrc/YbUl53z7hn4KQfm+H5f5i9bl UrYSQz/YYrO0+1DWQqXs5gK1teavn2bBnveLvvc848I/nEsWgazLfWcMX972tqQuF8ofysJiuMiD fmLU5RYJrCFsNE2bhZ1nTrpbyMEuj/STj/7Ooo+MhiAR8HLkdfEH7ErBVIcYtLy8KotK/r6XIeT3 4lWyXoWO1giv6fHXZOFK6Nf/xf+hp/wn8uu2oSQ1KkGUCTmM4iCQv8cUpxzGxRAwBAyBEyFAEEXw Mjs7KwGTbE8W9Nj6IFfOcT2GY8wWoATdBEwkHUKeuiVjR2f+geWJAMz5ZqbLsPf4vAxfZbV68ZOL hNOOk47znMytLV7K5YBgjoQVWx6SiAh5EszhGHd6/eQ7+oDC9fb2urHx8cyhvOuyHSbOeWdHRy44 pB/a09vjVt6sSHC54VrlPyV2IOCvWXqzTuqk6zNO8ok+CKaZmkXdDQMUDbY7BJO8e0+Rub2DIKrW ra7uX9Wda+iHZGNXpLpMmzE7O+eytnlcXV2TNqfR1zFky5PY+rGtrdUtLS35uqw9zPBEDogRbTEI O0Ce9fWD26SiH6i9PU79udJ7xQ/Nz0pYra9t+OkP4JY31Un9of2amp6SgHv/Nqm0d/x1dXVHqT8k LRkhQL2Fb1hn1wu74NAGxqDO9k73Vv5L12Xae7YFZovQbsHlRBTk4E6bkCBAZ/HGX2IRR1ngkZ0r WCdiQrbuJPhn68y8iW07mVbBVqEE7qzV8EISIoyM+NqP/WjJNSGYhqFbdP63f+1nvfzIrPKH24bm XY7w+cjA6A6IMhjlj4AlIfLH2DgYAoZAlSLQ29PjZudm3bS8vK9fv5ZMhhUHY3Zm1gcyDJuMQQRN PSLL+MS4W5CgmyGbEDtEzM/N+W0Fu7tP6OycQfAVWb0aR2t9bb04RWRqcsp193b7NQBbI/TOIT5T bAhcZmXNj3Z6bJliIHLNz8/Lx46/7s+doazH+Wmt9OgPDg25V69G3MLCoiQCEv3gJE9OTorT3CQr lcexFRIzHZIcW1iQ5IwkRXRoPzoDK+xEe1iPU7az3ONtdnzCTUxMSJDS6YNvnpck9rbc8MCNfUHE WXgd9dv+/gFJQjx345KgGRJdQST0pqenfbA9MDBw1CPKct3bgjjkU1NTEhis+F5bHkxQhc46Ojtc m+gvBvX29Pq2bU7akKHBQZlLz0gq6s+CH5VwTXAikMqbaN+GBofck6dPvT4GRRaIOjw5NeGYNtLX 15e3GP75BLlXZB79pOhneWlZdJGMNGArzCl5D5Ck6OyMo5+u7k5vH6xpRPuhiXDsZFXWDLh6daBY p/IGp6+/z2MyNj7m7ty5LexIqO56O2Zr2b6+fvmeP5F06JP6Mzr61mG3+g7GVmZEP+gvVsKqvaPd J2d43/Au1KkxqxJor0rdxo4Y5XMyCrIQJ/vhvrtJPPzKL/2s++Zvf8tNyBSGAZGF6RC0OWzhGU5n IOAf6N9L3LDOAufupNZb4N3C+ax1GL76Y3/c74ChQpAIgT8LUzL1AhlYSJL7jrN7BVt5spvGt2Xk Br/l+MdvflV+/6P7th1VfqU+ByTxgtw6TeW095LA4TmVGoVRSu6Leq1GHM3y1IiLipCVyxAwBKoW AZqvDz/80H388cc+afDRRx85dXBLCb2zOOp2N5bd7uayd4oJXHB26E3lJU7gQgDMubzn1hfllLKM vHolvS4rbuDqoPSC1flgG3muXbvmR2UU78354O///b/vewqXlpfE2Up6dBtlrnuXOOi8MH7u619P EgI5y8HjCeamJNDvkV7D9vZW0c+am5Eeu17RDTo6jGoapEe6sXxDmUk4vHz5QgL9ZXdd9NEgvXXT 4jwtLS66O3fvRNXPmvQeP3/+XEbM1Hp7px68fTsmdaDG3b1714+qOQyXcp8n4fDy5Yi7Ig5cV2eX 73WnPtGzeuvWTWGX/+gQygQGo6OjYitTbuj6kA9W5ufm/aK1N24M+4Ch3GU/7HnbMlLn8ZMnflQG CREWLJ0QGyYRce/e3ShTVFQ27PX5s2e+bSPQpLefpBFD2m/evBktSYQ8r1+/9oH+0NCgjJBplfZt TvQz64aHb/gRPipz3p9MxXgmmGzIaIjrN677Np42hpFOt+/cLiaO8paD55MIoS6TOByQRBqj4bBj RuEN37hRTEzEkIV6C2/WWII/Sc1xSTKiL87FehdSl59Kwmp5ecXdEP0wcob3AN9vic12SsIzFjEK 4tnTZ349GXwLRlqNSVvL+j937tw5cVv7jW98w33yyae+Y+Hnfv7r7stf+nKsohifCiNQ09QhPkmc UVYVLuqx2Et7UmNJiGNBZTcZAoZANSJQjiQE5VqXXqcZpmXIUFQWr+vukd5kWQsiltOl2FIeAv/5 uQXfS0ivT4/0bGsPmd6X9+es9EBBWeEjTlis4bBaToI3EkNMy2DdBd/bL9iU0k+5kxAqiw9sZ2e8 M0ogBxa6MKLeE+OTpAiYECigKIYF00MYDl+OIQc8CGwJEhj2T886PYT0KMcm6g9TMsCFQJM53Qx5 r4R+qCf0cC9KkooFZ5kiAy6MqolNvpdfEncM2a6XxBVD67GVUvUnDxl9+8bIJqk/yRSrJp9IrJR+ mKqysCijqmR9mba2dt8DX4n6w5oDU2KzJJx1VFynjJhJFovMQxOHP5ME5+RUkjBjWgSJeKZUVcJW GBFCHaIutcq0FEaLMIUoNsGf9o2EswDhuuSd3H3KttaSELG1Vz38LAmxXxfSptTkPw5vP0/7ZggY AoZA1SHAmgw6hLuSwuHodch8ZP4qSSQ+qolwhqtBP2DS1d3l/yqND8ESUwxiTTMoVV6CyOvXK7+V GvWHHlz+Kk0Ek0wxiDXNoFR56UlmOkSlybdvkoypRIIqXXb0Q1KIv0oTSebjjOCLIScjMoaHh2Ow KskDWyFRFmvqRSlhsJVqaWtLyWnXDIHzhkDlllo/b0iZvIaAIWAIGAKGgCFgCBgChoAhYAgYAoaA IXAmBCwJcSb47MeGgCFgCBgChoAhYAgYAoaAIWAIGAKGgCFwXAQsCXFcpOw+Q8AQMAQMAUPAEDAE DAFDwBAwBAwBQ8AQOBMCloQ4E3z2Y0PAEDAEDAFDwBAwBAwBQ8AQMASOj0BN5tLPx/+93WkInHcE LAlx3jVo8hsChoBHgMWjjAwBQ8AQMAQMAUPAEKhGBPbveJK1/1Q1Sm0yGQL5IGBeez642lMNAUMg AgKsoK3JBz75bmQIGAKGgCFgCBgChkC1IaA+Cp919fG36602PEyey42AbdF5ufVvpTcEzj0C29vb bm1tze8lfu4LYwUwBAwBQ8AQMAQMgQuJwPb2ltvYWJfOkxq3vbV1IctohTIEjouAJSGOi5TdZwgY AlWHwO7urh8J0dzc7D/5bmQIGAKGgCFgCBgChkC1IVBbW+caG5tcQ0NjtYlm8hgC0RGw6RjRITeG hoAhYAgYAoaAIWAIGAKGgCFwmRDYddZRcpn0bWUtjYAlIUrjY1cNAUOgyhGw0Q9VriATzxAwBAwB Q8AQMARcmIOwhIQZxGVHwKZjXHYLsPIbAoaAX09idXXFbW5uyVDJBtfS0lqxRS6ZJ7q8suyYWdLU 1Oiam1sqoyERYG193a2vr8lUlzrX2tIiC2lV9pWxvLzk3r4dc/fv368MJsJ1S/SzuLgk+tkRTFpd c0tzxWRZW11zK2K3LMra1tYmQ3wbKiTLrsdkY2ND6k+jyNLqbaYSwuzIGjGLi4tuSz5bW0U/MlVL F4OLLc+61J+VlRXPtq21zTVKfa4EkahFDtbOwUaoy/UVshXW8FlaWnJ8Njc1uZbWFtFPZfrDvH6k /rBVIu1sc3NTJdQjPNHPqltdXU30I5jU11emLu/s7Lil5WW3ubHp3z+0K5WqP0lbu+jfheiG+lwp ou5Qh87e1tpIiErp0PhWHwKV9SirDw+TyBAwBC4ZAjh+r1+98k55izjn83NzkozYdNeuX4/r9Eig MDMz4yYnJ11rIaCcmJjwjvrg0FDUAJNEyJvRtxK0rLr29nZJ0uy60TdvXF9/v7vS2+vEK41iJQSU K6KftbV19+jRZ+53fuf/dh0dHe5v/I3/IQr/NJPp6Wn3SmylU2RokGQVCRESRTdv3vTBd/r+vL7j nI+NjbnpqWnX1d3l2bx8OeL6+q64a9eGogZ1BHLPnj5zO5KUQTezs7NuWYKYW7duuW4vWxxbIdBe WFhwz5+/kHrbIonEFjcm+iFxhiwtEZNFBHLj4+PyN+E6OzskoKxzIyOvXE9Pt7su7UpdXbxV8WnL Xrx4IcnEdZGls5BEWxQ5bnh7yctGDz53183PLwgOI65J5sS3tLV4jEgA3Ll7R+pRvAQACcSxsXE3 OTHpOjrbfcLsleiHujQ8fCNq/SEZ8+zZs6J++P782YK7fuO66KfvIIw5nqHePn/+PEmESCJxZnrK bYj93Lp1W94DbTlyTj96102Ibt6+fevbFNra0dE3/n1848aNqG3tzs62+Aev3az4Bd3d3V5Q2tr+ /j43NDQY1VbSKNl3Q+C8I2BJiPOuQZPfEDAETo0AwfbLly8l0G92t2/f8sE1wcwbCbhHxNGI6Rwv Su/gWwksrw5clWA/cT77Jeh/9vSpe/36tbtz+3ac4N+Xf9SPxrh3717R4Zufn3evXkuyRgKqrq7E GTs18Mf84aMvvnD/7J/9c+ko3JVgocNNTU15p/SYPy/rbXNz8xLMvfRB/pAkhSB6/h89euSD8Ifv vhPNIX079tZNSID73rvvScIq6R2cm51zTyWYwX5x1GMQwdxTsU+C7vfff9/3EsJ/dHTUn3/w4B0f hMeQhV5KAqheSZINDw+LLmrc1tCWD/CePnvq3n34UGw3jstDIpE25MGD+8W60t+/6rBn5EI/MXqX 0cvLly8kKbQi+nmvWJeRb2TkpddXb29PDPVIL/+at4krV3olaXfLlx/5qD/YEKObGEUTg6YkeYd+ 7t277xND8CQZ/dlnn8uOBdvu9h3ky390BnWFBBG833nnHT9qB1nQD+8l8BkYGOBU7kRi8/GTJ36U zDvvPBB+Nb4tefHiuXvy5LF79913i/LlLczcrLxrJNlLwm5wcNCzuyZt7mff/9zjpfLlLQfPf/36 jZuanpHyPyx2SvT2LLovHn/ht9jkfW1kCBgCp0Mg/1b2dHLZrwwBQ8AQyB2BKenZpuepf6C/GOAT HNADtS09IDPifMQgnFFGPTSJE570Hidc6THt7unxPcsLMsQ8Bi1LMLewuCDOec++oIBebnow6f3f Fec4Bt27e9ePevj6z33d/Zk//adjsDyUByNBGiSIpQdMiaAJJ3RJponMSS9vDCLxQQ8uvXKagIBv d0+XBLydPlGzJb2XMWhmZtaPVCEpwzBliPpz9epV6e2v9zaNbceg8fExX5evXbtWDPBJOhDErUmQ xyijGEQwNy697Ey/6OxMRqnAl5EZXTISgR5enaKRtzwkDhcWFh2BfxjgX7lyxY/GoHeZ9i9/ShJT BNVXr9J7nIyOwWaGhq55PEgwxqDNzQ0/AoOh/d3dnUWW6Ie6PT0741ZX1orn8zwg8TwnPezoI5wK wvuHkSGMpolF8CIpz0gqEhBQUpcHvY3E0g/tBUlWplGBixLtSW9fr7ybFmU6z7KezvWTKRgkhHgn h1NB2jva5XubH+kUq63NtaD2cEOgQghYEqJCwBtbQ8AQqCwCJBmYO44jTPAfEsELCYDZudkoATfT Hlh7gW270nO1dZgyzmoMWigE08zVDgmcCGQIspgzHIMaxRHHESUBokFuDL5pHvRUsvYCw4JxhkNq a0/mTM/JNIQYxJQQgrnQKU741vhAF/3My7SEGOQDEwka0rJQfxoa6qV+Lfh1CPKWhWk7szJSpZH6 I7xDIpiBkJWRG3kTbcqGBLptHQfn0oMTMtCuxCCSENhKe3vHPnbUJeb6E2SxrkjehE2yjgptqupD eXZIQAfFat8IYCl3i18rZL8LTGApCoqmH6Y7EHQn9ScJ/MGC4J+kCFNoGMWSN21vb/mpTHWy9k9T 4/41bmjzafPmxJZi1B/f1sr6GKxdkl7jhjV40M+UJAZiEFPL9vSzxxH9sO7NpiSEY7W1e9ztyBC4 OAjsb4EvTrmsJIaAIWAIlERgY33Dr/2AQ1EjTnlI/pycx4FfF0cjb1pb2xBeu5mBts4fJ0mBA5Yn 4XCtwUeoXgKGNBVlkTUaLhMxVxoieMM2QkqSIzV+SHWMXuVQllAOjlU/BMJ5E7bCAoMCSKbdggt4 kLzJm5YkUGN0Tm3d/noMX8VkXer7lgy1z5u0vAR0aVJZViXIikEawNaVwGVVEqB5E2u6bG1tFnUR 8vOyiQ0xXQObypuK9SerfSu8BwiE8ybslcUooaQN2c9RbZkFk/Mm6gaJmZpa3oX72zfaO+Sjxx8d 5U1gT7IjnUyEr9YfkuAxbEVHLGXJwjlkiGEreWNuzzcEKoXAwTd2pSQxvoaAIWAIRERgS3p/kqBR HK9UYKlJCMShVzNv8gkGYVKbcgDhq7Jsb+/IFJF8e3LBgyG5nlKYcE6dZRYru0y0IT2SkJY/LHui H+cTWluyu0reRO8olC1L8kpPgoV8AzqSeDjhhCzp+pPIlwThG+v52wojiRKeB10axYne3hhJorXC UH7l6wUr/KNBlOowvJbH8cZGYivKN+Sh5zTpGF4r9zFBLqQ89z+fpSlZf2DH78iw/1r5vyn2WfrR ZLTKW37ue0+kLWfRUChLFj1HciBvQg59FyrfkGetvAu4vl6wp/BauY81qM9qU5K2VhIikkzMuy7T tu3p52BCUZNEak/lxsGeZwhcBgQOvrEvQ6mtjIaAIXDpEdiVkQcQsfb+vp8EGnWCCP7zJhZD85QR +Hv5RECcop2cZaF3bkf4AAiBQZr2MMk/2E7zruR3Tf5o+bNkYdQMU3zyJvgcRmo+3IO95Eml5ICv ykLwnzcx3B/K0k94Lu/ABRmUR8iX85Cei9GmwE/5KF/OKek5prLkTewwAGW1Kf58oakhMZw3Fe32 YPMW2Czy5lt/qJ97dfSgMIqV2naeuOzZ7F69VX5eMvkHWfU+vZbHp5ZX7TPkoW0KCau834XYidqK 8t0nS+H9qPKG1457rDo+7v12nyFw0RCwJMRF06iVxxC4pAhkOS2loFDHuNQ9XIuxCCPbGx5F3mmN 4RiL83WUc7TnPB8ldQ7Xcw6usyQuzoU+GCsUb0esvEXzNqBMsmQpniPAKYqWy4HaLPWuVN2LYSvH 5aFBRS6AFB5aTEQVdXGQ23HbnoO/PP6ZpKxHGwHTwPKmon4OxSS5ULwvR4GUR1Ybt3cuTBDkJ0xR lgxcNPCNkSRSOTJLqoLIRU3cZ95XppNH1tECViVlLoMsPF95ZLVvaivhfcdiG+BJWs7IELjMCOxf wekyI2FlNwQMgXOJAEMmmbvJAlpHOjBBCbMci+By8fC49xV/cIqD42wHF0cOcYr2OUmnKEzOP8k/ ZDpYgKJ+SjBHP3lDl/A42nGNJQtIkexQZ/0gcpjT0fJm/e4k547L47j3nYR3+t7icPaStpJ//4+X A+yPyEbFwKRYMQ7FJLkQQxY1xyyblbAzUafcFEMWtZ0sFSlUOkVE783jUwNq/2wYH1ZlDztfRqGy piTue7zKl7Ms6F9toJSthPftk/OQLyxmqVNsYowsOUQMO20IVAUCloSoCjWYEIaAIXAaBHAA2LKQ lzrDItVpOM6zanXhOHFq1K8Jf8c5qBhUJF9z+be4cFyGN5oEeUlugLm5eRIOL+X1vTvqkAcM1RmL gUnAdt+hyrDvZM5f6goL1pXiDSYxcCnyUAMNyq7mkyRN8rWVohwZduJFKsiXvQ5AIHQZDuvqkznb WfoJz0WRpbDgYchXi6jnshZ91XvK+VlbUyvTq7Yzk0Qqi85tLyff9LN0kc5ikJ+6Qe1W70tdLutX tVvlGT5cMUnqe771JwleSUZlT/0oypKxgGYoczmO1Qb8u0YeGJbcyyF1GXnrUzsDlYN3+hl1qd1t wuuqM2RRPYbXy3nMu1b9CeUbPl9tufjuDi+WOMZf8etICMiWhCgBlF26FAjkn46/FDBaIQ0BQ6AS COAgtbe3u97eXtnLu7voNBxHFrb/otcFZ0IdPv2d/16Ye88Ii7xJt+HMGhqNLPwRQKW3hyy3XPAo OpoZQa6ONImBSbnLdpbnNRRsQMsfPotzOKmslh4juFTss2VJ5t43ylai6kCHspbzGJuFB2aSrj/w 0XU0VN5y8k4/q6U52U42a7i44kTSMWuV+/Szzvq9pSXZ4lCnq4TP06CDrV5jUFNT0nZlzZ9XWdjW NG/SLZCVZ8jPt29iRdhSY1P+uDQ2Nnn2xSlWgTBqxypvcKnshwTRbGMLqY2GTNSWmwv2FF4r9zF1 NAnqD74L4cU7kuuKXbn5h89rbirUn8L7N7zmbUWmLvr3VCHxGF4v5zEJebWDrOlTqjO21T4JsV1u V1eX6+rsEjzzt/eTyGb3GgKxEbAkRGzEjZ8hYAiUFQF1HE/6UBwAHHDv2KQcHs7hZOAYN0uwlTe1 trb4hEiWk67nfNCXsXtGOWWjvC0iC8SWbGnSRbiam/fvJZ++L8/vp9X3WWRqa23zP0cX6eBFzzU3 N7n6QlBxFl5H/bZV9qeHMvVT2J0DJzdvwlba29r9cP/iwqpFpixit+UTWjFspb29TeqqrJqfsbDh ns2KfnIOXCh+W6tgIpSlH11tv60tsSd/Y47/0K5AWTv8qCytrYk95SiGa2pOgtxNv4Do/uym31FG 2tskqZW/S6rlVbsIy71ZqD8tETAhqC8li8qn9hTKWe5jklXgTxJc3zfKg/cgC5ySAG9tzb/dp26A jZZf5eBTz3FPcYpceEOZj1sL9TSz/rDrj7SBqsPjsiahY2QIGAIJAvm3+Ia0IWAIGAJViAA9ox0d nT7ZkHYycHboye2WYC7GnNwm6f0hIYIc6YBOtwDr7OqMgiIjS6AV2a89JBZII7Cit789UhAV8i8e M+wgMrWI800wvbEhW9npTiYFGVZXEpy6ugn8w4HM+Qh5pfeKd8BXU1v3kRxZX1v3vf0d0tsWg3pk BBK0srqyjx04sY0eDnqMJERSl9uTbVILO2WoQDr/uruLkVL5uzwkieipXV5eSY0Q2fXTxrjW2Rmn LndJmaFVWTMnJGyFbVybZQSJJirC6+U+rq9vcNgki/yur+3f8nh5edkHc909iazl5p1+XkdHh18/ SNvV8PqatHnYSKf0UsegK1eoy0wpTLZSVZ4kWtdEP4xQiBH4k2AgcUmPP9MFQiJZRUKxu7szSv1p aWmRxH+zn7KQTuStsT2x4HXlSl8oYm7H1FP0s7q8/13o9bMu+pHRlCdO+AavL0tI5KY6e/A5QSD/ N/I5AcLENAQMgcuHAI5vjYwumJqc3tsFQxzA6akp3xvT198fBRQcnSt9fT4JMTs3W+RJ4D87M+Nw zAiiYhC9TPT8z8/P7evNXVxakn3iN1z/QL+rlUAqBm1LQDk5OenGxyfcyKtXniVBy8jLETc5MeHm 5+ZiiOGd7+vXr3tnfGZ2Tz/0GiIf+untSQLyvAUiqO8Ru11YWJBAcs85XlxcckvLS35qUpOMyohB yEEP6oToIuxBnZ6e9t+vDl71AXkMWQb6B3ydHX3zpsiO+jMpdRkZ+/rjBC4Ejv0DA143i4uLRVlI hszPz/ugRRN9xYs5HXRLYqxLAqk54bu1pSObdt3U1LQkbDbc1asDvoc7J/bFx9K+Xbt2zTHff2p6 qpicIRkyNj7mR5uRXItBXj/Srif6WCiyJPhGNjBrK4w2Kl7M6YCESKu0tbx/SCAqzUm7ti6JiaHB wSiBP3z7BRNGIIyNjflkRCLLrpsYH5fzdXJ9QMXL9RNbod3gXTMl7YgS7cuUtLVtktjs7IyTZCWJ ylTP+cUFnxRSWajXvIf65J2NPZ2MgizEyX5odxsCFw4BW5jywqnUCmQIGALHRYDgZPj6Dffq9Wv/ 1yGjAJbEucDBwGnmeizqkTUtcIwnJyd8LzLTRWYl4K2X3pbrIgvOWQyCz/DwDff8xQv3Qv5wwrZk OO60OOh9fVfke5xggbJOTE65f/AP/mfvBDM6hakijND4tf/117zD/PDhQ/dTP/VTMWCRwL/HDUpQ 8EaC3A3pkWuSXmSfrKqrdXfu3PHyRBFEmNy6dcsRYD9+/MQHkvTMTUxM+h52dBeLWFfl/v377onI 8eTJE28rJEawW+ToijR6h/KSULy5c1Ns9qUsxrjr2mSKxtysBHNSp9DPyYOF06NIfWUV/KdPn4p+ rvrRKZOin9aWVsFlOJqtEDzeun3bPf7isfv880duQGTBdkkSDQ0N+fp8+lKe7Jfo44Yk8kZGXvme f0ahzUnCleD/zu07PpF3siee/m50Qlv79OkTX6cZnTIpbQ3J15j6oa29d++u++KLL9zjJ4+9rYAH ST3aml4ZKRGLeNdRT549e+aeP3/hGHm3JInN+fkFaW9uRtUP7xx2vHr79q2fgtEostHWsv7CbWn7 Yr0LwR57YNTM48eP3cDggNuVd+G46AcZ0ZGRIWAInB6BGnFeLC13evzsl4aAIVBBBGi+PvzwQ/fx xx97x/6jjz46lmOwszjqdjeW3e6mDAUWYpgwPf04gARWJCNi9fan4SN4IRHCPFwWuaM3JqbTtSfP rh9SjrPOEOUOCSJ0gca9e6r3qKZB5g03ln/uPQ4poxCYK41udD2CSiCxtLQszvqytw8WPNNFEWPL Qj2klx9sWGwPWWIG/WF56eFfXFhym9Lzz6iRjg6pyxKMV4KYqsPoFOba07uezGWPk0wMy4t+sBWm ZbATAj3wMROsoSyMyFhYWPRTZ5ABWUgCVILWZUg9tsIiokldTqaixZYF/ZD4JvBmNEIl9cOUjPn5 RZ8oog4zJaFS+uF9TFvLKIgWSeBRlyvzLnReP+iIvgASaKedZvaNb3zDffLJp77+/dzPf919+Utf jm1uxq9CCNQ0dYhPEmcUT4WKeCK2UpdrbCTEiSCzmw0BQ+AiIsC6D7HmaR+FH4F+z4mHeB711NNc r/FBE4GT0R4CBE4MXa4GIgHCX6WJwIDdaaqB6C3tvRJnasxR5WXkji70etS9eV5HPwRw/FWaWB+C XuRqINbiaerPf7HFo8qKfpiiE2uaTil5SNgxzaoaiCQIUx6qgfw0RXsXVoMqzq0MNbUnnbpzbot6 bMFtTYhjQ2U3GgKGgCFgCBgChoAhYAgYAoaAIWAIGALHQ6CmQXYhqo83vfd4UlX+LhsJUXkdmASG gCFQAQRq6tiju/K9yBUo+qVgWdMkQ3cbK9/zeynAtkIaAoaAIWAIHInAylatW1jdcU0yBWi7tsUx RN/oYiNQUyfJhzobBZGlZUtCZKFi5wwBQ+DiIyAvhRp7MVxYPZOAsPmXF1a9VjBDwBAwBM4dAiub koRY23bNrpCEsDUCzp0OTeDyIWDTMcqHpT3JEDAEKogA82qNDAFDwBAwBAwBQ8AQqHYEzGepdg2Z fHkjYEmIvBG25xsChkCuCOiLnE89zpWhPdwQMAQMAUPAEDAEDIETIqA+ivkrJwTObr+QCFgS4kKq 1QplCFwOBPRFznaJ+nK/HCW3UhoChoAhYAgYAobAeUKALVjxV6D6epsRf550Z7KWHwGrAeXH1J5o CBgCERHghb61teWTEOy3bmQIGAKGgCFgCBgChkC1IYCvwt/m5qb/rDb5TB5DICYCloSIibbxMgQM gbIiQNKBERDsJ04Pg5EhYAgYAoaAIWAIGALViAB+Cv4KoyCs06QaNWQyxUTAvPaYaBsvQ8AQKDsC 9iIvO6T2QEPAEDAEDAFDwBDIEQHzXXIE1x59LhCwJMS5UJMJaQgYAoaAIWAIGAKGgCFgCBgC5xUB SzycV82Z3HkgYNMx8kDVnmkIGALnBoFdWVNianrazc3OyoJRu66urtZ1dXe7vr4rMtUjbp52Y2PD TU5OuqWlJT/NpL6uzvUPDLiOjo7oeC4sLIosE37eKkNIW1pa3NWrV11DQ0NUWba3t92s6GZmZsYP X60V/XR3VUY/6+vr7u3bt25xcdFjwLDa/v5+19PT4/UVCxgcWWQYHR31c4uZkqT6aW9vjyWG58Oa LFNTU95u0RXDjJHh2rVr0Rdeo/6ofpCrqanJ6wYdxV64dmVlxb1+/dphMzplbGhoyGMTUxZshfoz Njbm63KdtCmtra3u+vXrflh4TGNhLjyY0L4hF7KgmytXrkSfTre8vOzrz9ramtcP7Ro2W4m2lrZt fHzc12XVz40bN6K3tdQZ5EAejmn3e3t73eDgYPT6o23twsKC5027gn66urpimqznpW0tMum7kPrT 3NwcXRZjaAhcJATietgXCTkriyFgCJx7BHCEX46MeGf0qjhaD9657/gkkHn+/IUjqIpFLFT1+PFj 76DfunXL3b9/z3V0dronT566iYmJWGJ4PtPTM+7p0yc+sL137567deumw1n//PPPHYFePNr1QQuB y1VJxty/f98NXh0UR3nCPXv2POqcWvTz2Wefufn5eXf37l333nvvuU7RD+ewl5iEDPAlcHrnnXf8 H/J98sknxQRJLHnQzZMnT3ww+f7777vh4WGflPj+97/vg6pYchDgUn9I4iHDBx984Lolmfjs2TOp y3FthQD3e9/7ng/ksFl0RBD16aefurm5uViQeD4kH9AFtop+7ty540iQIEvMukxQi80S4BJgIwtJ TWznzZs3Uevy6uqqLz/tO/p5+PChDyipP8gXk7BX2lUSd7QpyEOSBvtBzljEu/Dp06duRN6HBPvo Bz29evXKPXr0KPq7EF2QgHjw4IGXhbqMHZN4jUkk8JCFthb9UJex5Y8//tjXo5iyGC9D4KIhYEmI i6ZRK48hYAgcGwECAnpxcbpw0mtr63xP2PCNYe+M4oDEIoI5AkkCKHpYkIVewu7uLgnE3/ge1Riy bIkML1++9E4xPbcET42NTV6u7e0dfy2GHPCYnU30gxyMTqGXkJ7KwaFBH8zFTM7gjBNcEiS0tbV5 XJALHRHoEtjFIoIFnGKSIfT2MyIDuThHUEdAEYPoIcRuwWFAkkTYCvohcOAawWUsIhFEfQUHekux FYJcAikCFwKaGESAgn74RBZGqKAfdMUx1+hRjUHYK3ZLbzY9t+iHURDIxTVkiWUr6ID2lgQrASX6 YQQEbe+LFy+iBf+Ul/rKp+qEenPz5k3f5hFwx0rOYAfIAg7oBzmoz++++663Ea7FomkZDUjCChwY 2YV+qEdgRKKE92QsotzogAQ4dQZZaGOQi4RiLP1Qh0lsIsPt27e9fqjLJPIYERGz/sTC3vgYAjER sCRETLSNlyFgCFQNAjihU1OT3sHp7OjcJ1d7R7s/PzY2LufzD+hwqubn5r2zg8MTUldnlwQ0236Y bHg+r2OmpsAvPcWgqalR5Gv2IwHiBNy7bkJGPODstbfvn47SKfrBYccxjhFEkRwiiEI3BHEhEXxD OOoxCD7gn9YPASbnCC5jBdwELfQmE0SFRCIC/ZAkihUwoB8wIMANie/YSKzedsWfwB95QkIWercJ +GIQfAh0kSWcAkKgS9JVr+ctC8EctoIMyBISiSKIaQAxiLrBSCJslIBSCdn6+vr8lJVYyWetH9gF 7ZwSdQf5kCNG/aF+oB+CfdqQkLATznM9BjHijtEoJHrTbS22gy3FshXeL9QfMAn1Q70mQUOiNfbI phg6MB6GQCwE9lq9WByNjyFgCBgCVYAAzt3qajIfuK5+f1OIQ8pIhLW1Vbe5sZW7tKurK25re8sH LWGwAOP6hiSQIbiJQQwFhnCE9xNboSbnYiQhGF6/KvjjALM2Rkh1ohvOo8ONjc3wUi7HBC446vA8 oJ9CoElgEyMhok7vQf2IrYgsyBAriNJgOksWzqGfGHaLPcIrSw5NBGDXMaZXqa0Q6KdJg16tY+nr 5f6utqJ8w+dzDluJoR8CS/7QRbr+qGwEdDFI7UDtIuSp9hMjiRfW08Nk4Z4YtoJuqEME2mlZ9Bz3 xBjBgxy0/egibSuqn1hTZmjTIbXR0Fao37QnMfQT8rVjQ+AiIbDf875IJbOyGAKGgCFQAgGCFnq4 cXTSC1Byrq7QM7UqzlfetLKSzP0Ne1uUJ+eQB3kZoZAnbUsiZLMQ1BNwp4nEDIRDmjeRXKDXy+un tmYfuxr5XiuLhuIEbmzkP7Rd52ZnY5LoBwcdHeVNmgDKkkXPxQjoqDtaf7LsFlkIohS7PHHRQCBL Dg2q0E2MJEQp/agsMQJ/8FbsNXgLdRBTFuTAFpRnKEfYvhF85k2KvdaVkJ+eUx2G18p9TNumQb3y DXmoLau84bVyH1M3+Eveham21r8fa3xiIAYuWpezMOEcMvL+ybsuY6/6nsuyW5VP61i5dWLPMwQu AwKWhLgMWrYyGgKGwAEEcHhxNJwjoN3vePmbC6c2t/IPLDV4zZBCnC6kqfGybm/teNHy+oc1H7Yk sC8wPcCmtpAMIPjMm0i44KgjC47nfpLvhVMxZFEeB+VIpOI8TnHejjHcSsmi8u3Z9n7UyvlN5Tjs mSrLUfcd9vuTnNdgTnmGv9Vz1PWYsoQy6LHKEkMOeCof5aty8KlBrt4TXiv3sSYXlGf6+SqftoPp 6+X8rrIoz6xnx5CDtq2ULIqV2naWnOU6F7YXWbhwrhrqj5Y3xE7PlfsTHtqeZ2ESUz/lLps9zxCo FgQsCVEtmjA5DAFDICoCbM0J+fg2FeOGTocPhHOWbLsgy2HBPwE3TuDObr5JCHjsCg/gCDHYK34C VAxM2C4VSqkmORecVEfRX8jpH+WRhYmeA5M4uBxuAypLosd81zLRsirPw6BX7A67Xo7zpWQJ5dP7 ysHzsGdoeUO+eq+e03v0fB6foQ0o3yw+MTHJ4s85lS8GLlpe5RnKpOdC7MLr5TwOeSjfrOfHwER5 IEeWLJxDXsUuS85ynVMeh8mhsiBPnhSW9zBZ4K/ynkaWrOee5jn2G0PgvCJgSYjzqjmT2xAwBPYh QM/ESV7qx3Zhjn3jPnFO+CUKkxPKVO23B5mInEU9iV3lLMqxHp+3g34sIQo3nTfsTlI2u9cQuEwI VFO7cl5xD9vD8Pi8lsfkNgTOgsD+pZvP8iT7rSFgCBgCFUCAeaosJMa8Z+3ROY4YOpySDhX+/IiI wg9DZ0vvO84zT3sPCy2WJHIUh/RSlfzdCS8y3YK1FqAQA32MjJPwh+k1NPR6OT9LJZWSTrBElhj6 UR6ZmBR65HAo9b5y4pB+VikeKh/35O3gqhzKMy2nftf79Hsen6XKGspX6r5yyaVzxUO++mw9FwsT ygtP5aty8KnnYshyFA+VRbEL5Sz3sdqA8gyfr+e4R+8Lr5fzOOShfLOefxR2Wb856TnlgRz8pcuu 8ul9J33+Se5XHsoz/VvOc09axvR9Z/0e8siSRc+pvMflx5oX+CtM+TmJv3Lc59t9hsB5QsBGQpwn bZmshoAhcAABFo3SBatO4pjsrb6d7aTrzpwNDXvbuB1gXqYTjYWV9NWxCR+bnMPxksUyMxaLDO89 63FdXYKl8kw/b7cwRaKhsGNH+no5v6NXCQUygygvX5KDyFy5vJxy8CxdHf2wobfIo3ZYbt7p56ks iY72X1X59mx7//VyftPFDrPkgI/KovKWk3f6WbqtrfIMr6t8tA0qc3i93MfNzc3+kco3fL7KFwMT +Cof5ZslSwxMdKeQw4IuxUrlDeUs97HyUJ7h8/Wc3hNeK/cxwasueFhKP2pP5eYfPo/y6rtTMQiv cw55VY/htXIfa10+TA74hdiVm78+L+RRSj8ntRV9T/Auz3qu8rdPQ+AyIGBJiMugZSujIXCBEcAx 0j3F1ZE6TnFxvvnD2dH1IfR3nNsu7ETR0pIEFXotj8/WlhbvBGY56dvbSZIEZyfvJIR3NJuTrQW3 tg7uxKELqcVwjEl01NUnjpomPxR71Q/yNkZIEml5cRrTzjE64xz6OalDquU5yacGAlkLCqp+qA95 E3VHy5ttt9vepjWoyFOe9vZ2/3gtf8hLz2l9D6/lcdza2nqkLDEwQQiVJctW9FwMWag/tM2qixB3 rT8EZxqUh9fLfax1OUsWPae4lZt3+Dzf1haSz1n1R2WJUZe17aIdSwfGeg55Y+ACD2wlCxO1FdrA vN+F6EqxV12E+tP6o/YUXit1zP08l78Y9l5KFrtmCFQaAUtCVFoDxt8QMATOhEA6KDzuw5qaGh0O AU7XluwKEZI/J0E412P0+uN44VTh2LAwZEjq7KhDFF7L41gDuvSq7OC8uZnsFNLR3pEH633PZFQG +JMM2ZKtQ0PCKcQhbWxsco1NDeGlXI47Ozu9Y4wu0k46w2rBRu/JRYDgob29vf6b2kVwydsPDrze E17L4/jKlSv+sendBJL6s+X1FyPIhQeBSRYmKltHR0eUwAU7gNL1JzzX3d3t78n7H8oc8g35IR+B pd4TXiv3MUEu+tF6Gz5fcerq6gpP53YMH8qdZSt6LgYmYT1VGw0LjSzco/YUXiv3Me0s7yCtt+Hz aWfRG9dJ5OVNvH94F4JJVlsL/76+vrzF8M9X7HWrzpApdoucZ7Hb0/ouoRx2bAicZwQsCXGetWey GwKGwBkQqBFnpt87OnPzc/ueMz8/L9933dDQkHzmvwAizh3OzKrsf760tLxPlrm5We80X716dd/5 vL709vT4HprZ2dl9TiBO19raupezOcLoEMp3deCqaGHXJfrYKzHzajfFMb56dUAc9fxfY+jn2rVr ft/49L7wk5OTXj+JrezJmNcRDjgJqbR+cJQXFxd9UBkjiKJ82CQBHRiExJxngihk1ZEb4fU8jgcG Bnxiampqat/j+U4wd/369X3n8/pCQoS6jH40wIYXARXn0F2P1LEYBCYkAKanp/eN4MGGl5eXo+mH YI36gx7AIKSJiQlvQ4ODg+Hp3I6pGyTpWEsoDP4JCJEFe9XkWm5CFB5M/aFtoX0LA26ty/39/dF6 y8GfhAO2EhL6Ahuuo7+8CXsFf+yT9iwkZKO9ifYuFDshQTMzM7NvFA92g2wkE0+ahLDEQ6hRO77s COTvvV12hK38hoAhULUI9PR0+4Bg7O1bv1gUoxBwft7K9y7p0YzVowxAN27ccPXirI+OjnrnmCki U1PTIteid+BjBXP14hQjy/LyipuYnPDO8dbmlnv9+rXv+RkeHo6mz+6eLkdShEAycUh33YrINToq +pFgIlaPmOoHHTx79swHlxq04KDevHnTO6sxgCEQgB8B7suXL33gQPCAfjh3584d76jHkIWAmuCS 4A0dgQnB3YsXL3zPaUxbwWYJ7uGNDMhC0EKChASR9mrmjQsB971793xdQRaSMejnzZs3juAfTGLV ZRIit27d8nWHdgU5kOH58+c+OYEdxQgswZwgFh1gs9Rl9EPdoa0lqIyVmKG82ApB/9OnT4sLBFJ/ sJvbt29H6fEHExIQ2AM4vHr1ymNCHUYugl/kjEUE/rSn4DA3N+dlIdk7MjLi34MktGIROiAZga2A B7qijUEu7BncYhAJj/v373sZqL/IQR2iXmO/6CdW/YlRXuNhCMRGoEYqUmF5r9isjZ8hYAgYAmdD gObrww8/dB9//LEPvD766CPv7B711J3F0eItPGNsfMzNTs/6Yf8EEThkOMaxHQx6WMbG3krP2IJ3 cghWYgZQRVDkYH5u3uOCEwgO9CDSmxzLAVRZSMZMSoBLMIkD6PXTJ/oRp7jUKIiapg5X01jeaSME lDigOMMQ+gGTWD2nigmfBHI46QSV2DABHkElw6Zj09jYmA8mGbaNrRBQEkigq5ikAQK9twQMBDLU 41ijVMKyajKGpCZEUEmCSKc7hffmfUyCiCQEdRmi9/bu3bvRetm1fOiE+qP6wT7QTSX0Q70hwGbU DoR+sNlYo4gUEz5JQqAfbIb6Qw87ssRua9EPchDwU5cIwkk+VCLYxlZJzOhIDPRDwiZWsirUDzaC rWAzEG0tyRBkOin99E//tPv2t7/tf/urv/qr7itf+cpJH2H3GwIXAgFp62osCXEhVGmFMAQuJwLl SEKEyOGEsQtFjCkYId+DxyxGKVKIQ1ppAhPkqBZZcIyPQ3kkIZQvdsffcWXR3+XxeV71kwcW4TOT unw8Wwl/V+5j7ASqlvpTLXUZXKoBk2rSTzXJUi3tykXDxJIQ5W5h7XnnFQFp/2vqz6vwJrchYAgY AuVGoBqCyqRMBP3lLt3pnlc9mCRbs52uFOX9VbUEcpTK9JOt22rBpRoCbUWoWjBBnmrBpVrkqCZM kKVabKWa9FMtmKAfI0PgIiBQ+W6Ci4CilcEQMAQMAUPAEDAEDAFDwBAwBAwBQ8AQMASORMCSEEdC ZDcYAoaAIWAIGAKGgCFgCBgChoAhYAgYAoZAORCwJEQ5ULRnGAKGgCFgCBgChoAhYAgYAoaAIWAI GAKGwJEIWBLiSIjsBkPAEDgPCFTT3NHzgJfJaAgYAoaAIWAIGAKVQcB8lsrgblyrBwFLQlSPLkwS Q8AQOAUC+iLnU49P8Rj7iSFgCBgChoAhYAgYArkhoD4Kn7bQZW4w24PPCQK2O8Y5UZSJaQgYAgcR 4EU+Pz/vpqam/J7qbCtmZAgYAoaAIWAIGAKGQLUhMDc35/2VlpYWt7a2Vm3imTyGQFQELAkRFW5j ZggYAuVEgD3EOzs7XU9Pj6urq7OREOUE155lCBgChoAhYAgYAmVDQP2V5uZm33FStgfbgwyBc4iA Tcc4h0ozkQ0BQ2APAUZDkICwoY17mNiRIWAIGAKGgCFgCFQXAvgp+CvWaVJdejFpKoOAJSEqg7tx NQQMgTIhwGgII0PAEDAEDAFDwBAwBM4LAua7nBdNmZx5IWDTMfJC1p5rCBgC5wYBnIGtrS3Hp/ZU 6AJSsQvBuhbb29uercoSW4aEH5hsO+QBi/p6prtUJm+NXrZFP9siS6V7kZBlY2PDQ4Qs9fWVe42i m83NzYJ+6is6Goj6g91isw0NDZUxWeFaTfoBD3CBGhsbKzpdDDvBXtAPNlup9q2a9KM2i36wWbCp BPn2TWxF60+l9QMu2Eql21pt39BJpdtatZXkXXj6thZdGxkChkCCQOW8J9OAIWAIGAJVgADB7Zu3 o255aVkcc+dkzWrX3t7url4dcPWRg6mVlRX39u1bt76+XnCIa9zAQL+sedEtssV1kCfGJ930zLTX kAaWg4ODrrW1NarWcP7AZGF+wdVJIsSJD9fR1en6+/p8YBdTmOXlZffy5Uu/oJgmH8DkypUr3kmO JQuO7PT0tHv9+rUPJtEPTvq1a9dcd3d3LDE8HwIF5EAe5OCPRdeGh4ddU1NTVFlIDj1//txRj5CD gAHdoCPwiUkLCwveVggs4Y3Obty44deviZkAgC/1Z3JysphkJSGCLLRzMYl2jfpDPVJMent73dDQ UPRkHgsaY7fYDLKgE+ykT9qVmPoB/9HRUb9YIbqCN/q5efOma2tri6kenzAbGRnxiz0jB3Woo6PD 20rsxOLS0pJ79eqVW11dPaAf5IpJExMTvg6hH3hjL7RvYBPbVmKW23gZAnkjYEmIvBG25xsChkDV IoBT8fTZM3FEN939+/d80ESP4ePHj938wrx7+PAdcTjiNJM46I8fP3Ed7W3C96F3dpbFEXv0xWO3 tNTnndJYQI6Pj3vH+Pbt2xLUdknw4rzD/vnnn7t3RbbWaM7xrnvx4oUElateFzjnJCW+ePSFm5dV xt97771oTiDO8Pe//30fuP3AD/yAd0QJND/55BMfVN25cyeWehxO8dOnT92tW7d84IQjjMOOfO++ +64jsItBvv6IHMgDJgS16Oezzz5z/+k//Sf3wz/8w9FGRbDSPHwJEt5//30fyJGMQD/oCVxiBQyz s7NeluvXrzv+CFrAiPpz7949SXBejaEez4OgnyTEO++84+0CnZGoAZcf+qEfciyQF4NoVz/99FNv D6of2rzvfe97PmmEfLH0s7i46MtPIoY/bIYkGvohaUQyIhZRb0mGqH7g++TJE/fd737XffDBB37h 5RiyYBfUH5Iy8CWByDF2gj1TvzXxmrc81GVspaury/Ol/mhbi+6oQ7GI5B3+ADwHBgY8W/SFfr70 pS9FT/rGKrfxMQRiIBA3nRijRMbDEDAEDIFjIoCDgXNz7dqQd8ZxgpNeqFveMZ6cnDrmk852Gw4g PVDwH5LebO2Za5eelr6+XjchchJQxaB1cQBfiZOFA0ivOiMwcNLpZa+trXMvRU7kjUGUe056LNEP TjH40CN3/cZ130M2+uZNDDE8D4I5AhSSDTpcGozABacUO4pBjDwgiCTg1wAX/dAzR8/pM0mqxdIP vckkrEiGaK8g+iGgIugkwIpFY2Njkqxb8sGC2gp40KNMgDkzMxNFFPRD4ow6DG+1FRIP2Au6I/iO QQRs4MJoEP6oP9gK+kLOR48eRbOVN1JXGQEBb9UPCRCSACRosKMYRJIM/dDOa/0BF0ZAgBHJvVhb J9Km03aQNFT9IAsBL3qiLsci7ITtI7FZ9IIc6On+/fteb1yPQbRdJGHgT5um9Ye6QxKAhFqsdyG2 QgKCHS3gjU74w27AiLpMPTIyBAyB0yFgSYjT4Wa/MgQMgXOOAM4DPTwEC21t+4clNzc3yfl6cY4n pZT5B9z0OBEwNDU1HuiZ7OjoZKK7H04dA3IfrAm/TpnygCOohDPY2triHVJGJuRPu25metbVi37S U0DaZEoIQcSM6C9GwI1+pqamvFOOYx4S28OCE4FuDCJxRoCPUx4SzjHOMgEUwUQMInAE/7QsOOjg hKyxAm6CJBIg6SkGYIKMBC8xbIVECME2gWVYf9AHiRp0By4xCJuEX3qKDnUZXGhzGOGTNxHMkWgA DzAIieAbon7FIPRD8gw7AYeQqMu8F2LVHzABm7R+qMvIh6wx6g9lBn+1ixATplZxPpZ+aL9I6MI3 3e6DE3U4Vv2hzCSead/QiRI+A+dIhsSyFeVtn4bARUJgr1ZdpFJZWQwBQ8AQOAIBnDscHpyLurr9 TaGe29hYd+trySKERzzuTJdxZnAEcfbSgUt9fbLIH4FN3oSDt1Tg09jQeICdzgteWclfFqbIrK2t FvSzfz5/regLHW1ubrmN9fz1Q9ACNjpCJQRGMcFxjhHkqtObDqCQiXPIoPeEcpb7GD46ukAxCHmA FcmbGHZLsEawnSWH4sQ9BBR5k9pBliwkziCC/xiE3UJZsnAOHcbQD20ttgDPdPumsoFJjPqjfNQu Qj3ouRijmiir6kf5pmXhnhi2wnuQdxC6od6GlLwL6/y7MkZCROspmBxmK9ruhHLmcaz60Xob8jHB VXsAAEAASURBVMBueWfH0E/I144NgYuEwH7P+yKVzMpiCBgChkAJBOiB4g9KOzucqy0sBLkmiYi8 SXsjcfjSVFtb4+UjyNrZyTeIIkiDD5QliyZrYjij6GZ7O9mZI60fP0VE9AMeG5v5JyFw0g/DBNn4 I8jiL28qJYvqDEc+b9L6o+VP89NgRm07fb2c3zWQ1vKHz1Y5sOsYSQgdKq58s2RRHYbX8jhW7LOC XJVP78mDvz4THgTUWXKo/ag96W/y+tTylrIVvScvGXguAay2o6qLkJ+eU3sKr5X7mHaL+qG6CJ+v 56g7MWRRHlr+UBZ0hjzIm3ddxl5L6UdtOVZdDnGwY0PgoiBw0OO9KCWzchgChoAhUAIBnF4cDZya WvkLyX8rnNraSoLy8Hq5jzcLgfR+KRIuiWg13mnd3sp3/imOsQb+WWXUHTo0UZF1T7nOkWDY3c1O QngeBbBiBP7YCoStZBHnuSdvxxjein2WLHpObTtL1nKdUzkOe57KctR9h/3+JOfVBrICS5WD58WU JUt+lUXlzbqnnOdK2a1iFQMTrRda/nQZ9XwMWUphonLF0E/S1iZJZS2/8udTz2kgHF4r97G2F/BU viEPziFvDP0cB3tkUT2GcpbzGB78QYdhwrXjyMt9RoaAIXAQAUtCHMTEzhgChsA5RCDLUShVDHUw /D3p2DIINnekNz5v2t4prDsR8N3HsyDfjgTleRJJGQJ/qBSe+7DLSaAdxSTj+QlMCSgxZCkVRClO yJG3LOinFA+VJdFjvmuZqBzKM0NN/pTed9j1cpwvxSOUr9R95ZCDZ2hwFPLVZ+u5GHKENqB8VY7w U207PFfu46N4qHyKXbn5h89TWZRneE3PhdiF18t5HPJQvlnPV3mzrpXrXGiPWbLoufC+cvFOP0fL qzzD65zjL8QuvF7OY3hoeQ+TBX56z3F5h8/SROBxf2v3GQIXDQFLQlw0jVp5DIFLhoC+1HWoZtmL n05QlJ2BBPw5PPM0jxQX7zQ/y+U3VSXLYcmhoORqh8Gpsh/G4FFuoc+jzGfB4LyVt5rkjSFLDB5n sZ/0b2PLS/BdSYpd3kqVlXJelrJWCmPjW/0IWBKi+nVkEhoChsAhCPASx2liSORJe9HYbhLC50r7 XaEjVle47xARynJa11rIeph3CeWfGE5Ljaw/wVoY8AwxULlknIQ/jNGDU1t3uJOW6CuRJWvusMpb rk/lkYlJwXjAJAouGeuGaDlVvtwScspIPkthwm2hLMHPcjlUWbIernJwrdR9Wb89zTmdKx7y1efo Ob1Hz+fxSXuh9qh8Qz56LgYmykN5hnJwrOdj4FJKFpWDe/IOEI/Sj2Kk8ur3PD6Vh5Y/zUPP633p 6+X8rjagPMNnc46/GO/CsA09TBZk0zoWylnqmGepv3LS35Z6rl0zBM4jApaEOI9aM5kNAUPAI6AO CatX80LPchYOg6qhgdW3aQITxyZ9nz4ra2Xs9L1n/d7U2OwfsZsxBSGRI3G81EE7K7/Dfo+TWVcv yZmCs5e+b7cwNUVXtE9fL+d3tkj1OhVMVBf6fL7L/55i6Ed5HDb0FnnQTd76ocAqiw5bTlBI/lX5 snYhCO8rx7HaQFo3+mw9n97SVK+X85MtQSEtf/hslYPAJYZ+VBblG8qi8il24bU8jtVWlG/IQ8/F kEVtQHmGcnCsWKm86evl/K42oDzDZ+u5GHLQtqksWbjoObaqzJvC9kIxCHlyDnlVj+G1ch9r/dHy h89X2ZAlb7uFh9pBlix67qSY0A7xXN61J+04CbGwY0PgIiBgSYiLoEUrgyFgCBQd2eNC0dTUKM5A sk2dOhT6W5wdFmiEmpua9HRuny2tLb53Z2s7WQAxZMRilMiDs4NjlCfxfHUCsxykzcICjXpPnrKg G+TZzlhrAX2xcGW9JCqythItt1y6Xz2BvzrCygOcOAcmeTvG8Gxra/Oss/Sj5zo6OlS83D4JoNQO lG/IjHM43DGCqM7OziMxof5o0BfKWe7jUvrRhf3UnsrNO/085aN8w+t6TuUNr5X7GDvBFpRn+Hyt UxqYhdfyONbyHmaz8FTc8uCvzyQI1fqTlVBU+WLIghzgTzuWfhcmbW2yfXSMutze3p68CwvvGsWL T7UVMMn7XQg/xT7LbvWc3sP9RoaAIXAyBPL1aE8mi91tCBgChsCJEUgHhcd9QENDo2tva/dOlzp8 +lscLxwegrl62Q88b8Ixrq9vkGGaB7fh1J0zurq68hbDP18D2PTWY+C8KfIxeiSGLEyXARf0oA6f AoC+tra2XXOLOM9N+evH24EE3ciRDhhYvR5swIRAK2/q6+vzLLJWzeccMvT29uYthn/+wMBApixg hI7QXwwnnQAKHTHMOd0eqB2jnxjDyUmIoAPdNjRUBLLE1I/WU932MJSFcyTNtL6H18p9DJ/D6rLi FMtmwYRklNpFWFatU93d3eHp3I77+/u9PSjfkBHyEWhrgi28Vu5j6g+48N5Lt7V6DjliJVnhc1hb S9m13Sk3DunnqR0cVn9oT85it+m2Ks3fvhsCFx0BS0JcdA1b+QwBQ+BQBHqv9HrHa2ZmZt89s7Oz 8n03mrODM9Pd3eX3JZ9fWNgny9zcrHea+/qu7Duf1xfkIEEDJmHAjaO8urYqTldPFGeU8l29epXJ Mi7Rx16JFwQjRo0MDFyVk/kH/uhnaGjI60eDJqTBQZ+YmPDBbSzHmGCAwHF+fn6ffnCUwYXrBHwx iDITKE1OTu5jNzc35xMCJExiBC4wJ6BTfYTCIJvqLzyf1zG9ygQm6CIMdEnKgAs9vTECS8oHJsgz PT29LzmDDSNbLP2A/7Vr13zAjSwhUX9ICiBrDCIpRrlXV1f36QfbGR8f90kzDT7zlufKlSteP7S1 8FdCP/zR/oFdDKIu096jj5Cmpqb811j6wRbAhfaM+hIStqPXw/N5HZOYoS1FP2Fyhnfh0tKSr+cn HR1iiYe8tGXPPY8IWBLiPGrNZDYEDIGyIICTQXA5Pj7hJsXZWl9f8wHv6Oiod1RjOaMU5ubNmzJ0 vdm9ef3GLS0uuXUJEt68eeNWllfd8PCwTwyUpdBHPIQRGXfu3PZB5OvXr72jvrK84p49e+aH1g8P 3zjiCeW7TLA9NDjocIT5WxPnj4TE2Ni465ekDAmRWIQOsIcvvvjCJwAIYl69euWd0XfeeSfKfGnK Sk/6/fv3fbEfPXrk+ROwPH361CcEHj586O+JgQsO+L1797xuRkZGfOCAw46tEEjcuBHPVqjH/L14 8cInRQiysV/kuXXrVpQefzAnKYN+wAb9kCxCP2BCIgK86HmOQUxBgR+2Cn8CO+R5/Pix7/m+fft2 DDE8DwJqAlnshEAX/YyNjXldoZ9YbS31B34kxz7//HOfLAIX8CER8ODBg2iBPwE1+tH6ixyLi4u+ jSFRxTshFvX09Pj3DO8c9ILN0Oa+fPnSDUobTH2ORWBCsuj58+feXrEV6jWB/927d30iIoYs1GXa dhIHtK/oCRl4B2A/1B/sycgQMAROh0CNVK7C8l6ne4D9yhAwBAyBSiFA8/Xhhx+6jz/+2Dv/H330 kXeYjpJnZ3G0eAvPoNeSHtPNzS2/TgTOMgFwbAeDIGVGguzZQs9Ya1urG7w6GC3ALYIiBzh+9Azi jIIDvbs4ojhmMQn94JhPTk5IYmRLnD96TQekN7m0fmqa5HpjeddG0N5SnHOOcZSvX78eZcpBGnP0 8/btW2+7XEM3BHqxRh6E8lB/SNzRQ0jPrQac4T0xjtEJvaXYLb26JAJITMSYcpAuHz2nYAI2yEVQ SVKmEvoheEIWglzqLyMB0FEl6jLtbNLWbhb1E2tkSKgj9EOwTW87+sFGqMsnXWgwfOZpjwlssVk+ 0Ql1GbuN/f4BB5JUtCvgg60ygoVkfWxZeBdiJ9rWMiKB+qPraJwW69P8jraW+sN7SN+FJGZIIp2U fvqnf9p9+9vf9uX41V/9VfeVr3zlpI+w+w2BC4GA1KUaS0JcCFVaIQyBy4lAOZIQIXI8L+nYqHzv RiJLdcgBRjhflaaTYJJHEiIs/0lkCX9X7mPkgM6bfsqNQ/p5pp80Inu7UJit7GFTbfUHyapFP9Uk y0XBxJIQe3XPji43AlKna06exrvcmFnpDQFD4AIjUA2OjsJbLbJUixzgYrKodex9GiZ7WIRH1YJL tcgBNiZLaCHJsWFyEBPOGC4HcakmTA5KZ2cMgfOHQNxxtecPH5PYEDAEDAFDwBAwBAwBQ8AQMAQM AUPAEDAEyoSAJSHKBKQ9xhAwBAwBQ8AQMAQMAUPAEDAEDAFDwBAwBEojYEmI0vjYVUPAEDAEDAFD wBAwBAwBQ8AQMAQMAUPAECgTApaEKBOQ9hhDwBCoLAI2X7Oy+Bt3Q8AQMAQMAUPAEDgeAuazHA8n u+viImBJiIurWyuZIXApENAXOZ96fCkKboU0BAwBQ8AQMAQMgXODgPoofMbeIvfcgGSCXhoEbHeM S6NqK6ghcDERYF9z9lhnT3P2FjcyBAwBQ8AQMAQMAUOg2hCYmZnx/kpLS4tbXl6uNvFMHkMgKgKW hIgKtzEzBAyBciPQ1dXlrl696nsV6urqyv14e54hYAgYAoaAIWAIGAJnRqC3t9f7K83NzY5EhJEh cJkRsOkYl1n7VnZDwBAwBAwBQ8AQMAQMAUPAEDAEDAFDICICloSICLaxMgQMgfIjsLu7W/6H2hMN AUPAEDAEDAFDwBAoIwKhvxIel5GFPcoQODcI2HSMc6MqE9QQMATyQgBnYG193W1vbrn6hnrX1NRU sUUud3Z23OrqqkMm1rlAlsrQrtvY2JS/DcGi1jU3N7lKTXfx+llb82t+NNQLJs2NXqZK4LK9ve1W Vla8fhobGwWX5kqI4Xlubm56W2GRM+TAXipF2CzyYCOtra0Vqz/YCnOtqUfop5J1mTVqtC4z9LqS +lmX9o0/9AMm9fWVcf/QD/UHbJCjWvSjslSi/oAJ7eyatHHopdL6wWbRD7Jgt7qYYmxsqMPUZfCp pH4ot+oHLMCkUvUntg6MnyGQJwKVeQvlWSJ7tiFgCBgCJ0BgQxzzFy9f+IC7paVZHPUNR6B7+87t 6EHD3Oyce/3mtTg4Da6+rtatilPKHNJr14aiBt04f69fv3azs7Oura3NB3Q4Ybdu3XIdHR0nQPfs t+KYj4y8cpvCv1n0syH6IVF08+ZN75iencPxnwAeT58+9cEtAeXS0pLr6+vzssRM0KCf0dFR9+bN G9fe3u4LgLMOJoODg8cvUBnuJPEAJvAn+UCgS9DwzjvveNspA4tjPwIZHj165ANt9MN37PX+/fvR g4aJiQn38uVLnxxiFfzFxUU3PDwsdfla1KAOW3n+/LmvywRP6IsA88GDB471dGISyQf0A5E0o/50 dna6e/fuVUw/BLfUXWwF3Vy/fj2qfsCC+sOChdTlUD9gE5Noax8/fuzrsNZl6hF1mYReTJqenvZ2 C3+ty/39/b6Ni7mrhL4LaW/RhybR7ty5465cuRLdVmLqwHgZAnkjYEmIvBG25xsChkDVIoAz/uiL L7yT8/5777lacUZxMp4+eeI+/fRT98EHH0RLRBCkPH32TILIq94RBrTNzQ2R4zPfc/jgwX05UxMB y133SoL+2bk59/DhO8XFsyYnJz1W9+7edT09PRHkcD758UR00SA9cu9/8F4hEbPrvvjiifvss8/d l770QbTgZWFhQXh+5nVDMgbCaf+jP/ojr5/3338/mkNK8oEA90tf+pLr7u72shD0EuBh0zdu3PDn 8v6HugJPArg/9sf+mNeFrz8SVH388cfuB3/wB4tJkrxlQYbvf//7nt+7777rdQEW3/ve9/zfD/3Q D0XTD7v1EMwR6LNoLoR8YMJIGpJFMYgAigB3amrK64LAEsJ+PvnkE9++qf3kLQ+96+iCpNDDhw99 4I98YIIsX/7yl/25vOXg+VpX0I8m7dAPdZk6TdIqBlFXvpD3Dzs8YbOadCAB/N3vftfjROAdg7Su kGygrtDTj3zUqe985zvuR37kR6IlIsCDtpaknba1JMHRD4kr2r1Y9OrVK5+Qh6cm7ahPyBfaTyx5 jI8hcJEQsDUhLpI2rSyGgCFwIgRwRnFuhoaGfAKCHzPc8tr1az5YIPCOQTh79LTgANKzrtTQIN/7 r7h5CYAXFhb1dK6fq6trbnJ6yge39JwqMSKjSeR7/fqNd071fJ6fBHNMkxmQQI4pIQnV+ETNzs62 G5VgKha9ePHCB0nYihK9uQT89GLyF4PoKcUxJngMA0jsBh0RYBJQxCDKPCfJKoJqHZ5M/eE7vcvI gm3HoLdv3/oAkh5KZICQiZ5tAkxsKQaB/cjIiE/eDQwMFFkyoojkHZgQSMUgEme0YSRCNAEBX76j n2eS9CQpkjdhAwTWOppKRw3Ro039IQFL+xeDqD/ogLYtrR9wwU4YsRGDsAP0Q93VBAR8GZHBCA0w i0XgzygmAv+wLpMEwKbHxsaiiIKt6AgiTeDBmHcj32lvsJcYhL2iA9pZTUDAl3rMqBWuxWprY5TX eBgCsRFQry42X+NnCBgChkBFEcD5pselrrbOtTTvBdsI1diYDNGdEAdxd3cndznpfSNQampqPNDb 1Nba5kQIx/DUGOT5CL/Ozv3TLggcCLrX15E1fycdZ5TpD/XCt6lp/7oLBBAkaOYkyKI3NW8iKCGg IzBgaHBIBA8Evcgag0icYbvpaTEEdASaOM7YdQwiYENPaVkIGAhkwITAJm/CBgjm0A82GhLBAjKq rOG1PI4JkCiz2kXIA5wIWmLWZWxFp+yoLOiGc7Q5jFDImygztqA2GvLTRFqsJB76odwkhdLD+tEP thRLFmwW20zrB7mQL5Z+sBHKTBsP35B0ugo2i6x5E20tyRn48hcS+kEGcItBjHjAHsIEEXzBCZ3x 3o7V1sYor/EwBGIjYEmI2IgbP0PAEKgKBAgU+KuprZFREPubQoJKHMEt6TVbW8s/iMLxwtkhONBe XAWpvi6ZNZcE/vk6gTh4K+KgQ9obpnKE55ZX8u/J3WRRTFn/oVaSRHUp/aCbOtHb1uaWD7pDGfM4 JnABG/ge0I/oDOIedJg3qdNbSj8xEiLgoXwOk4Ve5xi9/vAg0NUe9lAHKht1LEavpdoKiZg06TmC yxiktqJ8Q56aTIshC8EabS080/VHZUMObCpvIpkIH7WLkJ+eQ4d5EzKofpRvyJNz3BNDFvRDMgrd pOuQnkN/MRJW2AFJEcqfthXFidEQsWwFnaiNhvrhXCz9hHzt2BC4SAjs97wvUsmsLIaAIWAIlECA gARnh3UWaotD/fd+oOfoWc6b1mQKBJR2uvw5CbaZirAnr781l3+2t7bdpvwJwwO9hDBUB3U9QmJm c2vTbUtQDyZ7UzGSYus5pmTE0A8OOEQSIk2JLDV+QTmC7rxJy5sli54j4M7bSaesJF20/OlyqywE OHmTBtJqnyE/PUf9iZGE0EBNyx/KoufUnsJreRwrH8Ug5KHnVN7wWrmPS8kBL2yItjhG/VFZVBdh WRUTvSe8Vu7jsLxZsui5GPqhTUGerLqs56g7MeoybRekughxV1lU3vBauY9pP9UOVBchD5Uvhn5C vnZsCFwkBA56VBepdFYWQ8AQMAQOQQCnC0dD/F//F97mZ5QX1oAkGM6bNrcTHjhZadJTTAvZ3s63 p31HeBDYH5QikUrl2xaHNG8iwKXMWfrxvFU/snhn3qTBq5Y/5KeOcawgVwO1w2RBNpU3lLPcx0fx UPlU3nLzD5+nPJRneC08Pkrm8N7THmuSKEsWDWb0ntPyOO7vtLylZNF7jvvM09x3lH4UF73vNDyO +xstr/IMf6c4xZCDdw/vIEj5ZsmigXB4rdzHe+9CEr4HW3/OIW8MXI5TN0Lsyo2FPo/3j45sy8tW srBW/vZpCFwGBCwJcRm0bGU0BC4BAid9oauD4aFJ+12BI7a7k/8Q4Z2jkgsF+fKWBeeOP6gUntsR MFE5vDCpfxL1JKDsRJBFbaUUJiF2KXHL9vUoHqF8pfArh0DHwQQ+ectxFI+YmCCLBpYcH0aK3WHX y3E+xD3EIP3s2LKk+Yffj4NdeP9pjo9T3qPq2Wn4pn8T8qi0fkJMsmTRc6FNpctTru8qi/IMn8s5 /kLswuvlPD6Kh8p31H1pmfR3nA+P0/fZd0PgMiBgSYjLoGUroyFwgRHAcaV3C+cll5d6OkGRA5bH ljtnWY4rR61MEcmbjiuLKD1vUY5lV8eW9wzSxuBxBvEyf1ppmcPAKYYsWb2maWBiyBHyCDGohCxp nod9Pw52h/32uOcVl1KYHPdZMe6LgUlYjkrjovoJZUofH+ee9G9O+v0oHooT9x11b8gbP0X9lfC8 HRsClxGBZEWty1hyK7MhYAhcCASYk8lCVczRPElPms7ppOOfvzCWVQcDgHRhyDzBUlmyePiBCcjH ApoZaxJk/ea053CmWAuDsRAhBvo8PZe3HPCDx2HOXTJYAynRT53/zPMf1c//z96bx1iSVPf+UV1d Vd3V+75v0z0zPbsZGZAAi3kIY8GzjI1/YD1vWIL3DzBYQg+Q+WdmvEge8B+W4WEhsB/YxmMYG7AE GGEJZmywxWAbzDBr9/S+Vu9b9V71i0/cOtVRWXlv3VuVGfdm3W/MVN+8kXkzTnxOROQ5JyMjrf5x WeTxxzGpuFB+PVnY14gd+4tItkhcnhyc3/KNXRFl1juHyVJvv+Wn0I/JYvW3svm0vBRMKI/64vRY ueRZsrwUsjRigjwpZbH6WpnGIytHvbEnPn4m22Gs9frhmpUni53b5LXvZXxaGfXksHzTYxky2Dmt DCvT8vkkz/LL7sumHys3liPeblUOFhrFXrHFWuNzaVsEuo2AghDdpnHVVwRmGQFelcV71jEaWjEc ayu1MxnslmETozFjx1aRj/cVvT1v7BWUeY9beOmCjBg7ZiwWXb6dDwOwd6536q/mM7GpsimYUNea EzVZlpoxWpvOmkIWe1Wc1d948Rm3EzOg4/1FbyMLgbc8WSyPldtb6QvTkdG413RRCwjF5zEuxi7e V/S2vZbT6h+f3/JoSyn0Y7JY/fNkyVttPz6uqG3KwdlptyxWX9NFtn4mnx2X3V/kd2u3eedMKQfj G+2RdRbyuFhAndcRl53gbuOFMbAy+c4f8qbQj9U3j4nJZuxMxjI+GS+svnmyWF6r45vZK4wTdv4y 5Nc5RaAKBPQ4RhW0JBlFQARyCWA4YSyQzFjKPTAncx7vIR97zZYZFHYY52IRSM5vRpHtK+NzwcIF oawbNycv+MgbK5AHo8XqWoYMnJP6Dg4OhtPnLUJmi7rNn187piw5OO/AQH8wfFko0wxyK4/v5Pf6 13emMOQwHEnU3wzhWBbyMEZTy2Iy2KfpZ8mSJZZV2ieOwIIFC8L5rdy4MPI4xpzyeF/R29SXtpvX Zi0POVIEIRYtWhSql7fAnuVZeyqaQ/Z8Vk7e4oYpZWEMRT9WZiwn+qH/cAztpexkbdbaRVye5Rm3 eF/R24zljcZaG/NSyELf4A89WLlWX8uj75i8tq+MT/oPbSVvTEE25IFJ2ddC6mbs89qt9SlrT82y sHohf6OAWLPn03EiUGUCCkJUWXuSXQREYMKF3IzIZrD0eqNq8eLF4S7UjesTnX+MHf6WLF6SxDDG AMSBxdjJGoFXx14RunTp0maqNeNjYELKvnps1E/rRj5el7l4Uc0pn3FhDU5AORh48MjqFUOOv8EF g55bX4OzFLML4xuDETmy+oEThnEq/axatSpUyozguIbIgiPXblnQDazQX6tGelyfZrdxkHBeaJ/Z gKJxgkkKJ5f+g4Nhrw2N68CsBBysVPqxcrJ9GZmQjzEnhX7oO+gH3WQdOuNkssa8ytimHNpL3usm LS9FEI+6rV69um5wBp1Zuy6DQ3xO9AOXvLGWPPoz7Rp5yk6MtQR0aSeUGyfTDzMfUyS7FlobtTIZ 75EFHsuWLbPspj5tfGKMSBG0bkooHSQCbSKgIESbwKtYERCBYgjEdxOyRstUJaxctdI7kM4dO37M H2pTykfdyZMnvSPR49asXTPVKQrZj3O0cuWKYNjwvKglXlF5+uSpcJdqxYoVll3qJ4bXoJ/pcPr0 aXfDO5KWLly8GOSDyVxvtKZI69atCw4d+oj1g2wYc2vWrPX55S9MiX42b94c3ht/5syZ8apjoB8/ fjwYzTgUKRJOI04SDMzBptyLXj/nz593y5cvT3LHkjIJiND/jh49GgIx5JHQF0GItWvXJnFcKJO2 Qjp06FD45B/ayNDQUDD2bf/4zpI2cKBoCzz7HTsv6Iq+jbOXysnFWeNu7qlTpyYEZ2g7OHnoJx4/ S0IS+vCmTZuCw40+LOHM0XZwxtasSTPWMuOCeg8PD4c+Y7LA49ixY0E3qfRDX6U/nzhxYkKglTGG IARyErRKkSgLfRw+fHi8OL7DBIeZ/SkSYy1tAf3Qbi0hC20H/aW6FtIO6K/0FwuAIA/9mv5NP+cG QivJgtiaCdEKNR07WwmUH9acreRULxEQgY4gEN9NaDUIwV2XjRs3uCNHjvi/o8Fgx5kbGjoRnBqb jpmiomvXrvPGzbB3ompGIM4BztwN7+hu8Q5wiru41JNytmzd4vbs2eP27d8fDC0cSpwFDLJ1iQIz yIJ+cB4xhE0/lzBOfWBm/fr13mGozdrg2LIThjFO/t69e4OxTrvDKMa53LlzZ7K7WjglO3bscC+9 9FL427hxY5Dn4MGDwUC/7bbbkjkuGODIsnv3brdv375wVxDnAUcGvaUKzKB7AiKUTTul7+Cs4Dzg +G/bti1ZYAZZKA9ZXn75ZYfzjcOBXHxu3749WWAGDpT34osvhraCTnBukQVeGzZsQNwkCYeb8ggS 0Ybp2+iGPrV169YwUyKJIL4QdIIj+cILLzjrL9Zu4JVqrLX2QF/etWtX6DMEQw4cOBDWOUqpH651 sKAfE0jDAUc3cIFXqsAMbYDyuA7v99cf2grtmGAv1yHGm1T6oWzaA+2EvsxYSxCBNszMHgLTHNNK MhuF38FZSQS6mUCPjy7a7b9u5qC6i4AIVJTAI4884r7xjW8E6f/yL//S3X///VPWZOTCkQnHXLl8 xZ07fy7cHcTgweDCwWrVwJhw0ml8YTjG+OIuC8bO/MH54ZEQZEqdKP/MWX9HbvhyeDMHj6Zw1w7D OXW6dvWaO32Wu7fXXX/fXK+fpVPqp2fAP1vcX3s+v0h5cZz44y47jhTTcdthTGLM4mTjTNFOabMY ximmTGd5EoghYMYn5eNwpgzgmTzWf7iTDJ926of2gX7oz8jFDCOCeKkcKGPCJ44td5W5m0v5yEFb acf4xth27lxtrDX9tHo3Oa7bdLfRD+0E/bDN2Ea7bUf/wbm2vox+6Mv8pdYPLAlSIQt9mUArTAjo pZaFPkMQhLGWaxH6YayNbzpMV/et/o72Qf+h7dpYO93HU975zneGQDZc/+qv/irZDJNW66zjRaBs Ar4v9WgmRNmUdX4REIFSCWDIWsJomU6aN98vzOX/2p0wcHAO+Gt3whheucI/e5vmKZCG1e33i1Su DY9eNDwsyU4cOP7anXCWUs40aFRfgjAp79rWk6WT+g/BOh6HSPX8ej0m5OO4pXocpZEc6Afnjb92 J/TDtP5UU/sb1Zcgc6rHURrJwT4CDp3Sly0YM5XMZe+nrTBziL+ZJntEi3MSWFESgW4mkP6WVjfT Vt1FQAQKJxAHIbhToSQCIiACIiACIiACnUaAx7RIBAbbMVus03hInu4moCBEd+tftReByhOI76pN dyZE5SGoAiIgAiIgAiIgAh1LgEdMLAjRjsehOhaMBOtaAgpCdK3qVXERmB0EmLJpiWeNlURABERA BERABESgkwiwzgbrS5DimyedJKNkEYGUBBSESElbZYmACBROgIu5LSbGIlZKIiACIiACIiACItBJ BOLHRTth3adOYiNZupOAghDdqXfVWgRmDQEWfrO3E8TvFZ81FVRFREAEREAEREAEKk2At7FY6oQF UU0WfYpAuwgoCNEu8ipXBESgEAK86sqCEPFFvpCT6yQiIAIiIAIiIAIiMEMC8UzNIt60MUNx9HMR aDsBBSHargIJIAIiMBMCvKbQ3jOvmRAzIanfioAIiIAIiIAIlEHg5MmT46ddv379+LY2RKBbCSgI 0a2aV71FYJYQ4FVXdldBMyFmiVJVDREQAREQARGYRQSGhobGa2M2y3iGNkSgCwkoCNGFSleVRWC2 Edi8eXOoEgs/Xb58ebZVT/URAREQAREQARGoMIH9+/cH6Xt7e92aNWsqXBOJLgLFEFAQohiOOosI iEAbCWzdujWUznu4Dxw40EZJVLQIiIAIiIAIiIAITCSwa9eukLFlyxa9onMiGn3rUgJzu7TeqrYI iMAsIrBx48bx2hw5csTdeeed49+b2eDd3RcvXXLXr133i1z2uwULFrienp5mflr4MTdu3HDM6PDx FL/WxYAbHBwsvIzmTjjqhocvuytXrjju3MDEXoXa3O+LO4rgEkyuX/f66R9wCxYOev20J4aODOfO nfP6GQ26gUu7ErN+4IJ+eOUbjya1K8Hk6tWrQQZemztnTnv0Q/9Blps3b4Y2S/9pV19GP5f8uEJC P7aAbmod0VYvXrwYZonRRmizfX19qcUI5aGf8+fPOz7RTTvHWsY2uNA+kGX+/PltYYJ+aCfDw8Oh /7RTP1wLGVOuXbsW2ivttl39hzZiizmim3aPtbQVxrXpjLXo1taEIAixcOHCtrQ1FSoCnURAQYhO 0oZkEQERmBaB7du3B6MaB3Hv3r3uf/yP/9H0eTD++A1G+eCCQXf61El3zZ9ny5at3lBI6WCOuqGh E+7o0aPByOnr73NHjhwOxjFBlpQOJsYfU0cJQixZstiNeIdu3759bt26dW716lWebboADcbbvr37 gvG3wOvj1MlT7sbNGw4mOLspE8/07tmzxzNZEgz0gwcPhkVRd+zYkVQ/tPNDhw6548ePO94Og5Ow e/fuMMUXAzel04Aj9+KLLzqcF7gQiHjppZccTFK+hg5HjjVhXn755eCs4LDAiMAZsqQM5sGCsgmI Ll26NMjwyiuvOF4njH5SBvNwJrkDS0Bk2bJlIZBHkGarnz2Weko4+kEWFhLGkWOsIxE0tsWFU/Rn 2gp9F/3AhCAe/Zr2CpeUATTGWvqL6Yfg2enTp922bduS64fgA7JwrUE/jC/059tvvz35WEvbYFaj jbVcj+jDt912W9JgHn2ZtsGi17QP2g5jLdfCTZs2NT3W8nvGStIdd9yRtI2l6FMqQwSmQ0BBiOlQ 029EQAQ6igCGPsYJBhROSLMJA3CXNygG/V2WO+643f+sJxgZ+/bt9YbGLrdz585kxvHZM+eCYbxh wwa3du3aUIX13tB54fkXQwDA5Gu2btM/bjSUd/HiJXfXXTvHDT4ciFe8MYYDheObImGQB6dlYJ67 4847xgy+Ubd3zz5vGO6dIF/Z8piDixOJ8UnCwfvxj3/sXnjhBXf//fc3bZDOVFZzoF71qleN3x3E yCUYQMKRSpEwyKk7enrwwQeDYU0eDgP599xzT3DyUsjCXUr6Pwu+EZQkEEP/hsnzzz/vfuZnfiaZ 848DBYO77757PBDDavg/+clPAgrGqhSBIhwo+g/OJW3FZmIcO3YsOFI426kWyCOYiB4oD6eW+iPf s88+G/JpKyZf2e2F+qOfu+66KwSGKA/5fvSjH4UgDU5iCv3QV9APZd97773jMzFoP+TTr1K9RYG+ 8txzz4U79OiC+iMf11P09sADD4zLV7Z+CMIQtNvqxzGb5ciYy1iLPPfdd1/ZIoyfnxsUBGOov81e IIhHu+Va2Kx+mAVhs6IIiiqJgAg41575kiIvAiIgAgUS4G4JjgcJ54c7xc0kjIub3vhav36dP7wn /ATja82atcEAjFezbuZ80z0GY+/osaMh4BHfPe7t9Q7/yuXuvHcizvggRYpE8OGsN7JWrlwxwSmA 8Tz/KMSBAwe9cTqSQpRwp/S6189aH4y55RT0uFV+NsbIyM1wJzOJIL4QZoJwh9ACRJTLd4xkppen ej0sdyZxomgn8fRkvhOMYx/BkRTpxIkTwbBmYVi7e4yeYIKBjjNF206RDh8+HPpsPBMEGbhjyV1m +nqKxNiDLNxJjoN13MXlrjv6IWCSIuHMMZ2dGQ+xg89rjWGDI46jW3aiDXBXm6BDfPeYNsN3nDO4 pEj0DfSDQ5nVD22FMd+cxbLlsXGDwEz8KAhjDN+ZqZEqMXOHthDrh75MUJwARSr90FZol/QX2qkl 2ivfCQDALUVi3GAMY2y1AATlMgOP/k07gk0ziaAKdaNetpB2M7/TMSIwmwkoCDGbtau6iUCXEMBY 4q4WCQOlGYPppp/Sz7G9c3r9OgPzJpDCuSQAcNLfXU7hcGPsDA9fDo+EZJ/VHpzv14Twxsupkycm yFjWFxwXyosdXMrCYZg3f8AbXdc9twtlFT9+XrjjQM31euj3j6bECYeqv68/yJHCicIpwXGkXWAM xwnjlPaXKgiBk0Sds/pBJvJwgu056ljOMrbpZxjWWVlgRDtGDtp22Qke8KddZPuPyQY3nOCyE04S gSIcFdpFnGgryGDPhsf7ytimL1MeDlOc6MvkoZsU+sFRYyYRjz3EzjYy2SNVYdyJhSxpmzGfOuMM WuDMijJHM1VfJjBG/7FyTQ7aDfIhZ4qAleknjPH+UZk40afoz+gHWctOjLXMDGGs5S9O9GVkSBVQ pJ9SXrb/oB90Rj9vtt3yCAeJR7JsdkdcN22LQDcSUBCiG7WuOovALCTAdHAMJqYeN/OGjCtXroZn NHvm9Dj+4oSRgUHGWgiXL9ee44z3F72N0YXTnXVwKQfDnXTZy1u2Ecj5zSmxckPhY//MnVsLBtgx 8b6it69dvR4cavSQdRZCXu+cEBBJcdefNkWi3KxjCSfyMJ5TBETsLmC+fmoBkhRBCNpKI1mQDx4p nCj0g7Odx8TyaLPN3rWcSVu2u+hWbnwu6992TLyvjG1jb+XGZViete14X9Hbxt7KjM9vebUxsHwn 1+qbpx8bZ5Cl7ER7tXZg5cZlmnx2TLyv6G3WKuAPOaxcK8OuhTjcKbjQZmFj7cLk4NPyCPSVfS2k PGsrVi55lshDButjlp/3yTH2mOhrX/vaCbOS8o5Xngh0CwEFIbpF06qnCMxyAjxnbY8y8AzrVAmH pOY01gIO2ePneMeShPFVdjKnPuvgUm4Ikvg3QSBr2U4uj6Zcv+6nl44FYbL1NmP5WgImLD45MjIa HHwr1+SBE38YgSmCENYGsgZ60M9YO2EGQrOPAVk9pvNpi5s1koX2VLaRDhPKMF1k62LypdCPOWrZ doJMlpei/1CeOWpWLnmWjInp0PLL+rRy8pwok8XadlkycF6Tw8rMlgUrHM8UbcVkaaQfG4+zchb5 Pa5vHheTL4UsjFsWoMteg/hu+kkhS6P+Y2PNrWt3kRqZeC7GNhvP8/Rjec30H2bWmE3ycz/3cxML 0jcR6GICCkJ0sfJVdRGYTQSYhv26170uVIlFo6YymMyhx4cc8yPHcQRDbGxyhB03vrOEDSsjawBS lM3RYKbEyM1yp5OPeMOLcqzMbFVNvhs+IFJ2wkgPsuQIE2eZoVimPKafvDLMMOaYRsfl/XY6eXCp l0w/yFF2EKKRHLF8KfVj9Y/Lj7fN0Yrzit62MvJksbwU7YR6NSrHZDF5i+YQn8/ksDLjfWxbfgpZ rN1ambEslpdKDpMllsG2TZYUgRmrL2VauSaHfTKemB4tr4xPkyXv3CZfClnQjeknj4nlNTO+seAn 9WKmJgvEKomACNQIKAihliACIjBrCLz5zW8OdSEIMdVz1806ac0eNxOIZuzUPceY1122LJzfyjAj K08mZiiUnUyO3HICjxqUhsfl/rj1TNNPIyacNYUsjcow+Rod03rt839hDomVmX9U5zBBvhRcrK3U 40F+M8c0+n0z++K6NtJRalkayd5JssT8Gsk8k31WRifpJ08Wy0uhHyvDyszjCzdjl7e/iDzkaFSG ydeMLPZWHAIQ2TVRipBV5xCBqhJQEKKqmpPcIiACkwjwGj4eyeBZTl7X1yj5ez63djfyqaPDbv2g 2K05mTUpJp3d5CtZFjOsauVboZOk8dNzSxbEFzlRlskyWE6zx9nx0/m0qdGNfptCjkbls8+MZmQp W55mmCBT2XJQRqNkTDgmhSzNlNHMMY3q1My+uIyYQfa3zeox+7tWvseyNPpdClms/EZMOKZZme18 0/m0MhrJkoKJyUEdGskSHzed+jbzm2bqixxlyzKVHMZpKlmYKfGf//mfoepvfetbm0GgY0Sgawgo CNE1qlZFRWD2E+DZ59e//vWhov/xH//RsMK9c2sLPvqbKi7rbgcDYyyTtzOUnWzBx7xykI+EsTOV YVQ7cvr/zvFrT/BHkVZufDYzvMqWgzIpowdZssoZl622I+9591jmIratDKt/fE7y+EPeFFzsWeR6 siCbHRPLWfS2lZEnB2VZvrEruvz4fFaGlRnvi7dN5jiv6G17O0eeLJaXQg7qZeVYuXFdLc+OifcV vW1lWJnZ81u+HZfdX+R3K8PKjM9teXZMvK/o7dr4Vj+Ya7JY2y66/Ph8Vl8rM97HNvlcf1LIYmXk yUKeyZIiCGHjeZ4sxsjY2ffsJ7MyeZUnr2G1x0Wzx+i7CHQrAQUhulXzqrcIzFICv/qrvxoMpn/7 t38L7xSvV01e/1UzMvKndvrc8NP+/oF6pygsn2dFSTYVNT5xzfCqrfw/dyxwEu8vcpvATF+/D7p4 Yy9PFsvLvjqtSBnsXH19c71+eoIcWSPQpspiiJrTZ78r49P0Y48gxGUYE+RIIcu8sVfoWbl5siBv 2Ua6lVFrn5MjRSafsYvlLHqb1xmSrMz4/KYznIWpHIb4d9PdtunWVm58Hnve3XQY7ytj28ppJEsK /ZgcVv+4rrQf9EZ7TSGLldGorZi8sZxFb3PtsXE0Tz8mn7WnosuPz3frWji5D1n/Rt4UXBr1ZZgg D/247LE2bo957dbyTIcxz3j7u9/9blhw9YEHHnBLly6Nd2lbBLqegIIQXd8EBEAEZheBHTt2OF7X eeLECff973+/buX6+/uC0cv6BlkjEGPnpl8Ekjvxg4MT35te94Qz2MH7zzHyzLCJT2V5873ziTxl JgyvBYMLQhF5C25Z3uDYMWXKgpHJXbGRkZuey8SFMNEPi3RijA5k3iVfhkxmGKMLjOA40XaQB2eh bMOYcnk/Pcl0Eb6M/WN5KYxd2sqiRYtCydZGY1nIQ3/GLt5X9PbixYuDE1tPDsrDgZrKYShCrmb0 Y9yKKK/RORhXSHmLG1pbsWManWem+2BPX83Tj8lB/6FNlZ2svnmyWJ7psExZGO87RRb0A//adW/i WMt4BxfGthR9mb4BG2sXsQ5MPxyToq1YP82ThT6FDI3aCm9i4VEM+L7jHe9IInPMS9si0OkEyrVo O732kk8ERGDWEeBO1y/90i+FwMLTTz9dt369/jGLJUuWBCc3a6RjdNz0r4hcsWK5NxzKHyYxAOcN zAuvA73hy47TFV6H6Y2dpcuWx9mlbS9e4u/W+PLsVWlWEMEAuPB4yqJFNUfY9pXxCXecy5vewb92 beJrUpGDv0ULF7m5fsZE2QlDEx3RTswQtjLt9ZDLl6fRz+rVq4Mxm9UPzgJGL44/7TpFQhaSMbAy eW0dnIyb5Zf1iYNLndEP7SJOJhtrxaRwXGiz6IB1abIBK3SGrCmCRDCwNnnx4sUYSXA24UKbNkd4 wgEFf8GBhQtObvatRXAi2euVCy560ulgzzXCXtUZH4B+aCOp9GN9OSsL7QZZkLORkxvLPpNt2uuy ZcvCNTP7ykn6FIFW9hMcKDsR6MBph0l2rLW2s2bNmrLFCOenzrQHa6NWKPpBFoKa1sdsX/zJgpR7 9+519957b7gxEu/TtgiIgH/sVhBEQAREYLYReOMb3+g2bNjgfvCDH7hjx47VrR7PaWJYcQxOdi2N uqHjx31+r0tl7GDorFm7xl31Bt9J/05xSxh/J/2MjgXeMCMgkiIRYFjsnW7ebR47dOfOng+G17r1 64IjlUKWtWvXuj5vINfedGIzEEbDLBcel0EWHzEpXRT0s3Xr1mAUM8PGEkYybQdHbuXKlZZd6ieO I2WdOXNmgvN/7ty58PgRjg3HpEg4jjhKPPMczyY67vsPbDZu3Jisraxfvz705f37949XHacXWZCR tpQi4ZisW7cuOJFnz54dLxKnEp3h2Ngd1vGdJW2gH8o7ffr0hL48NDQUAjaMkTigZSf6z+bNm0NZ 6MOCM3weOnQotFcLaJUti7UF04eVh/NNX4ZZKv0Q7KAsmMTOP2MvTvimTZuSBM5gQP/gWog+6Dck 9EPfJnBG/0qRaCuMG9QfLpYYX44ePRp4pQoSMa5jI9BvLQCCPIy1BCawD2hP9dKXvvSlMA7+4i/+ YpJ+Vk8O5YtApxLofdSnThVOcomACIjAdAjgCOBA/+u//mu4q/LqV796wmlGr9XuvmGA47BhfF6+ fMXfdb/pAxBD3sg477Zs2Rzu3k34YYlfkGPEG1o4CBhczIA4dPCgNwDnum3eAU4x1Z/qYQQu8nct caAwhn1GMLoOHz7icBTWrVubzDBGP9wVg8nFC5fcyChO5ZA3AC8Gp4a7q/VSz1y/LkJvfQOx3u/q 5XOHDuMcIx0HG6dh37594fA777yzoTFa75zTzccI5042DgLp/PnzDueb/O3btyfTD84JTi79B0Md PuiKvy1btgQDnvaUIqEfHIIDBw4E3dD/2cZ52LlzZ5I7/lZPZmXgRB30/RcmOLy0Fdry7bffnrQv 00cInKEjdMH2kSNHglNJECKVftAN/Zl2ane54YN+YJLK8UdHlIUMyIKjTV/as2dPeNyAx/lSjrX0 H+szyEbACLnQjQXWyC87UWecbtMJYxzthOsAY0oqx596IgfXQMq3sRb90H7uuOOOJI9VGW/qTdAB WegrbCMLwYmt/rpcr/8wJn/84x8P8j788MPJ2pTJrU8R6HQCj/nU4wdgu73U6fJKPhEQARFomgBO 0Qc+8IFgPHzrW9+a4CSOXDgy4TzMgjh37oK7ePG8N0QXBIMLB6sdiemvGH5Xr17xU8xrd8rqGTrl yjcanH2MKXM2G931KVMWHDmY4CzU9LNkyjvsPQP+ueH+2poFRcpG8IG2hZHMVGmcvPbopxZ8gAkJ OTDe25EwI3CecOxwZnDC29VW6D84CnwSmIBLu/oyuuGOKe2XtoLz2462gn5ggjw4cjBJNVsm2x4J DiEL/YigDE5eu/RDAARZ0A99J9VjTFkm6Id2wmMzzEYgMNGu/gML+jL6sccN2qUf+jCyIBN9Gf20 o/+gL/TDH+XTZqfqP5/61KfcF77wBff444+7N73pTVmV67sIdD0B35d6FITo+mYgACIwewl84xvf cL//+7/vPvaxj7m3v/3t4xXNBiHGd2hj1hAoKwgxawCpIiIgAiIgAoUTYIbab/7mb4aZEn/+538e gm2FF6ITikDFCRCD0JoQFVeixBcBEahP4C1veYvjUQyezeROupIIiIAIiIAIiIAIlEWAmZfM4vjI Rz6iAERZkHXeWUFAQYhZoUZVQgREII8AU8M/9KEPhedcv/a1r+UdojwREAEREAEREAERmDGBV155 xT3xxBPhDV133XXXjM+nE4jAbCagIMRs1q7qJgIi4G677TbH6tRPPvlkeKZTSERABERABERABESg aAJf/vKXwxtW3ve+9xV9ap1PBGYdAQUhZp1KVSEREIEsgd/93d8Nb1P4wz/8w+wufRcBERABERAB ERCBGRH4zne+4/7lX/7FffjDH076tpcZCa0fi0AbCSgI0Ub4KloERCANAVZhxzB44YUXgpGQplSV IgIiIAIiIAIiMNsJ8Bapz372s+63fuu33M/+7M/O9uqqfiJQCAEFIQrBqJOIgAh0OgEey/i93/s9 9/Wvfz28fqzT5ZV8IiACIiACIiACnU/gu9/9bgg+vPOd7+x8YSWhCHQIAb2is0MUITFEQATSEdAr OtOxbldJekVnu8irXBEQAREQAREQARGoT0Cv6KzPRntEQAREQAREQAREQAREQAREQAREQAQKJqDH MQoGqtOJgAiIgAiIgAiIgAiIgAiIgAiIgAjkE1AQIp+LckVABERABERABERABERABERABERABAom oCBEwUB1OhEQAREQAREQAREQAREQAREQAREQgXwCCkLkc1GuCIiACIiACIiACIiACIiACIiACIhA wQQUhCgYqE4nAiIgAiIgAiIgAiIgAiIgAiIgAiKQT0BBiHwuyhUBERABERABERABERABERABERAB ESiYgIIQBQPV6URABERABERABERABERABERABERABPIJKAiRz0W5IiACIiACIiACIiACIiACIiAC IiACBRNQEKJgoDqdCIiACIiACIiACIiACIiACIiACIhAPgEFIfK5KFcEREAEREAEREAEREAEREAE REAERKBgAgpCFAxUpxMBERABERABERABERABERABERABEcgnoCBEPhflioAIiIAIiIAIiIAIiIAI iIAIiIAIFExAQYiCgep0IiACIiACIiACIiACIiACIiACIiAC+QQUhMjnolwREAEREAEREAEREAER EAEREAEREIGCCSgIUTBQnU4EREAEyiYwdPKU409JBERABERABERABERABKpGYG7VBJa8IiACIpCC wDP/9RM3PHx5QlELBue7VSuXu62bN07IT/3lz//ib91zL+5yX/5/f1a36EvDw+6b337avebB+9yW NstbV0jtEAEREAEREAEREAER6DoCCkJ0ncpVYREQgWYIfOGJr7gTJ0/nHkog4tGPPuwDEity93dC 5nMv7nZP/uM/uX0HDrsPf/C9nSCSZBABERABERABERABERABpyCEGoEIiIAI1CFAsOERH2ywRFDi h//1rPvmPz/lHn38k+7jj33UMTuiE9M9O29373/Pb2gWRCcqRzKJgAiIgAiIgAiIQBcTUBCii5Wv qouACExNYHU024FtnHsSgYinv/cD97a3PBS+d9o/BEfe+IbXdppYkkcEREAEREAEREAERKDLCSgI 0eUNQNUXARFoncDdO3eEIMSlzJoRrNPw9//4Lf8IxKFwUtaOeN97fn3SYxvxcZyD4972lje6h14/ OWhAsIO1HViI0o5rRmJk+MITX/WBiNeMn5dzMZMDmdh+6nvPOCv//T5Pa0c0Q1bHiIAIiIAIiIAI iIAIzISA3o4xE3r6rQiIQFcSyC5YCYQf+oUsH/OPaBAseNvPPxQcf7Y/8sjH3f6xoET2uHf+8lvd 7/z6O3wgYNh9+nNf9MGGpzhkPLGmw+f/9itu1P/3P/05t27a4J78Gus81IIc4wfmbBBcINhx4sSt dS327T8c8pBp34Ej43JyPh4vyQZVck6rLBEQAREQAREQAREQARGYEQHNhJgRPv1YBESg2wic8IEF ggOkh8Yed8B5/7xfyJI1JOJ1InhU4/0ffjTse+SjHwy/+bL/LY9KPOq/czzpjX4GxAf8cZzXHu8I 5fiAQ/acrEvxkUceD7+b7j8ESQiAWKIMyiaQYnWyffoUAREQAREQAREQAREQgSIJKAhRJE2dSwRE YFYRYMbDp//iixPqhKNO0OGdb3/reBCBPIID73/vb0xYqBLn/h7/6AZvqLD0Cb+YZTYRlOBRC2Yu WOKxCdK7fLAgXvySc2aPtd80+/maB++fcOiWzRvC93pvA5lwsL6IgAiIgAiIgAiIgAiIwAwIKAgx A3j6qQiIwOwmQLDhef+qS0s8XkH68MPvda+OHHkLMnzXL1QdSGc1AABAAElEQVTJqzHjNOSDE5zn xKnTbtWK2swHHr/45j8/HYIT9mhH9hELO+fWTRvj0xWyPbhg4hs94iBHIQXoJCIgAiIgAiIgAiIg AiJQh4CCEHXAKFsEREAEmHXwqU88Mg6C9Rh4bIEAQRyEsEDCyZNnHH9x6nE9Ln7DBrMmbHYF52Cd BxIBDoIV2ZQNGGT367sIiIAIiIAIiIAIiIAIVImAghBV0pZkFQERaCsB1msIb6vwb5Zg22YQDPrH KUjMkLBHG+oJytoRHB+vHcGxz7+0OzzSYb+zcw5f8oGJFZarTxEQAREQAREQAREQARGoNgG9HaPa +pP0IiACCQkQdGAtCGYsxG+yYN0H0jN+lkOjxJoL/N2z8/bxAEa947eOrdMw1Tnr/V75IiACIiAC IiACIiACItCJBBSE6EStSCYREIGOJcAMCHubBG+wIN3tgwrkMUviqe//YFz28OpNv7DlJz752ZDH MQQyagtZ1n7LMcyOiBel5GAe1eBYzmnlkM/5s8eSryQCIiACIiACIiACIiACVSAwtwpCSkYREAER 6CQC73/Pb7hHH/9kWNuBV28SLOCVm48+/mfu05/7op8l8XTIY7FJZk28zx9viZkUBB3e/+HHwloR BCFIzHxgrQlbwJJz8rtPfPJz4VjeiMHaExxPMENvsjCi+hQBERABERABERABEagSgd5HfaqSwJJV BERABGZKYPTahSlP0eOPuGP7Vne7/8umVStXhCDD0iWLw+sy+/v6wndefbnNBxPOnr/g2Pcz997l X9v5m+HxCzsH57vbP76xcHDQ9fX3uVe/6n73v9/9a6EszrvaBxj4LWnDujXuoTe81i9t6ZP/547b trp3//qvhmPZx2Md9RK/WeDL4FERzmuJ3zFzA5kt1TvW9lfxs2fugOvpHaii6JJZBERABERABERA BGYtgcd86hn1adbWUBUTAREQgRwCIxeO5OQqazYR6BlY5Hr6F82mKqkuIiACIiACIiACIlB5Aj0+ aU2IyqtRFRABERABERABERABERABERABERCBahBQEKIaepKUIiACIiACIiACIiACIiACIiACIlB5 AgpCVF6FqoAIiIAIiIAIiIAIiIAIiIAIiIAIVIOAghDV0JOkFAEREAEREAEREAEREAEREAEREIHK E1AQovIqVAVEQAREQAREQAREQAREQAREQAREoBoEFISohp4kpQiIgAiIgAiIgAiIgAiIgAiIgAhU noCCEJVXoSogAiKQksBT3/uB+/RffNFdGr6csliVJQIiIAIiIAIiIAIiIAKzgsDcWVELVUIEREAE EhF4/sXd7qnv/8C985ff6hYMzg/BiA98+FG3dfMG98hHP5hIiuaKefIf/8l989tPjQdM7tl5u3vf e37drVq5oqkTfPOfn/K/f9oNnTwVjt+6eaN721ve6B56/Wub+r0OEgEREAEREAEREAEREIEsAc2E yBLRdxEQARFokcCgD0YMDg62+KtyD//8E19xT37tn3xwZKMPPPyG+58//5Dbd+CQ+8gjH3cnxoIK jSTg95//26/4gMXy8Pvf+fV3+GDGsPv0574YAhuNfqt9IiACIiACIiACIiACIlCPgGZC1COjfBEQ ARFoggCzIf7vJx5t4sh0hxBkYAbEQ294bQggWMmvfvA+9+jjnwyzG97tgwr1EsGKvN+/0c+AYNbH N/wMibe95aF6P1e+CIiACIiACIiACIiACNQloCBEXTTaIQIi0O0EuPP/w/961s8cOO1WrVru7rlz Ry6S517c5Rb6YMQWP+uAxO/2HTgcZiE4N+qe/v4z7tKly+GRjVc/eP/4Mdlz5z0mwbns9yZD3nHh pGP/cF7Sq19131hO7eNu/zgGMxue+dFPXKMgBPXl0Y2HXv+aCb8n4MLMCuqrJAIiIAIiIAIiIAIi IALTIaAgxHSo6TciIAKzngCzCZg1gENuCQd+tf/Lpsf8cTjtj3z04bCLIACLVzLzgDUk4kUsWTvi d/7XO9z/9fvjc+PgP+p/b4EMTsSaDjxSkU3vf+9vOGYl1EtDYzKvzln7gbypgggESixYklcGsiqJ gAiIgAiIgAiIgAiIwHQIKAgxHWr6jQiIwKwnYAEIAgZv9I81kHhEgcBAs4nZEP/n4feGAAUBh0// xd/4AMDuENxgjYb/b2xxS84b1mDw6zDY4pYEQWprOtQWvMTx5xyPPv5nYa2Gn33V/WFhzDxZhsfe 3DG4YHKwYNWKWhDlxCk/u2NsO+8ceXnUnQDGO9/+1rzdyhMBERABERABERABERCBKQkoCDElIh0g AiLQbQRwtHH4WVMhXvuAN2IQWPihf5yhmUSggRkSpNoCj7/p3u/XVGA7fhyCMr7r37hBgMKSbb/N n8NmHvC7d/kAwFP+8Y79ft0GHq8oO33BB0aoM+tEMKODAAQclERABERABERABERABERgOgQUhJgO Nf1GBERgVhPgEQrSGzNrIpBnAQG2p0q8NSMvbd20cVL2wszbNeyxD2YfEHywYAazMmxmxqSTtJjx w//6iQ+o1NaPsJ8SZKA8S8yWsMdJCM7wetKH3vAaf0xzr/m08+hTBERABERABERABERABCCgIITa gQiIgAhkCNhaDdnAQOawaX9dkPOYRPZkzHJ4v3+15pd9EII1J0gEIt7oAwAPNVgPInueRt95JMMC Lnbc0OtPTQhCxDNB4PKRRx4Pj5N02htBTH59ioAIiIAIiIAIiIAIdDYBBSE6Wz+STgREoA0E6s1g SC2KzXpgBsJ/+MUueavFpz/3xbA2BWtHtDIrw2Qn8EBihgOPevDXbGKGBAtiftO/onO/f0Rjy+YN zf5Ux4mACIiACIiACIiACIhAIDBHHERABERABCYSsEch9vp1ENqVhvzClPyRmAHBGhLMPiBoENal 8I9S1Eu8gYO0b/9k+Tmn7a/3ex7T4O0eNiMkPs4CHxf9q0OVREAEREAEREAEREAERKBVAgpCtEpM x4uACMx6AvZ6yqf9+gdxuuQd7+de2hVnlbb991/7lvvAhx8LC1DGhVgAwd6AEe+zbTsmu4Ami0sS WNi6efKaFPZbPlkD4qnv/SD8xfls2xoSU50j+zt9FwEREAEREAEREAEREAEI6HEMtQMREAERyBDg sQNmHPDYwSc++dmwBgOOOd/zZgdkfl7IVxZ/ZBHIj3/yc+FtFDw+sf/gYfeNbz8VHsOwQEleYawn cc/OHT6I8EzYzSMUyG2vF53qFZuce5Vfi4LjT5w6FR7BIOiBPAQy+L3NiKAA3vhB+vhjH52QHzL1 jwiIgAiIgAiIgAiIgAhEBBSEiGBoUwREQASMwO/4xx9IBB5+6NdjIOGc82YLnPGyE4GEDz/8Xvd5 /4pM1oGwFB7N+F/vmLB4pO2LP//Pw//bfeFvvzI2o6EWjCC48shHH57ytwQYHvVrTjz5tX+a8Hvy s6/otDdnxGVrWwREQAREQAREQAREQATqEegZ9aneTuWLgAiIwGwkMHLhSNPVwsk+4ddR4JWU8d3/ pk9QwIEmAwtmrm7x1Zgz+S2iN/N7jiG1i08oPPNPz8Ai19O/KJOrryIgAiIgAiIgAiIgAu0k0ENS EKKdKlDZIiAC7SDQShCiHfKpzJkTUBBi5gx1BhEQAREQAREQAREomgAxCC1MWTRVnU8EREAEREAE REAEREAEREAEREAERCCXgIIQuViUKQIiIAIiIAIiIAIiIAIiIAIiIAIiUDQBBSGKJqrziYAIiIAI iIAIiIAIiIAIiIAIiIAI5BJQECIXizJFQAREQAREQAREQAREQAREQAREQASKJqAgRNFEdT4REAER EAEREAEREAEREAEREAEREIFcAgpC5GJRpgiIgAiIgAiIgAiIgAiIgAiIgAiIQNEEFIQomqjOJwIi IAIiIAIiIAIiIAIiIAIiIAIikEtAQYhcLMoUAREQAREQAREQAREQAREQAREQAREomoCCEEUT1flE QAREQAREQAREQAREQAREQAREQARyCSgIkYtFmSIgArOZQE9v/2yunurmCfTMkY7VEERABERABERA BESgEwkoCNGJWpFMIiACpRLombfEe6k9pZahk7ePQE/foHNzB9ongEoWAREQAREQAREQARGoS6Bn 1Ke6e7VDBERABGYrgdGbzt244kZHR2ZrDbuyXj29PvigmS5dqXtVWgREQAREQAREoPMJ9JAUhOh8 RUlCERABERABERABERABERABERABEag6AWIQehyj6lqU/CIgAiIgAiIgAiIgAiIgAiIgAiJQEQIK QlREURJTBERABERABERABERABERABERABKpOQEGIqmtQ8ouACIiACIiACIiACIiACIiACIhARQgo CFERRUlMERABERABERABERABERABERABEag6AQUhqq5ByS8CIiACIiACIiACIiACIiACIiACFSGg IERFFCUxRUAEREAEREAEREAEREAEREAERKDqBBSEqLoGJb8IiIAIiIAIiIAIiIAIiIAIiIAIVISA ghAVUZTEFAEREAEREAEREAEREAEREAEREIGqE1AQouoalPwiIAIiIAIiIAIiIAIiIAIiIAIiUBEC CkJURFESUwREQAREQAREQAREQAREQAREQASqTkBBiKprUPKLgAiIgAiIgAiIgAiIgAiIgAiIQEUI KAhREUVJTBEQAREQAREQAREQAREQAREQARGoOgEFIaquQckvAiIgAiIgAiIgAiIgAiIgAiIgAhUh oCBERRQlMUVABERABERABERABERABERABESg6gQUhKi6BiW/CIiACIiACIiACIiACIiACIiACFSE gIIQFVGUxBQBERABERABERABERABERABERCBqhNQEKLqGpT8IiACIiACIiACIiACIiACIiACIlAR AgpCVERRElMEREAEREAEREAEREAEREAEREAEqk5AQYiqa1Dyi4AIiIAIiIAIiIAIiIAIiIAIiEBF CCgIURFFSUwREAEREAEREAEREAEREAEREAERqDoBBSGqrkHJLwIiIAIiIAIiIAIiIAIiIAIiIAIV IaAgREUUJTFFQAREQAREQAREQAREQAREQAREoOoEFISougYlvwiIgAiIgAiIgAiIgAiIgAiIgAhU hICCEBVRlMQUAREQAREQAREQAREQAREQAREQgaoTUBCi6hqU/CIgAiIgAiIgAiIgAiIgAiIgAiJQ EQIKQlREURJTBERABERABERABERABERABERABKpOQEGIqmtQ8ouACIiACIiACIiACIiACIiACIhA RQgoCFERRUlMERABERABERABERABERABERABEag6AQUhqq5ByS8CIiACIiACIiACIiACIiACIiAC FSGgIERFFCUxRUAEREAEREAEREAEREAEREAERKDqBBSEqLoGJb8IiIAIiIAIiIAIiIAIiIAIiIAI VISAghAVUZTEFAEREAEREAEREAEREAEREAEREIGqE1AQouoalPwiIAIiIAIiIAIiIAIiIAIiIAIi UBECCkJURFESUwREQAREQAREQAREQAREQAREQASqTkBBiKprUPKLgAiIgAiIgAiIgAiIgAiIgAiI QEUIKAhREUVJTBEQAREQAREQAREQAREQAREQARGoOgEFIaquQckvAiIgAiIgAiIgAiIgAiIgAiIg AhUhoCBERRQlMUVABERABERABERABERABERABESg6gQUhKi6BiW/CIiACIiACIiACIiACIiACIiA CFSEgIIQFVGUxBQBERABERABERABERABERABERCBqhNQEKLqGpT8IiACIiACIiACIiACIiACIiAC IlARAgpCVERRElMEREAEREAEREAEREAEREAEREAEqk5AQYiqa1Dyi4AIiIAIiIAIiIAIiIAIiIAI iEBFCCgIURFFSUwREAEREAEREAEREAEREAEREAERqDoBBSGqrkHJLwIiIAIiIAIiIAIiIAIiIAIi IAIVIaAgREUUJTFFQAREQAREQAREQAREQAREQAREoOoEFISougYlvwiIgAiIgAiIgAiIgAiIgAiI gAhUhICCEBVRlMQUAREQAREQAREQAREQAREQAREQgaoTUBCi6hqU/CIgAiIgAiIgAiIgAiIgAiIg AiJQEQJzKyKnxBQBERABERCB8giMjrrRG8POjY6UV4bO3BqBnjmuZ+4853p6W/udju5uAm3syz1z +pyjzSqJgAiIgAg0JNAz6lPDI7RTBERABERABGYzgZHrbmT4pA9A6HLYcWr2gYg5gyudm6N7Jh2n m04UqAP6ck/ffNczb1kn0pFMIiACItARBHp80uMYHaEKCSECIiACItAuAiNXzigA0S74U5XrZ6aM Xjk71VHaLwKBQCf05dHrl527eU0aEQEREAERaEBAQYgGcLRLBERABESgCwjcvNEFlaxuFUfl0FVX eakl75C+PHrzauqaqzwREAERqBQBBSEqpS4JKwIiIAIiIAIiIAIiIAIiIAIiIALVJaAgRHV1J8lF QAREQAREQAREQAREQAREQAREoFIEFISolLokrAiIgAiIgAiIgAiIgAiIgAiIgAhUl4CCENXVnSQX AREQAREQAREQAREQAREQAREQgUoRUBCiUuqSsCIgAiIgAiIgAiIgAiIgAiIgAiJQXQIKQlRXd5Jc BERABERABERABERABERABERABCpFQEGISqlLwoqACIiACIiACIiACIiACIiACIhAdQkoCFFd3Uly ERABERABERABERABERABERABEagUAQUhKqUuCSsCIiACIiACIiACIiACIiACIiAC1SWgIER1dSfJ RUAECiJw48YN9+KLL7rLly8XdEadRgREQAREQAREoJsJ7Nmzx505c6abEajuIlCXgIIQddFohwiI QLcQeOKJJ9wf/dEfuZ6enm6psuopAiIgAiIgAiJQIoH//u//dh/84AfdhQsXSixFpxaBahJQEKKa epPUIiACBRH46U9/6r785S+79773vW7evHkFnVWnEQEREAEREAER6GYCb3/7292yZcvcpz71KTcy MtLNKFR3EZhEQEGISUiUIQIi0C0EePzij//4j92b3vQm97rXva5bqq16ioAIiIAIiIAIlExgzpw5 7kMf+pD7zne+4/7hH/6h5NJ0ehGoFgEFIaqlL0krAiJQIIFPf/rT7tSpU+5d73qX6+vrK/DMOpUI iIAIiIAIiEC3E9i6dav7hV/4BfeZz3zGnTx5sttxqP4iME5g7viWNkRABESgiwj86Ec/cl//+tfd u9/9brdhw4ZJNR8dHXXHjx93R48edSxcOXfuXLdw4UK3efNmNzAwMOn4MjMof9++fe7cuXNhSid3 V9atW+fWrl3r2E6Zzp8/7w4cOBAW8WQNjf7+fnfbbbcFNinlQD/Hjh0b109vb69bunSp27JlS9BV Slluev0cOnw4PPfLlFvaCrpZvnyZFyPtOiMXL150R44ccVeuXAlrnNBWN27c6AYHB1MiCe309Okz 7rjX0cjoSGin9J8NG9Z7PmkDfteuXXMHDx50p0+fDgwIOK5evTr0odTrwKAf+rLph7aCk7J48eKk a9LQf3CIaCvwYRxZsGBB6Mv06ZSJ8hlTzp49G4q18W3NmjXJx7fh4WF32PflK5evhK7b3z/gx/xN bv78+SmRhLIY39AP4z9MlixZ4rZt25Z8fLt+/bo7dOiQO3HiRGij7ew/rK2wf//+8esP4xvXZNik TPSfoaGhcf1w/WGMRT959gGPe37/+993jz32mPvkJz+ZUlSVJQIdS6DHd6TRjpVOgomACIhACQRw FN/znvcEJ/YrX/nKJAOTYRFHASNw+/btbuXKlcFQZ6VrjNR777032foRGIAvvPCCQ6YdO3YEWS9d uhTe5oHTTV4qR4pVvnmLyPr160PgBpkI1ODg3Xnnnd7pXl6CtvJP+corrwQn6vbbbw/Bh6tXrzry cGjuueeeXEMw/0zOjVw4Um/XlPkjIze9fl50vXN63dZtW0NQ5tLFS27P3j3BMMbBTJVwcHft2uXb 6yqvo3Wh2BMncDQPe+P4Nv9s8tJUovi+c9QHiI4FoxwH2xwZnO8777jDzW1x5tGcReunJfvNmzcd 677wuXPnztBv6T/PP/98eFab9pOq/+BAIcuqVauC44RjiWPHWEMgD6c7VcKppN/iNBGQQT98Z2bY Aw88kHR8QxckxhACIAQ6d+/eHfpPq+PbTPry8KVh9+JLL3keq0IQkXYxdGLIB9KOh0Ae14FmU8/A ItfTv6jZwycdR5sgAE79V6xYMX79oe1y/UkVFGGMp80yrt51111hXLXrD44/OkuVGN+QBR6Mq/Qf Amlcl1P2H7vu0Ubv8GMZ8hAoQg5kvO+++3L7z9e+9jX3J3/yJ+7RRx91b37zm1NhUzki0JEE/Pja k/YWWkdikFAiIALdRuCrX/1qcOx/7dd+LdeYI9BAAAKjEwMdY4dFK7nLjrHOnZhUCSOLGRCbNm0K dyqRZdGiRcFIRsZU0zupN4Yxd3mYOcIdXO6IEZAgb5/fhyGWIhEMIfiBM0fgAyYY5TBCdzhTqdLQ 8aFwV3u9Z0IbCfpZvMgt94YpDt3582lWRSewdmD/gaCTdevWOu7M8YdDBRvuNKfSz1UfaDhy5FgI DhEogwlthNk7BIu405wqcScZ55++y51K6z+0Fe5koqMUCf3QLikfBwpnmz4EE2aIoB/YpEg4kXBB NzbbAP3AhDbC+IajlSIRDMFxo2zrP8jFuMv4ZrMjypaFIBU6QCcw4ZP+s3bNWjfgZ0MQEECHKRI8 4ML4xh9txq4/BAPYlyrRVwkKMduA/gMTgoqM+wTQUr1+kvZIkBkWtBWuPbXxbXXy/sNaUvQRdEM7 RQ76D4EQ6z95+mHtKa7df/3Xfx2Y5h2jPBHoJgIKQnSTtlVXERABhwH+N3/zN8Go4jnNvGSPYGTv 7DNdGYeOad045WUnjF4Mce7IscJ2nDCASBiCKRKGKOwwQDHQLWEUcieIfaleQ0YAAjZZJuiHP5ik cBhwXIaGTngjdK430CdO117iOZGGho4bqlI/mX0xfHk41D+rn9pMhGvJDN/j3rlnhkh2ijSMcB7O nD2XpP8AHGcbHhj/cYIJzgOBiBQON/2DcQMHO9YPMiELM0RSBUQoh4AH/YexxRKOFJyQM0VAhD6K LARksn3ZvjP+pUgELy/5PwJCtFFL8Fnkg4rwSBUQYfyiTWZnXnDt4Y/xj2BE2QkZKAse2f5j+qF/ pUg4/lyD6D+0U0t2bUzZfwj8wz+rH+Qy+4D2lE30cx7/fO6559xTTz2V3a3vItB1BBSE6DqVq8Ii 0N0EvvWtb4U7km94wxvCXcg8Gsw8IGWdBfIwyLjbgVFUdqIMnBfkiJ0FyrXntjHMUiQMcIzSekyQ wbiVKQ+OPwYewQ+cyDiRh3zoJ88IjI8tYhv9XL9x3cvSM0kW09nw8OUkTu75C7V2EDtQVkfu5JK4 w5oiXbxQK2dyW+kZ6z/Xk/QfHBOcBZjQNuJEHn2KwBltquxEP6X/xA6UlWl9OVUQz5xpK9fk4JM8 +k+KtkL/QUf04+z4Zm0nxZhCvRlnR/3aJQMDtwIQ5JNsDZMLCcZaAjO0FXjkjW/k0Y5S6gddmD5q RFyQDVlos7SXspO1Weu3cXkmW6q2Yv3Uyo1lQT541Lsu8xgGgfIvfelLSQJ9sWzaFoFOIzDxqtxp 0kkeERABESiQAIbmE088EQy8X/7lX849MwaeBRjyjAwzDFM4uRjoyINhk004VRiqzMhIcVfMmFj9 Y3ksz46J9xW9jZHOXUnqntVPbLij67IT7NGP6SIujzz+RryDm2LWjN25Nl3Eslge7anshAFOYAZd UP9sMlkuJ5DF2mOefpADGek7KYIQJovVP+Zi7TjFmEK5JkveuJJSFhvfrMyYiemMvpPCybW+Mcev 7ZJNvb21dpyizdIe+YvHslgeY5WirSAH4y2y8Bcn69v0HY4rO1nQxcqNy7O+TLtmPC4zNWsfWHvK ysKsQW6AsJ7Et7/97exufReBriIw2ULoquqrsiIgAt1E4Ac/+EF4lvO1r31teBwjr+4YVY0MGTPG UjiWZnxbmVl5Lb+ewZM9fibfp5KFc6cwRjGK0RF1t/rH9bK8JPq5XrsDaGXGcrBNPm+GSOHk3tLP 5Mt6j5+pQUIO+JWZOH+j/jNnzJm5keBxJmOS57igG9NbkrbS4G5xSjnQvdXXyo3bg7GyY+J9RW+b fvLkoCzLL3tcob2Oy5ITODM5rjfQYVFs6KPIYmXWO28K/VgZyFJPHvq7sasnaxH51gby5LA82JU9 1k5Vhsli7LJ1Z//P//zPhxsLLFRpwePscfouAt1AYLK10g21Vh1FQAS6jgCG5t///d8HB+lXfuVX 6tZ/KiPGjIypjqtbQAs7piojpSzmvFqZcTUsz46J9xW9jR7NybVy88pIYRjf9AEGUiM5kHVkpNy7 c87FTPJo1PLQT9k6uqWfOo7LWBCC4EzZqVH/iXVWNhPqaWXE5Vr9La+RvHbsTD9v6adxuzV5Z1pe o99PVd9UXGp9dKwvNxB4NATwyu3LsX7yREnFhLJj/Vi5JhPfLa8T2gpywa7sBJNG5TTDxG6CPPvs s+EtPWXLrPOLQKcSUBCiUzUjuURABAolwOuznnnmmbAI3IMPPlj33GZE1DugkQFS7zfKn6UEEhi9 zZNrxgCfOKW6+XOXcGQz4pZQbLtOqXGjXeSLLTdll290LWq0r6gax2Xktd+8vKLK7tTzwCTmkpXT mNhndj/fWbzyNa95TQhMPvnkk3mHKE8EuoKAghBdoWZVUgRE4Jvf/GaAcPfdd4cVtusRsenI9fZb ft6z3bavqM+pyjBDx54TLqrcvPOYLFZmfIzlNcsu/m2r27ERaOXmnaPdTJAJhwV5Wbiy3EQZtWfZ 85wky+NRiEYGdBEyUtdaGbdmZ8TnNZ1Ze4r3Fb3dqAyTgzJTtFuTJS7X6mt5dozll/E5Vf8xWVIy qVdPk6VsLrU+WjOHrcw8mXirffn951YZebJYXgr9NCrD5IBT2Uwoo5k2ELdtflNGasQkLm8qeV// +teHw7/3ve/pkYwYnLa7ioCCEF2lblVWBLqTAKtmMwuCxPOYjYwmnFczNGJDy8hZXt7K8nZMUZ9W RjwtNj63ycI75MtOJkve1FuTI2/l/6LlQjcYeJRp5cZlmHwpZLFAR54ctbxR39b8Gzv8KzzLTrbI ICv8Z1NtGrk35Of2NmXMZ3/fyndeV0r/YqJDLpcwpb32lplWzjudY+M2m5WF7/whq7GbThnN/sZk ycrB763N2jHNnnO6x1k5Vm58HsuzY+J9RW9bGVZm9vzGyo7L7i/qe60N1PqolRmf2+Tr75+8QHB8 XBHbjCm0R+TIk8XyymZCXRhDYUP9rVyro31nPE7Rf5g9QDJdmBx8Igt/yGHX7nh/kdvoxwIMxiA+ v+VNdf3ZsWOH448FRp9++un4FNoWga4hoCBE16haFRWB7iXAStQ8jjE4OOgeeOCBKUGYwZO3roDl 2TFTnmwGB1AGRqCVGZ/Knk3F8EphBJpRlRcQsbwUTDAAkQVjL8sFA9XyUsiCHBi9NSN9ovM/MsIi aSPBYJ3bN3nF/ViXRWzPm18LRPFmimy6MZZnOszuL/J7zSnpD9NArF3E5w+L+/k2nUIWC87RJsw5 MFnIQ284cxZMsn1lfMayZM+fss1SNuMgKW/xPMtL0X9sfMtrJ+ShM8a2FPoZmFdzck0XAdDYPyaf 6TDeV/Q29bXxzcq1MuBheSn0gxyMt/QT/uJk+mF/ioAIr7Uk5enH2izycr0sO1n/yZPF8uyYerIs XrzY3XvvvWE3rw1XEoFuJKAgRDdqXXUWgS4jYFMeV61a5TZt2jRl7ZctWxaOya5cjRFIHkZXCiMQ gxRjBcPGjBsT3l6zZ7JaflmfS5cuzWVCJm/nwPhbsmRJWcWPnxcnFyYYxVn9WB7czGgd/2EJGzgl lEW5N27cnFDCdf/mDGYlLFq00LMp/1K7aMEi5kW7q1euTpCDL1d8myWl0E+tnMWhPHMOwhf/T43T Ddfn7yam6D84JTgDrKxP2XGi7dCfaSd2ZzPeX/Q27Gm71m/j86fsP5RL/yHlyUIebXrRIt+eSk6U g37yxjfr26nGt8W+voxhef3n2tibXJYsqY2BZWKxcdSuNXFZtGH6lI2B8b4ytuk/jHHoJ9uXCUIg D+06RZCI6w9s6MuwiZO9OcOuUfG+MrZtHLU2amWYzrAPpgpCMObcd9994acsUHnmzBk7jT5FoGsI lG8ZdQ1KVVQERKBTCfzwhz8Mou3cubMph2Pt2rXBsDp58uSEKp0/fz4Y7suXL08y+wBDZfXq1cH4 On369ARZTpw4EYzRNWvWTMgv6wuGFw7BhQsXJgREMESRBcfGjLOyZLDzwgTD9+zZsxMMUh67wSBF fxisZSecAWTBIKdtxOnsubNBhlT6WbBw0C30DvWlS5cmBGd4FAMuAwPzEupnVdAP5cYJ2XDoVq5c maT/UPb69euDs0RbiRN9m7a7bt26OLu0bZw56o0c8St1kYE8nBbGlRSJYCwO5qlTpyb0HwIQ/NFm 2V92YnxDFpzc7PjGd/rXhg0byhYjnJ+g2MKFi9yly8OT+s9Z7yCiv6VLa8GbsgWinVD3oaGhCcGz ixcvhrZDm03h+FNPxlICEHn6Yf/mzZv5KD3Bn+uPXYOtQBx/HHj0R1tKkVasWBHaA9e9ODG+0X/o x80EWe+6667Q76nT888/H59K2yLQFQQUhOgKNauSItC9BDAMXnnllQCAIEQzCQNi69atwXk7ePBg MAQ5z65du8IdQvalShiBGDX79u0LzgpOy/Hjx93Ro0eDAZjK8cdh2LZtWzBI4YBhyh+PupB4vjWF 409ZBDxwTuBw5MiRoB+CI/v37w+OdirDGFlWr17lFnnnBTkwJpn9gHF65vTZ4OBOdUeMcxSTenx7 2OJP1eMOHjgQdENwZL/fxsnbtm1rMv309fUH5x8eR47W9HPp0rA7dPCQG/R9izadKln/oW3grOC0 0G5w8GhDqe6003+2bNkSgi+MRwQi0AvbOC633XZbkmntcGd82759ewgoMr7RTng2/aWXXgpOUSrH H1kIEjGGMb4RtGJ8O3bsWNAP/XjhwoUcVnpi7Nrq9UPQjrZy7eq1wGXf/n2+zbgw9tG3UiTGDPRD ez18+HBggn4Ya+GRcnyj/xAUQQ7kMf0w3qG7FDPOjDlM6Ec8WsksBNqt9SX6T6rADAERbACCh3H/ QT9mOzRzLbT2TT1+9KMfWTX1KQJdQ6DHX5AnzmvqmqqroiIgAt1A4Nvf/rb72Mc+Fqr6d3/3d8FZ brbeGF04LBg83JniDojdhW/2HEUcxzCNEYjRg8HCXUrkSHXnNK4DxjAGKEEZDC2mbXNnDsMsZYIJ +iEYgzOHAWr6QVetpJELR1o5fNKxGOY4tTjdI34diL6BfrfKG+427X3SD0rMIDB01Dv+w8OXQyk4 CdwhTK0fCr944aIbOjHk+8817zzMCc4msrSqH841Z9F6PqaVLHDHnf+4/xCAaMZZmFahdX7EWHLo 0KHQfzgE/eDMNXPntM4pp52N00//sfENHvRlHL2Uif5L4MGcXNoqszGmM71+pn2Z4NDQ0HE37AND rK66YMFCL8vqlgNEPQP+8Y7+mT3Sgn4Ya5ndRZ9hvCcokFo/Nr4xe8j6D/0YeVL3H1hwLSTozDWA 6w9tJWUwxPpGXv9BP60EQx5++GH37//+747Xhv/pn/7plI9xWNn6FIGqE/BjR4+CEFXXouQXARFo SODxxx93vIsbw5ZVqKdjwGGEYWzx1+6E4dUpcsCiE2SZqX5m6rjcahOj/k7h6LSc7FvnKGaLGRnc ue0U/dReUzr9/jOTIERMlLYynSBIfI4itunHpE7RD3J0giwzHd+K6ssz1U8RQQhrZzMd3+w8RXyq /0ymOBP9fOYzn3Gf/exnQ4D2i1/8YtJZYpNrohwRSEfAX28SrJaVrj4qSQREQAQmEMCQtGctmQo9 nQAEJ8Rp6QQDHVk6SY5OkaVz9NPTEQ5urZ10TputOf3TD0BQn6JSJwQgqAt9R/1nslY7hYn0M1k3 5Kj/TOYyk+uPvSGDWRU8BqQkAt1EoLU5q91ERnUVARGoPAEeGWA6Kynlc86VB6cKiIAIiIAIiIAI lEpg48aN4+dnrSUlEegmAgpCdJO2VVcR6DICrOfAom+krVu3hk/9IwIiIAIiIAIiIALtJsAaLDa7 xBbQbrdMKl8EUhFQECIVaZUjAiKQnADTG1l0jbRp06bk5atAERABERABERABEcgj0N/fHxbWZN/e vXvzDlGeCMxaAgpCzFrVqmIiIALcWbCFzlhBW0kEREAEREAEREAEOoUAb9Qg8Wpn3s6iJALdQkBB iG7RtOopAl1IwKY38m51XuWlJAIiIAIiIAIiIAKdQoDXbZN4dFSLU3aKViRHCgIKQqSgrDJEQASS E+B95qwJQeId4gQilERABERABERABESgUwisWrUqiEIQ4tixY50iluQQgdIJKAhROmIVIAIi0A4C TGvktVekxYsXKwjRDiWoTBEQAREQAREQgboE7FHR69evu9OnT9c9TjtEYLYRUBBitmlU9REBEQgE WJDSghDMhNDjGGoYIiACIiACIiACnUTA1oRAJtaFUBKBbiGgIES3aFr1FIEuI3Dt2jU3PDwcas1M iLlz53YZAVVXBERABERABESgkwmsWLFiXDzNhBhHoY0uICCrvAuUrCqKQDcSIADB9EYSMyGmm27c uOFYX4J3eRPI6Onpme6pZvQ73vJBYIXU29vb1qCKMUEWXjHWTibIMjIyEpjApX2yjLgbN26Gt7H0 zul1vXOJ8benrcADLrCAib2HHn2lTaOBCf2n1/ef3rnopz33Pug/cVtpZ1ASHshCamf/oXzGSNqL xjdo1BL6uTly0/fe9vYf2myQpUOuP3H/aedYS3u1a3unXAtn0n+WLFliTc+dP39+fFsbIjDbCSgI Mds1rPqJQJcSuHDhwnjNBwcHx7eb3cD4O3DggDtz5oybP39+MAb57ebNm8MaE82ep4jjCKjs2bMn OC7z5s0Lq2gzu2PLli3JgxEs9nnkyBHX19cXnFwCIxs2bHC2wncR9W3mHBihhw4dcidPngzrfZjT vW3bthkFnZopO3vMZa+fg4cPuxvXb7iBgT4fLLruFvrA13rPBSM5VcJpOXXqVFiQFecWw5jHklav Wu1WrFyeNACAPo4dO+4fiTrraLMj3qnz4rn169cnX5+F/rN3797guCAL68UQmKQvDwwMpFJPKIfp 1rRb+g9tg8XomI69bt260J9SCYMjx/jGnVeYMN7RfhhTYqcohTyXLl0a1w9jLUyQAf2kDhbB4+jR o65vbp+b0zvHXfd9ebV/vfPy5cuS6seuP/Rn2qoFI7Zv3568/6Af2gr9Bv0w5tNmGGtpxynT0NBQ uP7QdyibtsLCjowrKYOt6OPgwYOh/8DEAr9bt25tuf8wBtHOGTPtEdKUTFWWCLSLgIIQ7SKvckVA BEolEN9RwEhoJWGM4/QTgNi5c2dYTwIjg1d+vvjii+7uu+9OZghi8L3wwgvB+KNcDC8cy+eff969 /PLL7s4770zm6BKA2L17t9uxY0cw/LjTjlO1a9eugDdVIAJdoB90fNdddwUjHQMORxMu99xzj5tO 4KmVNmLHEnDY7dsF5W27fUcwJq9euep27d4VjPYdt9+ezHmhveLgYpCvXLnSl+u8kXzG7d23z42M jiQMFI0GOTCoaSuDgwTxRtxBL9vu3a/4NntHaM/GsMxP+g9tgjGAtkJwxvoPbZn+k8rRRT/0FRwV +gqOFE4meYw5BPNSJAtAsBI/4xsOP+XzekDGGvpPqjV0cCIZU9HPfffdF5jg7CIHn+gs1eymsz5g tm/fft9/1gX9UC79By6ML2vW1F6lWLaO7PpDQAT9EHDG6UWOZ5991t17773J9EO9X3rppdBHaBdc fwjqkYeOkCWV809f4RpMP2GMo//AiOsgbZqgVYqEfgjKEIynzfLmLfL2+XGWscbympWF8YdxCdYE fNA1dVMSgdlOoD3zImc7VdVPBESg7QQuXrw4LkOrQQhmUXDHBUcBA9CmtXOX0Ax4jI4UCTkw+jZt 2jQ+dZu7UBhhGGXsT5Fw3A77u/3clYMLhidc2MYIwyizKbJly0O9mQHBHTDKRg6MY4xTDDlkSZWO HTsa7gxyNxsZkGXe/Hl+5sFKd84HSc6ePZtEFNrl4UOHvQz9IQBR088ct2zZMrfIMzrkdcdd3RRp +NKwO+l1xLPOtBcewcDQXu/v9o/6YAjGu3Np+g93K3F0aRvccQz6Ges/6CZV/0E/OJG0EWY9wANZ CBYxxpicKfTD2MjdfvSzdOnSIAfthTGGcY1AFvKmSNSbYANlGxPGaxjRz/lLkW7evOEOHjjk+r1+ 0ImNbytWLPdjzALP60hwDlPIQnCVABFjKwEi2glsaMPmAKeQgzIY8+k/GzduHL/+0KfRD3IyDqdI 1JvAM8461z5rK8uXLw9jHO0ZBz5Fov8whsGAvktbIWiAfYCuuP60Yh/wW67pJK6hBCGURKAbCCgI 0Q1aVh1FoAsJ4DRbavWuOAYNRjgGRpxwYjCQuaPJHdayEwYJThKGDc52nDC+SKmMQO5qEwzhDiny xAlHF0M11VRScxyz08Yx5NAPjksKJwr9cKcUg9iMSOOC4w+nU4mMdByCq9eu+sdABifcmcRAph2P eMP27PnaK2tNxrI+h1jh3TsNPJISJ/oPTjjthBkkKRIzdXBccJziRDuGDf2nFYchPkcr2zgu/JnT H/8WWWhLyJoiUWeCddn+QztGFu4uExgoO1FnAkHoJzu+IQcJZzxFunjxkp8hc8Ut9OXCIU60HXjB JUWy8c3GeCuT/sM4w/Unvr7Z/qI/GUORJa//WHCEa2WKRICBNslYFuuHMZa2wvU4lX4oBzb05Tgh l9kHXCubTQQhrE7olbamJALdQEBBiG7QsuooAl1IIDbSsg7iVDjMmcZhyibyMEBaMTKy52j2O3XA +MIIzDr+5JGYtZHCiYIJ5ZixFNfBOOEIl524S0TAAx5WrpVJHvKhnxSyYBRzB7XHO7M4tHHCsCRv 2MvK3f+yk62BMrdOm6X8S94JLjvRRoIsXhdzPINssv6T4q4lZeDo0iay/Qc5yOOYFHcerf9Yv425 4FySCFKkSDa+5cli+rH2VKY8jG/8mS7ismycoR+nGN+sPfb1TQxAIFPvWFAiBRPaIrLQNhlD4mTj 23gfi3eWsM04i3OPHFlZGNv44xj6WNmJwAvJ2kVcnuWlGPMp1/oP7TabrP+0Ikusa3hz/VISgW4g MNFq6oYaq44iIAJdQSCeqWBGSjMVx8Cz32YdS35veSnuFFoZWQPQ6oHxgtEaB1xsX9GfVobVPz6/ 5Zm88b6it6kv+qHuVq6VEeelkIW7+bSXOV4Wyo4TsvEYQk3e8o30a34WBCmvrVheTYflPgZBfVn/ ARpZ/QT5PBcSszbKTnGbzdMP8nHXMYUTZe0xj4nlmbxlczFZrF3E5VmeHRPvK3rbyrD6x+dHX/zR nsrXz2iYBUH5eYEz3uxCSqEf6lpvfEMG0w/Of9kJOXCITRdxeaYz9JOCiwX9rdysLMiIHGU78Iz3 Vt96siCbte1YznrbyG56hWfZdagnh/JFIDUBBSFSE1d5IiACSQjEhqtd4JspGKcEQ6NewmAgxeev d+xM821appWZPZ/lW9Aku7/I7yZLnuFlcqRggm7qGcbU12RJwYRZEHGZ4Uv0D03F5I2yS9nE8Q9p rH3GhRiTmoFbv23Hv5nuds2Arl8Gs0ZIN/3rTMtOzbTZIIs3/MtOsCeZLuLyrE+laLOU24wsxi6W s+htKyOPCWVZvh1XdPl2PoZ7Y0JAMZtSyUG5yMGflZmVxb6nGGuNO7Jk5bHv9PcUTrP1DSvXOPBp eXArWxaYNCrDZGlFP/R/+x11aGR/xPXWtghUnYCCEFXXoOQXARHIJWB3K9jZShAiNjDMMIgLsLz4 uHh/kdtTlWGymLFYZNnZc2EcTZWaOWaqc0y1HwPNjDSrf95vpmKX95tW80bGglWN5KgFIeo75a2W mXd8XMZkF+rWL2rHlTvVlzL4867BuGF9S4JbW0n002Bac6yzFLI00zdSyHFLP7ect1taubXVjLy3 jp7e1lT1NR2VPb7V+sXUfTRmN70aT/2rqcowJlOxm7qkqY+I24CVG//K8uLj4v1FbjdT32aOmalM lIGOSFb/+JyWZ8fE+xptm40CyxT1aCSL9olAKgIKQqQirXJEQASSEogv5GYYNCNAfGyeIZGX18x5 yzzG7qSWWUbMpV45KeSoV3Y2vxl5s79p9bs5/GM2aZ2f44zX2VVQNudvroySBfH18bVtqlYp9NOU IMjcHLxmT5d7XDNlNHNM7slbyGy2jGaPa6HoaR/aSePKtCtR8A9T6CcuI++6Z3nxcQVXc/x0zZTR zDHjJ5zmBmVYOVb/+FR5efH+vG3at7Xx6fw+75zKE4EqEFAQogpakowiIAItE+DOAncV+LO7DM2c xIyBeseakdDKOhP1zjVVvpVhZWaPt3w7Lru/yO/G0MqMz215dky8r+jt2AiMA03ZclLIcqut5N9F JTiBfzvHrw1Rbrq1PobpIi7PgiRMNe/tLVeWnjlmpNuMiFiS2uMp5OQ9ez/xyJl/a9QGYk639Djz MuudwWSJy7VjLc+OsfyyPq2+ef0npSw2blmZ2fpaftlcGFOsX1iZE2QZ697IYQ7ohP0FfkE3pp+8 05p8jY7J+9108hrpx+TgvClkadQGTBbkSKGfRmWYLI3kzerCzpdiRkm2bH0XgXYSKNcaaWfNVLYI iEBXE8DAxhDgL8/YrgcHw8sMCDMo8o7NW1k+77iZ5FkZ9eQ3+Vp9+8d0ZGoki8lhx0zn/M3+BkMT HVmZ8e/IM1YpmNjq6FZmVhbvcgcDvXfuxFXu4+OK2u4fW6k9T5aRkdqjNHP9yv8slllmou+gI/y2 PB2ZfP1jbxwoUxZrj5SZlcXycADM2SpTFmsrWTko05jYMWXKwbmNS7tlaSQHcpp8dhx5ZaTQBhr1 n7G326TQD2XY+Gb1j+tseWUzoUxkgQ1lWrkmi32nv6foP7z6kpTnqJt8yFF2QIQymrEP7I03xqvR J/IjN+e1gESj47VPBGYLgXKtkdlCSfUQARGoHAEzkhA83m6mIrwXnpS3uBQLZGEoDA4ONnOqGR2D I41hYotyxSdDNuqFsZPCCLT65jGxPDsmlrPobeqKQYrjluUCD3t+3HRYdPnx+dAPxuPIyK3gh+2v zcIZ8brp88b85Nf+2XFFfc4bM9JNF/F5jYkZ8vG+orfhEQxwr4ubk953P1rrU77/mLxFlx+fjzZA X6X+2TEATrQhdJii/1h7zLZZ5LW8hQsXxuKXtm2yxOvmWGEpZYE9+qnXZm18M6fPZCzjc3B+bTy3 vhKXYXkp+g+OP+VQ96zDHY9vKdoKcsC+NpZNXBOIPOSh77TicMdcW9letGhRODzLhEzTD9efFE68 sc9rt/QfZGi1rcCSZJ/hi/4RgVlOQEGIWa5gVU8ERKB1AsuXLw8/yhrpGAgYGRhdKRxuDNJly5bl Otz2yjL2p0hLly4NDnfeq8fIw/BasmRJ6aJQDrKgi6x+cCp5dR13CXFwyk4DA7VyRv2d0hvXa2/K sDIxUMlfsmSxz2punQT77XQ+g5Hu2Vy9em2SIWuv81u6ZOl0Tt3ybxYvps7+NXX+lXlx4g0e1z2n vrFAUryvjG0cJLjQZ7POi+XhkKdwcmGCPNZv4/rSjq1dx/llbVs/tXYRl4N89B9ztOJ9RW8zvlFO XkDRZLOxuOiys+ezdmDlxvtpK6SUYy3jW3asNU60I+tjsZxFb6MfrnOMZVmHm+/0KZik6j/0EfoK HOJknLgupEhWTvb6E9sHFjRpVh5+qyQC3UZAQYhu07jqKwIiMCWBtWvXhkDDiRMnJjh0p06dCobh mjVrSp/2iZAYd5SF8YUscTp27FhwapA1RcLoxeC8cOHCBOcfAx3ZVq5cGRy+FLKsXr06TBU+efLk BP2cPn063BXbtGlTYFa2LDzasG7dulDm2XNnJxR3+vSpcf1N2FHSl8HB+W6ZN8IvXpyoH5yFc+fO BWdvwcLaDJ+SRBg/Lfrp7+t36CNOyHHDv9YUZikcF8reuHFjaCO0lTgdP348yLBhw4Y4u7Rt7ozC 5fz58xMCEdzFPXPmTAismXNTmhBjJ0YOnEvGs9ihQz84VuvXrw/9q2w5aAOMXzaGxOUhG852Kv3A Y4kP0l3yQZgrl6+Mi4J+zp49F8a2hYn6z6pVq0Ldjx49Okk/sEo1vgGBsgg2xNcfHGa+c12if6VI BP65xpw9e9aPcRfHi0Q/tBWuTytWrBjPL3ODwBhBqyNHjky4/iAbARGu2Skelymzjjq3CKQg0Puo TykKUhkiIAIikJLAM88843784x+HIt/0pje522+/veniMY65kz40NBTurPMdA+PAgQPB0Nm2bVsS JxeBcV4wAjF4MPrYJgCB4YUcqe4UUjaGHhwwQDEKL1265Pbt2xcMru3btydxXGDCHTr0gyOJ08Qj ADiZBw8eDAbg5s2bW9LP6LULnHZaCf3gWJtRjlOHXBfOX3BbtmxJcke5JnhPMIwvXrgYdAQjDOJD hw4HPugHpy5FQh/9fpYITu1l79SxCCXbOFXLfSBrw4b1XozWZof0DNSmY7cqP/rBaUInJPRDXyJA cttttyXrP5RN/4EDbdX0Q/9BpjvuuCPJtHbkYDxjBgLjG0FFZCEQwvhGIIT+gw5TJJw50wntk6DZ 4cOHQxtmfGs1MDOTvszdaxxc2gayMLbs37/f85kb+jKcmk09cwdcT+9As4dPOM70wzjPzAx0QbvZ s2dPCGQRGEilH/oP+kEWrgGmK9rwjh07WtbPhIq2+IUZPLRXrn3Wf2BC4vqe4rEQykI/BK3MPkAX BBfpy8jIWAurZhOBlK9+9auhD3Jde9e73tXy4xzNlqXjRKBTCDzmU4+/OGsOUKdoRHKIgAgURuBT n/qU+/znPx/O9wd/8AfurW99a8vnxsnG2GKKMoYHMwG4G9OKgdFyoTk/YJjGMMZRwEi3u0I4NakT d+JggtEFB4wx7ma2YqAXITNM0A/OP842TgN3wtBRq/oZuXBkRiJhmMMDRwGDcmDegFuxfEVbDMkb N657I/20G/ZsmLS8YMFgkCW1fgBKv6Hd1gJFvW7J4kVu+TTvVs5ZROBiegn9EDxDFvoPdymZDdDq lOnplT7xV5RP/8GZQi76DzND2qEfHFz6D3rCkWI8gUsqB9fI0GfoOza+4YgxG2A6j4TMtC8zvp31 slzywQgepxocXOhlWRnGf5O3mU+CZj390wuc2fnj8Q2dEHBmjGt1fLPzTffTxjfaShjfBmqzEtp1 /SEIwXjLNcCuP+2YeRDbB+iHgBn9p1X9cP367d/+7RBk4hxPPvlkuI5NV1/6nQhUgYDvJz1pbotU gYZkFAEREIEMAe7S8dfuhFGD8Zlqummj+mLsMV2bv3YmmOCkTMdRKVpuM0BbvWtbtBycj4UwmQ7c CQkHgb92J3PgUs0aalRfgg0EHfhrd+IuN7Me2p0sgNgp49tqHwDxkYd2YwnXnk64/nTS+Mb1p1P6 T6fYB21vqBJABKZJIM18u2kKp5+JgAiIgAiIgAiIgAiIgAiIgAiIgAjMHgIKQsweXaomIiACIiAC IiACIiACIiACIiACItDRBBSE6Gj1SDgREAEREAEREAEREAEREAEREAERmD0EFISYPbpUTURABERA BERABERABERABERABESgDwkWmwAAQABJREFUowkoCNHR6pFwIiAC0yXAwoWW4m3L06cIiIAIiIAI iIAIiIAIiEB6AgpCpGeuEkVABBIQiAMPrO6tJAIiIAIiIAIiIAKdRsDsFdkqnaYZyVMmAb2is0y6 OrcIiEDbCFy+fNkdP348lH/16tW2yaGCRUAEREAEREAERCCPwMjIiDt58mSwV65du+Zu3LiRd5jy RGDWEVAQYtapVBUSARGAwPz5892aNWsCDN5DryQCIiACIiACIiACnUSA2Q+rVq1y586dc0uXLnWy VzpJO5KlTAKao1wmXZ1bBESgbQRGR0fbVrYKFgEREAEREAEREIFmCMheaYaSjpltBBSEmG0aVX1E QAREQAREQAREQAREQAQqQUBBiEqoSUIWTEBzlAsGqtOJgAjMHgIYBhcvXnSsL9Hf3+8WLlzYtqmS PCd6/vz58Lzo4OCgW7BggbPFrFIThwdcmEYKk4GBgdQihPJi/SADTNo1lfXmzZueySV30+tp3rx5 bnDBfC/jrTe0pATEGiiXLl0K7YO20i79ODfq5Rh2V65ccX19fW6h18+c3t6UKMbLov9cuHDBXb9+ 3XVC/0E/pEWLFrVNP3H/YXyj/6CndiQb3+hHPErXzvGN/jM8PBz6D7K0q/+gH9oJsrT7+sO6BTwu gJ46QT9cC7n+cf1hvG1XYkzpBPugXfVXuSIwEwIKQsyEnn4rAiLQsQQw4GaScJz27dsXDECe08R5 eeWVV9zmzZvdypUrkwYAzpw54w4cOBAcbBwoFtwkALB9+/akBhiG6P79+4Mxak4CcsFj48aNSZlg mO/duzcYxThyp0+fdjgP6GfZsmUzUX3Lvz179qw7fOiwG5g3EByWkydPBGdu3br1bt78dAbyqNfP 0WPHAouFixa63jm97ujRo26Jb7/r1q7xbSZdAIAF1g4dOhR0ssA7CgRnDh8+7NasXuOWr0A/6QI0 9B/aLQ523H+2bNkSnJiWFT7NH9B/YHDixInxgBlyLV++3G3atClpAA397NmzJ+gHR86CNMhha+lM s5ot/4xF+eBCABH9HPNtmO1t27YlHd9GR0fckSNHHf0ZOXp9wOzI4SNu2fJlbq3vPz096SYPM5bt 3r07XHcWL14c9MN4t2HDBrd69eqWGc/kBwQfKNv6z9DQULj+0H/glDIdOXIktA/aLPIcPHgwrKOQ uv9gD9B/CBItWbLEETxDX1x/WN+hXTcIUupCZYnATAkoCDFTgvq9CIjArCNAAAODgjsc999/f3As ycNhIBCBcYrjkCIhw0svvRQc6x07doSycWZ+/OMfuxdffNHdc889wRhLIQtOJc7C3XffPV5/DPaf /vSnwShdv359EuOL+u/atSsYfvfdd994/QkaPf/880FnBCZSpKtXrwUDvRaI2RAcFQxSdIbhvvOu O5M5L0M++HH06DEfnLotGObU/+KFi+6ll192bmTUbdi4IQUSX8aoO+D7ykU/C+Keu+9yff4uO/2H trN3n3dm+vvCDIAUwhBMpK3gKNxxxx2hfaIf2iz5tOVUd7oJlDGGIIc5kjgxP/nJT0L/wYFJ4bzQ f3Cg6LuMb+ZIoh/GPe66pwrkcSeZMinP9ENABCaMb/fee2+y4MzJk6dC0I7grtWf/vMyY41nttEH AFIk+srLvs/Sdu36Q7k43ORz55/ARIpEMOSFF14I48mddzKW9YTACLp57rnn3Kte9apk+iE4Rbvl Orh27dpQfWbkPfvss0Em2k+KhH641tCfH3zwwXH7gP7DmEIALZV9kKK+KkMEyiKQLqxbVg10XhEQ AREomAB3TjHQMXTMQcH44m4/AQiMQQyRshNlYHhx14U7PZRNYhYEDgsGPHeGUiRmHjADA2fODHTK ZZYIBhcBCu6upkjIwXRc7tjG08fRF2wwEFOlw4cPhaJqs2Nql1T0xN2wS8OX/KvXTiURBceau7bM UEFHlpgRsWTJYnfs+DF3zQdMUqRz5867s0E/q0MAgjLpP+int3eubyuHfU75/YdymamDIxUHyNAP d5QJANhrfDm2zEQfRhb0Q9uwxHf6E/0HeVIkxjYcKAIhFoCg3HXr1oUxBlloT2UnxjeYkBjfLACD E4d+GN+4654iXb9+zQfJ6D/0l4n9Z/HiRW7o+JC74XWYItn1hzZr1x/KrfWf3uCIp5CDMrjWEbSK +w9jLt/pV8zqSZFoKwTwmAFhATzK5Tv9idk0tJcUiWsP4wb2gOmHtkubJYCXyj5IUVeVIQJlElAQ oky6OrcIiEAlCTCFHaMHAydOGMfchcLYwQArO3FHECMcpyl2FigX55+EM5EiYXhxZw6nyZwFK5dZ BwQgMJ5TpFOnToVgQ1Y/Nl0YJwunr+xEG4ALgQ8zRq1M9AUnZE2RqDOO46B/hj2rHzjRns+cO5tC lOAQ+ALd/EHWxbiV6D99fXP9DCPWiSi//+A84STRZ7PPjcOEfoV+YFN2op0QZOAOdlY/9B9kxZFK kagzbSV7N512TB5tiRlYZSd7BMT6bVye9e1UQSLWLrlx47pb4NssHOI03/flkZGb7kQi/Zhjn53N FV9/UgSs0A/XFxzr7PjGuhC0Y66VKRJtkmsM7SKrH8Za2nOqsRb95NkH8DD7IFVAJAV7lSECZRGY ONKWVYrOKwIiIAIVImAGBEZfNuG4YIAwM6DshCOA4ZUnh+UhRwoniueCSdQ/m0yWVIYxdcYQzcqC EWh5TNMtO6EbDPUeL0vWsUQ+/ghUpLirfGmsvr1zJ+vHZosMJ7jTPt430EXOGhS0FRxuZomUneg/ lIUe6umHwFqKgJW1R+srcd0tL0X/odxG45vJYvLGcha9jX5gb302Pr/JkWp8s/rOzVmYc+7YmJdC P4wnNr7RbuMUj28pZKFvMMYhR1ZH5PHH+JYiIE8Qj5SVI86jXae4Fhp7a6NBsLF/kA8ZrI/F+7Qt AiIwkcDEEW7iPn0TAREQga4jgNOCIUjKGoFxHgZa2cmMuzzDy2RB3hSyYIyS8mQxTinkaKSf2EhP Icu1a7XZFnP8gnVZJ5fvLGQXyxsAlvSPydJIP7Snso30mzdu+qDLSFh2Msskbj/XrpY/U8XaAEyy slgefT1FkMj6svWVuBmYzkzeeF8Z2yaLlRuXYXkpZDE58pggE/n0Hxt7YjmL3KZPWBl5slieHVNk 2dlzEZSxoJiVGx9j+jF28b6it6kvbJAj239MNvqOyVt0+fH5CMyQrNx4n8mHHGWPb7RHY19PFmSz Y2I5tS0CIjCRgIIQE3nomwiIQJcTwKhqZMiYMZbC8LJgiJWZVY3lpzCOG8lictgxWTmL/I5u0BFl Wrl550+hn5s3a8EqL0puIt/kzT2gwMybfro4KY+J5REcwIguM434NwzwloF66ZYsNXb1jisi34IL Vma9c6Zot41kMflSyAEDK8fKjblYnh0T7yt628qwMrPnt3w7Lru/yO/N6af8dTLonza+5dXPmKQY 32LuVq7JZN9TjW8mi5VrcvBpeXAre3yjjGbsA5M3llPbIiACEwkoCDGRh76JgAjMEgJmmFCdeHuq 6sVGTN7vLC8+bqpzTnd/I2MnPqcZ0HFe0dvN1De1HKaLvLo2I2/e71rJszIayYEO/f+lploZzRSC LM0cN31x7fwwmYrL9Etp7peN2mMsm+mxubNO76hmymgk7/RKnfyrWI6YQfbI+LjsvqK+W1uZ6nxl c5nQf+oEFGsypuk/zXBJoZ+4jLy2YnnxcVPpcrr7m2kDKeSgDNOP1T+uk+W1Kov9zj7jc2pbBGYr gckPPM/WmqpeIiACXUUAI8CmcJrR0AyA2AjI+53l5U3FbOb8rRwTy9LodzZFt9ExM93XjCypmZgu 8urWjLx5v2slz8po7NfjjLdy1taPRQ6TpdGvmzmm0e+b2Wdl1JjUD3jYcc2cc7rHNCojbjup2229 +qSQIy7j/2fvzWIsabL7vqh9X7q2rt67v6W7v2+GhEhxKEC2LL6Y8IgPNjSSAVkAZcPyi2TqjeIj KRAQIMsPskXxScsIsCTAA8jygyhBetCIFB9EmUNhOJxv6X2trr269r18fpF5MiPz5r11b9W9WUuf 6L6VkbGcOPGPJc85ERkZYpDnKUyXj2vWfa32CcsoZ36LS6zeZbFmlzaWa7UNnNaLXYhjo/6wD8BP vkzlMUzXaBn1pq+njDx/9dJuJB1laDla/zC/hmmaMK6Wn9efkFc48LMeg0stWhZnCFwUBMwIcVFa yvg0BAyBhhBgO6S+19zI1sjjBF4VMo5L1xCzVRJrGVpmPpmGFx2QlU972nstQ8sM6WmY8hvGNduv QiBlarlFZehhjEVxzQprjw9erMYHijj81iNAn5Ynxb6IFw3jjIqOjtZugIzahzL2C3eAKC/K72nr XSt/rT4b5jsv7VMGJtSbclB0tC1CLDSsDF60DC0z5AO/hmu6fHyz7sMxqmWGtCODmuAmZyPIiA6j mu6nL8JPtZV05U/7dtMZCAgq7lpmEJW0TTTeW4sJ5Wp9a/ECv60ey5RBnas55a+R5w95OENCDRG1 6Fcr18INgYuIQGulkYuIiPFsCBgClwIBPik2NjbmfyrA1FMxhAxNXyQIalj+k3/10G40jZZRtDKC 4MIPgYXVk1Y7PtOGK+JFMcl/xq0VPNE+KuBpuVoOeCh/ZfDS0xNhwjkIeQdvnI3Q3t7m9HT9fJpm 3nd3xe0Tnw0R0j6UsyBwfB4TQ0QrHWMHQwcLykVnQ2j7aH9qJS/aB6K2yC5xwwf9JRzvZfGSL0cx 0fGej2/2vZaj5Yb0NayM9lE+8uMYfsL5Tdsx5LOZfubQ7u4eT7KYl2j89MRzYDPLztNiblPswSDv lD9Nk49v5j24gw18aLlKX3lD6df5WONacdVPVOf5oCzC4Ie5p9UKfK3nj/LCtRFM4HlkZMTLKleu XGm5IQX+zBkC5wGB1koj56GGxoMhYAh8kAiokHSSyuv32Yt2UBCG0KBC0Uno15sHIRBhpugQMvig jgjyrV79gd+BgQHPdjVMwjQ+YYv+IARidEHwzOMCHsof35NvtaN9EHyPDhHSs4fWcQ+PPaLcdIry 32rXNxAZovbjL3aE5e3uRV82KcNYxdjo7elFi5T2yWIigVGbkUb6basdYxR+6BN55UXDaMMyFDrt j0WHyGqYpmk1LjqWtdywPA3TNGFcs/1gz3jWMRvSJ0znN9qw1U7HRn5Oodw94QXXJ/2p1Y75BF6o uxqEtMyzmN94/sBHnhfuGVPElzGWdWwUtY/2Hx3vilerrvrs13LDcgijvyq/YVwtP21rzhD40BAw I8SH1uJWX0PAEDgWgfHxcZ+G79iHDqGLbZMIIWUIXihHrIxouSEv+l17dnuU4UZHR73CnceEsvW7 9qRptUPAo85FmCAYs6WV9ilDsWQnRGQQkc/U7Wa/9sAnMxEsh0eGBZLWK1EjwyO8++G2Cz7DqW02 Onql1c3j6Y9eifrBzk72M7b78vlOfr2ifJYxflBwWWFEsc4rDPQT+gvxZSi5w8PDXmHb2NioaAN4 gQfGehlOy9Ezc8IyCWP8NKpEhTTq9TNGMXbQNvlPGoIJbmJiol5yp0qH4Zn+srmZnfMhqryNj0XP hVMVVEdm5jfmjnxfob+ihGO8KaN9MDBQDuMnr/yDCXMw7VPG+OHZgrFdyw1hZH4jrqxnoZaj86ry QpvBX1njR8u1qyFwUREwI8RFbTnj2xAwBGoicJqVhcnJSS98zc7OJt+Pp7DXr197IeP69eul7D5A uLt27ZpX/mdmZrxgCh/U7eXLl14Bnp6eJqjlDiH96tWrbm1tzakBhELxz8/Pu5s3b3rhq+WMSAHw gfL/7t27jHLJPYL6vXv3ShGMMS7cuHFDhPEjt7A4L+0SbdvmCi8oC2W1Dwrdtemrvj1C5WVra9Ot rq56BXdgoPUrubQ/ignK5cLCQmYFdXlp2e3LZ01v3brlFb4y+sqdO3f8WH379m0yfiiXezCDlzIc ZTFvMF6Wl5eTIlH6FxcXHYZP3YGVRLbIQ/tgiKB9QuWSPosSdfv2bT/ntKj4hCzzG/gzn+Xntzdv 3vgxzvgqwzFWmfcZL+/fryZFYgxZkfaanJxwvX2t371DwTp+eN6ExrO5uTn/PProo49Kmt+cY/zQ TrSPOowPjB92bZQ1fiiLZ8z79+/d0tKSsuL7K/eMHzUOJJEt8jB2+NE+unOIopAXGD+Mcww45gwB Q6A2Aq3fI1q7fIs1BAwBQ+DcIYDA8/HHH7vHjx+7r776yguFCIMIYghCKMFlOVZQEfRevXrlFQbu EcQQduCRVZeyHAoBCu6XX37plWtWn8AEgQzBq4wVMeqKgPfpp5/6tvn666+98ImygBBI+5QljMIL K4W3bt5wM6LAIZz39vW7VVFiUKzu3r1TmrINL1evTrstWcl9/vy577PtsrI7PzfvDQIolmU5VS5f vHjhxxDtsSsruAtirMJQMuJ3h5TDDYo9ShtGO9qHnQ+MH1YxGT8on2U5xghGhydPnripqSlvBKHP YrCBF1biy3CM27t37/rx8+Mf/9jzgjECIwRjXHeClcGLzh0otbQPWKBUotzRbmUqcxgMmUdev34l 7TTuOjoZP3OOOffa9PUy4PBl0A8ePHjgfvSjHznmNzDi+QNGzG8YKcpyzG8YdZlT6De0z8rKih8/ 8MizsizH+OG59+zZMz+O6Bv0WQzSaiwpgxfqTN/84osvKuQDeCxTPiijvlaGIdAqBNpEULIXkVqF rtE1BAyBM0PgN37jN9x3v/tdX/6v//qvu29/+9sN84JQzCo/ijeCBytlCDxn4RCOWblEOIcHlJiy lJZ8fVnJRZHDsU22jNcw8jxwz64HVgfBBoGU9jmJUnm49raIfENhO9s7bml5ySsLvdI+E+NjYpQ5 m82G7FZh9wOPdxQofmfhKH9RFEpWlzs6Ot24GCNO0j7w3j50OiUQ5YWxjLLNqyBnOX4YO4wh8NHx U5YBL+wHlK/zG3MJxgeUzLNwjGFtH3hA0T7J/NaMsZwdPyMyfoYahqStZ8i1dTeeLyxInz8YrsCC PlvGa0whD+rnucNcy/jh+cNce5L2UXqnuTK3MX7A56zHD5ho+5xUPqDv/+Iv/qJ7+vSpr8/3vvc9 b3g6DUaW1xA47wjIM6+tPBPmeUfD+DMEDAFDIIcAKz/nZVUD4ZNVsPPgWJnjd9YOIZjXVc6D6+nt OTe8sPpf1tb+WtijWE/481XKeZe+Fi8YP87L+GE3Br+zdrQPiu15cMxvZW3tP66+52X8nKfnD68T nZfxc5aG1bDvMH7Oi3wQ8mV+Q+CiIHA2yzQXBR3j0xAwBAwBQ8AQMAQMAUPAEDAEDAFDwBAwBJqG gBkhmgalETIEDAFDwBAwBAwBQ8AQMAQMAUPAEDAEDIFaCJgRohY6FmcIGAKGgCFgCBgChoAhYAgY AoaAIWAIGAJNQ8CMEE2D0ggZAobAeUKA9zXVhX4Ns+tFRyA4UznwygdMT1kx8usvJpUhqTd6JU3o j/P4S1E4YfqL02aS6Y1eSRP64zz+UhROmP7itJlkeqNX0oT+OI+/FIUTpr84bSaZ3uiVNKE/zuMv ReGE6S9Ma35DwBAwBAwBQ8AQuCwImBHisrSk1cMQMASqImBGiKrQXOCI1MjkAq/L3pygfhDLEMzd ahxXVaLVr/darKbVe66E5cIzt3rDVempX++hg9O00V30l7BceOZWb7gqPfXrvdLTtHrPlbBceOZW b7gqPfXrPXRwmja6i/4SVhQepjG/IWAIGAKXBwGVUfR6eWpmNTEEqiNgRojq2FiMIWAIXGAE+HIC n/DiV+a3zC8wZMZ6ojQ3AkWoWKuyXS1/mLZamnx4mMfo59Gxe0PAEDAELjICanhAVsGv9xe5Tsa7 IVAPAmaEqAclS2MIGAIXDgEe6Pv7+8nvwlXgQ2I41LNr1DtJlniyiasEpwvy2eQVd0e6Au8JVa7G Z+lzRxr5ZdJX5tOCjD5InBwfxdGuhoAhYAhcJgRUXtnd3XVHR9knzWWqp9XFEAgR6AxvzG8IGAKG wGVCgG+b43jAmzvHCFTXSzNMJ8kSTya6unpbJX02d6AeV0kfBQfGB09A7v35Iyo4VsksaZOYxJPl wOjHeFTBJ4uW3RkChoAhcPERwOjAzk3kFdu1efHb02pQPwJmhKgfK0tpCBgCFwgBW024QI11oVjN a8h6r9fTViZPR+/1+qHTP239Lb8hYAgYAucLAZNXzld7GDflIGCvY5SDs5ViCBgChoAhUBMB3UlQ M1EUSdJqyauFV81QUF4h/TiwkH5hhgLCcVBhcqOfAFaITxJrHkPAEDAELhUCZoS4VM1plakTAdsJ USdQlswQMAQuFgLNeKjv7e25lZUVt7297bq6utzw8LDr7+8/EyC2trbc+/fvHTz19PS4K1eueJ7K ZoZXW9bW1tzq6qrfQgoeIyMjTThMq4GVfkl6KFtYN9bX3fraujs4OPCY9PcPuL6+XtfeVmRfb4x+ Ja5xfn9BS1Z6R1L+odvc3HQb8uPNjM7OLjcyNOQ6OjuLcdGsmULiQH8J6ZMonyEfnxLineL19TW3 s7frOjs6XW9Pn8ck2eZ7Svq+pDw7YWBMn9eat7d3pK+sugO56RYs6Cu9vT0xJoVEoorUiEpr2piP 9tHxo32WMV22Y15i7KxL38X19fX5sXwWh9FxZg6YgA39Y0j67ODgYNmQ+PKY35hr4am3t9eNjo6e yfxG+9A2Gxsbnq+BgQHBZED6bNGc0lqoGMu0D9jo8wd+ynZgwjNweXnZz7X0Web8sxg/+vzhGYSj vzbn+dM4qvRVMAEbxg/ywVm0T+OcWw5D4HwgUP6sej7qbVwYAoaAIVATAYTQP/zDP3QzMzNe2UYQ /KM/+iP36tWr0g+Ompubcz/84Q+9cMy7owsLC+5HP/qRV2ZqVqLJkQiAX331lXv06JFXJBHCnj17 5p48eeKVhyYXV5UcQvGKCH9Pnzz1ykJ7e7s3jDx//tTNz89VzdeciKwBYG93z83OvnNv37x1h2IM kbPN3erKe/fqzWtRHjalSNI34rL005whnSIt/cjt7Oy4Vy9fusWlZc8H7eN5kz58cHgQkzop/ZST 2r4jPz5WBIOXL1+4bRk37cIugvqrl6/ckvB2dBjWpTa1ZsS+e/fOj2UUF5QFxg9jGwWvTMf4efz4 sR9DtA39+Pnz5+7LL7/0bVcmL7QHc8ibN2+S+Q0+Xkr/ga8yHe3D/MYcy/w2OzvrfvzjH5c+vx0d HUp7PHPPnz33GIDDixcvpN++dvti/C3TYRgCg7dv3/o+G97Tj8p08/Pzvq/wTKR9uGf8YEwr2/Gs +frrr70xhLJpH/otBpsyHeNH5QOeP8y9jKezGD9l1tvKMgSaiYDthGgmmkbLEDAELgUCrKyjbLM6 +fnnn3shkIqxQocQxE6EqampUuqKoIfQdePGDXf37l3P0/Xr173A88UXX7g/9sf+mOenDGaoO6uV 3/zmN/2qKWVOTEy43//93/d8ffTRR/7aXF4qleZVWV1HeWMV7M6dO7592CHy/PkLEUpfuuGhYddX 944V6KvSlSr3KCDsaohc4pFb/Gn6xaUlNze/4G5cu+7GJ8Z9HlZxv/jiS1G2nbt166asGOqjNqVT L/2Ygbjc9C71Rbzs7+17gxm7D+7evRPX/8gtdi65mbczsiOi101OTsTtk/Kf0kl5S8PwaVr8aZpa /G+JAjUz89b1yEFr16XforgcihHk9es3olTN+APYWHWPXKP042x1XjA8YChjvGr/ZPyg4KG8lDl+ MGiicDN+2MmEu3r1qh8/9Of79++3YPxUAsX8xpyCIeQnf/InfXuQihVlcGF+g68y3JKMH+YV2uTe vXu+yOnpac8H89tP//RPl7biPjPzTla2V3w76I6QsStj7kdifN472Hcfy/xWhmNs0Q6MG/qK7mJi lR0DMO0zPj5eBiveyEuZtImOH/yMHXj8mZ/5mYS/VjOEwQED1cOHD/1zh/IY1z/4wQ98H/rss89a zYKnz/h5+vSp35X4Ez/xE0n9aR/kBuSEsuSDUipshRgCLULAdkK0CFgjawgYAhcXAVZ6WPW5du1a ImBQG4QvBIyydkMgjKK0cL1582ZGQbl165Zf/WElswyHMre4uOi3SacKpPNb7FFY4JPVoea7VPGF 9v7+gZufW/BbyMGELcEYizhZ/OrVyDD0RpS9+lcLoa+/lPtoe3xleJQiCkeJm5+bdwNsTx4dcayI sW0bXibFOMNK+4pfba+kUw/9lJtQUU9DlRdiV2UL+crKqhsV5XZAjDPw0t7e4VCi+nv7/A6RtH3y /NSm3yg+87LTYEv6wqQoCChQGHM65NWQiYlJb4yYlb6SGjeUF66pq41Pmu44H8o9fQElN6IZ5eAe wxUrzWU4VmqZN1D01QBBufTfyclJv7qMga8MxxZyjJsoSvoFIcpFsdX5DUWr1Y55TXdiMJbV0Xdp HzAra35jbLDjDEVSDRDw0yOvD42NX3HLYixJx49y2porija7QsBADRCUhMEXAwRGG7ArwzF+MIYw x+v4oX0wijP/vX79ugw2krIYO2NjY0mZ0bx/VXZYLflfEtFCD2XxLLx9+3ZF+5Q5flpYRSNtCJSC gBkhSoHZCjEEDIGLhABGCBwCX+gQwghDQCxDIEVJQthBEM2/f6uCclmKCwo1/CBk5R3v2KPolcHL npx1wJkHCOH59ukUhQ6s1sVgwvbY07vagj59YHd/z5/90I6mHbju3m7h8dDzEgTnvLXpZxX1XNbg 9kjOpMBodni4589/SKP49Fu76+zpcrw2wpburNPys7xn09S60/xpmkOp8/v3q65L2qGzsyOOiOjD C/14fWNTtrfvp5mq+irpV01aEIEiTb9FUcmPH/oxfUXfcy/I3tQgHT86bkPi8EJ/hpcyHHMK5RW9 vw5/zG+VfaX5nDGfYNys1j6UiMJXhmP8oFRz3kHe9XRHzwFwK8NRDs+aPC8YA8CKuU3PFGklP9o+ jB3KDR33GCN4VpZhEMFoxjOGsUK5oSOsrOcP5epzLv/8CeWDs3hVJcTE/IbARUAgO5IvAsfGoyFg CBgCLUZABbxwFUqLJAyhC6G11Q5lAMEY4TPvCEPoQVkoY9VS65sXAOFL+VPc8rzWc1+vusnqKLhw 9oKWq8o6vGEMQHjOGyHqpZ/lVZXzIHfgBRP6gucjZ4TolF0IOHit7mrTz7z+UEgkYuZAhPOd7S1/ WGdHXG6UHPqCk+CCkL69sy2viITvk2v5imC+kKCy+Sh/r/nTdHs7e77P0hbhYX6koG3g70C2tu/4 d7jTfIXk/Q6V4ph6QhkbSfvkMuj4oX3KGD+q1Kd9NmVI5xlNk8a0xqfjVMsNS1H+MA602in21eYU nd/ou612m/E4LcYkep2K14xa7XTuou55XAjT9imjr2Bk5fnj59Wc4h+N72iuLeM8Box4OK1/2A6K E/26DINIrfFTpnwQYmB+Q+AiImBGiIvYasazIWAItAwBFBIEL5wKN2FhCIK4MgQv3W1RJHjBA7wg dOUVbuKa7WrxojhpmpOUrerscXn3ZAUdpaQtIxRHuTkAUSK8Ukm60NVLP8yT+skdK8wBoS1R/Llt l4L5JWl8WGQkop+wI6K2C/IG9MM8UXBeaY/ygcf+nmyfl+gsLhEFXss4lEgO1+OrIkWuFv2i9Nmw iA/CdmWnCrshVEnRdJ6+9Nc2wcn3WdmZgYEkxEzTNuuq4yLPC/Q1DOzKMELo2NCxEtZRx7emCeNa 4ddytNywDA1T7MK4ZvuVjyJMKEvnt1bPtfTHXTHQ4Yp40bAd32d9spb9wQih9dVyw8I0TLEL45rt j+YuzsYpNohQHuNH+W12+SE9NeZq/cM4wuAR7OCnlY65QutbxAt84MoYP62sp9E2BMpAwIwQZaBs ZRgChsCFQeA4hUSFDASeVjsVqLTMfHkarkJRPr6Z94qLlhnS1jBNE8Y1288Bh7hY1suRFyVXZEBw 03S5BKe4jYTLlMCRO5TzKSKncXqV0NiLzn94UKz4p7TwBXmzEXKn+YvSREo9ij/tUIxLROJQvkpR vFJYm34FO4UBEW+UIYXkUoT3cTrZDRG5ojrlsp/wVvuj9s9qZNToWC2+GeG1eFH+yuCDuui8UlQv 5UX5LUrTrDDlQ8vM09XwVuPCmKjFi44pdvC02sELP617vjwNbzUmlFurDygfIXZ5Xpt5X6u+ygv8 wk8rXS1MKDfkpZV8GG1D4DIgYEaIy9CKVgdDwBCoQECFASKKViwqMsQBxwkxSve4dNXoNxJebxn1 pmuk7JOkVUG+3rwibteZNE4nFy1DFYNqBLzC3RB9IZ5hJ3NTUczRkSj/mVBVpjWUe34oFazOSbhG +XyZGx+S/5Pio7RJUZAPxcUr/yGFXLo2qGXD0vtj6Pt8kjeTPXOTFExdfYzgk7rUr6hlc3Mnv0xg 5iYl1YBP+0pRFh3HxJUxfuopoxa/RXU4SVjIR4hBnlbZvOTLD+9DnsPwZvoT3TXtqoXkW80L9LWM Wu2jaQqZbFJg2AeKeNGwsnlpUvVORCZsnyICikmIXVG6fJjm02s+3u4NgcuIgH437DLWzepkCBgC HzACrJzo4VCNCATHCQEqcB2XrhnQ11tGI0aWZvBVjUajfHCuQ30uTieX6LUHURxRugtd9GlN0p2E fqQJH88XRhDaJ+Eir8Wo4QEu/KsaZIBhchxPn5TF/GteLRm6URkZvR8CiZO0Epk/PLNu+spvHfyD iU8mRo8iRyh8Risg3JHa54gvGlaUu7GwWv1RxzEU6x1njZWeTV1PGbX4zVI7+V3IR4hBnmLZvOTL D+/L4SUusbjbRpH07birhvw10+/HjxRSr7LbzLLztELc4SfsO6TV/pMPz9Npxn3ISzPonZSGtk+1 /IpJI/yShzNYkFd4HaoReaUaHxZuCFwEBGwnxEVoJePREDAEGkYAYYEHOr/jtlCGxMNT9FWgCOM1 rOgAszBdM/zKSzWhRMM1XTPKrEZDy9Ayw3RlYtIun3rEaZl5PrAFcCAiX2I4mSvSMoo1E+0D8JJP AR/8ovMiQl7qp5/lP18CdCJa9PV2qTOuEBc2YkhSzoaorjBUp+8JJ39q8894A3/qnnVRgPJHOuU/ m66IfjZFvXdh+9TKo+lqpTltnJah9Q/p6ZjSMRbGtcJfDy+aphXlK02tr9Zfw/Wq4a3mJXpW1JpX Io6iQ1+b1z+1nuGVcYECW9RPSKfhrcaEsrR9KFPLJRyn937uyZzPE8U3+69+nUPLDelrmGIXxjXb DyY6h2q5YRka1kj7QI/0ZfAf8mp+Q+CsEQglo7Pmxco3BAwBQ6BpCPBA5xN0/PDX6xAIVOBRITjM q2H5z3OFaZrl59NjuCIjCsIOP/jNf8qtWeWHdLS+RbxoWOv5OHLdIgT6rz2EWm6s8Com7f5TkNlP yoV1qe4PNedA2chsMdA0R9HnSuWWPqHCp9L2YWKa6OpEaNVHreYl1fH0lVZ0DdJnI7zS0tklSpRg ov0zpK9hfDIz3Q0R8gLB6vTT4sI8QfoAn6i+RZi0+UMxORhTjpETobsrJZv4qtBP4hvzaJ8tah/6 LOHMDY0oDI1xkKZWXnSspDHp+NZ5J4xrhV/nlaL37JW/MnhRPrR/hnXVsRzOx2F8M/2U0dUd9cei s2SUv97e7Cebm8mD0kLJVeVfy9U4rto+il0Y12w/fKhBJD+/afsQX0Zf4TPQuCJMCIMf5bfZOIT0 wv5YjRfSN9o+PDtVXmlkF0XIm/kNgYuGgEpGF41v49cQMAQMgZYhMDg46GkXCekcSIkgomlaxoQQ RnFBwKPMvBCoB2MiuMBPq53WV4XgsDzFSQXFMK65/shA1BcLpHv+M49SQlx9eONzlb09vQ0LgTX5 zLxaoFhLHxgY9EI65ebbh69E8GLCwOBATdI+sgr9bMZQQQ9jok+Egj27IbRfpCmOBJN9r2j39fUH X8/QemjK6vQ1RdVrwH93T7fvs/tyiF9WSBejmbTNgXy1pFOUm97+yMAmlpOqZE8b4TERJQlMsrw4 /wUc2gzhvwwjxNDQkK9OZfukX9rRNKet93H5mTNwZ80L8xvYM3/kx4/OKaXNb/3RnL8rn5jNu/39 KKyvvy8f1fR7NZ7TX/NzbRim83HTGQgI6tjw86rMcaHTNuP5VIYRQsdGUZ/VsDIwAQPlRftoiAu8 8DzWMRbG1fLn+3+ttBZnCFwWBMwIcVla0uphCBgCTUNgenraCxK8pxk6hA4+FTY8POwNBGFcK/wI 6OPj415A12+TazkrKyveOzk5qUEtvY6NjXlhs+hb7HzDHV5J02qHwDs8MuzbZ2NjI1Pc9pZ8116E wJGREdkxcZKdEBlyx96w2sXqKJ9jyysM29JPeCWEvtJaF30etL9/QJT7TrcpmIQCLUoVnxbsFYWv X4wQjbu8saI2BQTw0dFR+SLIgdvZ3skk5isue6LQjQgmnfFrNYkFKZOyOTesKE5MTPhP6uU/mcf7 17TZlStXSjPioZgwVsL2oab047LGD+Uxp9BO8BI6lFzGN+On9QZF5+tMX6Ed8vObzr1Xr14NWWyZ f2h4yK+kb25tSvtkP/NI+7CbaXJ8omXlh4RpH1y+fXj+0I9RthtVckP69foxiNA+lLm5uZnJxidC 6S/Xr18vZfzAB0Yr+Mgr//QdxnpZz0LmFMaP9lEFJpQPymgfLdeuhsBFRcCMEBe15YxvQ8AQqIlA XtCvmTgXicCDID43N5f53verV6+84HXt2rVSBC8EHYQ8tpnOzs4mygtC+8zMjBdGyxK8ULjhBYE8 FI7x87t9+3YphhkvbIoywCrdO8EELFhLZwVqcWnR7waYmpqKD4PMNWzVWyjwU4U7uzovL70U5uS1 j4mJcbcrZa+8j4xCJNwVpX9l5b03QESrcyejHxWaz1vIiqzODbhR6bNrIpCrcYYx8H71vTecTUxO OHYpVLp66OfTcJ+6PD5jV8ZcX2+fW1hcEFjjL4MIhosLi74vj1coc7XppyU17rt165ZXdhkvOifQ ZxjbKDU3btxonOgJcmA8Y95AeVtYEFxihwK1vLzsjSW6wqpxrbpieEGRWlxc9AYaLQeMUCxv3rzZ 0Ctsmr/Rq85v5MvPb7QPY6csIwRGoKvTV337LC+lY3l9fcMrm1NTk/LKRtH4abTWx6enfTBe0h70 F3Xv3r3z892dO3c0qOVX5nXmXMpWh7L95s0bP34w2Jfh6Cv0S8bL0tJSUiSLAowf5vyydkJQDuOH 9gmNm2DEc4hxTn8yZwgYArURaJOHclaaqJ3eYg0BQ8AQuBAI/MZv/Ib77ne/63n99V//dfftb3+7 Ib4RLh49euSFQAQOhEFWPhDKyhKMlWGELHhB8UZRYRcEQtn9+/dLOQ9C+eBx8eTJEzc/P++FPu5V gbp7924phhnlhRWxx48fuQ45W2BgoF92ALCCeeTbp1Fl7nDtrZLNXXk8YoDQq0an9yhti4tLoljO +9XJdlk9XJd+ogqUvt+tObNXpaNXjc3fa3jtK++zv3jxSgTjbc8L91uyO2Rs7Ipvr6yRpdiwUruE fKzyqVeNP5L22HJv375xmCBon10ZT/v7B15A5xWV4i9zaH69pnTbh65rYMNX+ujTp0+9AQTjIkoM itUnn3zicWqY4CkyPH/+3CvcKJooKhgCMHp+/PHHnqdTkG4oK4rkl19+6Q1WKHDMd8xvGG3KUiyV YYwytA+7Lxg3tBfjBkyY8xpx1cdyfVRevnzp+wdtwk6ApaVl2Tky5g2w9Jl6XVvPkGvrjl7BqTdP mI5dQ19//bU3IGJo5vnD7h0MEGU/f3je0D70V4wjuguP508ZO2YUF+Z3FgJQ/hk/9BGMVezA++ij j0oxnCkvOn5oF3au0F5qjG90/EDjF3/xFz3G9Lvvfe97vn5all0NgcuIgMiwbWaEuIwta3UyBAwB d1ojBBCiYKLsIqAj8LDFEsH0LBxCD7wg7LArAV4wRJyFQ2iCFxx8sKJ8Fm5blO21tXX/Cka38DAk CsxJ3k8+reKCcAwmG5vyeojozV1dwsuQnNEgX6Mo27HKzysZO9JP+DQor2DQX8rvK0fSLvKZXFFs 4alDFBjap7ZRpjpapzFCQJXxww4RruCBctuIUlmds8Zj6Cus4NJvUOLg5ywc5TOO4UfPIjhp+5yW f9qFVW6uYEL7nKTPnnYsUw/mezDR9jnJnHJaIwR8UD59Fn4wADDXntUKO2MYXljpp7/SRidpH+p1 Wqfjh+czRiva56x4oc9q+4DJScYP9TEjxGl7heW/aAjImG2z/UIXrdWMX0PAECgNAZQUhBx+Z+10 Feqs+aB8hNCzUpzS+svXKTiAUn7ZjQrp6nmathEf+dXFRp4MSb3Rq+wvkGQoTZWrtmkapRgxq3f1 0c9VUDPLtZI+51Dwfnu6BitlkCxxmkevRIT+JGGV8JBYLf7lnX8x3EXnhJySfsjSCf2MH3ZBnAd3 PsYP/TY6QA/l9qwd7cMq8HlwGFXPyrAa1p/2OS/PH4xUrT/fJqx9df95GT9weF7apzpaFmMInF8E 6t9bdn7rYJwZAoaAIWAIfHAIxAow9Q68uZsToAKxDMHcrcZxVYVc/XqvxWpavedKWC48c6s3XJWe +vUeOjhNG91FfwnT8Piqtz6B3nBVeurX+4hSSkfvuZJWacThmVu94ar01K/3cb48HR9MWqWh6exq CBgChoAhYAgYApcJATNCXKbWtLoYAoaAIWAInAKBvJJcD6kwjyrb1fKFaaulyYeHeYx+Hh27NwQM AUPAEDAEDIGLh4AZIS5emxnHhoAhYAhcLgRCPbtGzZJkiSebuEpwuiCfTV5xl3zlwROqXI3P0ueO NPLLpK/MpwUZfZA4OT6Ko10NAUPAEDAEDAFD4GIjYEaIi91+xr0hYAhUQYD3adWFfg2z6zlCIG2q mkwlyRJPNnmV4Fp6b4ZAkj/xZKJj9VlNEZpI7r2XcI3L5tM7zVFND4/ilYamNvqKn10NAUPAEDAE DAFD4HIgYAdTXo52tFoYAoZADoHQ8FBxCv7Brjs62MnlsNtSEcjr2hSuYfhVB8dfj9O8mi9/Dw0N w6/pCN6Vr1qYO9cIHO2unWv+jLnzgcC5GcuBEfx8IGNcnFsE9nbcYE+bG+6Vwz/7ZG14b0OeSaae ndv2apCxtvYuOaX5bL6A1CCrpSe3Xl465FagIWAIlIEAp3nzuTdcaIQ42l5xR3vR5yXL4MPKqAcB tQ4EloFa2QqTx4FHQqOCTGGGpIQjEfoSp0kJyNCpQV8+pRelzWRISGY8Rj8DR8VNFXyOdswIUYGV BVQgkBnLFbElBzAvmDMEjkNAPqc80HXoBroP3VB3m2vbW3dHO7ZR/TjYLko8s0Bbl3xyuPfKRWG5 ND6tl5cGtRVkCBgCZSLAN8QxPvDjG+e4o3359rsZIMpshipl5YVzlPdQgc/H58jkk/voONCTyefP Z8jHB/Q1qacThHv+JNCHx/mVjF/1DDNohOQPvCGbmeqGEUY/wjiEM2wG8xsChoAhcMkQaGsTWaWt wx0eHV6ymll1QOBob8s52YFrLouAGSGyeNidIWAIXBIEQiMEfu8O9y5J7S56NYo0zFBbL4qvt87Q KcrfAvqZYqrQz6Sppw4B/5m8Rr8e9CyNIWAIGAIXCYGjeMdMW7vsghBjxFGF5foi1cZ4rYaAvQJc iYwZISoxsRBDwBAwBAyBliKAQh0q1RSW0biD0jVtNn0kuGlckNzTqRYeplO/pj2GviaDvhcas+lP zb/R1wbx1+rtm0lmN4aAIWAIXC4E8o+Wy1U7q40hkCBgZ0IkUJjHEDAEDIFKBDhXgtc5eK2js7NT ViqqKcuVeZsZciTbNPf3Dxy7OuCDMy/OyoEJfIBFZ2eHX71pjJc8hkhd+TClWBxO2QcH+25/TzAR bLR9IlzyeY6jH6cPklW0c0gy7AM+T5RR26hdVrM6BJfIBRlPSV8RyV6z9MHiwPcTeQVJojo7OgUb aSNZZUtcFf6T+IwnyNcg//SRPXnfGUf59Ft4ybo66WczNXyHUUP7LX0EXs7KgYueV9PVxZxydutB +3t77kD46ZD5rcPmN98lDg8PZG6R+U3u6Cttgs3ZOPpsNOczv3V0tJ9ZX9HxwzXihXk/GLslAsT4 Ye5nglNeSiw+U1Sz5IN09wOTrDlD4MNA4Oyewh8GvlZLQ8AQOCMEopXUkxeOQjk3N++WlpY8kU4R Rru6u9309LTr7S33pOO93T335u1rt7m55YVi6jY6OuomJya84nDyWjaec3l52c3Pz3slygvookBd vTrlRkZGGifmd0MgyDYuzO7u7Lp3s+88HyiUCKZd3V1uYnzc9fRo+6jWXCf9imSav1rVjtxerPBv b2+7x08eu7dv3rpPP/3UPXz4MMoUkjgBfZH0a+MT0KfP0j6bG5teOAdW2eDrhoaH3NDQUEElJHMD 9CubKSg8pC79c1eU2+WVZbezveMNIYcS1ikK95j02+6enjR1SKJhfFIytXwYQl5Lu2xubvjxg5Fo ZHTEjY+PiTFCTi4v0a2trbuZmbfeUNQuSiWOOWVkZFh8FQD4+Fb8YQ5hHC8vLXsjHoaQbhk/09PX XH9/XyuKrEqTPjLz7q3bkH7LWBbWZH4bOZP5bX193S0sLETtI4o/Su7E5LhgMlCy0n0k/eSdW1lZ EdxQtttkXutx16R9enqD8VMV1eZF7Mrzhz67vr4hYxnjg3PDMt9PTIyXPn7ev3/v3r175xcGoudP m5ua4vkzLHyVZyxirp2XfrK4sOjbByNeZ1eXu3792snkA+ZBc4bAB4aAGSE+sAa36hoChsDxCCCg v3z5SpSoFXfv7j03LEocSu7z58/d148eufuf3hdBoxxBkJXKR48fuS4Rzj/++GNRFLrdrihVX331 tShVm+7evbulCV8IxE+fPnM3b94QwW9SgGzzgtjjx489bxhGGnOp0hXqoimN4lCMMq/fvHY7Yoi4 deumV552tndFkZlxG+ub7pOPP3LtIiyHSl0xpeLQtHzlL0gXeKH/O7/z2+7rr792KJdLS4sO48iA KCyREUISeyU/pZj1HU9f02eK1UCMODH9I+mfKyKgLy0uuTFRrjE6EIZR4t3MjCgLHa6vrz96C0aL DZTe4+gnRWY8SijILd492SkzNzfntre23PUbN6Tvdslq+4EoMjNudmfHTV+75rpEYPfM1IVPptCG blgxfST9E2Xyk08+ScbPl19+JUrvhoyfewKh1qMh0g0nXpc+8ujR12K0u+oND/CEkfPp06fSj2+J UjfRMM2TZThyb0WpnJud9/MHBkT6yus3b3xf/uyzh17hPRntxnIxvz15+sTPbw/u3/dGVfrNo8dP Sp/ftqTcly9fuqnJSTFQjfuxRfs8f/7CXZM+OzY21ljlTpGatliYX3Aff/SRG5SxzPPnxYvn7sdf fOE+/5z2UUPrKQqpIyvPwqfSPuwGvC/GVQyJW1ticJUxtba2JgbXT4RKOeOH8p49ey7tcMXdkHkF wwxGNMbPzZs3vTGijiqdOgmYzMoCxevXr92nn3wq8sGgtM+RtM8L98WXX7rPHn5Wmnxw6soYAUPg DBEoz2x4hpW0og0BQ8AQaASBra1Nt7i46FfVoxWWNr+CiqBzKNt034gCXJZbYmVbjA3Xrl/3igEK Eyti09euegVmwa/EtJ4bDB8okf19fd4AwaoTvCCw9/f3u1evXiVbzE/CTSTGosyGjtBsGCtQC0sL skL4XlYFr7rBwUFRLHtktX/Qr2yz2s0KVbq9NaJXL/2w9NQf8BERSqJu3rjp/sTP/qz7i3/xL3iF FmUh5TnIl+Qo8gTpcvQ1dRScxSIS/iVMhOKt7S2/st0r7TMuQjrGqh7ZsTM2Nu777uy7OVndlYNZ T0Jfmah6zfK/+n7Vra6uunFRqukb7FBh9xCK3frGerx6CLEgX1Xap4tgNXkTY0hu/LBiiZGRcV6G i5T8135XCGXrKi6YDA0NeoVmRww0ZTh2HMzPLfjdBhgOGccY7W7cuO6V3dcyllG0ynBv3r517CJi fmMlGV76pM9cnZ7y89t76UtlOJTsN6/f+J0YV8TYwCsY8EL79Ivxbm521mNTBi/sxngn/db3jWFW +KPnzw2Za5j/XgufZbm30j70F5R+dgIy7zOmMSSyK2FJjABlOPojSj6vL127Nh2Pn3ZHW40IRoxz +lEZblvm2rdiJOLZpzswGM+3b9+WGa0tlg/KGT9l1NfKMARahYAZIVqFrNE1BAyBC4sAr2GgTA4M DGbqgPLf09PtFay9vdZ/bgnBmC3TCKEDAwMZXkZHol0H0XbdTFRLbhCMN8QYMjDY7wXRsBBWUlGg SHMypwJbkYacDdvb23esKPPe9oAYIFInyosoC2znRgHel3Spq59+mifvy/Khsfdl9fanfvqnRMm9 EeASpD0K/Jqp8ForXQ3+hT6xvKqzL0aGgQHZ7RBsS8YAgJFmZ2dblPFqSm5t+oXsVgRG/KMsrK6+ l8/NtbleGS+Ri+j3iBLD6w9ra6upQlc3PhUF1hFwJFvrF12XnIuB8Sx0jCe2ULPSXYbCzdhZl50X wwXbxplneLd8aTl69SvksxV+dsfQV4ZFeQtdh+A0ODjgVmXFuQyDCPMbY5VdMfn5bTCee5n/ynAo ltsyRlCwUShD1y9jal94Lcsg4g1jMn7yO8vACWMecz5G4VY7noH0FV7T8buoggKHZO5lJ8+87NYo w7FLBSMD/aSrqzsp0j8bxYjH87isZyGGF7AZGsqOH549Kh8wH5szBAyB2giYEaI2PhZrCBgCHyAC KLk4Vl3yDkEQ4Zktqa12KAIIXgg3CFuh89vZJQzFvwwlyhsYRMFEUck7VjBxrHI34tLdCmHdVCEO KREmP/m/J9u3EUi54f3k0EUHMLZ7BYrXIuqmz6qv/4XUovLCkMQvURE74oH1IkU6ThPFQ4tfQkEJ hAGpP86b8J+hHxOJ00CfVw62pT/KIqnvKxEhSSD/o1XUdn/4IEpWwoePhoi4Y+hHeSJ6UQafKVef KAbjD+WgoCQH+sX0uUfB43UNr+RSvMcP2vXRT8s/3rctZw2gbLOFXIDIZGBMUTh9ifHcaqfjx4/b XGE6fjjLowynxsJCXgQX8GD3VasdfQBlWneFhOX59ilxflOlsQgTNUpsNDi/hfWp138YY09vzR/i yphixwrzfZntAx8cihm6Dg67lfZh1yBjrNWOVzFw0bjNlsbhuzjt19nY5t9xNgaulnxQRvs0v2ZG 0RAoF4HsrFJu2VaaIWAIGALnDgEEvJ3daMUYoS/vNGynhJUohHT4KVL8EQD5h8LAGQmtdrrVVQXy sDwO+sNtbzZmmIH/SqdhaKj8cITJT/5jhGDVuL2t8tR69ExW4MGEgxHrpk9G/6OsuEwuEpw4ZYUA wsmiCeJ00UUStgU/T4fEPoNkjAn5cIjFLg72dxDyyfFE/sgT3GgZko/3kfdEEcBo0aFfwQjoo/yz crcv/YQDIj3bDdCPePcZhIGY0YC+5y0O9nxIWShLieEsqAZhfI3BKy7UwdOB9jH0UwDq9ulqcWQQ iZmIc0dKVNSfyjBCeKOLlK3zR1gJeMFpmjCuFX6MM7iisaxKno73VpSvNJP5La6/hnMN5zfGe6ud Yl/UPhqmaVrJCwY65jjqz5kHecfuHRyvX7Xa7cruAuYNz0uwu4pyuSWceA6ubLVTpV7bIiyPMHhh vMNPKx3P49065IMy2qeV9TTahkAZCFRK2GWUamUYAoaAIXBOEUAhQdDAIdhUcxyo1mrH59lw1djQ cITFVjs+++hdASSK0+mUuViLTSpCQZWF8ek83zxaeZ8+bq9Yu0bR5nWNrKuPflJmpmjJq/D4xjMA AEAASURBVPchmYw/ukmDYv41X8JMHJAJl1x6nxJI9P2oimEEIWSQHxfqK/02IhH9jW9IGHklGAH9 SAwWiQu8EahJTJxLMsXk0phK+p7ROJgywiKSfDSatFnaV1AW6qWfUGnIgzIXOcxF2YrQfRJetG83 RL2xxIkinWXDE1E+eNWoDMfOmcgVMRPFJPy2kCGd33JNk5SoQ7zVvDDf69xV9IUFbR9NkzDYAo8f P7EireWGxWhY9nWzMEXz/DrnR2MlTzfqO55fmZNb7ZLnbVGXjTsK/Umf3a3ihz5AnXHaFkVlHZRg OCsq18IMgYuEgBkhLlJrGa+GgCFQNwKhgBD6jyOgAka1dCoDlSGQcggZrjr/ETdl8KK4VOfF+fem q+FWGR5qwL6WlUkqQmStP8EkiNTXCeLGQRBVftNU2nJpSKUv5ClIr/TJwOo9zivVkdf/FUE4ipF8 IZkgSTaicfoZUsENxfkdDnJVxU18aQoCJVGEXcxcAf9phmq+sGIB/QCfQ22fPImUMc9ZsbJQhX6e Vp33CS8Bq0nWgB9Nl8S1wKPGn6Lxo6xU9tnmM0IfUOy13LAUNdaUxQtlF2ES8RQ1XKt5CeeLIkwU H9IpdhrW7GtaRlGnTbHSebDZ5Yf0OIQ5csJLITtRIDubWu20DO2fReWl2BXFNieMvsjUWc0pTOxO O5mrhvXJqFkuQ+A8I2BGiPPcOsabIWAINAWB6kJuJXkVIipjohAVLRqhWY3WseG1JOIgczm8xAUq AEH56q2T3Th5HulqhHPhFCL/M8pAYhgQ0lF0gWKTo6NMV7MYhMn1tQEfFvMNH2GaWDKNYxPqVSXW MG8D9FPCsU/LjcFPyCaeihxRQBX+K1JXk7hD+hX8h5FZPk9EvyJTfQG1FJb6KDQxlXaMAmiaWEod pJSROpK2OkncZ48vprU8Q72UOfT4ivoU1YZcNntrMfFlJUVU67RReN3NmK1Ag3cJM1Xz0Yatbsfj uFCkGuVD04PluZq3qqJtEYbA6RGoPGHs9DSNgiFgCBgCZ44AW7E52RvXyHZe3mVPnEoUSUDqKXo3 NY1tjk/f/80o3AFpDS96tztI1hSv1lfLzBKNgNJ327Nxp73Lin16/kRVQV1YQYjTdMeXnqVfNb0m o6rqL0yc7zQ1E6cUNNmx9NMskS8SWRFifcn54n2iKDDaaq4F5elUu68zfZys6D32PGW40f50DJj5 rA3dt8eH6RX1Ffqx9uXW9Nssq1pfLTOM1bAyxjH9hH4Q7YgIuYj8ZfJS//zW4vUywUTnC61/iIwO Kfq2KothfDP99BN+la+TZdtH+1Mzy87TCvsjYwgFuciVwUunjuWM5TfihrNwcNX4i1I152901k1M SztGAelGMKHPra+vuffyVSHqcFDxKmFBARZkCFwCBFo8s18ChKwKhoAhcCER4IC1QfmMGJ/0akQg QPDSU8mLtmmrkFp0inqzgeqOP0VWfTtyJAX1yCcYW+34TBtO6x+Wp1tPNU0Yd7yfOvCrIuF6Amka Tmnn8MlwO/JRnDcSRuUOQb5TjUlp3qq8eC015oEL/ATsKH2fPwiP0sT0Ewk4TCA5uG0mfc+f5yT6 I/RRKvX0+ui1jCz/7JaGvQ7BJGMk8Kzm+C+g3wj/nFQPP56PpN7wI4XJ/ZEwg9lEFdB68Alq25C3 Sz4HigtfQUgIxPVEqfRfz0giWuPRAx/z8FKajp+i0/ZbwQ2fSsWFY0jL0TDlV8NbcdX5rWhOicqL 0OqOv77TCh6gSR/oiL+EVMiL78fyNYS4P7WKD+jy/NGv/mi/CMvTZ0EZzx/KiAxW9JVsz+WeIJ6t pfSV+HO/R8krIikqvOoEPxzg3MizPqVQvy9qn2j9tqZ8ED8v66FM/+PTq/3yiene3r5E/qgnr6Ux BC4yAmaEuMitZ7wbAoZAVQR4sCNEdccCd9WEBRF9IhDgis5a0MO6+vr6CnI2N6i7p9sLgUU7OeAN wYv6laFEIRzh9pND7dK66kF3PX0RbmlMNV8o0KIJe224WuI4PkpDfb0ALu2rh3/paQwouAjpKHOq aEW0j6EvtBIefFJNH/Gp9FMGi/n3ZCSRns+QpG8SfU9PWUuIiyIghpmurh6vTEWraJpI+JT/B4f7 vh+BXfTZzGL+q9GXzBIV0/SXgL6PSemhqGEQ4aDMyCiU5kVo591uvuDREysVUZlpmqiYLH2f5gR/ dIz6w0yl3ND5r3jE46fVigvlKi9FnzPUMJ13Qj5b4e/pjYyWRQdhapjy24rylabOb4XzbDC/ZXan aeYmX3trYKLzb3cJxl7mri5RYJlD1CCkVWW+Z1zhymifHnn++LEsK/Nq/FBeuIc/lPIyDCL9/f2+ 6P24/soHV+0/tCHP/VY7DAY4LTcsT+WD/vh5GcbV8nfKvAmOEZbpfForj8UZAhcdATNCXPQWNP4N AUOg6QiMjI54mvqJPy0AIZCwnp5ev3Kh4a26dslujuHhIS/sJKeDx4Vtb0WfwxwZGW5V8Rm6Q0ND XpHdiT/vF0by6TqEv+Ghenk5uaCIkBYJpNIWwdcP4IfP26Fw9vf3ncj4FNbJb85QxTu5aopi/hsS HX1ipaPX2vR9bEEh7CroF6MYyvTuTvyllJg+Sj+CMauVKjwnBgUtLrwW0A+jE38V/tlt0d834Pus CuSah5VdFLqenr7jFZcMfaXQ2JU6D/QP+DGrB9spBcZx1Ff6vSKl4a26siuLXSI6bsNykvEzEs07 YVwr/INDg57sdsFnHvk0J2OMHWStdsxvfWK4RJnLz2/aj8ua3wZkFZpdQvr5xbDu+pUVfS6Ecc33 t7kR+oE8a7ZlXg0dzx++AEE/oj+12nXJTjyedRim1BCjZfJ1DvgZHh4uZfxQDrsyGLd544x+OnW4 pGeh9oNa8kGf9KdGXGSwjXPUOwc3UoClNQTOIQJmhDiHjWIsGQKGwNkiMDEx4ZW2FX+mRCoRrK6u uh35RvioGClYAWq1YwVwbGzMK/jL71cyxS0uLXger1wZy4S36mZQlBIUgo2NjYxAinC6srLihdHG FJcUV79cX8h4mCZKAO4I6QjiK4qJJEMgXltbE972pH3GpH3Cx1slnYhaUThh8tPDLkkYB0WGCc2j V+IjPzszvKKQGEeCNNDxjjD5nYC+z56xV8T0xQA0MNDvdxf49qH8mD6K5q4YjgYHBl2v332gPOkV dgJ/EX1fsP4hrfxq8I+Q3iYK3er6epQppr8ufYfdEKNXRMnyK5ZBuUpewQ7pJ3GNe6auTgnFI7e6 tprJvLy05JXOiYnJTHirbtj5cWVsVN79XnehIY8+Q79l7GDoK8ONj417Ix3jlv6qbnNz0yu+k5OT xxuJNNMprsxvzLUYq/Lz28r7ZT+/TU1dPUUJ9WftkdXtARkjW1tbbk8UXXWs+K9L+2DAG4hX4zWu Vdcro6P++bK4sCDtk+7g2dzY9M+fyalJb3BsVfkh3ampKTES7btl6Svq6DM69169Wk77sItrdDR6 /mzFBnj44eyM9bV1/xoDuJXheP5gnFlcWJTi0vHD3KvyQU9v9MpT4/yk9BrPazkMgYuFgB1MebHa y7g1BAyBOhEIhes6syTJEHhu3brlXr167X8I5ax6vH792ivAN27cSNK22jM+PuYwfrx9M+MPT0OZ XJNDrJaWlt3169e98t9qHqCPwnDz5k336NEj9/TpU3cTDESRfPduxgvMt27e8saS+nlB20Xg4qp+ codacOgnjiLbvGGG7fQL84ty3+4VuA1RFN6vvHeTIjSP+hUxpclV/Z4Cf2JXST8tPxAGM8nkxrMd 0fzt3/5t94d/+COvvKE09Isx4Hf+w++6r7/+2iv+f+q/+lPu0/v3tUC5KrH66Pv0qrAn5So55cW5 blFyr1275vvou3ezouiP+hVmDmftkm3V06IsJNvaEzreI0VEdUnrVUlfQ+rhH2V6anLKLS0uuUVZ Ve8VgX13f9etiOI/MT4RrfR6gin/NemnkQ37MOJh/Hj7+q28BtLh+8qK9JOV9+/d9LVpb1hrmOgJ MrBL5fr1G2LE23TPXzz3Y5edGrPSVuzIuHPnbimKP6wzv1Ee4/jVq5cyZq76nQjMb0Oywl6WYgkv zK0bG+t+fuvskF1OsjMCBZc2Yn4r47UD+GBeuSHlPXv+3L1+88ZjgMFz9t07P8fASzROSN1ah0Hk zp077rnw8ubtWz9m2I3x4vkLb6hinJfleP6siQEPHHhdiPbgebQgCvg1wURfkyiDn9u37/jnz8uX r+T5fNM/d2ZmZvwrgnfv3pX7clQadgrduXPbPXnC+HkthrRJ2S0i8oH4eT5H8gFzagMueBw0kMuS GgIXGoFyRuyFhsiYNwQMgQ8RgfHxcb/6NT8/74VBlIjp6elkZ0J5mLQ5BKwlUeD4zc3N+RXvTz75 pJQtuWE9WQ188OCBA5NXr155nZVtsgiHKFQnc7Ei7DMjuIX3eYpRHAoDK3QoLZubsvokW8hRGBKl UuJTF9Krj36Sl6y4kFzuHmWOFVT6B3yRlsVljGBkG5N+lLqQFwmtg34GD1+vHI2UuFcSUFBYkUNR cEdtXmnhNZn0oE7JkKlPQO8Y+hleKLcK/2BxRZR/hHW/Ohi/roNyy6sAevBrwHrsDXhR+hleK3Mc F0Kb3Bbj2ZC00dLSopudnROjSLdj/DS2c+e4ko6PB4/79+/7Mfx25q0c0ulkDA+4zz9/KOMnOkTz eCrNScGupocPH7jZuVk/v/FKD3MeOxNov7Kcbx+ZP9iVMT8/5w1nzDOffvppqQou9e2U9rkryv/y 8oo3DrGDZkCMMjek/5Sx8y3EHOMZOMzORu1Dm6D0j8luGgyvZToMIqOyy4B5n9dBOHvhk08+Lv35 w/ihXyyIsReDGbtUMHg+fPiwNAOe4s5z78GDaCy/EKMi7cOuK8aQfw5oQrsaAoZAVQROKjVWJWgR hoAhYAhcFgRQUvh5hdIrQ6fUiE4IDEINwg2/iJez4QP2WUVlpUd3mpxc4AoUTu/VOum1EizUeh8r 6fmkHqt0V2TFnwMpOZwR4Tzl54T0yYZDIc+xklD0aeSPpLkjChS/fNqISPav59/nrZ9+qugrM3qN aQe3R8IPK3Gcsu7fMZayIlyiRBTtfZ4H+UMdvT/mx5MMCHIf3DbCP18wGZZtyyhx0ll8u3jlljJD F9weRz/M1oifwzjHpK/wO+vxg7GOVfXTj59GEChOy6r2XTGinTUvKPjnZX7rkvnNv8KDJVFcOp8U Y9jKUHYZ3Lt378zbBwwwQvA76/GDIQJDK7+z5oX2uXs3Gj/RtBZMZq3sGEbbELgkCJgR4pI0pFXD EDAEWofAWQqi+VqdF15Oz4cKbLEynGrD+Son95pDFWN48CuUosSkDuWBlJq6AfqaNSWW+pQMfHqJ U+j79JqJK07Lje7Cv/6Yg2rRSrYJ9DE8VLgm0q+gTUAV+tHZHFRa8SnM7QNr4lM9W0Mxp++3DRVX NfF54QMGjZfKZjJMKjGxvnL+MSnm0EINgfOJgBkhzme7GFeGgCFgCHwgCKhWrtfTVjtPR+/1WoN+ HUkqjQyaSa9CH307uE1KLApLItWTT6T3epV0HzR9xcmuhoAhYAgYAoaAIXBREShYLrmoVTG+DQFD wBAwBC4mAmjVx6+UJ3UrTB4HFpIpzJCQy3g0aZ6OX6aXwDBc7QJ+63YQoeEZwvENyfQXxhv9CA3F JoAzhMn8hoAhYAgYAoaAIXDxEbCdEBe/Da0GhoAhYAhcMATQMENNPfRTlXx8rnr55D46DvSXfP58 hnx8QD+fNIkqoK9koheCk5QZ/jWNxhp9QSIAJfB6iKriowDa1RAwBAyBy4qATYCXtWWtXpUI2E6I SkwsxBAwBC4BAuE7vaH/ElTtElShSNBCG1VXFK9xx13zWq2mbwH9DJtV6GfSKC+1rgH/mbxGvxZq FmcIGAKGwIVFIJ7r/aXCqH1ha2WMGwI1ETAjRE14LNIQMAQuKgKh4aHMT85dVLzK5RuFOlSqKT2j cQfsaNps+uhEf40Lkns61cLDdOrXtMfQ12TQz79+4Umdkn+jrw3ir9XbN5PMbgwBQ8AQuPAIxN99 cm3t6rvwVbIKGALHImCvYxwLkSUwBAyBi4jAxsaG/5Y4vG9tbV3EKlxinvMKOxp4PkyrXxweGpk0 ZXrN5zmOfpw+SFZBPyQZrlT5PEHGlInYF2QMkhl9BapOfDS5XQ0BQ8AQuEQIHMonphcW5t3s7Du3 tb3l9vf3L1HtrCqGQHUEzAhRHRuLMQQMgQuMwMDAgLt586avQbd8+93ceURAtfJAEW2ITc1fLZPG 10m/Ipnmr0FfP9dZI0liXzH6lSiFEDeMTyU5CzEEDAFD4CIhwE7NiYlJt7m55YaHh11Hp6lmF6n9 jNeTI2CvY5wcO8tpCBgC5xiBaDv3OWbQWBMEUq0TXbTSFYem6TR/kC7wngv6ymLKdODTyIDpwFsX /0oioJp6NTIgGngvDv20RuYzBAwBQ+ByI5CZpC93Va12HzQCZoT4oJvfKm8IGAKGwPlAIFKX88IX ofmwIn6DdKp355IZ/RjHC4lPrjHt1hAwBAyBS4TAUfKck3m6nkfeJaq7VeXDRcD2/Hy4bW81NwQM gToQ2NnZcXt7e669vcP19vbI9Wxst0dHh3K2xbY7ODhw3V3drruny7W1nQ0v+4LHzu6uR69HXnXp 7OqqA8laSZC60I6LNOSisIgW785uy3kf+x6TLtfd2+s6OzoKCjoZ/Syh6nyw62Z/T3jZ3fZ16BIe eoUXaSARKOVXPWtQRK1ENfgvoi/Jd3Z33LuZd27q6qTr6+v3fAWF5bwN0s/ljm4r+aev7mxLnz08 cB2CSU93r2w1zrVPEf+F9E8X6MfP9o47kD7TJf21p6dHmqeS59OVUl9u+i3zCv2mV/g4/fipr9yi VPCxK2M5ap/uM9sKfp7mt4MD2iea37qlr5xd+xx5PqLnT3v8/MmNn6JGbUEYfZXnz6GMacZPd0/3 uRg/vGp5lq9bNk0+YAo2Zwh8YAiYEeIDa3CrriHwoSBw2tcxELZevX7t1tbWXI8olAeiZCIo37x1 2w0NDZYKIwdrPn/2XBZIjrzytLm56QYHBt3t27dKVxrezc66+bk5zwcgbIuSef36dXmndaJhTKhP dBZ4qAyqQhySiyW0QGFFuX358pU/Tby/v98tCCYoUtPT06J09/nMddP3X7sgi/CRsCJlBuWF3OhK 1VEbaZxbXV11S4uLIpj3eB62hZf+oWE3fuVKpHSfgr7HJ8MHWAiTXHDwG9A/lD76SnD5/d///+Rg 1jfuO9/5jrtxEyNE4OK88N8o/Xrx2Rbldn5+zistPb19bs8brY7c5NSkKFPSPlX4r4t+UJV6vH78 PH/hxy/Ghy1597p/oN+fGVO2ArMo/eTt2xkZP92uXYwgG9JX6LNXr05JVZLOV0+1TpXmSA7DY35b fb/qDWYouhyQd/sO89vQqWg3mplDhF++eOkOpR9j6KV94OHmzRulz29LS0vSb+e9Ykv7bIvh6urV q25kdKRUpRtDCPMbzx/mNwx6B/sH7u69u/6+UYxPk35zY9O9evXSG3v75FmIMaJ/oM/dunmrdAMN bfPu3TsxaMpcKwbN9fV1MbReddMyfso0yofyAQZnjJt78rtz507p4+c0bWt5DYGzRMCMEGeJvpVt CBgC5xIBDBjPnj9zGyJ8PXz40AukhL1588Y9fvzIffLJJ6UJGjsiBH/99dduaHBQBNB7ficGysKX X37pvn70yN2/f98rvq0H8kiEv1mPwUcffeSuiIKNW1tdc189+toLyQjrjbjIAJHPoYpYqKXGYf4i K3Kb22KUeSY7H3rc3bt3ff1RclGqHkn7PLwvbcZKXaFSV0BflI3UUa7cx5ckPLyPk0N/eWXZzYlh ZkyMMJPj45K1ze0LL0+ePnV7O9tewcweNNYYfV9+yJ7WCQMIzhso2tzKyor7rd/6l+6FKHPsyuAe xQnlJUonf5VOwL+P0/DwJkffR/k/x/O/LfWekbECFgjl7WIcwk7yWhSZ169ei1HkRmIoUv6PpZ8m aMiHcv340WM3MDjg7t276xUVFLyvvnokY/mJjJ9PXWdJB8G9X3nvnosx5M6dW/4gOiqCQfELGcu0 E8a8Mhzzx8uXL93y8op78OB+otTOSj9+JHPKp59+Wtr8hoL7+PFjP7/d++he3D4H0j5f+fnt4cMH PqwMXJaXl8VA9NYbp0ZHR32RGEieP3/ud65cnW5sfjspz/758/SZ2xRl/8GDB4nBF96++OIL6bMP pH3KMYSza4fnzKCMnwfSFkwihNFPvvjyK/eNb3xe2u5ADHgYZjC+T05Oeng3NzcEk6/crhg974gB LZ3kTor+8flon9cyvy0uLrlvfvMbiXyAcYR+S5uVbcg7nmtLYQicPwTOZi/v+cPBODIEDAFDIEGA 1acVURhY3deVUrZuT01N+ZW5mZkZv6qaZGiZ58gtLC3610GuiYKir4JwRSBGQJ6TXQllOFYEF+YX /A4MFdApd2h4yI3IDwWGbd0nd7FSnRBAM85oxz7m8PBIFP8VtyWKLu3D7gcsBp3dXW509IqsvB+5 pZUlaZ88vfx9Mf2kzEzRklfvAzKsJqO4oMRiJELpxnXKFmF2qqxL+2xubfqw9E9MSOn5iGL6yW4B 0lTUBwLyi+n0y+6PP/7Hf8b9Odn58J3v/Fk3Nj7mKUd/Tk8/JRYXGF+K6K+trjt2Qly5MuYNEKRp k8pwz2rhsoytiPGU/4hOlNL7M/TT2EZ9b9++cTt7u25KlBZdKe3o6BTj0FW/i4eV1TIcRobXb177 1y/Gx9NdQ6xyj44M+/GDQaIMx+oxK/4TE2OJAYJyJybG/Xh6J/NbYrxqKUNHbubdjN+BwfyWtk+H 3zHD/IaiV4ZDsZ6bnfPtMyJfKFDHV5ZQKJeXl0r7dKJ//rzn+TOeGCDgB8Wb+Y4+XZabmXnr24dn n042zHcYnHflla+FxYVSWPGKvxiZ+/p63fhYOrfxqtmEzHVLMg+zaFCGw3DGvMHupbx8wG648uSD MmprZRgCrUPAjBCtw9YoGwKGwAVFAAEDoQcFIXQIHJyBgBC/u7sXRrXEj2C8LEI4Rgd/xkBQyrBs 98e9F2G1DEedt0Xo7OvvE2UhqyEODg55AwRp6neBNu8zZWkW05GzFwSTdTESYYxI2kd2A/CPrdyd skX3/cqqbP/Pt0999NNyg/TsNlCX7BCQ97XF6MKZFPSL/FkUffEW6o31Dc0p17DOAc0q9NXA4Ank MA+Iei/C74MH992nsjPmigjpHXKGSZRP/jaBfkTseP6P3KG8nrLi+wivHCRO+O/s6vQGm3V5fYXt zJWuCv3KhHWF8PoUq/28158fP/2ivLS3t/nxU2mwqot8Q4kYG1vyCtGgrGDnx0+/KLoo/Ri0ynDs kmE3xIAYykKHcQZj1qrwyrvurXaMZQwNnDGQb5+B/gFf/MJCOUouBqC9/T2Z3wbkFa+saNzT0+tf RShnrhUjqqz4SyepWE1H+ec1BPoSO+Ra7eiTGOO7ZNz2CgahY+6lHy8sCK8lOPoJu5ool51V6uCh V/osr0OU0z5ODB6RkRsDVej0XBXah89tmjMEDIHaCGRn2tppLdYQMAQMgQ8CgbW1SJku2qaNoIEA z1kIrXbsPmBrO3zkFRcEQwRVhJ0ylKi19TW/Gl9xsKCAoK8brJOmbhco4T5PqICGRLLh7LbgPXoU dI8BSWPDAIo3B4iyQle5KyNLJy2hSngYDH3ufVjMt2DPytuhBLaL8tYmCm3oOruiHRqbG1UMM3XQ z9gsKnZCxKVVCU/sDsRX4f809DN5A/p7u3Kon+DPONGdO55T4YP7TmkfdkPowaaVOzy0XiGaJ/Nv b+14oxW85BXL9o52P6Z4vx1luNUOJYq6Fs0pXZ3Rwa5bJe2EqDW/cQgjO3zK2JXB/KaHYubnN4yJ Zc5vnBuC8+XmOkOH9BWcb8NcXLNvDw4Ee+GF2YR+Gzowot8y369Xm1fCDKf0R4eW7kVjOcZASbbL ocjsXOE1iMq5VlM178q5O7ii8aPPH9qnjGeh9oNCXmL5oDGDfPNwMkqGwEVCwIwQF6m1jFdDwBBo OQKsnvLOOA6FNu9UsSpjpZCVH1yyqp1hJlr9P5QvD5QhBOrOgiJeOuKVw92d/O6DDMMnvMkq9yiM HL5I7b0lIqAqMrpXLBHkWdWsz2XpV82jxQXGAwRwnC/X85PmRkCHx105n6Gqop0mj6pCGQH9MLq6 v5j/2C6TZivgP42s5SumX5Ejpr93sCdKrBx46ftENi84yWmM/gBC7dv5Nqyge4oAjCE4P2azrCQK HmO9jFcPdIzq/BFWS5XNnYrdO2Gq5vn35PUUnJYbUtawMuc3FNpKl85vLTcSiVKv/bHWnK9pKnlt XghlUF8MDnnDDKUkc+2pXn2rj1944XnojR+53SE0GeOZsbPHHNdip0Z/jDB5xwGi8MgYa7URAjy0 H9TqK7vxGMvzaveGgCGQIlA5mtM48xkChoAh8MEhgAKrggxCVjXXcsFYCtYyioRR+FL+VMGpxmsz wjmZHVfMSwQUK9yNO7RufjXA9vFRGtoHpV7rTnl848Ff40Da70BWcyNXB31Jn/CAl/ICdpS+pxeE I4Bz6zFRhnx+woiAkrygAM/QjAIif0CnGv2IB/LBT5whpi+BkSO4gn+N5MppDEFhgffk9OEnLSOk zy4hWPTR/g93pOcmUizxkc47bir4z9KPEjb+V40LtA//QufDYkw1XRjfbH+t8aNjSsd7s8vO09uv MZaVFzXE5vM2817rq2XmaWuXV6UvH9+se3qo9oEiXjRM+W1WuUV0+EKIjg0tN5MuBqUMXsIysqMH jqIQ5loM4a12KS8FnCTjOH12t4ofnj9p+1QvJeW3epriGKlfZRWLk1qoIXDBETAjxAVvQGPfEDAE ihEIBbjQX5w6DVUBIw3J+lQ+OC5dNtfJ7tQYUl0oibgpg5ekDAWgoEpJmoK4bBBivzoI1iDqk6Vp UOpx0SF2sV/DfAx/RBhVJdfTPoa+F2LjNP6i6bP0E/KUJ8I37ROlSGO0Kqr0ShLhBtcg/YQk+ZSf rDdNEqSJk4Z8YYbIuvA+yEuioKgkTwP4UF9vdEhIhfQlUm5JkvRtkh9DnyQncfX2R84XabVL6luE b1x4vfyehldWcrV9atE5H5hEYCXY1WL4FHHQV+xjfbYqtVbzovNK8UBkqESYKL9VGW1CRFqGlFnY b8vnpZCNuK60TavbB0yiOa4YYOWP3WANuTgjF312NJTfEhsCFxABM0JcwEYzlg0BQ+B4BMJtz0Xv blajcJwAoKJF/gyAavROFX6cRBwrl2FdT1VejcxJfRWAgrQc9FefqzddJbWofUSt9sqU0omvMW+k UWG9kkKdIZ5Wjn6SVcKJ4pLHI7735gnx04QkrXDH0a/IEAfky6uWLi40Sp7nIH8fEKmXfg3+dQzV JFWDBc9Nhn7AXwPeesfFscOsgTKrJtX61gDl1H22auFphC+jjgqXwUtkfEp5q/RFYLWaF5pGy6il YMKfNmMlr80KOa6EcjDxdU3m82qdtkRe6uyz2o7Nao08HZ3b8uF6r0g1wgdpeaUEV/Qam9Ju9Pp7 P/ih+19/+W+43/o336/IOi8HihL3j//pP6+I25Bzpoj7zX/wTyri8gF/9Zd/zfEjjzlDoFEEzAjR KGKW3hAwBC4EAjzYeY+UX7qiczzr4TuntQTSorMRjqfeWAoto9rqjvKn73E3Rr2x1KrQFfMSiV7K b32UVVwjdegPc1eGI6vF8lqaMJNMbiRB9iDCTII0X2G5pIVGkCcOipKn4WDCAZAZTAIdgjj6YZbf mNix9NNyMktvAf1i3KJ8bXHn8MnjIrP8n4a+5K3Cf0esuESYxGVoRwV58bMzI+pPAQ9JqxCWo5/E NeZJ+ywUs2XBn7ZbY/22MR40dcpLlg/iUz7KEMlQeKJytFzlMeQlnAfD+Gb6FfciPiJeotI0XTPL ztCSAart4/teJjIN8WmygzmX8vS3GHLDXV55irrCnvKbT9G8+/C5Eg7hqIS0H4fpmld6lpKWUdRX NCyaazMTZJZIE+4YF2psr8QkLUD5TUNq+/bl1T4OVI7Oa0mxrZ2rduw3Hn7q5sTY8J/+4IcVCf/o y8c+7vu/+x8r4p6/fO3jBuSrI8e5KfnMb9791r/9vvsbf+vvuvkmf14Xmv/4n1UaTfLl2/3FQUCO VzdnCBgChsDlQwDDg372rREjBMIDB07xnmu02p7FRmnp98Gzsc2949OCOH8OQgFpVaz4bGirnX6J QusflqdhmiaMq+5HWETY4qp+UodCZOgnji9idHkh/fAwPvARpSBOxoGVCIa0X/TFgcbpJ8RCpTXD htx4sm2Oz2LivDDKH36xknJ4EN3z2cPwk3KN0E/w8TShJ4VllCDlxbMR/4mYPSKdzyJ/oqA0PqHj PTFNn7hu+hEzKUnloFO+9IAyEGGi8TDAlnf5eYzaXbf/IkTEYwV/Phu8nc7RV3B+HOfIaZ9FmePT oa12Xd3KS44RKVh56SphHFNP5pXtnYOk3LDuCS/xFzvCuGb7dX6r9uqHzm+NzSuNc0l/1b6i9Q+p 6KtdXfKVolY7du3xhY7dau3jB5a0YQl9pbur28+1KPmq6Gv9YYMfz8tOmeNa7Xp0rk1es0tLVP7A jrZspaO+amCoJR/oeK+HF/jH0NYtn1/leVE5Q9RDpTLNgHxOe1KMBM9fvqmIVMMEOxh+/OUj97kY LNT9WAwUuM8ffqJBVa+/+it/rSLu+Ys37o+EZrNdK2g2m0ej1xgCZZjdG+PIUhsChoAh0AQEQqEp 9NdDenAw+v530eFSHDCHoNNXxypBPWXVStPT2+0/SRZ96SErmvjD2kR4QTjTT5TVonXauD75Pjtu v+CrE4pTX3/2u+n1lRnWK1JWq+c78p/l7BaFDgF43x/4mKb2wqgIqV3yecxUcWmMfkpNfGQNs2tk LOf29fV6/RmDFQYQ6RhJhn356gJZe3p6NVcSlwQcQz9KFzAQ0E9oZDxBWgHI/wOovMvI6UF8I/Sh SdYgO0EocxhdDhSTTILoEDs2S6jSR57U5YjlbtN09fkYo6wq0z9VkdScPkzHj/DbajfYP+iL0C9T hOXpwbL9A8evPIb5Turvi8vRckM6GjYwcJKxHFI63s/85tvHf40o29h+TonbJ2vEO57uSVKokqtz WUhDD63s7YvmwDCu2X4U6b7eXv/p37xBhPlNDzgto316ebYwloPDGLW+YMKYQmluROHW/I1eBwej 8cOOgbzTNvPzsZ/D8imae9+fPAsrD2JW+aD/VPJBdiychvtv/dRP+lclXuQMERgavvEgMjzkjRQ/ /kqNEFE8r25gAMBgwfV7/+JfuReyWwLHrgk1Dmi6+cXFKO5FFJd/VYNPbLNbAjr/SV4ZOc6FZZCX 8rR8zUv49//Df/Q0oQ0v5s4/Aq03X55/DIxDQ8AQMAQyCIyMjDi+S64CuUYiBPLpLRTL3t5oFVzj WnFlVRnha2VlxX8GLVTctre2fZHwWoYbHhp2M23vZIUu+rxfWCavvGCYUeNNGFfdj6AVa8Peq5qx XitzspGfHQ4I4FvbW/4b9Z2dqWKwK5+UQ/kdHBxx0e6QxuknbxkgzOZYSTj2MuKR6/NCeqfbl0/U SdeIXZQJXvhI50Bs0CIS/huh740aSjcRrnNMBbcJfc+oRiiByF7gQ32Q/IGmRjdCn8pUwYedBYMD Q+796nK0g4d3UuL0KDMoDH2izCWGM2VT0iT8x+nz+Hs6DfxBgRoY6HcbGxtisDqQHQ9dSW7GNvyM jPQHW/GT6KZ7Bgb7/YrxVjxuwwLghfEzImOsDDc0NOSWl5Zd0Wc4GcussvfLKmqrHfMb7QMffOYx nN90nilrfhsQxbJDttoXYaKfJx4ZKad9hmVOX5Y5n7ZQ5Zu2wCiBcZM5sIz2YYyi2G9sbPo5Lmwf xjGG19HRkXLGj8z5GKwYK+AQvo6y458/7W54uKT2kXIWRdGuJR8wxzXidNePz6NzciMEqqS9e/uG j0FxvxP4MQz83J/6WTcn9WBXxJ/5+Z9LKKD0k4+dFLjv/Yt/7Xht4+f+yz/hFX3C2GFx5/ZNeT3i //FGgf/7H/2fEvd77nv/778i2ru//Xf/vr/+vf/91xJaGAi+mzuHAlq/9iu/JDTH45zZC3TmF5Z8 IAYTXsuAl7/yP//FOOy1C9MQSBm/8PN/2v2lv/Adn8b+nE8EbCfE+WwX48oQMATOEIGJiQnXK4aG RbGm6yoY7MwvLHjBY3JqwgtErWYRQWv66rQva2FhPimObaCzc7OyrbrbwWsZDsV/fHzMra2vu53t naRIBOWlpSXPR2Orc6p9ogxDDsmrtvRFMgTj8fFxaZ8+Nzc/J0pttDLGDo0V4QND0cTkRPAKRAP0 KZ5ClDXxJk7JwKNP0yZKS5cbE0y2d7bcuuDiV9ulfHh5L0rEwMCgGwkEY2+AaIC+h4NMnh+Y41fs DkXJXhHFcnZuzr17J8Yi3i/e3XMzMzPu3ew7UToFG7+KmPLfCH1KrZf/sfEr0jd73JK8E3x0RPuI iC2Kw8ryilSl3feVom3TNekXV/vY0OvXr/s0C4sLno8ow5FXIlh1ZnyV4dhqTb+kn6DUqUOReb/y 3g0ND8mvHCVq7MqYGIoG3fLycrBD5Mjfw8/01at+hVt5bNWV+W1qcsobiPLzG+3F/DY9XU779Mju g5HhUbe1teU2xWiljh1n71ffewW3jN1vlDs2NuYNQbOzs/J532h+I3xF+gntc/3m9VKeP5R54/oN P6fO++dPNP/w/JmXeYZdEGW1D4axSRk/LA4whtRF7bPqhoYG3ZUroxrc0uvo6Kgo1QN+ng3lg/fv o/ZBPmj4dZlkak88TakD50Lgnr96k9D7Tz/4Q+//XHZCsFOC8yF0twIGCPy6SyLJJB52LfyP/8Of db8qBoNv/fRPhlHe/+f/u287jBE/91/8CX+P8YH7SXlG4tipgHGA+H/09/6Wj/urYkjYlPJqHYL5 9/52RAca1AeaaoBgx4MaO+CLONL/mf/659y//Df/3u+MIJ+584mA7YQ4n+1iXBkChsApEUAZPalj BfX2ndvu5cuX7vnz515p2t7Zdu9m3okgNOkF55PSbjTfoAhX09NX3ezsnKwAHYmwNSzKwqIXxO7d vScrYo2tuDRavqZHaUShw+jw6PEjd/2GrLAIxDMzb/1q3Y0bCMZF2rtSqHbVPHqtli4Nx9gRYTLr y++Vra/bsh2TVebpa9dyq4RKV68pnQpfHUnyFooro1dE2d/3Sj4r6xxatilKJrtlxmVlJ1nxp7AT 0E8zBZnp2sEtpLekXf7d9/+d9JNZt7G+IcqL8CIru7/927/j/uAP/rObGBt3/823v+2GK1ZzlZBe hVgBfcrIl+nDKv60+V1C46IwLImAODe34Lp6umUldU/aZ9Mr4qw6F7qAhcL4EwSy6n9D+urc3Lyw 3ybjZ8gbzVZX19zt27dcXwkr/sr2NVGoN6WfPnv21CtvGEGYU7jeu3M3s7qreVpxpbxbUvenT57K WH7srorRgR0AM35+m3CTU5OtKLaQ5pWxK9J3N/38RgdjbC+IsZfdK3cFk4aVucJS6gu8KvMsO93e vp2RsTvhd0bMyXjibCHarizH8+ejjz5yT548kefPC8dYYrX/zdu3/vkzOVFe+/D8uXZ92s3Nzgse HX6ux+iMMfru3XLb55rM7XtihOG5jJ9+/ObNG8/X7du3SzPMqHzw9OnTRD7YEaPvjPQbDEgY1s6L Y5cBP31lAr7+6KtHfqcD4T/70z/hX43g9QbOhdBXM4rOg8AA8adjA8NJ6scuCcr8K3852sEAjT8t Oxqeye4Gdkjwyoju1qiXPjTZJfFXhaYaXCgDXl+8eu3psstDd3XUS9fSlYOAGSHKwdlKMQQMgQuG AFs7P/30U78KxupLhxwW9vHHn/jtwydTtk8OAMo/25LX1tdkNWzZb2efnr4mim7rXwkJuUYh8JjI is/mZrRayEoZim24PTbMU59fDUb1aaGc7s9qbo+sLqPUserPIZFXRADkXdzodPmw5Aboa1Kyh+z4 ZXoJ85+8iGlLPFuUr16d9Kun2/Kqiiz4e8MQwjtxFX2lQfq+JAxqnpeYoZCvmBUUpT/5J/+kYLEX J9VEklf+03/9Fm7K1yi91kE/LiYyUOiN5uc+wIeDQUelv/aK8YFVfxQHjDP0Y1aca/YV+FMX0tew Bq/gj5I9NDjkVtdW/atNGO7gpUwFF7YxSH3yySd+58OmGGRYdWcleXhkSDBp/bkUIXQo+w8/eyjb /pf96jL946OP7nkls6LPhhmb7Kes6zKHjIyMxvPbSmzUvFH6/IZyeUeUWRRs2gZD9pT0HV6JqNln m4wJ5CjzG59/7pbfr7hVmW/h7VPpO7y+Ura7fu26GxsdS/oKr5hhBNCDn8vih/n0nhhnMCCub6zJ Lp4jzwfPavAp0/nx8+BB1D4yr1D+WckHx9WbXQ28TuG/ViF9GmX/z/+33/bZeKUCBZ3dERghfiwG CtxdCc+7iXhHQz68nnt2WGAs+JYYPfhiR+imJqOdEs8kTaNGCAwY8F9kHNFdHuzg4PUNc+cPATNC nL82MY4MAUPgnCCAkn8eVjUQ1BFKw/eDzwoihC1We/h3chdqwlDJa5v5+FxJkpxV/mgL+1Auktt8 /gbo55Mm1OMIf4npxxfebR8WJSrdTB+UH3g9qRPQF0tGwkXkCYjGXtql/q3RcSYlUwf9hIE8K/mI GB+Utv6+ATF8FB1wqAVL5sDrSVWlnxTUsIfxg+IUns/RMJEmZQAXVv/5d9YOpc7mt2wr8GlflNqy zhfIlp694wyTMnc9ZEvP3vXK2RDX+q5lA8/gjvHDORT8ztrxNZvzMH6Ow8GfC/G7sgPii8jAQHrd 6YACT/zvybkQf0l2D6DUc89ugmY6fd0DY4e+DpKnr+c+5MNr3S+IYaPa+Sgazuse5s4nAmaEOJ/t YlwZAoaAIXCJESjSNENttCi+XjhCOmGeMLxJ9DNkqtDPpAn5qeYP6GTyBuGh0SaTphrNMDygk8kb hJ9r+mFdzG8IGAKGgCFQCwHOb/juP/vnfgcEB1FiYNBXF8jHjgHiOTeCXRKt3DUAL9/6qZ8oZLdo 90VhwiBQDQ1BkHkvEAJmhLhAjWWsGgKGgCFwORBA4cWFWnDoj2Kjv5o2m56t0ukCfpgXv+bJh4d0 1U/ayvQV9JMkQjP/+oInFZaltLlqRvxpGqMPHrg68YkS219DwBAwBAyBBhDA6MCOhznZNYChITRA QEaNFL/1b77vqX7j4Sf+2sw/U/GXL+CjmUaOKambHqwJ7dDp+RaNvuIR0jB/axGwr2O0Fl+jbggY AoaAIVCBAMp4qpBnFdF8Yk0bppfc3gKhcdXyaHio6GqYXpWGXINkFfQ1GdmSssXv8wQZlWxy1YxG P4Ek46kTn0weuzEEDAFDwBCoFwFev+BTnP5chtxOBIwU/Dg7AcdXM5rtoI8xgDL4okXoNuRcJw6m PIljFwdODShKgzKoL+XmjS6axq5nj4AZIc6+DYwDQ8AQMAQ+UARUeUcRPYnT/NXyanyd9CuSaf4a 9H2eioxphpBERbIwMs2S+iTe6KdwmM8QMAQMgUuNQMVDoim1DT+5yQGUeYcyz7kNapDIxzdy78+g kAzf+xe/5T/LqedB/E9/4c/6Mn7tb/1df1AmX+zgs51//Vf/N/8pTU1XrSx2OnDAJQaLH0teHF++ oDy+kvGb/+D/8l8BgSZlYHD57+WzoebOLwJmhDi/bWOcGQKGwCkQiFayIwKh/xQkLWvTEUgFrmJ1 vDg0ZUPzB+kCb7jbIhOcECgOTaKT3RpBusBbF31lMSUa+DQyIBp4jb7iE0BmXkPAEDAELhsC8VTH pRXyih5Eya6A/GsLQMmnOnF3b1V+FcNHNPCHz25izPj+f/g9MQz8E7cpX5rBYfz4tV/5Je//zb// T9zfEEMB8byq8au/8tcK+fKJ4z980QNDxXf/6T8Xo8O/TqJ++Zf+F/dz8ulQylOaRP7yL/3lwq9m JBnNc+YItMl7qRmR58w5MgYMAUPAEGgCAr/5m7/p/uE//Iee0t/8m3/T/fzP/7w72pXPeu2sNYG6 kWgNAjyO8opnUVhR6fWkK0pz5A435osI5sKK8uaS+Hczyue/ErM8X9xfbP7bB6aKKmVhhkAGgcON ucz9Wd20dQ24tu6iL8OcFUdW7nlFYFc+YfzX//qvuJcvX/qvfvydv/N/yNdZhs4ru3Xzxa4FDo7U 8yDCjBgTeGViUgwQRUaRMG3or5VP46qVGdI5C39bz5DMCRe/XZuFnRjb2uxgymahaXQMAUPgXCGw Ke8ZzszM+G+98813c+cZAVWQ8wo8PBeFFdWlVrozpH8kfNViLalKrUQ1+P9g6CdAmccQMAQMgUuD wMHBgVtaWnTz83NuZ2fb7e/vXYq61fraBYaHgduN77qola9W3KUA9BJWwowQl7BRrUqGgCHgxALf 765di75r3i3f8zZ3/hA4ktX5Nq+hhwq4Ktwhv4SJyyjcRemiZPq3bvrJhsDQYCD0M+UpVfiI/Edt Mf+ZdDFfcRpfPaMfGGKagE/QFOY1BAwBQ+AiI9De3u7Gxsbd+vqG7IAYdh2dHRe5Osa7IVA3AmaE qBsqS2gIGAIXCQF70+z8t1ZkgMjzqQaJUIuPw/wlVmLz2Qru66affutTqARKsrIC7bDYhJ3EE5Su YTH/3kBRQCikZ/QD/KrgH6QwryFgCBgChoAhYAhcbATsYMqL3X7GvSFgCBgClwSBWGlPaoPiHirv GlEUls+racNrPs0x9DPFSF69D8lk/OEN5cb0NV/CShyQCTf6CTwKdAafNNZ8hoAhYAhcNgTYtRc5 uar3slXS6mMI5BCwnRA5QOzWEDAEDAFFgN0UGxsbjoOjurq6/CseHR1ns1WS90bX19flfdED19fb 4/rkncq2trOxI29vbbut7S1/ind/X7/r7jnJ6y5IWqGmGfq1BfLXME+bOxAsVtdW3c7ujoOPwYEB 194BJtAK00Kncfrk8i583UJewfC0eMUi2EGxt7fnVqV9KLavv9fzE+cOLiFPAT910A/LCgjmvCLK ShGcgbItv/aOTn8wWHeXtI8WV4X/eukHhNKya/EvDC2trLiZt2/dN77xjTRPoa8KPoVpGws8lPGz 5sfPvuvtlfaR17VacQp9PVzt7Ow4zqzBDfQPnHD81FNS7TTMb/Cxvb0dzW99fa5T5rmzcDq/ce3t Odv5jfmeMUT/6JO+0nVmr/Px/NmUcwJ2XGdnpxsY6HcdMqbPwh0eHrq1NZ4/+66vr1d+PH90UimX I/AAFxyY9Eh/OSvHM5n+0tnZFbfPCeUDpj5zhsAHhsDZzGYfGMhWXUPAECgfgdO+joFg/vz5c6/0 T8jnptbW1tyTJ0/c9evX3eTkZIkCmChxS8vu1atXbnBw0P9evprzfH388UdeGCwLXQTRFy9eeGPI 2JUx0UeP5P6lu3Jl1N26datBTPICbKiAhjWqDD86OnSrq+v+NHGE0H75zS/Mu7ei6E5NTcn7tVeE QCP0w/Jif1gsijuKNk4Fb64SzCFii0tLbvX9ezc0JO/zihFk7t2sVxompJ+gUCV5IgrR3zrpJ9XI GT0SUhou1729fffm7RuHwj04NOQOxDCyKLgMSL8Bl3aMVjn+66afFBh7avCPArey8t7tiKHqP//w h+4PfvADwWaw2Aih/Nein49r8H5J2ufVq9eiJPT5Nnrz+o3blXa7e+eOH08NkjtxcsbP69ev3fLy smP8dHZ1+j7MuL4jvKBoluVQnJ4+eeoODg/c2PiYH9NP5P7atWk3PT1dFhtSzpFbkFPy37x5I20z JG004GZnZ70x7d69u6XObzwzMJZhrBoZGXG8qz/zdsa/pz89fdW1yX1ZblcU7afPnjn6zNhY9Px5 9uy5PH+u+edPWXxQzooYEXn+YHigry4uLnhjxL2797wxvDxejvw4ZjxfkTmehQH6DV9DOJPx8/Sp fw6Py/jBQPP48SN34+YNNzU5JdNs/vlTHkpWkiFwURAo74l3URAxPg0BQ+CDRwDBLzJA7LvPPvtM FMsO/5UNVlDfvHnr78fHx0vBaVN2HcDLxMREouhjBHn06JEYRZ66+/c/deUcvHkkSv4bUaBWfJkI o7jR0Svuyy+/8gL7jRs3WiB85YW5aHXQKy2DA+7mzZtemdsVhfuttA2GCHZF9MpqXX0uT79KLk0W Kt+S9L0YH94LJuPyqTEOF0OpHx0Zdc+ePxMDwIK7evVqfavLVehX4SYIjjLui+Fhfn7e7QsO16Ud +nr7hJcjtyKruISzWjg6OipBkl7LCqhU99aZWJMJPq9F4f/3//77bk9WTVF2MUiwS6XYacbi2NOG Ykzk03cYpm7duu37J2Pp6dNnXsl7cP+BYHOSnTyNc7a4uOTm5ubcp59+IkruqCeAgsn4oT/fvh3x 1zjlxnKw2+Dlyxeyg2hX5reHyfzBmMbIiHJX1vy2Lsob7UObUH8c/sePH/v57cGD+56fxmp4stSL YgxZFoUbPjCI4Dgo8KkomzwTUDDLcBhDmNsPpMyHDx8kxqm+vjmPVZfsbBodHSmDFb9LBmPIFZk7 7t27K2W2yWcVJ2R+e+7b6LPPP0v4k8iWunezc3783Lt3zxtmKGzsyhX3xRdfeVw++uiehLR2PqFM b0yUOY4dGd/85jcT+QCDOOOnU3arlDV+4Oekbk76u/+ihOwKM2cInAUC5Zl1z6J2VqYhYAgYAidA gJUfXsOYmJzwAgYkWNlAsGC1EkWC1fjWuyM3L2Uh9FydkpW4eHWFFbrpq9MiIG55XlrPh5Nt21tu aXHZDYmiwmqlOvwIxKxmIpQ17tDqc5p9BZE0zdHhkVf8WW2fmJj0Cj7fqOCVA5RsFKxF+dzZkWAW uTRvBVkNYDVeecCLP5BloZ+4wHsoq8j0FVa0UVra2jkKs811iVLLTgS26m6w7f6E9KNiY950Zc3z l3AT8Sn0t7Y2ZQV3TdpmUFYseyNeZOcDuzO6xQCxIqvve6J0hlVphH4j+KCs/Zlf+AX3nT/359y3 vvUtGUMC2jH816If1LZhL8o9fYK+ouMHo+LU1KTgsecWFucbpnmSDLyuM/vu/2fvTYAsW6q636yu rqFr6Bq6q+d5uCPIIPhhoDI4PSBQUUHx+x5iPBxC5GnoJ0g4AN74VJTg6eNCBCrh/TQUw6vwIBRR UCEUgcsg3KnneR5r6q7qquqqeuuX+6xTeXbtc+qcqrPzVFWvvLf67J07d66V/5WZe62V02W//GL9 +jkDEufQ+vXdvv3Q58QIIyMjMpNoVPqz/qIDArp9YtDRvzETYXr6bgRWZOaBYLJmTbPv35Qg8hkQ edG/XRdHXoxA22CE34/2B/0bjufurm43PDLsHWoxePHfn7Hbvo6Gs2NwWHFPnV7qTL/qyoHj+aJ0 E01+JpV2Hs3CA47wyanJaPKhDV++dNnPxKCf19DW1u4djEPiDB4Vh1aMMCptZ3Bo0G2W2TG6PFP1 g9bWNv9N5tuwnMO16zfdL/zqe9z//ujHi2yePnvePf7//aN8s5b3keZPfP1J97n/+LLMfrxR5D3r guekW+hvuZc3q2yrJc5mQqwWSVo5DAFDoG4IoPii5DGiHgYUwNbWFm9cTkxM+rXl4fN6X7P+lqnb KDptsg9EGLpkejsBgyJGYDkKSmf/hr6iMad0O2VGAnySBqV94YAlrda8/lZ6ay7NhBiOI8Mj4iYQ +cj0ekJTwYHA0gdkNCzPcRglvMy9W5aCGshJZj7HJG1oDIyuAABAAElEQVTCp+Y/934SPy5nut8R x0uPGJWqjGqxOmTqssdEFFamdheDZ0d5qpx/8o6mLeSQuiV2Wuoqa6QxHFmaMrdXyKx3kDCyzXIR HCI4JIoGv88ylWHqNkkSRPpLvc/mn+nR7LlAYJR7Ttallzz3YQH8NVmtvzjvmIWBgyo9WwjjH5kN ySyWrVu2+pk8teZfS3raxh2pL5s3z5+qDVY3bjAKPxhleQhGLtjgsAoDzk144fmd8QlZxpOvikj/ hqMO2aT7N/a8IdCG9KjlkNd6X4+JU5OZOziI0ssumFWFE4K6sknkl29g+Z0YV9ImQmcvNKmvYIWz ir080s/rzReGP9+X5LtXOluI9kN94VvJbC918NWbB80v2Q9pSpxEffPaKv08S9CotzqDRd/L43do eChTPwAD1Q/0uM086OeV56f++fPuc1/4sjgA+93Lv+O/5UVmSfniWHjfB/7U5/Hz/9d/Fz7Lz0i9 Ko6WD33kLxek98EHD/oZIQsmtAR1RyDfL0zd2bUMDQFDwBDIHwFG/QlFwzIgSRwOCqZ5V2dwBy/X eIkhgKLO6Eo6eN5E6WFGAAYFCmGeQTfSY9QyHZoLcZom/Xz+vRqx858sFDMljhAMf8JcmZP8ML75 wxhn5HlJ8vH2tfKpv8pdcj8+dsfXBWZAYDT4kNjlsilks1fMqSeZoYr8y79X+mRWFHDWkM/ILJGi ARXkD04zUmcZ7WWGSDFNaTbJXYH/rEclcUH+JU4GnyiNV+L28I8WlX8J5apv2ECVGUvNaxNZhC+u 9fJxMntn0s+UmKtLYar6XbORKyFrY0GMPMJ4pBFIbadrBZd0UF5wWOFczDPQv2HoZvYpIh/aFP0b MpxzrOXDEXQI1JV00LqBEynvwMbD1Mk1Uva0Yc+99rXwm7cT4s4d5DMj3581QV+bIJBg0uT7WZZc 4ZTIM+CEIJT7JvMsqSvSB2pfTGQOQTfFLMcL+gFtjKU8Kym8/odeJfvkbHcvfuG3LFu2P/SRv6qa t4cfOOge/YN3lU3/jnf9vne4DMiMMAuNQcCcEI3B3agaAobAMkWAERU2GySo8hmyqnGcyJB3UBps dpgVmCaLgj4hyqKOHGalq0ecLrXQ8od5apymCZ+Vvw6t0fA6fGN+PLLB6ULZi8ZvIRnKJ/rnjCjO jGrOhfn5JM+y4omT4O3ogjFdiEri9Z1ZPzOEdN5g8A/1PeqOPJD/cYiUGv7V5z9XPnlHFesCSwm5 hBfwuCt0eNTs05GeFElicc34mSMYfMwgSWKTd30+ojRXyt+nKf7DexKC/P1ElCRW4jVf/eWB0tT3 NHGYJozTdAmn+qTWX20/3ihJZaWOGKZN4+hjtkieAScRIctA0vaDMRcjaDvNMv7neInXv/m2klFw 7d8mJ6bmzZTISL74KKn/in2Ws0NlpmkWT2jhN6mLOFAJSjd8S7GKwQt8JA4g+tXSBsQ9UTgpcJzk 7IPwDgaPCf1qKvhlcMIMmOAASPOaSr6k22r1gxjyWVJBMl5mBsSrv/flGU+WRxTLKp45fMw7DlhO Uk3YJHs1ZQXyYhnGm7/v5VmPLS4SAuaEiAS0kTEEDIGVgQAGXTWB4yHzDsoLynhWQAnEfsQAzTuw FwMhW8FL+EMhrT7wjhqhes3bSV5JPuF1EsNoP6Fk4ocm879N3kmRYFd7/nP0EzqemOavNz5boSMG Q0JSnEQqDH4LITlUY8bPUJjzI+nzhfMv4uPzlPQFupq/57UQrbgkufNvkD+Ku9wiH+rLXD7cyLMF 8i8RSfEmyD8hWmAroVXM08cmC1o88QAf5b+YZSFtklGQfyHnWn+K7UdopttQYkQljGu6WvOvJf3M dFKerPajcTiJYoRiXSmRW0JZeYmxpl1xT8tGMaCqUF85wSPPgGQwXglafn9T+EfjYsiHGUtV8VLi ZA25rd91WN6SZhuQgNeYdcX3GQF9LrX+UK8Vu1SSut1SZ6uhEWJXG3Gp9BntsrY8SlOzhIEZBOz5 gOHNDIHX/+D/UZpI7liK8fn/eMK9+Y0/7HbvSjZhZTkDm1e+6AXPdX/7iU97JwD3L37hc91PvvFH /IyPdN5vfuPr5P0dJflj9H/+C0/493kAD6/+3pdVPeuCMjz+iX/0y0SYucD1UgLv43R52Uv/21Ky sXeXiED28NoSM7XXDQFDwBBoNAKqOMKHju5VwxNKYKWg+oEq0JXSLvVZUdlRovMyTB6oUTHvcR0j ZqrYiJNRs9pDiDflCe/TuTGOr8/LgCLRPlXRmaTpyWvh/Eso8mr4uj702czxUuTEa+r6gsTK/wkb ios+K2RUKX+lFTJQkn8xgVxIRvK/Z6vITBKtr/tnhXSlSm7AU9n8lVaQlihuU1E+ZcBDcCkMJlz4 NJn/pDJL3Wa+UiGy2jZabboKpBZ8VE3biMNHOaGVFiFGn1Jt/7ZAl1zK+CLu4EOx91W0Qh5Fniuk WcojZk0lvJS0nLksCwwu9J2ae2HxV4pJ0rNk5ZPwOJcuK0194pRGGVQSIiLHvOWTOInKl0n5q5UP 1VcQrzpVylOp/gnG+7vf+wFv/D/knQ+v8kea/kFhX4Uwp2vXbvp0t2QpiQZmHvzDP3/O78OwZ+d2 cV4k739OnBXvee//694uSxoIzKB4uRj0pIdeuNkjDgicGSzx4n3+OJkDHp6V9NUEdTrw7lIDPDKT 4g2y/MRCYxGwmRCNxd+oGwKGQE4IoLTo2mdVYKohtZACoHaR3wegmgyXkkY1YiU6L6/kgU7Rnfe4 jhGqJFXKspo0c+/De0Fl85eqvunvXEq9YlJ/k0t851lKnjdWJK9kbwjyWUT+CaRkUnx9jn4hStNI /lyWGknKvzyR/8FEcfH867sL5u9fnjPytS5omZQpzyY0xOGhJYZGKn9/MqfGeR6qz19JLYZ/JTXn rFB8CrkGtwvlr3xU+1ut81HlU22+i0pXlF/5t7OWApRPvbgnvqzwUlpp52UWFRNfSeaxIBHJgyqg y3q56jiqoGK/ACzp1lc1jWoTrvH72rDMrswbhQcx5KNLlrwc4Cdoqwl3CZPVtrMyJaoquiifSqml ouSNC1+gSvVRxVaLfgDPLJFi/yD21qjnzB+MdwxuZje8Olh6wCkYathXglSf/f573uFnDnDP3hFv f9d7ZWbFBT8zgQ0iNTC7gHw/L44HpfcPn/mcn03xrnf838UNINn48q2/+m6f9l3iHKkUcGLg9Hjr W/57kYdK6Rd6xowOZnM8dH9lugvlY8+XjoDNhFg6hpaDIWAILEMEcDwwJVL/qmUx3JisrCIomXEW eN5BNyHLMrihrfxF4YWN4jxNVbP8beGfJC5ro64wVem1arTyrr8kj6y8594iWbMsLUBp07LPPeVq Lo85xbiG/HkdIsoaWWrQbKDh08gmcYU6IOOoBWGQQRKUP6+wimFBKG6XUGX+vji85NOT91z+PkP9 R4j4ddFy71erpPOHGfkfTBJ+JCEXZFdN/gU6i+KfdyHlZ8mU4b+a/AtpavnROkD74b8w+LiCkGqr t2Eu1V8rjay2rHFZG0VWT6H6lM2FtUxKN3xT45Tf8Fm9r6vt3zRdvelrfvQnVWPiG5C+Wf9f2nFS b7NH9VU+uoFo/TmYy7GlsGEqMaWtZy4G7LSdzb1Z/6siL9qxBiQUk6R/S3d+QcI6XIab3GawUqRQ S52Ff5a0oKuw5xGOjnqFZw4f94a7OgQ031pOv8CxwF8Y9uxMllu87KXfFka7hx444O/DmRC60bcO CpGA/N71jrfJko4fLnk/faPLMFi+UWnpBDMrwj9/NHY6M7knP2ZCsPlmukwZyS0qZwTy16JzLoBl bwgYAoZAFgIoaXpcVy0KNYoMm9T5TbmSufQl2eusivSRfyWJ6nTT3pYci3Y3Y604iotuGta+Lt+d ySlOK7yMsq/A/PXZ0wWcFrdDuipc+lsZPM6oR1Fn80kwQAkuBtGUUQzZyLNlrW40qM/1t5h6/kUV SQoeAf+urwOeppq4hQyIY32ypGoRR0VRSa8x/4RBfUl/JZaMg1s2xsQRRXTWshmO8ESxpR3MV3A1 I/2dn7/EJCFIolHzf+cngq+E4cKzFP/FPOa/Wny0mAtto7RZv6dJMOzi44QPMCkaOIshUuU72ja0 /whf0zal/IbP8riGFwwCpRvSYCNVQgxetH/T/iPkI+zfWtu0LYcp6ngt7adFjl4mZC2bgRcCR73m HVpa1vrvD3sOZS2JUaxiyAcazEDIbj+Jk4S+LYYT3H/jBlneNv/7k7TlWcFNvg3h9yAHYal+4DfB rKAfaHuvloV1chw4p51wbG4tsygq5Y8jgFkQGPB5hU0D2Rs/hvReI0s1Hvvox/wyjYfFScGykIfv P1AVXzqT490yi6JS+IVffU/JY8qMkyMddPZHPZZ1pPO2+9oRMCdE7ZjZG4aAIbDKEVi3bp13QkwV TskIi8vu5Sg66+Rs8rwDhv9aMaaTjSdLLbe7U4mxgLKjU1Xz5Gdd+zqfvRopIa3pAk7gtriQKPkl lnWZjDAE2lva3O0pMaJEFmuLpxoko0koqTigSk87qD7/kiG/0Cj20wCEKb+uIWGuo6Pd+wK8QwSF tDBbhEgcRxg0bel6oqyQxQL5F59jBPm0hRfC9yQbjgNtkWNc1zTd8s6ZYhlIJ+/OiEGzppkz7BOj AtI+aD4L5K/J/e8i+GfRip8tohkpXb0Pf8vlH6ap8nrdOpGPP7IVWYQZi3yk7hBH+wlnP1WZdc3J OsTIIOjJO2EGehpCR0eSJnyWxzW44IRQuiGNqUK/svi2HOZW+Zr+DaMumX6OfOYqhm78G6t/0+N8 szBRJ3C74JZ3YHYV35Yxf/RyqcFNfaWvIWC05h1axTHD7Bw2tEVGzfKfBpwhOEnoU3J3EgnRjo7k uFhO4kgHjUOGeTshoE3buH37tpzAlJxiEvJT1A+WUldKu6ow+5qudQZCo0f8mYXBzAOWgDALgaUV hJd/x7f5zS1ZGpEVdBkG6fiCMNOBMDae7FnBL3GcgBEuCSFNVp7MgoA2s0AajQk8WpAZxQaCIWAI GAKGQCkCG/o3uJGREb9GM3yCgjEpx5ahhLS15z/7AAdET896d+PGDeFlwqmiDE+3bsu56eIM6esr nSYZ8lvP656eHnfp8mV3Zzx9dN+sKGRjYn+vreFc9FKDIzQ+Ep7Tz+dKwohld2+3u31nzBtS64Uv DeOyrhaFdNOm/pR85oybhfIP7CDNtvBbyMP/JPyta2sXI7bdH9XJqHLoEBkXBSkTkzQrRSrz8xe9 K4Fm3uiePpDncokh193dKXV22CvHyEqVcY4qxbBqa213XSjyAf/V5l+CSa38C3+z4rgp9QGU8l9d /kWgqr7gCEraz9DwsMcgdEyNeyNvxs+WiuHE6+zq8Abb7Vtj3vmh8kGAd8bv+BkZ69evr7psS0nY 29snfcpNX1c2bJgbycRphnOC0dgYDhH6N2arYdCl+zdvaEi9D/lbSpkXepdRaGbFsC4/HSZEPsir r7c3/SiXe/q0GzdvelnobD4I4QDm+8N3oLMz23CrJ0P0X11dXW5Y2s/ExGTJ7Ji7k8nxnRvFAIzR fmgbzLhI2q04VaVta5i4M+77wL6+Po3K9bfPtx++yaV1hW+A6gfqNKmWkWTeHKnpG+sT1NDmVIxG B3j5ednTgQA/n/rnz4tD4Mt+pgZ7RWQFTtMg4DhQx0WY7rG/+rjjjxkP1Swv0VkQ6SUkYZ52HRcB c0LExduoGQKGQCQE0iOftZDt7et1Xde73HXxnKMEq/Fy5coVfxb5jr17oyheKL6bN28WJXDEXbt2 ze3YscMrwxgL8NIuo7gDAxtrKdqi0+J42TQw4K5evepu3brllVMyY7QFJXXbtm0lTpLKhLIs2cAw LbFKS3NiE87enj43IpjgnOno7PCzRRjBHR4aFGUUx0yfx6n0zeryL30n624unyYxWqgrN65fdyOj o25A5EGYFAX91ugtP722dmNuLv9SGIL48EEBynaZqdItBgMOoXExJKGLYjsqfE1MTohjZlPBSRLk UyKGID4j/ywksuNmfZ2gft4Vg+nkieMyejvtxqTOPPnUU65DRgi7urrdli1bktdLeMjOcSmxW7du 9RhcFX727N0jWUFQRtWuXhOjqsVtVT4kNs/QIjN4kMGFCxfc0NCQr6PQw5AZ5L63p7h8LE8+yBvH TK8Y1PABPkn/Nuv7Oxyte/Zs8QZ53nzQv4HJqdOn5/Vv16VN0b9t3rwpbzZ8/mCA8/n6DWnL4oBW hxDyGb016uXVIqP+MQL9F+3n6pWrrl+ule41wYSZNLt37RM2cm44hYJu37bd43H9+jXv6IQu35/L ly95p8BmkV+MgKOVunruwnn/PVSHA8siBoeGXX9/X7T209XdJe2nRzC4XKIf4DiaFEf4jn37/Gy8 mnCh+80hsPyBDSRZmhHODii3Z0K9WUj2dPi03yuC0zMIe+T4ThwSzxw55nkrR5P9IrL45BhRjhNl mceL5KhQ8lsowMdXvv6UXwKS5/KUhfiw56UImBOiFA+7MwQMAUPAK+B79+51J0+edMeOHnPre9e7 KVF2RkZG3a6dO4sGRAyoMCZ3797lTp065SZEIV4n9xiWGAvwGGNtsJYTg21SDNpjx44VRygHBwfd xo0bxRkyoMmq+FWNK1Skw+swC01LHGmaxLjvcCjH50UhPXv2XGF67JjHZNv2HQVnSGhU67v8pgPp lMYcDzixxEYqBLkoJuFabvyzZHSU6crDYtAxhZwd5cfHbnvnyIDgkmw2WHxZMyyMhutt+fw1RYHg 3G3xKskbOhtFBnfvXvQGDPWE4/7GZZSwRwzO3p7eOcdMwH8xm6RAc7fFK+WdCF9o/6QSPjhlPvuZ z8ro6R2vRG4URxn7Vvzrv/yLLAtZ4/bu2ete+9rXFihUmX8hda0/jHDvEXpnzpx2x4+f8M6ZUWnH jFru3bs3mL1Sa861p8ehiNF0Wozu0dERP5o7NDgkBs16t3PXzjn51J51TW8wirxr1y534sQJd+TI EXGk9Un/NiHG3ajbIRvO4aSIFTD2d2zfLu34rDfg2sXZiROA+oV8whHvvHnCocusoYviKMKJSFtm dhEGL86SWAGDe58YssiHOrte5MH3hxk9u3fv9vKKxQtLUHYJzfPnzrnjx467ddL33hZsGPE/cPCg XwoWi5dN4pC6I33K2TNnk/YjTuDBm0PifOiS70HioI/BCzNmdu5M2o/qBzhcGSzYEVk/WKi8L37B t8gSiOPuf3/072TJwv/wyTHsOTIzRkB/+crXn/TLMNgHYkBmzhBwCjCAobM1snjZs2t7VrQc63nc x++W59U6FJ4QBwSOGJZ2lAtzyzW+rchnubQWXx8EzAlRHxwtF0PAEFhlCLAW+YEH7hdl55YYcmL8 y0jzdjFwYxr9CilKMIoWThCcD4wiM00XZShmYO8BDIPxsTvJchAhzkwRDL3awpwxm7yHIZqO0xzn xzOC2iMjUUxxHx0RhVhGB5kWzEwANq6cHxbKv0AjSAaNkhDeBs+Ysrx584AfGWMqO8ZT16YBGXVi 6UP4kuS2iPyTd4IXS5jiZo4G+2XsEOX4jiw1uCNGCyeJMLrrNxUNeQmvfdbV5V8t/zvFwP3pn/np hNN65T+v3NVH9DGzqeshX1emZEr7lq20n66oBi7cUqcw/gc2DiTtR7DZs3ePbz/z6lv1xVtUSvqx Bx54QGY13ZYp7mN+H4Jt4tij34sZKDdOTEaWcYKw5IAR70b0bzgdtu/YLs6zjX4pBG0ZecXYHyON OUsuHnpI6qw4nJmN0S7L/7bLTDidlZdOn+f9hv5+cWIin2H//dlc+P7gLIkZqCs4YXAIMRuPsG/f Xu9YbET7efDBB/w3GfmgH2zdui16+1kIf/ZjYPkDyxmYCYATAGO7o8w+DAvlV+tzZl9wPOgHxenx Vtk8EqcBzgB44pk6RmrNt9b0n/rM57zDo9IJG/8gS0RId02WqqX3mKiVnqWvDoEsba26Ny2VIWAI GAKrHAHWujJSp1NzG1lc1k/3izLY+CCbcorywN/Sgxq/c4Z0LXmyAWWfTMMtH2rMfx4b+n45CvLc G/RNXvnMNODCLJaQfzkOQscA/og1Umc7xCnEX4nXIzODOf4zHxMZmf9SPkLipU9qvWM5RP+G5dB+ ZGO7urWfWlEoTY/hhjOGv0YH+rcNy0Q+tOPMthwZJOSzXL4/OLyXx/cn2RiyEY6h+eJfPvKZz9tc zM+/5X944/+J/3rKG/7MBni1LGVgo8hwtgFLN5x7ld/oUd9myUNWeLEsg2AWQ4fMXArDJonj5Ikk r+TJy2QjSE7EYA+I0+cu+Pde/ILnOhwk4RKRMJ9K18rn3jIzJdLv4nRhKUjIUzoN968RftgQ1k7O yEInn7gm8fTylbdgCBgChsCqQuDRRx91jz32mC/TI4884l71qle52clRNzsh50xaWHYIZJub2bHz mQ/SBZdhuuzoJHbm9tUwacZ18HZwGSbMjs6ODd9LroN0wWWYLjs6OzZ8b3XkL86VznjT4efjZzEr BYGF23KckjS1dLqm1lpniMXhzagsLwQmZB+Jt7/9He6cLHlZv77b/dEf/ZEsiZrbcHl5cWvcLBaB prZu6RO6F/v6qntPnKwyZGLBEDAEDAFDwBBoMALJJAGM6jAQm44Ln+t1kC7JSB8Ufy3/Ao4rEp+i GO3CEDAEDAFDwBAwBFYBAuaEWAVCtCIYAoaAIbCyEVBHQ5aFnBWXVdpK6RqYvxxPWV2olK4C//dM /tWhaKkMAUPAEDAEDAFDYPkjYE6I5S8j49AQMAQWgQDraTWE1xpnv41HYO5s9DlZZc98wAiXP7XF PeslN5mFqTp/ViX6vzCbNL3gGaT9Yy64zuC/kMbvG2n5JzglYCXXS8EnyM0uDQFDwBBYPQjItyT8 nKyegllJDIF5CNjGlPMgsQhDwBBYbQiYE2J5SrQpU9tSDaxg4Ps0hTj/Q7ymqVyuqvMPHFaJp0Py T5MJ74vsFC8CRjSuwD8OCst/6fgEOdilIWAIGAKrCQHVUfhUZH+3VlNprSyGQIKAzYSwmmAIGAKr EgFOTpiZmfF/XFtY7ggUjPYimxjzBYO+GMdFVlz63ZIXCjfpNAvkX0JG3tX7MJuS6/BG+ZSX9L0i S4WIknjLvwiPAlaCz9xTuzIEDAFDYDUhoA6IWdFXcEDo/Woqo5XFEMhCwJwQWahYnCFgCKx4BKan p4tOiLt376748qy+AmQZ7QuVMnwnsFKzlkOUrt2QjIP0ZclUkX9TIQ1LLMIsS2Y7lCNg+ZdDJokv g0/ll+ypIWAIGAIrGgEOKpyRv7vTd2VlYNgPruhiGfOGQEUEbHiwIjz20BAwBFYyAjoDghkRFpYb AqEFXy1vZd4pidYb/a02b9KVeackunBTldMhTbsko7mHJdGW/xwwdmUIGAKGwOpGAKcDsx+am5vd mjXN4j43J8TqlriVThGwmRCKhP0aAobAqkLARhNWlTitMIaAIWAIGAKGwOpHwHwQq1/GVkKPgDkh rCIYAoaAIWAIGAKGgCFgCBgChoAh0AAE5mY/mAeiAfAbyQYhYMsxGgS8kTUEDIF8EajHTIipqSk3 NDTkbo+NubbWVtfb0+vWdbQL4yXz5/MtiM991o2P33GDg4MOnjo6Olxfb69b29ISgXYpidnZGTc8 POJGRkZk6uga193d7dav75bppPF92uz1cePmDTc+Nu7a29uFjx7BphHycSKfcTcqmMDTunXr3Pqe Hj+9thS9/O9YenT79m03Ojrq1sgU386uTtfd1S1VNnaddR4L6skdqbstrS1SV7pETuvyB2Eehbn2 Mzk56To7OxvcfkZFPiOeSzDpkbrSiPbDvjn0KdSXVulLqLOdnR3CV+y6ksjn5s2bvs4gn17hpRH9 m1Yd+tmvfOUr7oEHHnD9/f0aHfWXujo8NCzfn9uura1N+rf1vu5GZUKI8S29c6f0+9Mr35+WBnx/ 6N/o24ZHhj0M9G3w0ojNHOnraT9joh+0in5AO+7ooH9bRPsx30Psam30lgEC8bXGZVBoY8EQMAQM gYUQQNE5euSoGxGDu1eUP/aXOH7iuDt//kLkjaNm3eXLV9yxY8c8XRwhKISHjxzxDpKFylHP5yhd R44cc5cuXXadHZ3eGXL58mXh7bg3HupJa6G8RkdG3aFnD7nbt24XldBTp06KfC4KTnH3ALl27Zo7 c+aMX8nbJU6ZcZHPqZMnpe4kivJCZanXc4zK8+fOuysiExwh68SgvH7thjt16pSbEoMmZhgTw+nU yVPu1q1brqu7Uxwya9zZs+d83REBRWRl1tOk/WDA9PX2FdrPUXfz5mBEPhKnzEnB5OLFC65NnGZd YkBduXLVt5+JiYmovIyJ4+7QoUO+D8G4XdvSKvXktDtz+mzk9jMreFzy/RtOTfo3nL6N6N+uXr3q nn76affZz37Wvf/9/4/76Ef/2t0UI7MRAcPWf3/EWdUjmPD9oR2fPXvW1+OYPNG/HT161H9/MLRx uB6R78+NGzdisuGmZdNG+nj6EZyZ3V3r3dWrCW/wFDMgn8OHDnv9oLu7oB8cP+7lU48BkJhlMVqG QKMQsJkQjULe6BoCS0CAD+6lS5e8MsCo0Y4dO7ySsoQs7dUAAYxtjEpGevbs3VMc0WbTqNOnT/lR qYGBgeCN/C5HR2+5Cxcuuu3bt7nNmzf7EZ+e3h6ZETEpPJ6VGRptMjsj/9FlDLjz58+LATfu7r// fm/kUmrq3zPPPOuf7d69O8qIFCOEZ89J2dvb3N69ewqjyMnmXufOiYK6rt1t3LAB9nIPt2QUGeMF 2WzcuNHTY3bI6dOn3SVxBnQIPrpBaq7MiGF/5coVP2q6Z/eeYp1Y19buTohDBMfRzl07o8iHunJR 6iyOh53SNzWJcUlYI7NlLly8KHWn3TuOcsWjkPmIOKtw4m3avMlt27q10H7WS/s55essM2iS0cv8 ubki9YSZVQ9I++ns6vIEu+T38OFD7pwYVvsP7CvU5Xx5uSuj/GfPnvEzmfZIm9UZB21tre6oOGs6 xHkVq38bHBwS+Vx227Ztc1u2bPEFX9/T7Q1u+rd16zqkv23NF5BC7v/4qX90GNwbBzb62SGNMib5 /pw4fiL5/uyZ+/6wceFJacv09wMb43x/bsn35/yFC0Jvo/8GMcqP0+qUfAfpa6m/zNKIEeg7hodH 3f79+/ysA2h2yUyvw4cP+7a8f/9+X6fz5gX5UHZmX4T6Qas48hiooM7Gaj95l9XyNwTyRMBmQuSJ ruVtCOSAwJe+9CX3jne8wz366KP+79d+7dfcO9/5Tj+q1SilKYdiNjTL69dv+NFSpuGi+Gno7+/z 0/4x9lBE8g+zYuBe8SOTKDXhlNPNm7d4R8TVa1fzZ0MoMPLDdOkeUUAZZdeAArphQ7+Mit30mGl8 nr8Y/TjiNohiPDeNvcn19fXJtNg2PxNAJhHnyYLPm/Z2WZyByIXp4xq4p+4wpfv69esanevvpNBC PizVCZ1SrSKfHpmuPCIjqsgwRmCK8h0Z2e+T9qIOCOgy3R+HDIYejooY4YIYUMyMwSk1136axGE0 4EdWr0VqP5OTE+6q9Bt+Sn3BAUH5W1rWeoOK6eUsc4oRhmSGDs6zfmm36oCALry1i9OK9kXdzT8k s7xw7oZGG20a+eBopf+LFd70pv/T/c9f/Z/uzW9+s9CP48TMKhszDCakvuA4C78/PT2JfC5Em+0l s4guX/LLuhIHa7LMgBkrmwY2SfuZ8XU6qwz1jpu+O+2uyawHli9RTzUwUEBfy8zFoeEhjc71F0ci 9NLyWY98xKmJfhCn/eRaTMvcEMgdAXNC5A6xETAE6ocAHv8PfehD7id/8ifd7/3e77nf+Z3fca94 xSvc5z//effII4/49Yn1o3bv5jQ0lEzBbROFPAwYMSg9LIeIoWRMTEzK3gujXhENlVF4QtkhsCwh RkDpwnBkbX86MCPB7xUh6//zDhj+lBk8GHkKQxK31jsoJu7kv/yAKfQYl9BNy6dFeFuzpskvRwh5 zOsa+RCyRiXZz4RwS5auxAjsA0FYu7a0rmC8rBWsmMkyIW0o7zAtjkKWHVBP0vJhdB1jV+t13ryM jt4Wo21a2u38UeOk/cyKEyLO8p1hjDVpR22tpf0bGKwT/nDwsQdN3oH+bXx8zDti0vLR/QbYEyFW CB1msWjOpzPr99vhW5Nuy76fWdssDvApmamR//IDvnHsF0I7bhHnYRiQD/3boBjkMRyKOFHp+2kr c87EhCP2Y4CHWxG/hVCupB/EcviGMrFrQ2ClIWBOiJUmMeP3nkbgk5/8pMPwQQHgw8vI7xvf+Eb3 nOc8x6/Z/NjHPnZP41Ofws8WR4yZTp4OqizHWIOKrDFcsqbzwwfK2J2JO35EN81nve9Z409obi5V Rn3cmiRuXBTWvAN4TBT2N8CgDQPjdGsKcWNi3OQdUNJnZpJlIGkDBgUdI5cZMzFmzbD5I4FR5XTQ OjshdSXvMCvGwKTUW/Zmw+mQDvCCMcFMibwDe3PgHFsj7ThtuDQLb9J8fF2iTuUdxseTtpF2zEB3 bXPirMG5GSOMjyV00oYltJsLxibOgbwD/RvGo9bPkF7Yv83M5C+fkHYjr6em7vpvPPU1XWdpVIrV WKE+5ckrzkLahu/LUm05aT9N/vnUZP6zZthbhtCc0b9pPzMmzjP6lnxDdfqBOSHylYLlvjoQmK8h rI5yWSkMgVWHAIYM0/xOnDjhHnvssaJhw/rzb/3Wb/Xl/Y//+I8oI/SrDtygQDMyxVRHdlS5CR4X FcNJmSqcd9DZFll8JLST6bGTE/krgSjHhPmKMcZmwsdUhCUqyAbF2CvpKcVYIou8xJCPOheyMPH8 CT/KbyKv/P5VXlQWISWtP9QnnAR5hmnJf0YMATGhMuuKOmti1BVtP14+eByCoHyAh2IXPK775WQF Q01lpvzWnXgqQ20bikH4OKwrYXwe11rerPaT0Cv0K5Mxlr7lUcLa82TZwV35I2ThonXlbqE/rp1C 9W/4/oK2nOUQKbQn+re7smFk3kFnTjUVvjUhPeLgkXaM0zHPgH6g/YW2lZCeykzrdvjMrg0BQ6AU gflDWqXP7c4QMASWCQKMhv/oj/6onwHxQz/0Q8XRca8EFIw/ZkjkPxKwTACpgQ1VDKp5BSOqmoCy mHdQXsrxjx7IwM+UTM+VsxByZQfli5DNS2IsqPKcJyPU70TRlJHsTELKS/6KMW2PkI2JxMuzOX59 0nz+EUxmVPkuGAdZhOAFB8H8uRJZqRcXt5CTQ2U2E2H2wXRhBB35KF0tlY8rYKV1W5/l8Vuprmj9 idV+5niZX1LlJcbskGr7t7tSV+YvYpnP+2qIoR2rIa2yCMulNVkN4fBZva+n7+q3ECM/O3ffp0SY qVKsK/NaMv1swhz1Wrq3XIPysRCRxbcfKUsZrBeiac8NgZWGgDkhVprEjN+GIfDVr37V/fmf/7k3 /nEIoCDw0eNjg9eb+z/6oz/yv5ys8Id/+IfeAEnWTq7x15qe35/7uZ9zDz74YLE873//+91p2VEf 77p62EkHLTaiZOOul7zkJX7WA0sxNECb47IIz33uc72TQp/dy7+hAhdeL4QJSlWloPrBQukq5VH1 swV4UW1lwWRVEyyfsJryVpOmPIVqn1SWj+YyK8sk8g5FY2EBQjHko/twav3MZAlGcmZGXES+r4OP TF4K1kyMulItjaIDJxO0OkVWgbvWpzpRLJNNde2iWuzKEKkuekFMMmtQdXnXK9WCPNaLUJIPuCfY lym7egMi8IXDsnJIeGRJWt6hqvrou7d8eUnkU760KrWq+A2yUR0F8apTJXhsl4bAqkTAnBCrUqxW qDwQYI0fO8+zs/uTTz7pFQWcBXv27HFb5ei38HhCNpRimQS7xH/xi1/0jgp44hg/0vHbKzvWh4G1 nuzkzjngfMDIg3yf97znFTeo4kMVOiB4/xvf+IbnZ+/eve51r3tdmOU9fc16Vj3HvJapkWvQAioE VXHUUVQh6ZIfqWKiBub8DBNudIru/Of1i6mmvNWkWSpHc6dhVM4pJi9aJ8pxVJRjuQRLjafOFqpt RV5kj4qyQ5pL5aHwPgo05cWAyeSlYNjEkA9HglYTYvBSTb3N2s+jGv5rSeP5oL4sYGDG6FOK7SKz olCq5MECXXItxa85bVnWas6puheoi+BS1ohVuVVZt6ujmp1qrg4ICgAx79OYoBOj/SiNivIQ3Ip1 KrtIS45FP6hUH5W/OewWJomsR+R0HE7dwBGpM7gWftNSGAIrGwFzQqxs+Rn3ERH4ru/6LsffJTmS j2URbKrFxpDvfve73UMPPVTCCeed//Zv/7aPIy0zHAhvf/vb3cte9jJ/nf7nF3/xF93P/uzPure+ 9a3umJzV/ra3vc07FbI2JdR32cjsL//yL71j4jd+4zfczp079dE9/wtunZ2dXplTBaYaUPzGbGgZ ohiUVQQlI90grJo8F5tGZV9upFb1UU5iyDs0y67sBGbnpIOO4LYU0qSf1/MeJVPlmWXmqsyYgZR3 0DqgNEvpJdzBb9Y65tK0S78r8pIpn4KxIBs0KnZLp5idAxuDeiNKrJYs+ejoqvKbnUt9YrX9IJ+0 jIpxIp+1Eeqttp80H5RU21QMPqC3VjaX5YSFSrxkbUDLu/UMoXyy8tX+LWszz6z0ucSpVZlL5vMz pX3ijJqWfRYy5VMAJUZd8fLhWygBGEp8EAU+kv44zwVennxx+WkWJhpHn1KL8Z/kXNu/6Ae6l4rS zcpBN3jNepaOA0OOvV63rl1Oz1mHGzedxO4NgVWJQHXDBKuy6FYoQ2BxCDATAeOWoMswKuXErAcN fGwqBXaAvn79unvnO9/pXv/61xc/vFnv4AT58Ic/7C5fvuw+8IEP+BkTWenu1TiUOY6x5OO+EO4h RqRtLRiwahyEzzUufXxamKZe13q0IuexpwMKEMY//La15W9w69GcWv6QHzUsW+XYw7wDcsUoofxp XIhT/mLIp6VlblmWMFRSdKYosz+C5zfjRJGSxHW4aW1NVs1r+cMsdX1yi/BRS1sI86j2ml3z+cNq yVoSo/ylZ3RVm38t6dQRBc20wQAmiIwTVrJ23K+FTjVptbxa/vAdjdM04bM8ruf6lfn72uhJFDF4 KfJRxnGm/VtLS/5GbnmcS9t1+XT1eUKfwl/Sl82nrXUlRv/WJn0KI//QVLpaSj/TSRpQ0h/nLx++ 4wStn8pHEpe0b5wm1cw4Ct+t9drrB4Wjh9OYKC/8thf642rzZyCBIz+RawyndbV8WTpDIE8EzAmR J7qW96pEgA+dOiE4ppFp/+UCiu7NmzeLj5luVyl85CMfcQcOHHAvfelLKyXzyzuYAcEMi//1v/6X e/jhh30cyz/SynbFjOxhJgKdnR0+PmsZB3GMVKlSlJlBnSJbxOGFUsLGkzrbQLOenEjqHXUxb8UL ml2dXZ501oZoU4U20FlIozzm8ctoV0dHuzdwwSUM1H12jUdRZOlU3sEr6TK7ALrpDcvYeJG4dpFf jFF/RtEImXW2sHFtR6Fe54qLYL+ugH26rnj5iMwwXNQAzZMX2ijYg0m6X4Q34nDe6CyFPHnp6pL2 I9hkfS80Tr8refJB3us6E4NOj7oN6XHSDu0nBi/0b3xPOV0h3b/psY+x+rcQg0Ze823plPbj+5TU qRMYvd6hKPLR/jhPXnEqe/nIJsxpgxs+cLTSftoiOJ99+5HC6ilNYbm1z9Pvdvgsj+vO7rlBqHT+ Rf2gxu9Pyayx+b6nNBm7NwRWBQLmhFgVYrRCxEQgNHD44OCIKBf+9V//1R+pqc+Hh4f1ct7v8ePH 3Wc/+1n3Uz/1U279+vXznmsEysknP/lJ97Wvfc390i/9krvvvvv8o4sXL/qZEV5J0cT2uygENm4c 8Ip4WrYYLsxAQUmMMRKFAuj3DhGZj42V1rPR23Juuiij/f39iypjrS9RJ1vFoGZvlFKDbtbdunXb K6uV6m2t9Cql7+3t83ODJ1Jtz7dHWaKEwgp2eYe1MmOmu7PbK+iTkxMl5O4UnEQ9qb1fShLV8aZL nFHMQGCJVtpgGL895utzT4V+pY6suJ6eHp8dvIThrvSXtKF1MkOJupR3oK9myRw01VBRmtRjprz3 9KwXbPJXhXCIdK7rkLZyK2Vwz/pvCMsfentK9wlSXuv9Cx3KfHt0tCRrHAFj42O+/cRwstJGqSs4 7NL9G3zQv23YsKGEx9g3pX1dHOraZ9walT4+CMwCwGFFW29rz7/94CxkbyucROlvId9B6sumTQPC YeUZnkERFn1Jn87sHI7qpB8JA20ZXjf0x6krffL9of2kMQn1g8U7wc0DEcrWrlc3Avl/eVc3fla6 exABFKfwA4NSmRWYAfHHf/zHJUZiuZkQKDp/8Rd/4V75ylcWnQpZeRL3pS99yf3TP/2T3z8CA5U8 +fvmN7/pDZAYxlc53pZT/FKUR4zp7q5ud02WxoSG1MWLF7wBvnXb1ihFxYhiOQ/Tyq9cuVI0XnA0 XZFlON2imPWLkRUj4HTZIryg8IXOtOHhETGsRt327ds9nzF4ocyMBF69ek2MS52JNOsxQhmNuTfK ps2bpMiz7vq1G375BeVHPjeuX3MdYnSmN6DNCx/WIG+WvWior6OjI0Uyt6V/Ghu/7U/XwWkSI2Aw 6Ma8avzTHtkoVn7c1m3bRInP33ChrNuEFst32MtHR9sxemlPjLJv9vLLHxXaMHUFI/LGjbnZcaNi aNKeMObWdeR7zK6Wsre3xztnhoaGg/5tVpb2XfX9247tO6LIhzrA/kn8Jv1bYoBhZLIBNP3bxo1x DEuwOX/+vHv22WdkM+kvCf3rHq4nnviKbPz8TXfo0LPzHFmKZ71/cczwd/XqVW90a/6Xpc7iYNyx Y4dG5f67U2ixXOnypctCqyAfmc3EPcsdN2yI4wSnjvCN4ftzU2Z8akD/GhkZcQMbN7r2wmwwfZbX L/3Ghg190l6uBO3HSR2+7B2e6Ad8h2oK5nuoCS5LvDoQyH+oaHXgZKUwBIoIYOSHo0R8ALPCJz7x Ca8wvOY1r/FHe6KEl0t79OhRf4rGRz/6Ue/tz8qPuKeffto98sgjPs373vc+n4x88cCjxP3ET/xE uVctvgYEUHj27d/nzp0/J5uEHhdjpcMbD2DNchkMrFiB0R9ocmrKoUOHfd2buDPhFcBdshFpLMOS 8mIoYWCfO3tOjG5R0gWniYk73uiPaSyw+eGevXvEaDgnx9MedR2iFE4x86BJ5CNyC52EecsJ+eze tdtdEYPh1KlTfnrynTvjMprc7TYObIxizGkZGfXHSLly5aobHhrxtOFl06bNMkoYx1iAF9oPxstF cZSdFkzoL5lGzQare/bs9nVXec77F/ncd/CgO3vurDt8+Iinffv2be+AUAdF3jxo/ozqIx+Ml5vi iMBxhHy2bt1SNMY1bb6/TV4O586dc3x7cBrxDeFv/759rrMrmW6eLw9J7tSNffv2Slu+4I+aZonT uGBCG94hRmeMWSpazq997evirLroDUgcmfxh9P7nF77oZwMkjtZklo++k8cv7Wfv3j3ujPT5x0+c EGemtB+RzYzsDbR//34vrzzoZuXJ9+XAgf0OB/yzzx7y7efO+B1v8OMMYflIrMCsP74/fHtw3LHU irbMKWKb5AhzcIsRoLNL+nz8Bqof4Gxlj6L7Dt5XcSZrDP6MhiGwUhAwJ8RKkZTxuWwQ4MMXGjmj qSmtMIpi9/GPf9z91m/9lj/pgndQ8PTIyLAwfETf+973+mUYlaaeMoLGzApGR8qFBx54oNwji68R AZxNe+Q41XFRuFifvEbW/iN3ZBk7MPJyUAwpRrmpRxhWKO+xlK658jZ5gwllUGeIwAf8xA7Q3Lt3 nzcS2AeCdf3w0hD5iBG3W0YFJ2RJxl1ZP41Dpk3uY8sHehtlRJCZPMwQYWPItvYt0WaohHUAR9F2 mYXAtG0UdJaKgEkj5MNeGAcPHvBteVrkw+h7u0xnj2ngKjYDYiwhH5bvzEwjn7aoThnlg5HaXbt2 +XbM/jL0b7SfRsyk6+5eL/Lp8NPbMTI3t25uSP/2gz/4AwpPw3+ZvbNP+rdxWYZH+2mkfHBS7d9P +xl3tB82KW7E94f+bdOmTX6WyITUWWY24QhhllHsvpb2s3fPHvn+FOQj941qPw2vrMaAIbBIBMwJ sUjg7LV7FwGUaIxCDWknBEoUDogXv/jF/sQKTq/gHYzHrOUYn/70p/3H/Qd+oLIChNH1nve8x6dV 2ulfFFwL9UMAI8U7nPLf43BBpjEOUAYbHziNg83I8l+TvFBZUTzDtrhQ+jyfM6rdIX/LIdBXNMIx lC478mHKNn+NDuy5sDzaj1tW7QfDib9GB/q3mDPMGl3eaujTfqJsJlsFM+gw1n7SQMXZADlN1e4N gdWCwPLQmFYLmlaOewaBcAO+tBOCvRmeeuop9653vct76DGSdPQvXEsPWCyhePzxx/0siGqMKaZb 82fBEDAEDAFDwBAwBAwBQ8AQMAQMgZWIQI07p6zEIhrPhkD9EQg3mws3pmTaJEdnPv/5z5d1rvs8 YUYPdJMipibrcWzsL/D3f//3fuO0l7/85fVn0nI0BAwBQ8AQMAQMAUPAEDAEDAFDYJkhYDMhlplA jJ2VgUDohAhnQnzuc5+TDZwuure+9a3F2Q84IXQmBEsy2OiKqdJsNMhRm7/5m7+5LKa254U8G7Fl 7YVRDT2m6NZz5kdTc5tsJlV6LF01fFgaQ8AQMAQMAUPAEDAEDAFDYDEINK2Jv3fWYviM+Y45IWKi bbRWDQJszKeBjSVxLvD3gQ98wL3lLW8pzoIgDUs3dLMv9ovACYET4yMf+Yj7zu/8TvfCF75Qs1qV v1/72tfc7/7u7y5YNta/8qeB692yMeTv//7vF2eS6LNqftN5+XeaW11TS4ebnZIz6C0YAoaAIWAI GAKGgCGwbBAQHSjQg5YNW8bIkhBA73RrG7+P1pIKkcPL5oTIAVTLcvUjEM6EYIkFJwX89V//tV9a 8YpXvKIEADbbCmdCkPZLX/qSHBd32L3//e9flIFdQmCZ37BB58c+9rHoXIZOCF0OAxNN7b2uaW27 nBwwFZ0nI7g8EWiavL08GTOuigg0tcU7FrdI1C5WHALLpS03tXW5ptblsJHwihPhPcdwk2t1Y1NN bmR8xq1hD98WqTvW362aesAMXCcDYBbmI2BOiPmYWIwhsCAC4caU7PFw5swZ97d/+7d+M8r0Dt9s OKkzIdgHgs0p/+zP/sx9z/d8jz9eakFiOSeAp3//9393mzdvdvfff3/O1OJlz3FibPxJ4LokiBNC 9u0vibKbexeBplZborPcpd/Uak6I5S6j5cDfcmnLOCCszi6HGrECeJhpcacvXnfHzlx0/aNyDHeT HB9s/d0KEJyxuFQEbGPKpSJo79+TCIQzIdgT4k/+5E/cfffd517ykpfMw4NZEHoEGks2PvWpT7mb N2+6V7/61fPSNiLi/Pnz7p3vfKf7/Oc/3wjyudHkWECOLN24ceOyOK4wt4JaxoaAIWAIGAKGgCGw YhFgiS+6Cr86aLViC2OMGwJVImAzIaoEypIZAiECzIRgir9uusixnB/84AdL9jQI0+vsCJZifOIT n3C/8Ru/4bZv3x4madj1f/3XfzmWlOzZsycXHtiA87Of/eyi8mYWyetf//pFL1kJl2EsigF7yRAw BAwBQ8AQMAQMgRwRQFdhwEqX7uZIyrI2BJYNAuaEWDaiMEZWEgLsN8CpFyMjI94R8WM/9mPuoYce KluEnp6e4jOO73zta19b1mFRTBjhAufDV7/6VcesgYcffjgXijgS+FtMwNkT7u2wmDzsHUPAEDAE DAFDwBAwBJYrAiyLtWAI3GsImBPiXpO4lbduCDC7AScEyzBe85rXVMxXZ0Lw+zM/8zNVG9bk/y// 8i/eiOckDZwFHAH61FNPuaGhIXfw4EH3vOc9r+z0PWYhHDt2zB+RyUkTz33uc11HR4e7dOmSO3ny pD+p48knn3Q4SQ4dOuSuXr3q94UgTb3Chg0bHE6alRpYQsOpJoxUME2yUU4RlBT2HyEwWtLIKZuK Cbxw3GwjMYEXZiTpKFLjeJmRE3KmHXJqXiMjWmtZ7Sg7nTcggAe4gAW4NG5G0KzHhPbTzEjf2mbh qTGrQJFLWFca2X7AA14IjWw/0J+amvLtx/o30EgC8pmemZbW29j2Q531vCyT70/YfuhXGtXX0r9R bwnL5VvY6PaT1Fz71xBYWQiYE2Jlycu4XUYIMBOCD8/3f//3L7jBJEY+H2w2o6zlSM4//dM/9ftJ fOYzn/GbR37Xd32X+8IXvuCdD2yG+Vd/9VfuzW9+s/vhH/7hEmRu3brl/u7v/s597nOfcw888IBf +sGSiK985SvuF37hF9wzzzzj96bAyXHhwgW3adMmf8/mlHv37vWOipIMM25wjuDU4N3lGFDglhJQ /nDiDA4OehlwT9i1a5c/dnUpedf6Lse64jRCCcQRxUabzBLBsRTbmGKzTxxhLS0tvk7jGGFpUex6 gBLKfibXr1/3s5LAhjZG/V3szJta5aLpx0U+56Qd3Z2669raWsRZNOW6ZPbPNsEFJTlWoM7fuHHD b8iKcUv/xGynTQOb3IaN/YJPPAcA8rh8+YpsxDvk6+yMGHU0yW3btnl5xcIEOrSfU6dOecOF9sOy OOoIbbmtLe6xadeuXfP1lvZD3aAtb9myxW3dutXX31i4YMjRv7E/EZjQv1F/6FPCmXsx+OGYa5UP +yeBCTwgn9j9G3jgpG9Z2+LWNK9xU9KWN8l3sb+/L6p89PtDe6aucs/f/v37o7cf5ENdod0gH/p8 6gx9LfU4ZmCghO8PbQfa1BX2fqJfob+LFZDFuXPnfPsBE3X87pFlrbHbT6wyGx1DoN4ImBOi3oha fvcMAjt37vTK43d/93cvqJyQlpMnfuRHfqTqD6Uae3/wB3/gvvzlL/t9Ffj4/8qv/Io3+DC+nnji Cfdv//ZvJU4IPsq8g7Pid3/3d92LXvQizx/7Vvzar/2ad0LgDOHvb/7mb9w3vvEN95a3vMW97nWv q0l23/Zt31bccLOmF1dAYpRxjH4cEDhxmMGCknHixAl/tCpLb3BCxQjInFkqKDrQRfHCsHz22Wfd 0aNHfb2KZehSJ48fP+4OHDjgFT+MfowqZtsQYjkikAXywYn24IMPeiUdoxdDBlxYWlTP2TyV5IzD 4bjUC+jtPXjAG00TdybcsePHvNJ+QGYrgVOMQH3FMYNCziZnkL15c9CdOn3azczORJOPc7OeD04C oq50dKwTA2rGnRPejh8/IXX2vmh9B+2HOkH7oa7gnNH2Q12mX45l6CIf2soeMVRoK7RbjEzi6HNi 7ROkDojLly/7/g2jCfo4tulraD86ey/vesv3iuOqkQ9ObTDB2IUPfpFZrPYzJA6z06fPSPvZ6uUD XdoPuNC/bN4cx+Gu3x8cInx/cDhj9MIHsyCf85znRJMP5T5y5IhvI9QLvj849YhDRvASy/inrfAN pp3Qx1FXwIjvIHUap1WMgHxwyuAMoc6iCxB3WvpZ+hqNi8GL0TAEVjIC8dyGKxkl490QyEDg3e9+ tzf2d+zYkfG0NAoDn2M5USiqDf/0T//kvv3bv91x+gaGHidy/PIv/3LRkGAkGCUJ5SQMf/EXf+F4 98d//Mfdi1/8Yp8GY4A4ZlJo4H0UGvL4ju/4Do2u+hclNZbyXjVTdUqIcYvBzcwQ3ZcChYdRQgLK IEpHjIChgNIHbZ26zSgUihhKGaN2MQKGG4oXcseAQvGk7nANRjwjTYxAuRkRQz4ogPCBcoyzD6UZ +cQKly5d9CODjGTD45PdLgAAQABJREFUA7y0r0tOZhmWenRz8GYUVlDCGZlraWn1DqJEPmtcX1+f 61nf7Z0CkxPJcp68Gbp167a7LjLCEcIoLjMw6Cu2CUaEc+fOy79x2g/1EkOX9sKsBy+fQvthSRvt PEZAPhgptGHqCnjACxjRfpAdI84xAjPlMKAYQea7Ah/UF9oP/Rq8pL8refFFW8XZoLO64AWHBEYm 7ZxvX4wwPS39xumzXj5h/7ZhQ7/Ip1v62YuCSbKEJm9++P7Q71NPcBCBCfVFdQ3qUaxAXaD9YODr 94c2DS/wST8cI1AvcRrCA7S1/XCaBEs++Q6iK8UI0GEGKXWUtkvbUf2A35j6QYzyGg1DIC8EzAmR F7KW76pHAEOQj081gQ8TBkot4Q1veIOf4YDnn489e0JgdGngo4sCHW4oicL26U9/2vPFSPHv/d7v uXe84x3u13/9173C+ba3vU1f98odH3WUP5RRC3MIgC1KT3o0ECOGP4wXRljzDjiKUPKoZ+mRfYxL AiNBMQJlxsmQGJWlI/soyiiq1NMYQTFJywcFFQMGTHBG5B2Qz+DgkFeI26VehEFxunk9jnzAHn66 ujD65+RD3enuXu9HCodHhkMWc7tmlpY0INeZ2luGttPSstbdujXqJu7k335ow/SJ1AvkEQbqDthg 6JIu74B8cDKoURnSw5BBdvASIyAfnAzp9oNhRxwzNnAM5B0oM7hQL9L9m840i+UkGh29JTKYdN2F ZZZh2elT6E9u3IjTltWwD48Chx/qMXoHgwox5IPjjLqi/WqIiTp/cZbECDjO+Oaq0a806eto2zyL 9S2EDn0GbTkMtB/VD2I5FEP6dm0IrDQEbDnGSpOY8XvPIICyQWDaI8YfsxrCwP4OKNFM/dOA8sIH EscCm2WiPPHR5i99/jTTthnlYFnGYgKKEB9/PryrLVA2QpbjiDiUM2YnoHDkGVA0Me6hExqW0EQx JKCcoRCln/uHdfwHTKCTJW/FCYMib4cWxhPYU16lq8UkDv6QD6NV6qjR5/X+RT6MjraILJqkLYYB xyPtc/zOuOeH6zzD6GjiAMqSj8ahGOctH+qIH5EUWTRn9A3wQp0eG5f2055v+6GeYDxi4KbbB3wg EzAhTbou1VtWYAI22m7D/DWOthwj4FAkKN2QpvZv8JJ23ITp6nFN++HbhuMjSz7QUNzSz+tBP8xD jca14iRLB63H8MIsiTwD/RvYU950n0Ec/Yq2MdUR8uKHdopxjx4B3TDAG3/IkDRZdSlMv9RrHGME 7cvC/DQu1reQ7xwhq8/Q9kMadaSFvNq1IWAIzCEwv7ede2ZXhoAhsAwQ+M///E+vpLEJlAY+/GxW iSeeIz/5QKMIoEyj1OGEeOlLX6rJM3+ZBcE7rOkk8C4zAHSqcOZLQeQXv/hF933f931BzOq4RMFj hI6QVgLDOGSQd0CWhLQCqHRRSlFaSZe3QkpdIVTCRPlV/vL4pbzIp5ySrvzF4IX9IKgva0QO8BMG +GAZgvKbt8NK5ZNVVzQuwYRR/1JeQ76Xek152f8BCmlMyJtTMggTk/kv3dE6kMiitMwaR78Xwwmh /YXWTw9C4R+NU37DZ3lcKx2tFyENjVN+w2f1vlY+tPxh/tQd/rT95GvkcvJQUh/XpIxteNI6q20s 5LPe1/Rt5fo3z0uBvxjyobw4dFUWYVlVZsiHdPnKx3nHJfSVbpoXeKQ+0R9znVcg/0r1VvnTNHnx YfkaAqsBgXyHZlYDQlYGQ6CBCKCMsG8Daw9Z96iBzY+YZvzyl7/cj1axDwTLNnBKYOww8yEdUBbY yJKPKOHpp5/2nnpdzvG1r33NffKTn0y/Vvb+Fa94RaZCUPaFFfIAg0QxymJZFZxYCik8KM00Pxof Q+GhLhJUyQp5UT40Tfis3tfIhroMTaUb0tC4GPJhFgRBaYZ8JPGsSpj1inz6Wb3vMfx9yFDAlT9w K6arNwOF/DBaKu33oDNGpu+W7mWTBzu0ZUKlOstzTcd1XkFpqCxCOspfjDoLXeoBoRIvMdqy0sji I+RP03mmc/iHT+LdQn3EoZgOyt9UoT6ln9fzPmmjSf9WKd+8MYG20qD8ioHypPe0d61P+iyPX20b SjekoXHwkfQ/4dP6XtOOK9FQXhS7+lK33AyB1YWAOSFWlzytNKsMAXZNZwrofffdVzK1XDdbY5+I 07JJFSdfvOQlL/EbNuFU4L1QMWAq/Yc//GF/bKd+QFm2wTTlffv2eRqPP/643whTpzYuBGXeI7sL 0V/ouSoDpAuvF3pP8Sn3nuZVyVGxEI1qny9EQ3kJZV1t3rWmC3Ep924MPkJMtPxZ/FTDb9Z7tcTN FBx6lfhInBD57jkQ0phvQs2ViHQhfnNP6nc1R2O+4RJSiSIf7xAJqc5dhzLLGxOoVlPeatLMlWBx V3PyqdwvxuKlUilURnn3KyEmlfiRxhOx/WRzopjEkE9IQ+mGXGlcmC58Xs/ramhUk2apPEFD+wst f5inxmma8Fmla32vUhp7ZgisNgRsOcZqk6iVZ1UhwJGDTI9NL3tgHwjWpn7sYx/zo3wcvcloGn/v ete73G//9m+7n/u5n3Pf8i3f4pdqsD7xe7/3e32cjrq9/vWvd+973/v85pXsD/HTP/3T7oUvfOGq wU/LSYHC64UKGCoDWYpEVtxCeeb9vJbyLZaXEJdyecTgoxztdHw1/KbfqfVeDf6CL6LM6xjjZR7V KZr8q6ORMyNSHiltVaWKIZ+qGIHn6sCrNrvMdNXQqCZNZuY1RFZLo9p0NZBedNLl1K8suhB1fjGG fEIafPfCe4qj38J0fJ2L6rOrhkY1aZbKGzSUjpY/zDMrLnxe7lrztLpeDiGLX40ImBNiNUrVyrRq EODozA9+8IPuBS94QUmZ9u7d6z70oQ/5PRw4Vo3lGhpwTnAqBssz2KiJPR50Y8rwA/fd3/3d7v77 7/dHK5If6VZT4KNeaQpnubKGGGWlUSWj2hkjWXlUG6c0lGb6PY3XdOnn9bzXteJKM8xb4zRN+Kze 16ESWGnkKwYvc3Ule6YDzgns2zWyN0S+YW4TO5VFSE+dJEw1b27Ol5emNaqkz40YlvKSYJW19j5M V4/rSnUgxGlOjvWgmp2H8hLS1ZQap2k0Pq9fykvbyWo/MXnRfktppsur8XnjQp+i7UJplvBSaN7w ocZiyfM63iAblU9WtspfjDpbST7KBzzG4KVSHVBe4COGfCrRUF4q8ZuWK/kx2wd9heUelfJPv2v3 hsBKRsCcECtZesb7qkeAPR5e9KIXzSsnH1vO7eYvK7Ar8/Oe97ysRyVxnLetZ4+XPFgFNyjYullW LdN5UbxQIHhHFYosODTvrGf1ilMaWcYCNJS/vDelhFYlXpQPTUP6vAJ1HxkpzZAOcYpVDEx0d3Sl meaFvRHgt3lt6c7yYbp6XbcWjgDO4mVmJtkDgJ3/2Swzz0Dbocx3y7Qf5a9VZJh30PoIzXR90TgU fjW28uRH60qaD2gqJpomTz7IG1zY2LDRvKh8sviAT43XdMTlEXwdqNR+ZpP9VmLIBxrUR91kMV3e WJhAF17ABppKV/nRe9p7jPbDCR2ErG+58gcf9D15Bmiog0ExyKJXy3JV8iFP6jm/lfLNomVxhsBK RSDf1rpSUTG+DQFDYMUjEH7Iw+tqCqZH02VtLsVoBYpZ+lz7avKtNQ2GNEqJzugI34c3yoWyE0MJ 1PJmYaJxmibks97XlBWFFMMtjQt46AaAKsN60w/zQz4ovTMzc84PfY6yzCaQa9e2iDKfv8HdXlDS VRbKB7+KiSry4bN6X4OHV8BFFtPzNvIrnDwj7Uf5rTf9MD/qAG2V8qf7AHCiDiHDGO1H62O6zsKv xsU60k95ydrQNiYvYI98ytVZ7d/U6AtlW+/rjnUdPkttK2H+Ghej/WD4Q4eypw3usH+LUVfgA+yT vqx0I1ni4Ie2U4vBHeJayzXHuBLSmBCn8uH7Q33KOyj2WfWW9gMPtdYVsCTob95lsPwNgeWAgDkh loMUjAdDwBBYVgj09/d7ftJKOgoCSgZKVwyDG4W0r68v0+AeGxvzPPI8Rujt7fUGd9bRcMSheDFz J+8AHXhBFmn5YFRytj0jShg4eYe2toTOrIyU3p1KTmJQmiioxPf0cFJN/oqxV9IFm4mJyXmKLJgQ ent6/W/e/+jpPHcKR8wqPZwyU4JTS8GRpPF5/WIggQttNm28aBwGeQwjF0zgR9ttWGbqsdbrMD6v a22nWi9COvBH+1FDK3xW72v6N+hkORSVN+2L6007nZ/WA6UbPqeuEGL2tfRv6b5WcaIeaRsL+az3 NfLhO0dflja4uadNgUms9kMboa2AQxgUJ74LMYLSSX9/Qv1AnSbV8sO7FgyBew0Bc0LcaxK38hoC hsCCCGzZssU7GjgGNVQObty44RXDzZs35z7tEyZR7qCF8gUvYbh8+bI3auA1RkDpReHktJZQ+UJB h7eNGzd6gy8GL+x7goJ8/fr1Evlw4gujYuyTAmZ5B5Y2sJcKNIeGh0rI3bx5oyi/kgc53XR0rHN9 ooTfulUqH4wFTsfB2Ovs6syJemm2yKe1pdUhjzDAx1051hTMYhgu0Ga5GW2YuhKGK1eueB62b98e Rud2zcgouLBJcOiIoO4MDg56x5oaN7kxUcgYPjAu6c9Cgw750LbZY4j2lXegDtB/aR8S0oM3jO1Y 8gGPHnHS3RYnzJ3xO0VWfNseGvZ9W1ek9jMwMODLfunSpXnyAatY/RsgQAtnQ/j9oT1xTx8bazkn jn++MUNDQ36vKxUQ8qGu8H0KjzHX53n84hjDaXXx4sWS7w+84RDhm533EqI8ymV5GgKxEWh+t4TY RI2eIWAIGAJ5I/DEE0+4b3zjG57MK1/5Snfw4MGqSaIcM5J+9epVP7LOPQoGR6Oi6LCRZwwjF4Yx XlACUXigyTUOCBQv+Ig1UghtFD1wQAFFKbx9+7Y7LUfEonDt378/iuECJhhIyAdDEqOJJQAYmefO nfMKIHul1CKf2clRsl1UQD4Y1qqUY9TB1+jIqNu9e3eUEeWE8SavGN8aveVlBEYoxOfPX/D4IB+M uhgBebTKLBGM2nEx6tiEkmuMqn5xZG3fzka6tTmJmtqS6di18o98MJqQCQH50JZwkHA8caz2A23a DzhQV1U+tB944hjmGNPa4YP+DKcU/RtORXjBEUL/hiOE9oMMYwSMOZUJ9ROn2YULF3wdpn+r1TGz lLbM6DWbOVM34IW+5cyZM4LPWt+Wwana0LS2zTU1t1WbvCSdyod+npkZyIJ6c/LkSe/IwjEQSz60 H+QDL/SpKivq8IEDB2qWT0lBa7xhBg/1lW+fth8wIfB9j9l+cFqpfoAscC7SluGRvraW7w+OlI9/ /OO+DfJde8Mb3lDzco4aobTkhkDDEXiPhCb5ONscoIaLwhgwBAyBeiPw6KOPuscee8xn+8gjj7hX vepVNZPAyEbZYuQSxZCZAIzG1KJg1Ew04wW6aRRjDAWUdB0VwqiJHRiJAxOULnBAGWM0sxYFvR48 gwnywfjH2MZowEGEjGqVz8zoxSWxhGIOHhgKKJRt7W1uQ/+GhiiSd+9OiZJ+040JNkxa7uzs8LzE lg+A0m6ot4mjqNn1rO92/SKjxYQ13XMnANX6PvLBeQYvtB+cZswGqHXKdK10s9JDn/aDMQVftB9m hjRCPhi4tB/khCFFfwIusQxcxYc2Q9vR/g1DjNkAi1kSstS2TP82JLzcFmcEy6k6OrqEl401z9zB adbUujjHmeIS9m/IBIcZfVyt/Zvmt9hf7d+oK75/a0tmJTTq+4MTgv6Wb4B+fxox8yDUD5APDjPa T63y4fv1pje9yTuZyOPxxx/337HFysveMwRWAgLSTpriDIusBDSMR0PAEDAEUggwSsdfowNKDcon f40OKHtM1w6PhW0ET2CCkbIYQ6Xe/KoCWuuobb35ID82wmQ68HIIGAj8NTqoARdz1kO5MuNswOnA X6MDo9zlTliKyZs6EJdL/7ZJHCDieYgJQSat5fL9WU79G9+f5dJ+lot8MiuPRRoCKwCBOPPtVgAQ xqIhYAgYAoaAIWAIGAKGgCFgCBgChoAhYAjki4A5IfLF13I3BAwBQ8AQMAQMAUPAEDAEDAFDwBAw BAyBAgLmhLCqYAgYAoaAIWAIGAKGgCFgCBgChoAhYAgYAlEQMCdEFJiNiCFgCBgChoAhYAgYAoaA IWAIGAKGgCFgCJgTwuqAIWAIrEoE2LhQQ3itcfZrCBgChoAhYAgYAoaAIWAIGALxETAnRHzMjaIh YAhEQCB0PLC7twVDwBAwBAwBQ8AQMASWGwKqr5iustwkY/zkiYAd0Zknupa3IWAINAyB8fFxd+XK FU9/YmKiYXwYYUPAEDAEDAFDwBAwBLIQmJmZcdevX/f6yuTkpLt7925WMoszBFYdAuaEWHUitQIZ AoYACKxbt85t3rzZg8E59BYMAUPAEDAEDAFDwBBYTggw+2FgYMANDw+73t5eZ/rKcpKO8ZInAjZH OU90LW9DwBBoGAKzs7MNo22EDQFDwBAwBAwBQ8AQqAYB01eqQcnSrDYEzAmx2iRq5TEEDAFDwBAw BAwBQ8AQMAQMgRWBgDkhVoSYjMk6I2BzlOsMqGVnCBgCqwcBFINbt2459pdobW11XV1dDZsqyTrR kZERv160o6PDdXZ2Ot3MKjbi4AEuTCMFk7a2ttgseHqhfOABTBo1lXV6elowue2mRU7t7e2uo3Od 8Dh3QktMgNgD5fbt275+UFcaJR/nZoWPMXfnzh3X0tLiukQ+a5qbY0JRpEX7GR0ddVNTU245tB/k Q+ju7m6YfML2Q/9G+0FOjQjav9GOWErXyP6N9jM2NubbD7w0qv0gH+oJvDT6+8O+BSwXQE7LQT58 C/n+8f2hv21UoE9ZDvpBo8pvdA2BpSBgToiloGfvGgKGwLJFAAVuKQHD6fTp014BZJ0mxsuJEyfc rl273MaNG6M6AAYHB93Zs2e9gY0BxYabOAD2798fVQFDET1z5oxXRtVIgC/w2LFjR1RMUMxPnTrl lWIMuZs3bzqMB+TT19e3FNHX/O7Q0JC7cP6Ca2tv8wbL9evXvDG3des2174unoI8K/K5dPmyx6Kr u8s1r2l2ly5dcj1Sf7du2Sx1Jp4DgA3Wzp8/72XSKYYCzpkLFy64zZs2u/4NyCeeg4b2Q73FwA7b z+7du70RU7PAF/kC7QcMrl27VnSYwVd/f7/buXNnVAca8jl58qSXD4acOmngQ/fSWWQxa36NTfnA BQci8rksdZjrvXv3Ru3fZmdn3MWLlxztGT6axWF28cJF19ff57ZI+2lqijd5mL7s+PHj/ruzfv16 Lx/6u+3bt7tNmzbVjPFSXsD5AG1tP1evXvXfH9oPOMUMFy9e9PWDOgs/586d8/soxG4/6AO0H5xE PT09DucZ8uL7w/4OjRogiCkLo2UILBUBc0IsFUF73xAwBFYdAjgwjhw54hXA5z//+V4hJg6jCkUD BQPDO0ZglOXZZ5/1is3Bgwc9bYyZp556ysc//PDD0UbqcDhgLDz3uc/1ihflZ0TqySefdOCDIhhD +aL8x44d84of8tFjzXAagRWY4DiKESYmJsU5ddJt2jzgdu7Y6UnCH/Xn5KmT7sEHHxBM4hgvl8U5 hdMB55SWH2fNoUOH3ez0jNuxc0cMSITGrDt18pQbF0fec57zcLH9XL502Z06fco1t6x1PWJYxQg4 Ew8fPuzb64EDB3z9xGA4dOiQj6cuxxrpxtimjj700ENuw4YNvvjIh/ZDndm3b1/U9sMo7gte8IJi +TH+6d8wvmP1b/BAW8Fwu++++zwmOESeeeYZ35aRT6zZGdev3/Dth34Ww5IwJjN5Dhe+BRiYMQJ9 KfUTQ5f+TcuPAX706FE/K0Lbd9780H6efvppX1/vv/9+Xz/hCz6QEfUn1uwzyo/hj3zUUYYTgG8h TjX4ixGQDwMSOKte9KIXzfVv0n7Ahe9RrPYTo7xGwxDIC4E4mlFe3Fu+hoAhYAjkgMCNGzf81G0U HVWwMK63bt3qFUIMcRSRvAM0ULwwUEIDHyWHexQwDIcYAWOJETCUX0bmNHCNQQWfjN7FCJQZ58eW LVuKDgjo6j3Okljh3LmzwkOTG9g4UCSJfDCqwOzatevF+Dwv7ophcEmMfEYI1YCCHiOVvb097srV K27iThz5DA4Ou9Hbt8RQ2FTSfgY2DYiBu1ZGl+O0H8qP0Y/TYdu2bUUDHyObtoyBFav9YLgxasus HWY+aEA+zNyBD5Y4xQjMGsKAon8LHTCMsNPf0b+BWd6B/o22ijzozzTAA/Khf6PPiRGmpial3Bdl hkqX9G/dRZIdnR3SntbLEYo33JQYujGCfn+Y9aAOCOjSp3CPIR4rMFOHbx+8qIMZHpAP/X0s+Whd of2AgwZm5GHwgxnfhBiBtoNDEUxC/YDvD0tDYrWfGGU1GoZAngiYEyJPdC1vQ8AQWJEIYBCg9GDQ hQGFA6Wd0bsYBjcjgkzdRklnHW4Y1BGAQREjoHhRZl2GEdJEMcTIIk2MgMKJoQ8vYUA5xqhj+jAj Y3kH8KAuwAtrtsOAvJgBAa8xwpCUeWYmWU+vxoLSBSfq8+BwHPncvCllFnodHaV1lvbTKjJidg8O gLwDzjvwp82m5QMmtCvaD+nyDtRJnFK027R86GfgAcMmRtAy027DQD0mDmMOGeUd6N+ghWzS/Zu2 7VhGLvu53L07JX1Kh2+3YdnhjbZ17UYc+Wg9KPf9wTkTw2GFfFjKhHzS7Qdjm/qi38oQrzyu+bbw jaFeQDcM9Pk4zWL1tdDJ0g9o16ofxJBPiIFdGwIrEYHSlrwSS2A8GwKGgCFQZwRUgdBRjjB7DBcU EAyKvAOGAIpXFh8aBx/wk3fA2CakFUDiwISAcpx3QDEGF/hQukoTJVDjYvCCowPlt0l4SRuW8McM CU2jPOb1O1bAXssf0tHR1PEIdbbYNkQWWXtQNK9t9gb32Hic9oNxjyyy5bPGO0OoU3kHrY9Z8gnb ct58kH+l/k150TR58kM7BvusPkX5iNW/qXyUblhulRlLM/IOYf+WxoU6rHExvj84ClU+ioGWX9sU jtgYDl+d5ZDmA340jjob41u4UF2BhxjtR2Vhv4bASkXAnBArVXLGtyFgCOSCAEYLhj9BFb6QkMbF GMlVGqpkhXwof/AbY9RSZ35k8aKYxOCjknxCJ4Ril8asnveTsh+El4PMeEgbudwzEwJ+707lb+RO TiZ1tpJ87shyjLyV9Om70+KYmfHbTqYxASvlb3Ii4dcDmNM/WgegmeZF4zCyYiw90PajbSUssmKi /IbP8rhWOko3pKFxmiZ8Vu9rpZGFCbSIp/3kbeTSJirJR51p7P+Sd+Dbo+XNwkXlo/zmyQ98gA18 pNuP8kbb0e9lnryo00XphrSUP/jIu3+jPir25XiBN00T8mnXhoAhUIqAOSFK8bA7Q8AQuMcRWMgg UWUshuKlvCjNtGg0vtG8KB/Kb5rPet6jZKIIQlPpZuUfBROZok0QVjID8fA7XUiXmahOkUojCxON Azf+8gwzcsIApwyUC8rL9HT+jhmtj0qzHE84IvIOlXhR/mLwQTmXCy+V+IDPmLhou1Ca0Neg7Vv5 1fg8frWNZvEBPY2P0b+F9VHpapn13vdvEfYPUV6UrvIRYoJ8VI7h83peQ6OSo0P5U37rSdvyMgRW GwLmhFhtErXyGAKGgEdAlQFuwuuF4AmVmKz3NC5Mt1Cei31eSdkJ84ylHIc0s65j86GyyOIlhnyU RiU+kKH8n2tIaFRDBF6qSbd4djV/MFkIl8VTqe7NSvUx5E3lWF2ui0tVDY1K/C6O6vy3Qj5CDNIp w3TpZ/W617qyUH5541LSfso4FBMe47SfanCJIZ+QRlZd0bgw3UKyXOzzaupADD6gofLR8odl0rha edH39DfM064NgdWKgDkhVqtkrVyGwD2OgE5bBYasdb7l4AmVAFU2wrQalzUVM0xXj+uQl0r5hWWt lG4pz6rhJTYmKousclXDb9Z7tcQpjcp2PcZ4LbnWnhY+lJdKb1eTptL71TxTGgkm5R0emq6aPBeb phKNsO7ErrflyhODj5BGiEGapzBd+lm97ivJJ6QRp38rUCxfZf2Up3upLYd1IKuuaFyYLpRbPa+r oVFtfVoKX9BQOlr+MD+N0zThs0rXmp5y6nWl9PbMEFgNCKxdDYWwMhgChoAhkIWArq2tZVRiIYVX lYyF0mXxU2uc0lCa6fc1vhYnSzqPau+VhtIM39M45Td8Vu9rFDT+oKl0s2joZoxZz+oVp2vFy/GB IQ6v1SjQS+VJsc/iRePYo6K5Od+xh0Q+0Lgr8plfKuVF+Z2fon4xlepsSGW5yCcGJpQbOowsqyxC LDQuBi9KQ2mGfHCt8Zou/bxe92EbVZph3lqPm8VAlBYdPqr7tRqh5b5Zyp/W7bozEGSouCvN4FFR Nkl7zxcT6Gp5K/ECv3m3ZWhQ5nJB+avl+0N+yBt9RffhKJe/xRsCqwmBfLWR1YSUlcUQMARWFAJ8 1PVosXIKXVaBUDJU4cl6T+M4oizvoDSypqKi7PCHApM+3i4PvvSItixeFBOOJ8s7IB9V8JSu0gQP 5S8GL21tybGc7IOQDvDG3gickLFWeM47tLYkvOjeECG9GdkoktDSslbqS76ffdoOjg78D1l7Q6h8 tD55xnL6R+tAIotSjwh8UF/C9p4TGz7bkJc0HcVE23v6eb3vlY7SDfPXuBjyUT7S7Rh+wv5NsQv5 rOc1fWhra9J3ZfOStJ+21DG89eRB86JvU+zBIB2UP02Tfl7Pe3AHG/hQupq/8obRr/2xPsvjl2M4 CWk+NA5+6HvgN89Q6fujvPBbCybwDo7IlPwtGAL3CgL5aiP3CopWTkPAEFh2CKiSBGPhdTWMdnd3 +2RZm0sRh6KjSlE1+S02DUogykzWJmTwQblQ5PMe/YF/zmcnlMOEZ5qG67wCShpOF5TRNC7gofx1 dXXlxUIxX+SD4js7g5KebFKpD7mHxzYxbtaK8Z93WNe5zpO4WzglI6Q3OZXs6h/DWUXbaG8TB53I YmqqFBOJTGRGmghOPNoo/FAn0saLxiHDGAad1kednVUiHxkBJWia8Fke19pOK/GiafKgr3mCPe1Z 26zG80uc9m95G5bQ07aR7lN4NiW8ENYVDGF/k9M/9CfwQtnVIaSkGtG/8f2BjzQv3NOmeB6jLWvb yJKP1h9t74pXXr/67Ve6IR3iqK/Kb/is0jWytWAI3GsImBPiXpO4ldcQMAQWRGDDhg0+TfrISZQu jt5CCYmheGEc9fX1eWUvfeSXnkPe39+/YHnqkaC3t9cb3GlMyJvj03CEkCbvgIJHmVUWIT0UY479 Qz4xDEtmQiQOETmmbrL0hAWOzESxXN+zXljMd3QODHrW9/g163ekfqYVWpVZb29fCFdu1719ST2Y mLhTQuOuHN/JX7sYnzHaDwZuT0+Pn+KcNhioJ9QXnscwctevX+8Nttu3b5dgwg28wANtPUZQOnrs YUiTONpPrUZUmEe117RRnB3IJt2/6fGdGzdurDa7JaXD8Ux9GRsbn5eP8rahP/kuzEtQ5wj6N9pw uq5QXzHCcd7EkA8OBujgrEob/2BCH4x8YrQfvi18Y5RuCDn9G89ifQuVjvarygsyg7/FtJ90n615 2q8hsJoRMCfEapaulc0QMAQWhcDAwIBXvq5cueIVMM3k/PnzXsnYtm1blNkHKHdbt271xv+lS5eK xiUKy9mzZ70BvGXLFmUv11+U9M2bN7vR0VGnDhAIcn3t2jW3Y8cOr3zlykQhc/jA+L98+XLJKCr3 KOp79+6NohjjXNi+fbso47Pu+o1rIp9k2ja/8IKxEEs+GHRbt2z28giNl/HxMTcyMuIN3M7OZEpz 3jLCMMG4vH79eskI6uDNQXdXjubcuXNntGnHu3fv9m314sWLxfZD+bkHM3iJEaBFv0F7GRwcLJLE 6L9x44bD8akzsIoPc7pAPjgikE9oXFJnMaJ27drl+5ycyBezpX8Df/qzdP924cIF38ZpXzECbZV+ n/YyPDxSJIkzZEjkNTCw0bWvy38JHoS1/fC9CZ1nV69e9d+jffv2RerfnKP9ICfkowHnA+2HWRux 2g+0+MYMDw+7mzdvKiu+vnJP+1HnQPFhThe0Hf6QTzibCH2B9kM7x4FjwRAwBCojkP8c0cr07akh YAgYArkgsJSRBRSe/fv3u+PHj7sjR454pRBlEEUMRQgjOFZgBBVF79y5c95g4B5FDGUHHhl1iRUw CDBwDx8+7I1rRp/ABIUMxSvGiBhlRcE7ePCgl83Ro0e98omxgBKIfGIpo/DCSOHOHdvdJTHgUM7b 13W4ETFiqH979uyOZmzDy+bNW9y4jOSePn3a19k1MrJ77eo17xDAsIwV1Lg8c+aMb0PIY1JGcK+L swpHSY+fHRKHGwx7jDacdsiHmQ+0H0YxaT8Yn7ECbQSnw4kTJ9ymTZu8E4Q6i8MGXhiJjxFot3v2 7PHt59lnn/W84IzACUEb15lgMXjRvgOjFvmABUYlxh1yi2nM4TCkHzl//pzIaYNrXkv7ueroc7du 2RYDDk+DenD//fe7p59+2tG/gRHfHzCif8NJESvQv+HUpU+h3iCfoaEh337gkW9lrED74bt36tQp 346oG9RZHNLqLInBC2Wmbh46dGiefgCPMfWDGOU1GoZAXgg0iaJkC5HyQtfyNQQMgYYh8Oijj7rH HnvM03/kkUfcq171qpp5QSnGYMFwQOHBgIlptIQMYySo84Gp7ExPjWW0hHxwjSOC0XU+Hxh5sUZw 03xAH4VY5QMmjDjXGmZGL9b6yrz0yAdMWD++TuSzfn23OGUaM9kQPJAR+HR1dYujKtkvYh7TOUdA H0ww7FDcqSeLkQ9srulemhGIUUv74RfHHYZlI9sPM4roXzDywCWWAy8UOfIBE+oK8gETDLpGBNoP M0T4hQf62sXIpx5tGQcVM1Z8++nsch2LmEHU1CYybU32FlosntBXgx8scEYstv0slgd9jxlmzNhR +ejyCH0e85f+jX6F9kOdxTHSiPZDmamz1BdtP4tZZkb/+KY3vcmdPHnSf9cff/xxL+uYmBotQyA2 AtJmm+K5MGOXzugZAoaAIbBEBBj5QfHjr9EBJ0jMEbBK5UXp46/RAcVzOckn5ghyJewxsmPOkCnH C/LBmOSv0QHjjen2yyEsp/aDMclfowP9G7NDlkPACdIoZ0xY/uXUv+EEWS7yWS79G7JaLt+fsN7Y tSGwUhBozDDNSkHH+DQEDAFDwBAwBAwBQ8AQMAQMAUPAEDAEDIG6IWBOiLpBaRkZAoaAIWAIGAKG gCFgCBgChoAhYAgYAoZAJQTMCVEJHXtmCBgChoAhYAgYAoaAIWAIGAKGgCFgCBgCdUPAnBB1g9Iy MgQMAUPAEDAEDAFDwBAwBAwBQ8AQMAQMgUoImBOiEjr2zBAwBFYsAmzqpSG81jj7NQQMAUPAEDAE DAFDoNEIqI6iv43mx+gbAjEQsNMxYqBsNAwBQ6AhCHCcGME+7A2B34gaAoaAIWAIGAKGwAII3JWj ndFX+LVgCNwrCJgT4l6RtJXTELjHEODs7Zs3b/pST0xM3GOlt+IaAoaAIWAIGAKGwHJHYGZmxg0N DXl9ZXZ21k1PTy93lo0/Q6AuCJgToi4wWiaGgCGw3BDgnPfNmzd7tjiD3oIhYAgYAoaAIWAIGALL CYE1a9a4gYEBNzw87Hp7e11zc/NyYs94MQRyQ8D2hMgNWsvYEDAEGokAIwoWDAFDwBAwBAwBQ8AQ WM4ImL6ynKVjvOWFgDkh8kLW8jUEDAFDwBAwBAwBQ8AQMAQMAUOgAgLmhKgAjj1atQjYcoxVK1or mCFwbyNQj486G0WxVpP9JVjSsX79etfR0dEQYMfHx/10TXhqa2tzfX19nqfYzLB+dXR01I2MjPhp o+DR09PTkM0/2cRrcHDQgQ2YdHV1uc7OztiQeHrwcOvWLb+xGLyASSOm1SKf27dv+z82ZEU+3d1d wuPcaTGxAKKuUlfYk2Vty1rX3dXt2tvbY5EvoTM2NlZsP1pnG7FMi36JtkNdIbBsjLbciM1zaT9M AQebtWtFPt3dvg2VABfphvZDXwtP1BGmpTdKPsiGNkSgP+nq6hT5xB+zm5yc9PIBG/3+NKJ/o87y DaSvZb8C6iz9WyPko98f+hUCfX6jvz9gQ/tBP2iEfDwQ9o8hsAIRiN+rrkCQjGVDwBC49xBACf3m N7/pLl265JViDKpnnnnGnTlzxtXDwVELovDwjW98wxsLGFAog/DGb8yAAvj000+7Y8eOeQWUtayn Tp1yhw4dir6rN4bc17/+dXf58mVvaKOoHz582J07dy66fC5fueKOHDninSEYUMiFewy8mGFW5HPy 5EkvExwga5vXejyOHj0uO69PxmTF19Ujh4+4a9euufa2djc1OSX15ri7ePGi8BF3qZS2H9o0RgIb 1j755JPFjWtjAUP7OXr0qK8b0MSIoz+hX4m9eS6GE33KhQsXfP+G8U/7OXHiRPT2Aw/wAk/0bzdu 3HBPPfVU9P5tdnbGHZe+7dTJU46+DfmcPXtW2tPp6O1Hvz/avyEf6snp06cd9ShmuHr1qpcPfSwO iEa1H8pMv8of8sHZi3z4JlF3YgbohfoBzhn4QD6x9YOY5TZahkA9ETAnRD3RtLwMAUNgVSCAQoFC zujGQw895DeN2rlzp9u1a5c36q6I0RkrYMhiGGzbts3t3bvXbdiwwd1///1eWcegial8wQfKMZhs 2bLF/z344INeKY2pfOEQwhGCQvyc5zzHbdy40e3Zs8fzAx/Xr1+PJR4/on3h/AVPm/rBqPa+ffv8 LIjTp89E3en8wsULfpR9//79vs5u2LjB84LMzp45Fw0TnAw4pxjdP3jwoOvr7/P1d6vUGRwCg4ND 0XjBWUWd2Lp1qwMX2s99993n6w51CMMqVsDYxqB74IEH3Pbt292mTZt8/UU+OI9iGS/avyEf2i+b 4tG/UW9xEsFjrIDDAfmE/RvywZlH/0ZbjxUuXLjoRmUWxP4D+71sfF2R+js0NCzt52wsNryT4dln n/XfH+RD/4Z86ONwssbu344fP+7xoH7Ay4EDB7wzD2cAszViBfoU6gt9CvWFektbog3HdJ7RfsAE Z5DqBzt27PD9y/nz511M/SAW9kbHEMgDAXNC5IGq5WkIGAIrGgGUcKYoY2jjiNCA0YByjDGBIpJ3 wCjBaOMXoyWcso3SgwKYjCznzYnz0+pRfpkmzWilBvDgFBIUr1gOEfBHPsgjxASllNExRpdjBOSC 0rlmTZN3PihNeIIXZh9cvnxJo3P9ZST98pWrfkp9Wj59fb1ucGiwuAQgV0Yk82tSTzAMBjYN+BFL pYczgvaEsRdrNBcDF1o4IcK6ghHD6DLtK0ZAPtQV2g/TtjUw2o5hh3EVa2YTtFhyQPsJp9T39/d7 5wyGLtjkHWg/tGVm7dC/aWCUG3nRv/E8RqDvYtZOeklXS2ur65d6OyhLRcbH4jistC+ljobfH+RD /0adBrsYAVrIh2+hth+VD9/AWPKhPvKtw8lLG9LQKvKhr2UpD/U6RtC2imMolA9OK9UPYrSfGGU1 GoZAngiYEyJPdC1vQ8AQWJEI6EgTCl8YUMKIw8CKMQLEKCCGCYpOqOzAk649RfmKEZiRgWKFkpUO zEjA0IvBC8o3o9sowtANA8oy/CGfGNPboYEzZM2aZtfaWnoMLPWEdeTDwyMhi7ld+6Ufgk2bLH1Q Y0GJKU66jlrj8/odlpFjQrr9IB+M3snJCXFY5W/QYSRRVzBU0u2HegI/1NkYDkX4oP1ouw2xhxfq dYz2A136FOiFzirlB/5oP/zlHejfcIZkyUfrDlP/YwRmo1APsvo3+CPcHIzDC2WmDWu71fJTX+GF fkf3FNFnefyqfGizobMKWvBBH8y3MoZDhPbDNwb5pPs34nhGmhhBl9lpHVWaoX4Qq69V2vZrCKxE BMwJsRKlZjwbAoZArgiogpc2XCBKHEqXpsmTEQwBDJcsPlBIUXowgmMYUVpe6KaD8hdD8aKs4IIC nOYFPDRO+U3zWs97HFEov8yESG9cB3/8ocjHGBVTo3Ht2vLyGYswkkvboE4KIL78abypK2A2Pp7/ Gm4MS/jR+hnyou0H3GK2nyxe1MCD3xhB22kWLxqnafLkR7HXNhvSUvlQl6gveQdfZ4WIlj+kp3Ga JnxW72v6C5wMYV+mNMK4GHWF2SH0XdqXKR/8Egc/ym/4LI9rNfyz6gq8EOjzYzhE9Nui9SIsL3Hw EEM+IV27NgRWIgLmhFiJUjOeDQFDIDcEMEjUaFTlJiSG4kVA+co76PKGLD6gDS8oPLFG/aGZxYti EoMPDBLFPs0LfGhcDF4mJ5L10GkHhMoGfuB3+m7+S3d0Zo6WHx40MFODwAwENt/LM0yL0UKZaSVa L0J6yp/yGz6r97XSgI80L/BRlI+0+bxDyEualmISo85CW+lUMug0TZrXet5r/5aWjdIgnv5NsdP4 ev8mNCZ8tiqLkIa270nZXDXvQN+m5c3CRfmLIR/4ABv4ULpafuUt7I/1WR6/6mRN8wEt4uAH7OAn z4B+oPLJ4kVxiSGfPMtpeRsCMRAwJ0QMlI2GIWAIrBgEFhoVVSVDDeE8C6YKldJM09J4VYrSz+t5 r7gozTBvjdM04bN6X4OJKsaV8o4hn+mCQa/lT/MjerEoxbNueiZ/I7dSXYEPArzwl2eYEaMF+YgL IpOMYhWjrigNpZnJkESq07Hc83rEV+JF+YvBB2XRupJVLuVF+c1KU6845UNppvPV+Lxxob5W4kXb z/R0nH0yKvVvsTBBFpXqgPIRYpeWXz3vK9UB5QV+k76nnpRL86qECSlDXkrftDtDwBBII2BOiDQi dm8IGAKrAgFVBihM1ohFuUIupMRovgulK5d/LfHV0qg2XS20F5NWFfnFvLuYd1QWWe/GwIQjMQlq pGTxwUkR3ibPflinWDX8F8qu2nQL5VP+ueIOJo2WT6X6GPKmPJcv1dKfVEOjEr9L5yDJIeQjxCCd f2xe0vTD+5DnML6e18U2mu07K5LKmxfyVxqV5KNpiozlcBHWgSxeNC42LzkUteosQ/lkvaSYhNhl pUvH6Xv6m35u94bAakRgbtv31Vg6K5MhYAjcswgwcqIbVdWiECykBKjCtVC6egBfLY1anCz14Ktc HjH4CDFRWWTxE4eXxI9fNGCyGJEZAZWdFJkv1RjJdOkFrCfJMcSuRgJVJ4cG/zEjAgdMudBUWMdd 7nk94ivVgbDuxMJloTJV4nehd6t9HpY1xCD9fmxe0vTD+zi8FCiWr7Le25h3W/btR4ggm0ryCeUY YlXP6xB3eEnTVP7S8fXkQfMKedG4RvyqfMrRVkxq4Zd32IMFfYUlUrXoK+X4sHhDYCUgYDMhVoKU jEdDwBCoGQGUBT7o/C00hTLMXDeJI04VivC5xv3/7J3JbyXJVv/T5XJ5HspTzXN11et+8wMhEEgg fquWkEBICNj0BpBgw4YdLFj0hj+gJTZINGxY9FsgIVYgEOv34PXcXfNsl4caPVSVq2z/4hPXX1c4 nb6+1zcj7rV9ott180bmzfON74kTec6JyMyih1KFx5WxLSxbOSWq13FlyNzqHJIhmeFxKTlBrzh4 yJTcZmE5uPaQzq1wOIg+OdC+9kyGEGfZ2+3tlTmFrbAgD97a2+Ne9r3zTRC1xQoQ9R9xVzYP4flk o0WcFB0X1pW9XQ2LOJGNlS07f75asOiY/G/L/K72qv35c6s+NpbKtaKa/VSQVex4+2Rfvh31fOd6 pfGt6Hfqy7E5Qbb0g0zJFSZ913is+lifekOJ5IZyVCfuwn1lb8MJbaZIbihDdfXoh/NxfAr8IVbb NgaazUBcb6TZrTP5xoAxsG8Z4ILO6+b4Y7vWgkMgh0dOcPhb1eVfzxUeU9a2XhdXlETB2eEPvPlX uZUlPzyP2luERXUpcOCgyzmWXOGED+knBZYO91pO+EemnM8QCw+BZH97R+39T7+v97Nz7TWCeU44 j+5lrzjQcS/72JpPdAS6CNsifMIb7it7W322SD/goB689QQMO8UoLGp/eB7VadwJ98XY1rhSdJ99 SizCIZsN2xqOb7F5wUaxZcpKwfNbhK+ra+Mrm0O8ZW1joxrfJDc8t/Qj7sJ9ZW+DQwmR4vFt1e+P rR/apdfJFnFCHfiEt2wewvPRV9TerbBwfL364Xolf8UnckOhtm0M7FEG4noje5Q0a5YxYAzsbQb6 +vp8A4ucdB54iCOiY2IyQeCCw4PMvBOoBy/iuIAndlF75QSH8sSTHMVwX9nbBIw4bDiAkisZ1IkX 4dW+GJ/d3V0+iCXZkHdI4Ym6rs6uLMWsf4/rB5TX7on2+fL6deWBevSV2IW+qATQm01vBVl1+qm8 8q+zqys2FB+44NDTJ/L6oe9gU2BNkYTo7+/37VX/DBuvB8vqmHBfjG31g2ZjYXyDe+kibKtsO9n4 1lMZ85debX4Dxps3lbrunu4QYpRtJc/pr/mxNqxLM75VbAMceSzSGdcnBeVRCFk7qWyjqM+qLgUn wBEW9dGw3WBhDJSNhfuqbeev79WOtX3GwF5hwJIQe0WT1g5jwBgojYGjR496R4L7NMOC08GrwgYG BjLNbIb7y97GQR8ZGfHBkt5NLhlPnz71m2NjY6qK+jk8POydzaJ3sfMOd7ByTIoyOjrqxSwuLm4Q RzBH3eDg4Pps4oYDSv7S0XHIO6QEB/lXsvH6QRzL4RE4iZ8kok8SwCy6/pkPuNEZrxk8fPhwyQwU n2501PVJ54gvLi5sOIBXmhLQkawigRO7kICgr9Av8vrh/msCKzhJlcQjMMFW8gHHwsJCUvthTKHN YAkL/Ya+gv2kSigODQ15PeTHN429R44cCSFG2+4f6PdjxuKLRaefja95RD/Yz9hIZdyJBmLtxOiH ktcP1x/6McF2vUHu2qnr+mA8QT/IzI+1jG/0l+PHjyexH3BwzQVHPvin72Drqa6FjCnYj/qoSA39 gxT6kVz7NAZ2KwOWhNitmjPcxoAxUJWBvKNf9eDcThweHPHp6ekNwcu9e/e843Xs2LEkjheODk4e y0ynpqbWgxeCp8nJSe+MpnK8WF4KFhzy0Dlmm7/Tp08nScygKtqMI/7w4cP1d7ZTD0c4o2fPnuVr kkJf4LWX0zPT68ELzvn01LSfZVfCJDYYkkAkz0iShc4xTjvfx8ZGk+lncHAgG+jrzx4/fpy9cTOD FOzx8ZMn7nWlK9nJkyd9UBebE85/6tQpH+BjLxoTsB9sm6DmxIkTKWD4BB59heBtdnZ2XSYB1BPH C/1EM6zrOyNtkHhB3qNHjzbYDxzRd9EPAWjsovENOfnxDf1g46mSENjPkaNHvH6ePK4keME1P7/g 7Wd8fMzdsnGIqugF/ZBURB/0FxXGO/rumTNnVBX9k3GdMRXZKgTbDx488PbDmJOi0Ffol9gL44oK 4x32Mz4+7vuL6mN+0i+xH/QTJjfhiJUQ2Dn9yYoxYAxUZ6DNXZSrPQu4+q9trzFgDBgDLcrARx99 lH388cce3Ycffpi9//77dSHF0bp69ap3QElK4AxSd+7cuWQz/gLMrO2VK1e8M0igopnTy5cvJ1kK KxxcLu7cuZNNTEz4JA3OKU4hjmglsIw/4y8s6OPatWt+ZgynPdQP3+spK3MT9Ry+6Vj0c+f2nazd OZ7M8C+4wKXLfRIAp1gxI0Doh2AOZxhHmdlbn4AYHcmOn2DGMt28A8+hoK8QxPU7LK+WXjsHfclz gj3VWw70H6/3J+vHwwG2TH9lhhJ9kVS7cOGCTxStH5hgg+CNZCY4SC6CjYCGxBmBVqpCMAsnrKhi BZNmccEBnpSFIPL69es+cIMXxhT0c/HixbrHt0ZtecIFllMPp5z9VJ4lNDc37xN4jHH0n1pLW2d/ 1naocgtOrb8JjyOYZczXyi5W8xBwnz9/Prl+sBf0Q/9khQz64RaMd955p+5nH4RtrHeb8Y3AX/ZD oE//JVFFsiRF4kyYSdahH7hhwkKrrfAP6rUfrl0ffPBBdvPmTb/y5JNPPkm2ak3tsU9jIDUDbjxp syREatZNnjFgDCRhoNEkBCBxNHACcTBweHCQUzo6IVEECWDBOSWwBQtOYTMKThMOMU4hOFIG2mF7 kY9DDCeN6KfRwAVMYIAXgjscdJ410Cz9MDvHH/yAI8U926FetI18+slrl4BoP3jAYenZsf00koQA j+yHT/orvNQTVKpNZXzST/hjfCGoq/chdmVg4BzohzGFvqIEDYmRZhT0gi1jP+hmp/ZThi2XYT+N JiGkHxLOjC3ST7Nm2NELWGQ/9NtmjW+h/XD9YXxrFhZdf9APnOzEfmiPJSGaMeqYzGYy4Gy2zdYL NVMDJtsYMAZamgEcC2aUW6HgfLJEtxUKQVOzAqew/TieqZawh3KLtnE+d+KAFp2r0TqC7GYlhkLs 6AfHPHP/N7uY/WzWAPohiOOv2QX97GSFTAzcrWQ/rXL9Iflu15/Nva1V9LMZmdUYA63PQO1ry1q/ LYbQGDAGjAFjwBgwBowBY8AYMAaMAWPAGDAGWpgBS0K0sHIMmjFgDBgDxoAxYAwYA8aAMWAMGAPG gDGwlxiwJMRe0qa1xRgwBowBY8AYMAaMAWPAGDAGjAFjwBhoYQYsCdHCyjFoxoAxsHMGuN9ZJdxW nX0aA8aAMWAMGAPGgDFgDBgDxkB6BiwJkZ5zk2gMGAMJGAgTDzxg0ooxYAwYA8aAMWAMGAOtxoD8 FfNVWk0zhicmA/Z2jJjs2rmNAWOgaQzwNG9eJ0axC3vT1GCCjQFjwBgwBowBY2ALBkhA8Lpc/BU+ rRgD+4UBS0LsF01bO42BfcbAysrKevKBd5xbMQaMAWPAGDAGjAFjoJUYUOKByRL8FivGwH5hwJIQ +0XT1k5jYJ8xECYh7MK+z5RvzTUGjAFjwBgwBnYJA6yGIAmh2zJ2CWyDaQw0xIDdKN0QffZjY8AY MAaMAWPAGDAGjAFjwBgwBnbGgFZD7OzX9itjYHcyYEmI3ak3Q20MGAPGgDFgDBgDxoAxYAwYA8aA MWAM7DoG7HaMXacyA2wMGAO1MFDGzMLLly+zhw8fZvPz81lnZ2c2PDycHT58eP1ZE7XgKOuYJ0+e ZNPT09nr16+z3t7e7MiRI1lPT09Zp6/5PDxfA06ePn2a8fDPgYGB7OjRo03h5MWLF9nk5GS2sLCQ dXd3e/2go2aUZ8+eZY8fP/YPF0MvIyOjWVdXZ3Io6Acc6Iflvf39/dno6Ijbbk+OZWlpKZuZnskW FheyzkOd2dDhIddf+t2S4/TzH+hnamoqe/XqledkbGzM21FqUtAPdow9UwYHB70tHzyY3h2DC+yH 8e3QoUN+bIOXZpRwfOvr68vGx8ebMr6trq5kMzOz2fPnc46GVT++jYwMu7EuvX4Y3xhrGd90/WF8 S71kn2vp8+fPPRauP+inWfbDrZUzMzPZ7Oys54HrD9fCjo6O5N02bz8jIyNN8w+SN94EGgMlMJDe EygBtJ3CGDAGjIHYDDx69Cj7/PPPM4KGc+fOeefi1q1b2fXr15M/wfrmzZvZtWvXfMBy/vx5H/yD jaAqZcHp+vTTT32Qe/LkyezYsWM+mPrss88yEjYpC44oHOCQw8nQ0FAGT83Qz+07d7K7d+/64PbE iRMe05Ur33qeUnJCgHDl2ysZfRfdHD1yNCP4/vrrb5Pr54lLgnz99TfZyupydsr1lYHBgezevXtO R7eaYj/ffvvtuv0Q8H/xxRfZxMRESvX4BNWXX37pg6jjx49np06d8sEdNkWgmbIQVAQOXAYAAEAA SURBVP7iF7/wmM6ePeuSZiPZ/fv3s6+++sqPeSmx3Lhxw49vJHgZa7FpeEo9vjHWf/PNt9kTl8Qb Hx/LjrnkKvZz9er17MWLtOMbSRnGNxIAcIJ+GN+uXr2a/OGFXPewH8ZYsJB8Bltq+0E/yCVxxjh7 +vRpn0DDfujPKQv9ArnyD6QfeCpjAiRlW0yWMdAsBiwJ0SzmTa4xYAy0LAMEczh8zPhcuHDBz5iO jo56B4wZGIKpVIVZbZw9AhZWHDDLjvMFHpz3VM4XM1AE+LxG7NKlSz6gYxb38uXLfnYZRxWHLEUh 4UHbmeUngGJliPSDg5rSOUY/sy4hcsIFlWBAPwSYYLt9+3b26uWrFJQ4GavZHZcMef3mdXbh/AXf d3v7ej0/6OWWD/7TPHkdeXdu38l6e7pdvz2ddTtOCDBJXD19+szPqCYixSdkZD/MmNJXwMFMO4kj VoykKshj1QE2Q0BHH2GbQn/WK4Vj4yGZiC3TV0ngMc4RRF28eNEnzh48eJAskEIWs/2Mb+gE/Zw5 c8avaoITVgOkKASOt2/ddsm6V07+Wb8CoteP/+ezZTfmYVvYWIpCPyCYRS8a31gBwbhL8hXOUhVd f7CZvP1wjUyZPCMBgzx4wH7ghz5LUgQsqewH/4BJga6urnX/APv5zne+48eTlP5Bqn5gcoyBGAxY EiIGq3ZOY8AY2NUM4BQT6OJYhAVHEMeDGboUDg+BP7JwkPPLpElIEOylmi2cm5vzs4IkHlgarMKs MsE3zmqqgIGgEv6RGy5NZmku+kmVhFh1+qGv4AT3O9lhIehGf9Mz02F1tG2Cp6dudo4gruPQ26XJ LFMeOjzob4mYm5uPJj888czsjNPP6+xw7tYY9NPRcdAvd4ebFIWAgD6Sv00H2wYDQV2KwnhCXyF4 Cu2HW2boKyQTdYtGbDzY6uLioucE+SokRUhMMKYQaMUu4p8xJD++oR/Gt1S2jH6wn/7+vqyru2u9 6dzG1D/Y75JHc9nC/OJ6fcwNXV9IyoT6IejW+JZqtp2EB/qh36pgT9IXSd8Uhf7ICi9shVvvVMBG XyE5Qb9OUcDBtY6kTFikH/THrWhWjAFjoDoDb68+1Y+zvcaAMWAM7BsGcDIoOHxhwfni3mkcVv5i FxwZHCsCSZytsBAsUFIFLswYExTkOQEDWAgoUmDB+WYpLM55HgvJAOrQDUFW7PJirR8g9+DBjc9c oJ+AMdVKFQIol63axInXT3elr5BISlGePqmsLsjfpw0fHR2HXIC7lEQ/BC60mT6Rtx8SAeBRv47N i+TIbkN51NGvU63KYExBXhEWkljYT4oZbsY3VoYUjW+ybY3FIV8xtsGxurqSdefGfGQdcn2W8uhx 5brgv0T6B70wjnKtEQcSVbGfDh/gMgbGLqyYkX4Yz8Ki8Q39pEgowgncwAnchAVbBkMq+9F1LkyG gCf0D1KNtSEPtm0M7DYGLAmx2zRmeI0BYyA6AzheFJy+fCHgpCQJct1sC86VZIZYwIbTg6NIciB2 UXuLOFFdisCFttJm2p4PLOEgJRaCXPQDlvzDFsHBHys23iSYVX65tmy9vaDPipMXi/GXtrM6BP04 QtZ1EfbNg+0HfDCRYtWMZNB+dBQWbIo6AuEUq5qEpciWVadjQpwxtmWnRfYjLLL3GPJ1TtpLYCmZ qudTOks1von7A2vje4hF+HRMuK/sbcYUkkD0TcmVjLAuBRa4xzakC+HgEyz8aTwO98XYVlCf5wRZ wqf+FEN+eE7ZBnLzRfhkY/n99t0YMAbeMrDZgt7usy1jwBgwBvYdAzhVOMaUIicDx4uSYrmlZBTh AIOwpFqVgcwiLMKBAx27oB8FjZIrmXwXPnGnfTE+l5Yq7ZXMUEYFW8VJf5MgSSTu2wocY70Zg+dF MNsbs7xZfuPtByvJ6we5wgeW2MUnQ5C5ZrOhPHRGPUkk9adwf9nb6o9bYUGe8JYtO38+9ZWifqs6 4c3/tszvklHECXJUL7xlyg7PxXjP6hyK2h/uV4IxRT9BBn9qe4iDbdWn6CuhfvK8CAf2E1s/tFvX OMmlTkV14ABPzBJef/KcIFdYxF1MLHZuY2C3M2BJiN2uQcNvDBgDhQzIGWBnuF14cFCJk1GtyPFI 4ZAKy1b4VZ/C4amGRThScELAwB8yJbdIXymw8OaHStk4yy48xL7ks2I7xjwwTzKKOFEMzjE6ThjL /lxZqejHWV3hqYVvZRs7K/xxnZVqq2SGPw/rdFy4v+ztWuxHx5QtOzwftqP2hhzoGNWlsB+1VzKF QZ+qjx3kwsnyciV4lUxh4FP2I7zhvrK30Q14tirClwJLKENyhUvfw/6kfTE+Qyz58wvLdtzlf7eT 7+Coph/5B9Xw7kSu/cYY2IsMWBJiL2rV2mQMGAMbGJCTsqFyiy/VHIwtfmLVawy0EncpsKy6gJui IGWNhtwHQXmuKsLXWtubAouaV83uUuConZP4CqoFSy3HiNvYn62EJXZb6zl/Cl4ko7r9NL/PCp/w 1sNjvcemkFELplpx1HqcZIpLfarePo2BvczAxied7eWWWtuMAWNgXzHA7JkeIFXPrJ5mMrYiS87F dsdt9ft66reTISy6D7Wec9d7rLBIZvh71aXAETppkhti0bbw6nuMT8moFlCDN8QcAwerDoSl2vlT YPEy/CoIrYjIIVqLnWrBm/tl3V+r8R72nWrH1S10ix/INkK5OlR1qTihvciUXOEIP4U3rCt7e7v2 Cl9sLCQRDxwoXrkTttn37eoZx/DwHW1Lxnb62Y67HQnP/SjkHTyhnYT4UmCpRUaIL9eU0r6Co5oc 9dla8AoUv+FBo3ogqVYqab99GgN7lQFbCbFXNWvtMgb2OQM8dI1XZvG093ocgvBhbXIoQipVFx4X 7i9zW28X2MopERYdV6bs/LnUXsnM7+d7Chw4gDjH4CjCorr809yL8DZad7C9kscves6C8FXwxr/U Sj8rjpd84RYJCg/dq8cW8uep5bt30l1A57RTePjK2jMphLfwoJIq1R/VJ/KnVb2Oy+8v83sY0OXP KxwpOEG22iu5IR6NNdXwhsc3si0ckpk/l/DF58WNKWtvH5LMEIvq4uPI/Nim8S3EoO2UWNReZEqu cOhT47G+x/rUa22LcKgO3mL3WzhREkJywzarTn073LfVNufjTTX4KnzGHqO3wmH1xkBqBuJ7Rqlb ZPKMAWPAGHAMcGHHEag3GOV3cniKnGPd65l/fVoM0oVDMkMZYMPhwWFpNhbhE94QZ9nbOJnolbZL rmSEdSmwdBzq8P1MuhAOPrlVQ/rpyL1eNTyurO3OtVfoLbuH2uXLsntYJKXTvWpQDnT+mLK+86pS HwgU6AcZ0pnwliW36DyyC2Sii7CwOoo6+lLswAW56o9qf4hFdcIb7ouxLSxFK8RUlwKLcBSNs7Ip xrd6Arqd8IVN6Dqh9ofnkX50TLiv7G1k8EffLOJFWFLpB/6li7CtqtN4HO6Lsa3Xyar9oQzq4Ave Yo9vnF/cl6kf+Sqx+3rIm20bA81mwJIQzdaAyTcGjIGWY2BwcNBjKnrgI3U4XqyyiF1w0pkdQWY+ iBK2/v7+6I4X7UQORXL9l7V/VDcwMBBWR9kW9/AhuRKEU0gdjmIKLD3d3X4GFSx555jv/PUys+X6 S+zS4/oJZangDSVLrypP/u/tqxwTF0ub77PIeJ1LiEhnPnHW3RMXhjs7gQtOPX0iHzDoSfrdToea 8Y0JaGhoyJ9eT/kPZelNBxp3wn0xtjV2bYUllf0wvhE4oov8+KaHUaYa33p7ev24oTfehLyrH2sM DPeVvY1tIAc+xIFk0IeVJEnRV7AN7AeZkissfAcPx6DD2EXjeVFf0XUghX5op7BIbth26nSNCuu3 2873/+2Ot/3GwF5gwJIQe0GL1gZjwBgolYEjR454hzT/rm8cDAIGnJ0UjhfB0eHDh33b9J50NfTp 06d+c3x8XFVRP4eHh/0MEJzkHSawwEcKx5hGjoyM+LYW6Yd3xRPw4QjGLizhHnIJK5INCiQlc2Fx wW8edrylKPRJ+sui00+YEOFWEbDwmkH1pdh4RkdH/dM6wRIWOCJ4IQju7IwfuCAbLMikX4Tl+fPn PoiCk9izp8glIUK7kZu3n/n5eR/sKVER4oyxDSe0OT+mwBNYwEFwGbvQXxlXkJvHAg4KY3GKojG9 Mr5tfM3jC9ePGU/AmqJIjsZ4ySQAx4bQj1aRaF+MT/oI9oHc/Fi7uLjo+/HRo0eT2A/6oc3goL+o YEv0FfqSH3e0I+In158i+wn9AyYPrBgDxkB1BiwJUZ0f22sMGAO7lIG8o19PMwimcX4nJyd90MBv mfW5efOmdz4uXLhQz+l2fCyOzunTp33wf+vWrXXnCwfw3r173ikeGxvb8fnr+SFJhrNnz3oncGJi wjugcAJHOIHnz59PkpgBM4kX2v3gwQOPB13jKMMRON955516mtbQsadOnfIO8P3797M3azO6zDA/ fDjl9aOESUNCavgxTjj6QfbMzIzTD7frrGSPHj12Ad58dubM6ejL2gWTgGF0ZNjhmM3mFwgmKytF 6CvM9J4/dy5J4AKeM2fOePu5c+fO+swygRxYCLDQX4qCfrARAhVslz7LH7oi2Dx58mSS1VW0lSCW ds/OzvoH4oGDwA6OGHPgjM/YReMbs+3h+EagiW0TjKdKspJkoN3oZ2Ly4br9oJ95hwf9oMMUhX55 7NixjHGWBxaiHxKLcIT9pLr+0FZkcfuB7AcsXH9u377t+xFJiBSFvsK4jmzGWq49YHn06JG3H66T qQJ/koknTpzwOEgqUsBz9+5dryc4Q09WjAFjoDoDbc6IN94oWf1422sMGAPGwK5g4KOPPso+/vhj j/XDDz/M3n///bpw45QTLODkMAND4MJsJs47QVbKwiwuCRACTC2PJZDAaU4x46+2crnAKYeXUC6c EDCkCFyEhaQDgQr6gRO+4yyfdYF4vc7oytyETrujT/Qz8WDC95H2g+0OyxvnoA9mOOipAheAox+C lsmJSV6Y4QtX+PGx0Ww0UbKqIpVnP7zxSYjZ2UeOg/Zs+c1y1tnV6Z33ncyyH+g/rlPX/UngQoBA cEuSClsmOURgSd9JWQj8wYKtyF7oJ1p9lQoLQS32MzU15cc3vsMFY0rq8Q39EGCTAGBcwZYZTwgs w3GmFm4ateWnT55mkw8n13WzsrySHTl6pO7xra3T3SZ3aOfXCa4/JCEYbzW+cR1qln7oK4wt2A/6 0fUn9fjGGyRIQtBfsR+CfxI2qVZkqA8il+sg+hEnXH+4Fup2DR273SfX9Q8++MBf4+H1k08+SbZq bTtstt8YiMWAs9+2NGndWC2w8xoDxoAxEIkBnKtzbsaWQAXnmO84g82Y4SBoe++99zwOnFOw4Pgo iIlEwabTIo+ZSZa94jhRwJHSERUodHHWJRxwQJvJCXjQz7nz5/xKiGUXtBzsOOg5aYZ+cGKZqYMT EhAda1jEW6rPdvfmEAJrgn2wYDfNsh+Sh5cvX266/cA9tkOATSKEpBGc8Je6ENwT5B8/ftzzIv3U G/SXgRv9ML7BCcGlxpTU9kNbhg4PZf0D/T7Q5gUvPHy2GZwwpqIfgmvZD7w04/qDfi5evLhuP+qz qfWDPGyHIJ9EiOwHrlJjQQ8khGQ/9BFwNONaWIYN2jmMgWYwYEmIZrBuMo0BY2DXMCCHq9mAcbKY CeOv2QUHDMe0FUorcXKoBXSDTlrFGabPmv1sthLsZyerQTafqfGaVuoreutA461q7AwElM1IPBSh JvHAX7NLK11/WqXPopNWGd+a3T9MvjGwEwbspqWdsGa/MQaMAWPAGDAGjAFjwBgwBowBY8AYMAaM gboZsCRE3ZTZD4wBY8AYMAaMAWPAGDAGjAFjwBgwBowBY2AnDFgSYies2W+MAWPAGDAGjAFjwBgw BowBY8AYMAaMAWOgbgYsCVE3ZfYDY8AY2A0MhA+qCrd3A3bDaAwYA8aAMWAMGAPGgDFgDOxVBiwJ sVc1a+0yBvY5A2HioRlPFN/n9FvzjQFjwBgwBowBY6AGBuSvmK9SA1l2yJ5hwN6OsWdUaQ0xBoyB kIGFhQX/PnHqXrx4Ee6ybWPAGDAGjAFjwBgwBprOwMrKSvbw4UPvrywuLvpXsjYdlAEwBhIwYEmI BCSbCGPAGEjPQG9vb3by5EkvuBVecZaeAZNoDBgDxoAxYAwYA63MAKsfjh07ljFxMjQ05F+x3Mp4 DZsxUBYDdjtGWUzaeYwBY6ClGFhdXW0pPAbGGDAGjAFjwBgwBoyBPAPmr+QZse/7gQFLQuwHLVsb jQFjwBgwBowBY8AYMAaMAWOg5RiwJETLqcQAJWDAbsdIQLKJMAaMgd3LwKtXrzL+2tvbs66uLv/Z jNbgpOh+0c7Ozow/PcwqNZ7Xr1+vP2eju7s76+joSA3By4OTpaWl7OXLlx4DnKCnZpTV1RWH41W2 /Ga5gqXrkIPR1gwo2Runn5euz1LgpFn6ybJVZztLXkfefhyWA03Tz6rvs/Rdbs/ClptlP2/evPFY 6L/NtB/6Rzi+0VcOHmyOW9hK4xv6gRdKp+srB5s8vrXS9Wd5uTK+Ndt+uBZ6/bg+S79tVgntp5n+ QbPab3KNgUYYaM7VphHE9ltjwBgwBmpgAKe2kUKwcuPGjWx+ft7fp0mwi+Nz/vz5bHh4uJFT1/3b 58+fZ9evX/cBQl9fX/b06VPveF28eDG5A3b79u1seno6Gxwc9O0AG/eznjhxImlQh/MHJ+gEffDw UYKHs2fPen3VTXIDP5ibm8vu3Lnj9NOR9fR0Z/Nz81lPb3d2/PgJH/A2cOq6fkqfn5yczGZmZrKB gf7swIH27Nmz59nY+Fh29Mi400+6xY/o4u7du/4+Z/oK9sMf/UR9p67GNXAwffTmzZu+f/b39ztO nnm9XLhwwScjGjh13T998OBBNjExkWHHBPxPnjzJxsbGfL9NmRQhmLx27dr6+IY9YUPYz+joaN3t auQHcHDr1q0N4xvJGcba1AGmt5/pmayvv9cnNJ89m8uOONsZdzaU2n6uXLmyPr5JP3By+PDhRuiu +7e6/vDsgoGBgYzv6KUZ+sF2ZD8kVx89euR0M56dPn3ajXfpxjfsh+sPYz/PccBf4JkO586dy0ZG Rurm2H5gDOxHBiwJsR+1bm02BoyBqgwQzOEA4pT/4Ac/8A4XdQSaV69ezS5dupQsEQGGr7/+2jue JB2YUeZp2p9++mn2zTffZN/97neTzXQTVN6/fz9777331ttPQuTLL7/0Ad7x48f9Z1VyS9hJ+7/9 9tsMR/BHP/rRevtJkHz11VdeZwSbKQoz/TijBG4nT5KIOeBx0X9uXL+Rfefdy8mCl+mZaZeEeJhd uHB+PRFDQuSK67MrboXGCYcvTVnNbrugcn5hMfvue+9mHW42GfshAIcr7CeVflglQ18h8YFcAn36 DX0W+6Evpwp0Z2dnfbANDgInCoHL559/7gMoAqkUiQjshwQEtsv41tPT47GgH/otY0yqQJcgDpnI k35IYMEJ+vne977nkxMeYOR/SN7BAckptR/7ueq4eu0wnXQJtBQFW6Ht9F1df5B779699fGNZECK QvKD6w+B9uXLjGVtPuDGptDRj3/842T64Q0SJBO5Dh49etQ3/8iRI9kXX3zhE5z0nxQF/YDj8ePH 2U9+8pN1/4C+Ay/vvvvu+vUxBR6TYQzsVgbSpQ13K0OG2xgwBvYdA8yU4qDj4ChAwfkiyMZBx9nA EYldkMGqA2ZZeNMHsinM+PAdBx7HLEUhGTI1NeVnwuSgIxfnlD9mp5jpTlEIFpiNYwY5vNUAxxRu SJakKhMTlb7A7JdmStET3xcWF7JHzlFNUQisH7oEBDPI4UqD3t4evypianoqW3IJkxTl+fO57KnT D7PHJCAo2A+Bd3v7QddXJl1NfPtBLkkzAin6hgJ89MPqHVbRYF8pCoE1QST6od+q8BYf9IX9kJBI UbAdAij0oQQEchnv4AYs9KfYhfEN/ZAUYTyTflghwljL+EbiJkV58+a175fd3T0b7Kevv8/bz6wb czgmRdH1hz6r6w9ypR+SrakK1zr0g71IP4y5YMOuGItTFPoKkwD0VyXwkIv9kACmP9NfUhTkcC2k j0o/cANH3OqVyj9I0VaTYQzEZMCSEDHZtXMbA8bArmQAJxynhyXTYcHBwOnAiccBi10IXHB2CAxw tsKiRADLUVMUHGNm5uBEzqjkEkSxj8RNikJgQrIhP5uOc4yTyvJuuItdSLrAC/rhfuCwoC94ejST Rj9wT7KqzyUdQv20rfFEIPHkWRr9+MDE2U8Y4MIN9tPRcdAF2/P++RkhXzG2aTNJBnSTx0LfoQ9h P9h67ELgolu7Qv0gl1lt+msqW0Y/JBny9kPwTx32Q9IxdqHN2A9jan58E7ZUSdb5+QVnP0tufOv1 /SJse7cbU8A6O5vGlpUYC5OJ4GF8oy/DmZ6JEOIse5s+wlirZ6iE59f4xu0rKQptZrzFVrBbFWwJ 22bsS2k/jBnqo8LCdUD+QaqEiGTbpzGwGxl4a8m7Eb1hNgaMAWMgAgMkGSjhLLvE4KgT3KRw0pFB sgOZ+SJsOKMpgiicQAqOVr4IH0FW7IJjzIwxzqfkSiZ14IOPFE4guiE4IdDPB5Y4yvy9WnIPq0ww q6xZ9PaCviKeFt3tEbEL3PsAaU0XeXlgwX4WX8THgv3APXrI64d+Qj3HEMDELuqP0kUoT7YsHYb7 YmwLi+SGMqhDhymwiPuiMUV1qcY3tbdIP8KSenyjf4ZF4xt1KbCQWCbwB4c4EB7q+GN/ioR8rdef FNdC9ZUi+9H4JhsTX/ZpDBgDmxnYOMJt3m81xoAxYAzsKwYIkAgsKXknMKzDgY5dcPAoeQdQcsFH kJXCCayGRTylwEF7feDvglzJFR+hk44DHbssLVWC1wPuORD5IBdsej6E+lNMPNX0o/6z5BIisZ10 3g6yvLzi3wuS54T2CwvP0ohd1AeQmceiOnSTIkkkLPk+G3KiY2LzIjuVLkJ5qks5vhVxgr6oRzex k0TYBMnCUBcbOHE4KLFxIAM7Ro7aT11YpB/pMNxX9jYy4AYsefuRzrCfFLxo5Yfkhm2lDnxwF3t8 wz+QnW6FBWwp9BNyYNvGwG5kwJIQu1FrhtkYMAaiMYDTW82RkTOWwvGSDMnMN1r1KRweBdKSGWJR nfCG+8reRjfoCJmSWyQjBZbl5UqyykEpLNSDd8UF5bHLiuPElwIw4umNSxDgRMcsK+5Vpa7RW4oQ lpU17rY8sIQdSi5I5lanVN/ean8Z9dWwCF+KPktb1F7JDdunOh0T7it7W+2VzPz5Va/j8vvL/E7y jCKZ4blV9/p1/Fu8sE+NbyEGbb/FEn/1jvoAMiU3j0PjsepjfYZY8jKEDd5ij2/IqMU/EN48Vvtu DBgDbxmwJMRbLmzLGDAG9hADckxoUri9XRNDJ6bod6qr5ohsJ6PW/bXKUIBT63l3clzIy1a/T4Ej 5ES6KMJTC96i39VTJxnVcIB3pUpQXo+8rY6tyKjs3SIfsvbT1aoO9Fbnr6ceLPznrK6q3a2sbJ2o qEdetWOln6JjQp2Ffaro2DLqqmHR+Ws5Rsfu9DNsa8hB/nypseTlh99jjyu+z9Zko4nspwYsKfQT yijqK6oLjwv1VuZ2LTJqOaZRTMiQDan94TlVVy8W/U6f4Tlt2xjYqwxYEmKvatbaZQzscwbCpZJF 9/luRU/oBMjZCI9VXXhcuL/M7VplaIlumbLz56oFS8h5/vcxvksXReeuBW/R7+qpk4zqMQPBeD1n rf9Yzl+bjMhAHHRxUmnF1omGjcfV3+ZGfxH2nRRYapFRyzGNtjuUEXKQP28KWw6x5OWH39OMb6HE Lbb9ioAt9pVUXSsntR7XCKxQRlFfUV14XCPyqv22Fhm1HFNNRi37kCE5an/4O9XpmHBftW0dn8Lu quGwfcZASgY2P+0spXSTZQwYA8ZAJAa4qOvezXpmJbZzAuRkpHCMJUMy81SpXsfl95f5XYkcyQzP rboUONCrHDbJDbFoW3j1PcbngQN6SGdxsE1ywsHd9OyK8rG8fT5GESeqO+DAtLfHnXuo6AcZbwrv yljHUvCA07J5qbU/bmfzZeASFrU/PKfqdEy4L8Y2crZaVp4Si9ormfm2ql7H5feX9Z0+qz4gmeG5 lWRs98+GiJvIE5atrlnCF5sT2q8xVDI3cvJ2zBN34f6yt4Wl6LzCB47YWDi/rj/VsFTDm/8d5+P2 DfyVFLdW5uXbd2OgWQzE9Uaa1SqTawwYA/ueAZw4XmfG31YOXRFJOA9y8Ip+J4eH15bFLpKx1XJk YaGNsYueBF6ERTwJb0ws6AYd0Xa1X/L4Lnx6f7v2xfg8dKjDn3a54DkLFWyr3ik+mCDgFvfSRdhe 1XU4vDwsM2ZBNwcOtPkbMvL6Qa6wHHJvYYhd1AeQmceiIJygop6AYaeYq+knZZ8Fv2xZugjbJCw6 JtxX9nY1TpAlncXGQhDY0VEZz4s4WeU5J67ExiEZyKHtar8XvvaP8Im7cF/Z28iAmyIswqbxuGzZ +fN1d3f7KvXPcL/sG96qJQjC3+x0m7FC44V0EZ5LvNSjH37DObmO69zhOW3bGNirDMT1RvYqa9Yu Y8AYaHkG5AwANNyuBbje/130cCkekoaj09fXV8upGjoGxwunpOjBbNTRLh3TkKAafsx74SlFnKhO x9Rwuh0fgtNLm3EA9UYInQw+hEU61L4YnziN4Fl1zzfIO6Q8tBKH+ZALbg52xF902N1TcdKL+srS 68qbKFIkq7CN7i6HBV1sepDfqtOZe6CeO6aLYyIX+iN4ZCuhOPoJOoOTFMGl+mO+z4JJdSnGFOQJ S9Gsa0os2DFJINks2FSkM47BxmIX5FCK7Oe16yuUnt4e/xnzH/oiWBjL8gF3WJeirygoBkdeR9Rh P+BVsi8mL+qzeRzIVB28xU5CIE/XOcmlToX+AwYdo/rtPtGtFWNgvzFgSYj9pnFrrzFgDGzLwOHD h/0xeScdRwEnHecsRUCHgzc0NFQYcOtd5exPUZBDwKBbXEKZvM4Px2twcDCsjrKNHPQjXYRCcIzB h27qmYkKz1HPNishkLWy4pz0XMBNsA3GgcF+d8q4S7jB3N/n5DhuXr7c/BpO6ihDg2n6ysDggJen 1x76L+4f3s6B405iprs7/uodgleCF2w2HzBg2/QXgjn6dewCDhKKetVgKA8sqewHubLTIlsGH326 3iAqbE+t24xv8I8elPzQb4VNY7HqY32iH/qL5IZyhG14aDisjrbNWMvYkX9NqniCNwXl0UC4EyOn p6fHJ2by9gMnJCHQT4ok0cDAgLcRbCWf8EVn2HDKayG8V/MP6rUf9G3FGNhvDMS/8u43Rq29xoAx sOsZOHr0qJ+Nmpyc3BC8TExMeCf12LFjSQIXHKsTJ0545+vhw4frvOKw3Lt3zwcLYE1RcAKPHDmS PX/+fINzTDJkenra40wxO0db4Z9ACf2EztvU1JQPaM6dO+c5i80LtzacOnXKO8Uzj2bWxYFpemra J0KOHTu+Xh9zg1nA8bGxbH5+bkOgi4P+7OlTHyz09cdfvUMbx8fHvf3MzMw4/bx9JeizZ8+yN26F yKlTJ5MELmA5e/YsH5v6CvZEkHXmzBm/P/Y/zBbTb7Ef/lQINB89euT1kyqIwo6xZ+w2DOjQF4EV fRpuYhfGN2Qxe5wf37BtbJz9KQqyRkdHM8az+bn5dZHw8eTx42xkZCTJSggEox/w3L17d8NqCPoJ wf/58+eTjG9gYSylj6APFX0nqZbKfkgqcy186sayJ0+eCIrvO4+dfkiGjLnxL0WhL2A/6CdMzmA/ jLfYeYrVISnaajKMgZgMxF8jGhO9ndsYMAaMgS0YCIPTLQ7Zshrn6tKlS9n169ezb775xjunOBsk IU6ePJkdP54msAQgzg6BFEkHHGK+E8wRvFy4cCHJbSEiioAAuV999VVG8oMgAucUBxBemM1NUQiQ 3nnnnezKlSteP8PDw975IwkBDoKJVIXEyyknc9IFtStuRreruyd7/ux5tuxWR5w7dzZZsE176ZdL ro/cunXLc3DAzezOTM/4We3Tp09zSJJCPyA4uXPnTnb16rUM/Sy5QHPWOenHjh5x/SXNigwai71c vHhxPWBgFQD2Mz8/7+sJ9lIV+iZBytWrV32ihsCKPstsM/05xYwybcVuGTuwny+++MJjUSKAQI8k UqrC2MG48uDBAx/QMYNMUAlPBNspVjSprQSPjLF37t5xfXYkaz+I/Uz7PnTi+AkdFv2TfvDee+9l X375pdcRHIXXn5T6YcUFfeX27dt+fEc/JAJI1ly+fDlJskqEYz/005s3b3r5XAdIXmHDqRLPYME/ wF7xDUL/gGshY3BK/0Dc2KcxsBsZaHOOuq0B2o2aM8zGgDFQlYGPPvoo+/jjj/0xH374Yfb+++9X Pb5oJ44wsy4E3jgeOIMEnamCbWFimJ6bm/PBE04Ys95gSRlACQtLgpmR07JynFSwENikLgQMClhw 3Heqn5W5iYah45Qzw02w0Omc4iF3S8KhQ50Nn7feE9BXfJC94GZz3dW9t6836+/tc8+liD+znccK F4+fPPaJkQPtB7NB11d6nf3spBzo33niD05IOhA8gQm7ITHSjNlKZpHps+ABF0kSVkCkSkCE3DOr ji0zziEfHNjzbh7fGrVl9IMdY8/oh6AbHdWrn7ZOx+MhbsXaeWGsn52dbbp+aAHXQLD48c2t6mGs 5TqUunD9YXxDR+iH/kq/5fqcupRlP9jfBx984JMrtOWTTz7x/KZuj8kzBlIy4K4zbemtNmULTZYx YAwYAw0wQLDC7FizC0EBjjB/zS444yln4qq1V0vcqx2Tah/BCn/NLvQVHFn+ml0IDMbH0s2qb9Ve OCFY4a/ZhWQdK3VSrtbZqs2sMrDxbSM76KdV7IeZ/lbQDwyRcEh1a8xGjWz8xvWHBCJ/zS6tYj/N 5sHkGwM7ZSD91NVOkdrvjAFjwBgwBowBY8AYMAaMAWPAGDAGjAFjYFczYEmIXa0+A28MGAPGgDFg DBgDxoAxYAwYA8aAMWAM7B4GLAmxe3RlSI0BY8AYMAaMAWPAGDAGjAFjwBgwBoyBXc2AJSF2tfoM vDFgDBgDxoAxYAwYA8aAMWAMGAPGgDGwexiwJMTu0ZUhNQaMgToY4GF0KuG26uzTGDAGjAFjwBgw BowBY8AYMAbSM2BJiPScm0RjwBhIwECYeGjG6yMTNNFEGAPGgDFgDBgDxsAuZ0D+ivkqu1yRBr8u BuwVnXXRZQcbA8bAbmFgcXExm5yc9O8S5x3nVowBY8AYMAaMAWPAGGglBlZWVrKZmZlsYmIiw1d5 8+ZNK8EzLMZANAYsCRGNWjuxMWAMNJOBnp6e9Xes8z5vK8aAMWAMGAPGgDFgDLQSA6x+GB8fz54/ f54NDQ1lBw9aaNZK+jEs8Riw2zHicWtnNgaMgSYysLq62kTpJtoYMAaMAWPAGDAGjIHtGTB/ZXuO 7Ii9x4AlIfaeTq1FxoAxYAwYA8aAMWAMGAPGgDGwCxiwJMQuUJJBLJ0BW/NTOqV2QmPAGNgrDOAY PH36NJufn886Ozuzw4cPZx0dHU1pHveJPnr0KFtaWsr6+vr8sk09zCo1IPh49uxZxjLSwcHBjFtf mlHQz+PHj7OFhQWPASzN1A995c2b11lPd2/WP9CfNUs/3FeMftoOtGUD/QNZd3d3M9Tjn8fCEuPF F4vZoUOdDkt/U/UT2g99pVkPgeN5NU+ePPH9Y2BgwNtzMxSE/dBP5ubm/PgGJ4xzzSiMb7Ozs9nr 16+zftdPwNJM+6HfIr+/rz/r7mme/dBPGN/QC0v1m3Vr4fLystdPK1x/GN+wZfRDP+F62KyCfuQf NFM/zWq/yTUGGmHAkhCNsGe/NQaMgZZloNGZBRy/K1eu+PszR0dHvSN4586d7OjRo9nJkyeTOsjT 09PZrVu3spGRER9s88DN27dvZxcuXMgIYlIVHqAFJ3ADDwQMfCdoOHv2bNJ7WQkSrl+/7oNaeMEZ RD8nTpzIjhw5klY/7qFiDycms4HBSsA/NT2VTUxO+H6S0kFedfq5c/euCyrns9HRkWx1edX3m66u 7uz0qZPZwYQJNALtuw4LgQLJu8WF+WzywUQ2fnQ8Gx8bc1327St0Y/ffhw8f+r4BDvQxNTXlecF+ CGJSFeznxo0bvq/SR9vb27OrV6/6JNH58+eTJgBevnyZffvttz4Rw/hGIIW+uDf91KlTSe1H+hlz /aKrqyu7f/++19e5c+eSjm+rqyt+XJ2fW8hGx0Z9t7hx84ZP5B0/fiy5/dA3sB94IVGk8Y2xN2UC jesPsgmye3t7M/TF9ejixYtJ9YNCrl275icG6KfYD/ZEYgZbTpmgwX6++eYbj4HrD+MdnBw7diy5 /aQav0yOMVA2A5aEKJtRO58xYAzsegaY9SHAJWh47733fHBNUgMnB0edWSmcoBSFgB/Hi6AFRwun FCf0q6++8gHM97///WTBC4kPgv3vfve768Ebgd1nn33mnTESEeCLXZg1xeHDEQeLHHLwySkdHh6O DcOf/8WLl9m9u/ecTo74BAiVBA1Xr1z1jvu77767ji82oImHk35lSBhco5+vvvo6O+gc9tNnTseG sHb+Vc8Js6Yb7Kdj2tezKmIoUfCv4ACdyH6wJZJn/P3whz9MZj8EcwRwcELgQiEB8Omnn/pxhaAu hf0wvmEn6CdsPwkaxhoSAanGN1ZigAWdkIihIJsAjyAcfKlWN01Pz7jx7anvJ0pODQ0OZd+4ZM2K S1AwvqUoXGvom1x/fvCDH6y3nxVNN2/e9PpJNb5hP/QJ7Ef9E/2gGzD++Mc/TpZ8fvDggU8gXr58 2eNBF9gR1x+uB9SnKNgP1xrs55d+6ZfW/QP0k9p+UrTXZBgDsRiwZ0LEYtbOawwYA7uWAZZ6MtN+ /PjxdQeL4IBZDhxiZuoaXWlRCznIwPHCGT19+vR6gELQzXeWpbIqIkXBGSWIYjZMDjpyWQWBUwoO ZodSFGShn/yMoPSFg5iioJ/79++5JEObDyYlE/2MjY95/TDrnqKwKuXh5EM/0x+ujiGgHBk5nE3P TLvZujSvqn306HE2Nz/nA0s96R37YXa5vf1g9iCR/cA7wQn2Q99QgM8MKrZMEMFr8VIUZJHAxHaU gEAu+iERgW0RkKcoJBL5I/APb7/Ajkm0Mr4RaMUuFfu57xOYjGcq9Bn0lXJ8W3r1yo9hPT29G8Y3 bsUYOjyYzbprwquXrwQx6id9gZUprOoKEzDSD0mbFNcfGslYyngW2o/088pxlsp+6I9ajRHaD4E/ CRJu5aFPpyjcdscrNVkxFI5v6EcreVLYT4q2mgxjICYDloSIya6d2xgwBnYlAwoc8/fS44zhtBOQ 44DFLnoOBEFTfqkpy2IpKR0vAt08J2DgmRA4XSmDKHSRfxYFPIGP1SMp9ENgyTJpsBw6tPFZITij bW0HkuqHwIRbLxRsoxtKd3flmR3Pn6cJcnHSKXAQFt+PXRIP3RBgxi4kH7APbCdvP/QTAgiwpggY sA3aXXR7DliENTYnnJ8kK31FY0gokzrsJ4V+GN/gZSv9CGuIL9b2vGszeHoKnv/Q1en6seNr9tFs LPEbzot+KEXjG1yR7CVJEbvAB8leEiF5++E74x4JkxQJEXBgp9gKcsOS2n50zUVuWBh35R+A14ox YAxUZ2CjJVc/1vYaA8aAMbAvGJCDp1mOsNHU4XThqMcuBAI4glvhwOkBBwFM7EKwTck7gNQRXFLE m/8S6R/4wAkHh+RKFHwIH4mi2IWgEu6RmQ/8qWOFBImbZYc5dnmx1t6DByu6COWp/ywuxOcE2/Dc B7rYgKXjoOfsxcv4SQhsAzz5fgIe6tAZNpYiCaH+WIRF+kkxptB2yQln2amnCIvsvVIb519xL5sN pUg/0mG4L8b2wrr9bL5LmdU7lBcJVhIxXmh8y/NCf1X/UX+KwYXOCQ7G28pYtjFckH5IxPIXuyjB necEueKE60+KhIhsQ7YStp26VP5BKNe2jYHdyMDGUWU3tsAwGwPGgDFQIgMEJDhelCKHR8Fmipl2 zUbKyco3Eyw4PDiLsYtkFGERTzomJhaCfhx1iuRKXuikizvti/H5eqmCgxUP+QctgoV68L55E39p +6u1QCDPCe0+cKCSmHi19Mr1l7gJKxIuy8sr/rGTspWQe+FbehU/cJGNIjOPRXXYe4okhGxD7Q85 kU3pmHBfjG3ZhuSGMlSXAotwFHECJnTG+CY9hjjL3EbG0qvKGFqERXWysTJl58/F2LbV+Max61gS rMSDd7hBZt5+9J3xLUUSQkkXtT/kTfjgDbwxC2NFNf2Il9h9NmYb7dzGQCoGLAmRimmTYwwYA7uC AZyqakVOhhyRasc2uk9YJDN/PtU3G4twpAjmcDLhRTLznOi7Ekn6HuNzeaWSXNgKi4uhvFOs42Jg 0Dmr9RVwUFZW4C6uk77i9FNJdBQ/oFRcpegrkiGZFRY2/5uir1TXT4Ur4d2MsNwaYSk6q7hKgUU4 JDOPR/Xxsby1C8kMsch+lpfjr2jabnwTvhR9NuRdcsWLvguv6mN9CovkhnJUR39Snwr3l7nN+asl OoRFeMuUbecyBvYaA5aE2GsatfYYA8aAZ0DOAF+KZk+2omk7J0bnreaIbHXueutrlbEd5nrlFh1f C5bUOKSLIrwpsEiGgpQiHPAWeXLOiUVGLcmFWo8rakltdcIBJ9X0o+NqO+vOjpJ+in4dYms2FuGr hlfHNPoZtjXkIH/e1Fjy8sPvsbFgOuu8FOfO1uGsH7deU+4G55eMavpJEeSGvBdhUV14XLlsvD1b Le0Vb29/Vf4Wba0mZ6ec6Hf6LB+5ndEYaD0GLAnRejoxRMaAMVACA1pWzKmK7t3cSsR2ToAckO2O 2+r89dTXKqOeJEs98us9tla89Z53q+Oli6L9KbBIRvX4n9syihCWWcezMLYXIrxlSs6fCxn8V+Fk 68RIKix5fPoe9p1mYxGm1DhCDoRBnynGlFrbmwbLWsu37rJk1aLbMpyIl2brJ+S9CIvqwuPUf8r+ TCGjFsyhfoqOFyfSYdExRXVqH5/1/rbofFZnDOwGBiwJsRu0ZBiNAWOgbgZwBvTQrFpmUSRgu4SF nIztjtP5GvmUDMnMn0v1Oi6/v8zvkiGZ4blVp2PCfWVv46Txh0zJLZJR9NC9ouMaqVOiayscBOIk B9rXnsnQiKztfqvnPhRhUR3PqGhvj3vZP8DzMXzWBf1sRi0s4m7zEeXVqA9I5lZnToFFtlGERXU6 ZiucZdWrvZIbnld1OibcV/a22iuZ+fOrPjYWgr7q9lNBVrHj7ZN9+XbU8z0c34p+J07EXdExZdVJ hmSG51Vdhbu4Ywpyq9mysNBP4C9mgZNqSQJhEd5asHA+bq/BX0nxXItaMNkxxkAKBuJaa4oWmAxj wBgwBgoYwBnQq9/qSULgxMiBKPqdlp7yKq7YRa84LLr/l/aBBQcm/yq3GLiqYRFPOiaGfJ2zmn7g JCUW+hf8q08II5+VZbsV/bQXvLEiPLaMbfXHoudPiJMO92aKykM0y5BYfA7aSqJD/TN/lLAIb35/ md8lo6KLjRkRcFBP4CJ7L1N2/lzCovaH+2XfOibcF2Nbdiq5oQzVpcAiHEWcqP9gX7GxIONQ5yFP w8rac15CTvQw1861Y8J9ZW8zpqg/wkG+iKvYnCBXr+GULkIswhaOx+H+srd1jdt6rF31vKHLmIX2 wgtFugjlCV89+oFLxiHxHZ7Pto2BvcyAJSH2snatbcbAPmZATtJOKOjr6/M/k0MenoOZChwdOUXh vrK3cWRwSLfCgTzeVR7b8UKO2lvkeAmfjuH4WAVnTe+Fl1zJQueq6+3tVXW0z66uTu88EqTkgxe+ 45B2Hup0twNtfm1m2aD0znq9sSM8vx5cmkI/9MXuri4vfvNbQVb9TB/H1OOkh22pZ1u2QZ9QcKDf U0d/IRDWjK/2xfhUf5QuQhmq6+/vD6ujbdeCRcdEA+FOTB/AnquNKdJhTBycu6e7x4sotJ83lbfg pLAf+EAO/TXPC/1VdSn0I9tApuRKD7Iprk8pbFnXZNmKcPCpOnhLeS3UtSaPBQz16qcRfyWUb9vG wG5iwJIQu0lbhtUYMAaSMHDkyBEvh/eOhwWng1dv4RBpFi/cX/Y2Dt7w8LB3SPWKMsl4/vy53xwf H1dV1M/R0VEfrOU5QSjvTSeQO3z4cFQMnBwHDyw46fnXCOKMwtPAwEASx/iQSzDQF8CytPa6ThHw 8mXl9XbDI8OgVnW0z6GhIb8UmfbnHVrq4G1keCSa/PDEIyOj/paMFy8Xw2rPEYkJAr6enu4N+2J8 IaAbGRnxNpt/ZR79mMAK3lIELvRJAmrsNq8fsGA/YElRsB8KdhsW+jF1YK03iArPU+s24xttZlzN j28LCwv+NBqLaz3nTo8bHBz0OlhcfLHpFC9cHSuIxtZ423RAyRXSj8Z4nR6eWLKPblIkrNQnsZ28 /YCD/nLs2LEk9sN1EDz0E+SGhVe9skIh1bVwbGzMtzl/LQz9AyVNQpzVtvNjQrVjbZ8xsFcYsCTE XtGktcMYMAZKYwCHB+d4ampqg/N179497zCfOHEi+r2nNIbg6OTJk341xMTExHrwQvD04MEDHwCn crxIuoCF4ODp06frXD979ix78uRJdvr0aR9kre+IuIETSKAEJzjDKpOTk56zc+fOqSr65/Hjx51T vJo9nHro9FNxjnGSpx5OeT7AmqLgoBMQ4JCHwQtOOzoCR6dbuZGiDB0eygb6+rNHs4+yNy4xRMHJ fvz4cfbGvebw1OlTTk9p3I8zZ8744IW+IUcf+3n48KFPVJ06dSoFJX6pNeMG+pmdnV2XSSCD/RB4 EginKCQL6Q8zMzOb7Ie+Cyf0p9iF8Q1OKPnxDf0QyNGnUxTae/TY0ezFyxfZ40eP10XOzy94+xkf H8s61pbhr++MtIF+GN8Y48NEK32YQPfs2bORJG8+LeM6AT5YVMDAtZAVEIx/KQp9BVvGXkL7kT1x HUyRmKGt6AZ7pc+GyRn6LN/p0ynsJwXvJsMYiMlA/KtMTPR2bmPAGDAGtmBAAccWu6tWM4P6zjvv ZNevX8++/fZbHxzgXBCAE+BqpqrqSUrayRLTixcvZjdv3lyfBdMMJvUplsKqKTicOMVXrlxZ54AA F6cr1YwlWHDwLly44PXz9ddf+xUY4EI/1OMkpiro5+zZMxkBwq2bt1yg35UtuMCl41CHS8ycSpKs UlvRwes3r7O7d+9mJAJcGit77vTDagAFezo25icBw+kzp7P79+9l15wNoQ/s58WLxey0C3D7+uLf KqP2oZ/Lly9nt27d8rbMdyVpqGc2PlU5evSo5+H27dseA8EdWEgInHWBJbylKMhlHLt27Vr2xRdf +P5BMo8Aj/pUKzJoK30DmXfu3PHL6tEPYwpj8Pnz5/1nCk6QccQFsq/cCiYC7rn5OW+76Ae7Orq2 Oi4FFtp+6dKl7MaNG36sJTnFKi+wML6RJE9VWL0DFsYUxlpWYXD9IVn17rvvJh3fSEjRT0mAgAGe SGzCBwmKVPaDXNmP/APZD302pX+Qqh+YHGMgBgNtzlHf/OSbGJLsnMaAMWAMJGTgo48+yj7++GMv 8cMPP8zef//9uqXjaOEQM9tC4IszmDLoDwHjhIIFZwcMBAo4Q80oBPsK5JitTDUDlW8rly9WZaAf gkn0o4eG5Y+t9n1lbqLa7pr2oR8cY2YJ0c/AQH+y2f48QFY/oCNKb29fklsf8hj4jn7oJwR2Bzva XfKhf0f64VwH+hubccVusB/0RGBF8NtM+6GvwI+W1qcKoOBSBflwQn9hfMOO4aYZBb2wIgT7YdUV trwT/ZRhy4wnWmrf29Ob9fRWnhdRDy9tnc7+DzX2jA+NbyRY4YIxfyfjWz24tzqW1UOPHrmVTU4/ 9BH0QzKrGYX+yrgCP/RZbKgZ9kPb6bPyDxhTdnKbJvr94IMP/EQDOv7kk0+S3NrYDN2ZTGNADDib bbOVEGLDPo0BY8AYyDGAk8XS2BTPOsiJ3vSVILtVZlhw+vhrdsHxbCX9pJyhrMY9s8n8NbugH4KV LM1dBlWbS/CW6taYqkDczlayH4Ie/ppdGN9S3Vq2XVsJspuVjAmxaXwL65q1TRKkVfTTKuMbumiV 60+z+oXJNQYaYaA5acxGENtvjQFjwBgwBowBY8AYMAaMAWPAGDAGjAFjYFcyYEmIXak2A20MGAPG gDFgDBgDxoAxYAwYA8aAMWAM7D4GLAmx+3RmiI0BY8AYMAaMAWPAGDAGjAFjwBgwBoyBXcmAJSF2 pdoMtDFgDBgDxoAxYAwYA8aAMWAMGAPGgDGw+xiwJMTu05khNgaMgRoY4KFeKuG26uzTGDAGjAFj wBgwBoyBZjMgH0WfzcZj8o2BFAzY2zFSsGwyjAFjoCkM8No3il3Ym0K/CTUGjAFjwBgwBoyBbRjg 1af4K3xaMQb2CwOWhNgvmrZ2GgP7jIGlpaVsbm7Ot5ptK8aAMWAMGAPGgDFgDLQSA6urq9nCwoL3 Vw4ePJitrKy0EjzDYgxEY8CSENGotRMbA8ZAMxno7OzMhoeHPQQu7FaMAWPAGDAGjAFjwBhoJQZY qTk4OOj9laGhoezAAbtTvpX0Y1jiMWA9PR63dmZjwBhoIgPMLlgxBowBY8AYMAaMAWOglRkwf6WV tWPYYjFgSYhYzNp5jQFjwBgwBowBY8AYMAaMAWPAGDAGjAFjYAMDtkZ5Ax32xRgwBvYKA2XMLLx8 +TKbmZnJ+Dx06JBfMjkwMNCU5ZJPnjzJnj596h9c1d3dnY2OjmZdXV3J1bW8vJxNT0/7e1jb29uz vr6+bGRkpCmcvHjxIpuamvIP9IKL/v7+jOWszSjPnj3Lnj9/7u/n7ers8jg6uzqTQ0E/9BXuMWZZ L31lePiw225PjoVnsTx+/Hjdfugr/f197kGx6ec/0M2jR4/W7YdbtXp6epJzgn4YU+bn5/0Dc2U/ zbhl7NWrV96WNb5hP7qFLTUx9Fn+uB+ePsuY0ozxbXV1xfWTx95+WCbf29vrbHkwa29P7y4zvtFX 0BPXH/Rz+PDh5A9a5lqK/WDL9F/0A45m2A/9AztmvGV8AwPXwmbbT0dHh/cPuK3CbqdIPXqYvN3K QHpPYLcyZbiNAWNgXzGAk/PVV1/5YAGHi2dMXLt2Lbt7925WRoKjHjLv37+fXb161Ts3OFwEd19+ +aV3xuo5T6PH8uRu5JKEIBmDUzw5OZl9/fXXPhHQ6Pnr+T36+eKLL3yAS+CE43fjxo3s9u3byfUz +fChk3vHYxhyTujLVy+z6zeu+6CqnjY1eiz6uXXrlk/MEDz19fb5PnL16nXXZ141evq6fs9DYa9e uertZ2hwyAVx7dmdO3ez+/cfuPOkvVUK+/nmm288foJb7Ifvs7OzdbWp0YPRD3Y8MTGxFtwOeV2B hURAykKSCvshGaLxjb7DGJd6fGNMhRcCOfQDF4y9BJspy8rKcnb92vVsemraB7cDbnwjCXDz5i3f Z1JiQS+MtYuLi14/JCFu3rzp/1I/uJAxnj7KGEtfITnCmM91IGUhAXLlyhU3jtzx9sM1CBumr9Cf UxbpR/ZDwgzbwYZS20/KdpssY6BMBtKndstEb+cyBowBYyACAwQLOBQkHr7zne+szzwx2/Ltt996 Z/nEiRMRJG8+JcE2gfW5c+cyycQR5HVewshsauyC40uQjwP64x//2HODTBzBn//85945vnTp0jpX MfEwM4gzyozc5cuX12WiHzjBITx69GhMCOvnRj+jH8rEAABAAElEQVQPXJB7+vTpbHx83NcPutUY 165d9Xrrd7o56IKr+GU1u3fvnn/C+rvvvrs+i9znVh7gpN+6dSe7dOmdda5i4qGv3Lp1Ozvk2n3x 4sV1UdjT9evXPbaxsbH1+pgbrB7Cfs6cOZOdOnXKiyJpRdBLX6EPkbBJUUiGEFj/5Cc/WZ9FZub0 F7/4hcfyve99L4l+NHaQGMJ+NHMLF59//rnHprEmNi8PXQKPfnv27NkN4xv2jX4Y2+g3sQuB4927 zn7mF9yYf3ldPyRav/jyK9+HsB/3wufYUPxqHRIQ6IMxVfqBB2yZ2f9jx45Fx4EAxjeSH9gONkTB frBjbIj+m0I/yAUHqzF+8IMf+AQ4dax8+/TTT/21CfsRV+yLVeQfsFIm9A/QF4k9rj+p7CdWG+28 xkAKBmwlRAqWTYYxYAzsKgZY4k+wTaCEo6HCLB2OBo4zjkjsgmPMrCmfR44c2SDu+PHjPhEB1hSF mSYCKJy+0Olkho7gmxkpOEtR4IREBJyE+tGMLsFeioJeHjx44Gf5kR2W0dExv3SZVRIpyqtXS9kj 56ATtIXL2JldPuxux5ibe+7+5lNAyWZmZ/zKi5HRkQ3yCOjoL/TZVLO5JCDoI/mkB30W/THLm6LQ X+m32E+4jJ3EGUEdwR4BVoqCHGZw4SQM2ggowQYnrBaJXeCfsZRkSH58Qz8kS7CvFAX9cBtGX1/v Bv20O/0cPjzkb0dYXEgzvrHCgLYzxof6IeHL9SflajwSRPRRVuCpYE9KuKYaa7neMm5gP2HSHWz0 Y1ZepVo5w7UO+0E/4fVH+qFPp7Af6cM+jYHdyoAlIXar5gy3MWAMRGNAy7Rx+MKCw0EAzhJZnNbY BRncJ42jxV9YFMiwP0VBDsthwwSE5BL0ElSmCKIIXAjYCFzCYBsscAQ+kiEplrcjgz+wHDy48ZkL BNs8+4B7qVMUZvxdVL2JE2R3d1X68fz8XAoo2dMnDosrcBCWCk8Hve2kSFgRuBAsgCNvP/QdAjz6 Ev06dkE/yJHdhvLAQr9OZcvYKfLy4xuYwIduGONiF8Y35BTpR+NMijGFdhLE8jyIIk7Ujx8/SZck 0rUm1AH2Q1KRABe8sYvkIFMcSCbfsR8C/xQJRWxDfTYM/MEDFjBgyymK5OT7inRGn2bcsWIMGAPV GbAkRHV+bK8xYAzsQwbkQISzUKIBR5CS4h5UggECl3wAhXxw4PToGOpiFrVX7Q9lCZ+OCfeVvQ0f BC+0vQiL6qTDsuWH58NJx/kFS/5hi/SddvdHIPzGzWrGLgrqD671z1CeOFlcjD+Tu+r48Akgx0mR /XS4RBHBxGKCVTMEA8hS/8xzgt7AmmJVk4L6IiwEeRQdE+KMsS3bKMKiuhS2rLFL/TNsK3UpxzfZ z1ZYwJZCP4wpW41v8CGbSqEfbIPxFpnIDovqsB3wxi5Kuqj9oTzpDE6w99hF9lMNi46JjcXObwzs ZgYsCbGbtWfYjQFjoHQGcLo0s1PkZMgZY7ls7IJDSinCQb2w+KCPioilGhbh0DERYXinWNzneQGH 6lJgeb1U6QP5BATt95w4PMvLK/4vJiecW5y0uYAhX/RmjNevl5yTvpLfXer3N85+CAQIWdQvQgHC B5bYRX0AHHks9BPqsHVsPnYJseRlpeyzyBYWBW8hnpRYhCOvG+FRvfq26sv+pL+qP6r9oQzZ9+vX 8W/BI6hXe9X+EIvwibtwX9nb4PC2HIyrkiFs2E+KJJ6ucWq/cPAJFv7Aq2t3uL/MbcYKtXcrLMhL oZ8y22XnMgaawcBmb6UZKEymMWAMGAMRGZDDVIuI7QISnUuOYi3n3OkxwiKZ+fOoPoXDUw2LcMg5 y+Ms8ztOsRzjaudNop/VSvCq9ufxOL/YY112T92PW1bXne8iLOCgrKy8Pa5SU/6/q04G+nGhQeHJ hW/FJWdil2p9NpSt48K6srclQ+0Pz6+6VPZTLVBLiaUaJ/AjLLFtmf5KsjCU6b+s/SP7WV6On4RA N9XGN3GSoq9UkyEcFe5ij2/ZeuAf6kXbwiLuVB/jU312q3MLy3bHbfV7qzcG9hMDloTYT9q2thoD +4gBOQM0OdzejoJKALX1UfWca+uz7M0923FXdqur6SIFFgJuioKU4vYRlBfvKbO2tvbGx+Ik+GbB SbP1U0uwDdjauGtMW7XIqIa3MenFv262fopRNbm2OHe2DqoWPa4fvMMNyWi2foSDZhRhUV143A6b vO3PUtvGVoBoa7X27pQT/U6fW8m3emNgLzGw8Ulne6ll1hZjwBjY1wwwi6MHvVWb0cmTVLTEMjxG Dsh2x4W/2el2rTJqPW6nOPhdLc5R0fLuRmQW/TbEIV0UHZeCE8molmQAb/UkRRH6euu4DWWb6Mmd soJl++PqlR4e72W4VRAr2zjr4i78bdnb1WSEfSfsU2Vj0PmqYdEx+9V+1P6tPmvhbqvf1lK/IWFW LWHobTmB/Tg59M+wj+bbEZsT5IUywJK3E+ELj8vjLOt7LbaRx1eW7PA8tLWanJ1ywjMv8Fc4v84R yrVtY2AvMmArIfaiVq1NxoAx4Bng6dV68nytlOjBbBxf5AyoLjyu1nPXe5xkbDULpPr8k8vrlVPL 8Xp4nmSGv0nJCQ4gDikyJbcIi/CG+8relmNc9JyFCr5K4N9+YOObM8rGwfkOtlfmFAj+84XbMCgH HG+xAwZ/foKotRURm7FUlr2rb+f3l/ldfaConyBH9TquTNn5c6m9khnul03pmHBfjG21t9lY1F61 P99W1eu4/P7yvrsxZe3tNkWcqK7ooa/lYaiciTFF41vRudex5N6WVHRso3VhP5FcnVPfGY9jjynI 1DVOcoWDT9XBW2ws9EUlISS3CIu4C/dV26Z9+Cp8Fp232m9tnzGwWxmwJMRu1ZzhNgaMgaoM4Cxw UScRIaeh6g/WdnKsHB45weHvVKdXyIX7yt4GP6Xo/lIcFf7Aq+PKlh+eT5wUYVFdChw4meiWtkuu cFIn/aTAAifwj0xkh4XvJCfA294RPwlxyL2alLLiVgDly8pKpe6QewtDPbaQP08t33lVaXu7cy0K 9MPvpTP1p1rOudNjZKPIzOuH1VHU0ZeUTNqpnFp+p/aq/eFvVCe84b4Y25JTtEIsJRbZqGw2bGvF firjm7gL95e57cf8jsrrZNX+8PzLzr4p4i3cV/Y2bdX4VsSL8KXCwvglXYRt1ZiH7cTWD3L1Oky1 P8RCHRgJ/GMnIXxfcTqiFOlHdfXoR9hpI3+x2xByZ9vGQDMZsCREM9k32caAMRCNgdAxKnJcqgke HBz0u4se+MjryHC8ent7q52ilH04Mj09Pf5J2zgqYdFr0QYGBqIHlsjt7+/34oseEqc6sMQuOOh9 fX3e6ZRcycQBRGc4isKrfTE+e5zD2O5WIKAbOZ+SQ5/jj2O0SkH7YnyqPy4VvLVl6VXlLR7wFr+0 ef0gJx/kwhM6w8lWABoTD5zQX+gTef2AAzw4/RwTuwwNDXkRsttQnuo07oT7YmyrH0huKIO3EGA/ KWyZ8Q3upYsQh8beVOObOJHcEItesdvXH99+sA3aTN/M2w99WNcy9acQZ9nbXHsI7NGP5EqGknjY cXit1f6yP9VesOSL6qTD/P6yvwtLUV/Bpur1D9C1+E2RSCmbDzufMbBTBiwJsVPm7HfGgDHQ0gyE gUXemdsO+NGjR70jnn8XO04HTgYBboogijYMDw97uHpPurA/ffrUb46Pj6sq6ufIyIhvM+8/x2kK C/ey4oimCqJGR0e9+Py72NHP4uJihpMY6j/EWuZ2u9PP0NCgdyD1CjmdX31n2PGWotAnaTNy5dAi l9UY8wvzrj8fyA4fPpwCSub14wJZcSCh2A4BA8FCCvtB7tjYmA/mXrx4IRj+8/nz554n7IugO3Yh IUK7nz17tsl+6McEH6n0Aye0GQ7CwjgJFuyHADR2ob/SV5CbH99k20eOHIkNw58f+yEpMj+/4G0m FLrobIrAcmQ4jS0z1lL0TCNhwXYYZ9BPPTPt+n29n/QR+iRypQ+dg3GWpMixY8eS2I/0w5iipANY uBbRd+hLKa+FcJPvs6F/AN5aC23QNZV+ZishamXOjtvtDFgSYrdr0PAbA8ZAIQM49SphUKa6ap8E 0ziC09PT64EUDtfdu3e9w3X69OlqPy9tH47O8ePHfcB279699ZkxHNGJiQnvjMphLU3oFiciyXDq 1CnPx9TUlHea4GRmZsY7qGfOnEkyIwY8AheCx4cPH3o8OHA4pg8ePPDB3Llz57ZoRfnVOOE4juiD GVOwvHr5KpudmfVJGSWRype88Yw44SdPnvRByuPHjx0ObhFZcYHMU8fRYnbixHHPzcZfxflGsH3Y JWceP37ik0IuVPABP/3mgEuGnDx5KkngQuvghITH/fv31+2HZAh9h2AO+0pR0A/jBoEKfUWBB7oi MXHixIkkq6toK+2m3z569MgnIsBCIgD7YcwBJ5+xCzJoN2N1OL4R4Eo/JExSFGwYLOhn2tmu7Af9 LDg8JEPQYYqCfmg3HBDooh+uYfRhOGOsTVXOnj3rx3VkM8aCBf3QVwi0UwX+tBss6htce8BCogb7 oT+nSJzBO+MbExWTk5Mb/AM4QU/YTz2JBOmXc9PH6vktv7FiDOxWBtpc5984pbVbW2K4jQFjwBgI GPjpT3+a/d3f/Z2v+au/+qvsj//4j4O922/ilN+4ccM7gThbmunGEUo14y+UOKI3b970M0/MqOKI MRN24cKFZIE/WLhc4IziHGv5P7zgkKWaERMnyL19+7ZPgLB8WfrBAcSJr6eszE3Uc/imY5lRVgCH XhZd0N/X3+t4OZZkxlKA0A+BPgmQru4uH7AsLr5wAc2oC6LG3fd08w7Ly29c0u6ed9J73az6qyUC mBWfFKhnllBtO9C/82RB3n74Dgb6SqoVGWoHgQt9hX5CwMEKDQJOkiUpAn/hIFi6deuWSxQ99sv/ +U4ADiepEpvCwqquO3fu+PGFQJLZbj5JJta71L9RW56cfOiSM7NePyQmsJ9RZz/jTkf1BIdtnf1Z 26HaZ8PFhT4J+Ln+wAVBL7pBRyQgUq2YERbGN/TDNVH6YfwHS0r7YXzDfrj+YD/oB1vm+kMyke+p CkmQ69ev+ySe9EPdTvRDG/BPaBd9/p//+Z/Xn4GRqj0mxxhIzYC73rVZEiI16ybPGDAGkjDw7//+ 79nf/u3fell/+Zd/mX3wwQc7koszyMwpAQMOcT2O6I4EbvEjHDAcUfAwcwiWlEFLCAtni6AfTAqm wv0pt7XEvxFOGg1caC+coBscdXRDf2mWfsDAH/qBF7A0oyDfY3n9xr2Zo81h2bn9NJKEoO2tZj/0 W/oMfQUdNaugH7AwroElZSAXthn9gAM8GlN2Yj9l2DLBPrYMpp1y0mgSQtxozEcvYGn29Qf90F/5 24l+1K5GPsGAfmQ/zRxrwYGO0A84djLWkoT7oz/6o2x2dja7dOlS9o//+I/eBhrhyH5rDLQ6A278 aGuOZ9LqzBg+Y8AY2PUMhEsz8/em19M4OVz1/CbGsTh8OOf8NbvgCIf8NhNPK3HSKlh26gyXrUf6 rNnPZlaxHz3tf/PetDWt1FdSzqpXY5mAslnJmDwuEg/8Nbu00vWnVfosOiljfCPpRQKOgg3sJJHh f2z/GAO7jIF0azN3GTEG1xgwBnY3AyyRVGkkCaFz2KcxYAwYA8aAMWAMGANlMsDKDt1OSHKyVRJg ZbbRzmUMFDFgSYgiVqzOGDAGdj0D4SvmeIaCFWPAGDAGjAFjwBgwBlqJAW7nIBFB2cnzclqpLYbF GKiHAUtC1MOWHWsMGAO7hgFWQuj+WVsJsWvUZkCNAWPAGDAGjIF9w0D42t5w8mTfEGAN3bcMWBJi 36reGm4M7G0GuLdS9xjz9GkeMmbFGDAGjAFjwBgwBoyBVmGAB1OqpH7zluTapzHQDAYsCdEM1k2m MWAMRGeAh3lpVoGZBl6FZ8UYMAaMAWPAGDAGjIFWYYDXjqqkfj2u5NqnMdAMBiwJ0QzWTaYxYAxE Z4A3FWhWYX5+3r9PPLpQE2AMGAPGgDFgDBgDxkCNDDx8+NAfyVsxjh8/XuOv7DBjYPczYEmI3a9D a4ExYAwUMMBKiNHRUb+HJIQ9F6KAJKsyBowBY8AYMAaMgaYxMDMz42XzZowjR440DYcJNgZSM2BJ iNSMmzxjwBhIxsB3vvMdL4sEBLdkWDEGjAFjwBgwBowBY6BVGJienvZQmDixlRCtohXDkYIBS0Kk YNlkGAPGQFMYuHTpkpfL66+05LEpQEyoMWAMGAPGgDFgDBgDOQbkm7Byk7d6WTEG9gsDloTYL5q2 dhoD+5CBU6dOZTwbgnL//v19yIA12RgwBowBY8AYMAZakYHl5eVMD6Z85513sra2tlaEaZiMgSgM HIxyVjupMWAMGAMtwMDY2FjW29ubvXr1KpuYmKgbEc+SYKnk4uJi1t7eng0NDfl7Ng8cSJ+/Bcfj x4+z169f+1ePcu+o3v5Rd8Ma+AFcTk1NZc+fP8/gAX5PnDiR8VCt1IVbbODl5cuXGUtZh4eHM3Se uvD610ePHnlOWHXDq2HBwT2+qcvS0pLHQt/Fne1x+jnq+soB139TF+xmdnbW66fD9Y+BwUGno8PO 0U5vP+gHLPCDfpplP/QPgg7sh37DzCdLsOm/qQv6YRZW4xvjCbacuqysrHjdoCP40b3x/f39qaE4 +a+zmZnZbH5+wcle9ePbkSPjbvxvzvjG8wIY3xhfDx8+nI2PjycPVOmn2A5/BM0k9o8ePZo1Rz9v /PVHr7Xs6enxfbYZ9lOGfyBO6ehnzpxJ3t9NoDHQTAbSewLNbK3JNgaMgX3FgJw2Gn3v3r262k6g /cUXX/hA+/Lly97RIZHxzTffeEe5rpM1cDAO+pUrV7Jbt25lvL7r3Xff9c7fl19+uaPESgNQfLDy i1/8wr9p5MKFC9m5c+f8q0//7//+z+9r5Nz1/pZA7uuvv/bBG8/+wDm/e/du9tVXX/ngrt7z7fT4 1dWV7ObNmz6wpL+dP3/eO+nfOp09nKo89Xyn5673dwQr3377bTY3P5exCui0c2qp+/yLL5M/mPWx CyjptyTvzrt+Mjo65jm6evVatrK8Um/TGjr+2rVr2Y0bN3ySCvsh8Me2GRMIsFIVEiCffvpp9uTJ Ex9wcLsYScX//d//dUHvfCoYXg4JTeyW8QUcJ0+e9EHmZ5995gPNVGCQf/XqVT++kRhCPyQhsOOd JI4bwY1+vv76G6+Lky4ZQ1DIq51V18i56/0tyQc4IPmg6w+r+Rj3SdSkKtgH9sMYR2IVLCSewcZ4 m7KgH/on9nP27NmMlQNwQT+mLmUhCfL55597/0D2wzWJcaUe/dy+fXsd9sWLF9e3bcMY2A8MWBJi P2jZ2mgM7GMGCJYp9SQhCAxwDggqcXY6Ojr8qgMCzLm5uaS3duBcMdsPDpxAnFJmoQi6wZjK+SJY IJBjuSiOKIECfzhO7ANLPc5XI12SwJqkDCtTCBTQjxIABFcpb7158uSp1wFBHBjQD8HUYYdt4sFE 9urlq0aaWsdvV9eD6nNnz/nZfmYHSUYcdImA27fvuIA7TfDPbOkdF6D0aZWM00//QL/HwkNiUyZn COZIKMKD7OfYsWM+qUg/0YxqHUTv+FCCNlYSKRGCfhifmFm+fv16MvsBA0ElrzBGPjiY1WZ8Y4UG OFMlZ0g0oCPGN1YyYT+sxuD+eMaUVG81or137tx1Olj2Y0p3T7fnhUQrY15lTEmTsGIcJfBHP/DC +IZ+CHZZ/aXl+zvuiHX8kJl6rj+nT5/2OgEL9sMYRz+hv6QqcMLYTz8lEUK/RT+sbKI/p7r+4B+Q OEMn6Ef2g35YVVTP9YfzUOj3tMWKMbCfGLAkxH7StrXVGNiHDPzoRz/yra4nOMUxZtYFpzgsBL0E DDhlKRwegns5nHrdqPDgCBLsEWClKDi/OJwKtCUT54mECPzigKUoOHlwQ1AZFpxCEiOpZlC9fiYm nQPZsemBYvDE/lQB9+LCi+yZ0w+z/AQKKmwPDg26Gd1Fp780s+20mb457FbuhIXl/vQXgs4U9oNs ko/cNpS3Zb4TeKayH2yDcQO5oX7ARl8huZkqoUhgSTAHlvAedOyH/gNOAq3YBftAFmNqfnzjO30o lS2jH8a3wcEBH1Sq7eiHfstKlbm5NPbDmE/bSTaHhcCbgJs+zf4U5cGDB76/khBRoc9IX3qoovbF +iRxhn2wGpAxXoXxhH6M/ridJ0XBfsBDIiYs2A76oU+zv5bCyigKiUnaYcUY2E8MWBJiP2nb2moM 7EMGfv3Xf90vCSfg+NnPflYTA3JmcI7DgvNFAIGDgRMfuyCHmVpksqw9LHLEUr16FDkEDcz65AuO F/ymCKKQQ7BAcJDHQh06I4GUYgb15YuX2UunowMH2nxwHfKCcwwegkswxy7PnrtX0Do5+T6LXPRD mZtLM2v5/Nmcl9fRsfE+euznUMchn4BgmXvson4AJ3n7oe9IPykSItgxgaN0EbZddalWZZAwpE9q DAmxUAdvKW4PYXwjeMRW8vpRPwZrisJ4wUqhzs7N45uSRk8SYEEv9ANvK7mxVtcf+lGKFQhc47BT 9JMfa+EE+2HMT5EQQQ7c0C/gISzClsp+dJ1THw2xyD+oRT/wy+1zlF/6pV/KeL6FFWNgPzFgSYj9 pG1rqzGwDxlglv6sWzJJ4XkOtRQcYwpOVr7IWdYx+f1lfsdJwfGSzPDcYMMZI2BIEUQpaCziRHU6 JsRZ9jZtpc20Hec4X4QlhX6WXi/5wAUseccYHPyBd9n9xS6aeWuv0mdJmsQuKyvLTj9uFtBzUmw/ 9OkUSTz1R/SQ1w82RR19KUUQta6fXDIRfci+hTe2jiSnyH6EJYX91Dq+pdCP+uOBggdQtq/VvUiQ eKY/0lfom9KF+kNYJx1qX4xPXVs0poYyZFPohuNiFyWVq2FRf4qNRdwXYZHOdEw1LKx4IbmPXr// /e9XO9T2GQN7koHNHsKebKY1yhgwBvYzAz/5yU9883mQ4XaFoJEgiVLkZKhOQcV252tkv5w7OTb5 c+G8UGpxePK/rfe72qv2h78XDuEN95W9zWoMJV0kVzL4LnzCq30xPl+/riQXJDOUUcHW5lePvE6Q hFhaqiyfbytIQuhtFK/dk//hL2ZZdvfUYz/0zLx+kNvmVo1QUiz3V38swoHOqA/7kwcW6R/1x62w IFbHRIKwflrxUtRvVZcCSzUcgBVXsbHQX9exrI2p62QFOFL0WYJ6xje1PcTBtuqFN7+/zO/iHZnq F/nzYz8peNE1Tu0PcQgbnMQe39CNrj+SuxWWsL5om2cskThhBQTPubBiDOw3BiwJsd80bu01BvYh Ayx1xGHg/tXt7jHezomRE5Ridm47GcIipyimasWLZIayVLcd3vA3O90mYPBBrnOMJbfoXCmwrKzq vuxKYJ3HQTxDPmtlJfbtGHBSSS4UcaK4Ch2uRk5CrKzpx4VLjo7NvAjfslsxEbvU0mfBoONi4pEM tT+UpboUfVb2g3zJLcIivOG+sre3a6/wbXdco7jgRO2VzPCcof2E9TG2wQGerYrwxeYE+eKEbcll m8J3/kLuKnvi/Bti2UpC2Le3OqbR+jL1w9s1KLzdiQcbWzEG9hsDloTYbxq39hoD+5ABkhA8cZ17 Rre7JaOaA7gPqaurya3EXQosq2vJBQUpxWRtHVAUH7+z2lrbmwZNpQ1VeUkApJbABaS1crczzVR+ VYuMWo5pBEM9v20lLPXgjn1sCl4kIx/4h23TMWFd2du1yqj1uEbwpZBRC75acWx3HM8K4nXXlN/5 nd/ZlOSpBYsdYwzsdgYsCbHbNWj4jQFjYFsGePr8b/3Wb/nj/vM//3PDDE/+x0VLLMNj5Fxsd1z4 m51ubydDWLa6XWOncot+JyySGR6juhQ4QsdcckMs2hZefY/xKRnVcGjGMIb8t+d8u1y6cBJ1LeBP gYWHdEpHRbyoTty9bUP5W9VkCAdShbd8BG/PKNsI5Wqv6qrh1bGNfoZ9QHLDc6pOeMN9ZW9v195U WEiW0W+3KyF32x270/2hDLW/6FzbcVf0m3rrwj6Qx8J31aXAUouMFHYMjmpyauWEWzGuXLnin/vx m7/5m/Wqxo43BvYEA5aE2BNqtEYYA8bAdgz86q/+qj+E2YdqT9EOH9YmhyI8t+rC48L9ZW7rqexb zegKi54OXqbs/LnUXsnM7+e78BbtK6sOBxDnGBxFWFSXhJO1B9ZJZthG4avgjX+p5TWhFN2WEWLh FgnKAcdbLc58+Nt6tw8cqDzw0Wmn8Kfi6qB7un7sov5YZD/SDxh0XEw8YUCXl7POScGDVvPHlvFd 7ZXc8Jyqk72H+8reFo4i/SArFRZvo2vcy1bCtqbCgUz6ica3EIO2hUXcqT7Gp2QgU3LzchhPqvXt /PE7/a43URT1FWETdzuVUcvvsAuNoZIb/k51211/fv7zn/vJkO9973v+FbDhOWzbGNgvDMT3jPYL k9ZOY8AYaGkGfuVXfsU//Il3fH/22WdbYsUh1auyiu671fMXeGd77MKr+8BT9OAvnDEcHpwiveIv Jh7JqMaJeIuJA0cTB4+2SxeSF9alwNLZVXldnHQhHHzy7AWeBQFeOfPh/rK3u7oqrxYs0s+yeyAl pdPxRn+KWXx7SYg4/RRhkc66Cl6FWDYu9QFkKjiQDNXRl8Acu4RY8rJk3ynGFGTr1ZxFDzcUFuHN Yy3ze7UxRTbF+LZdQNc4pjZnG5XXMctWwnOqHwtvuK/sbYJt/uivkhvKkP1Ih+G+srdpLwG3dBGe X/bD/vj6ybK+vj4vfitO4Ascsce3Wv2Dan2FNvzP//yPH3f+7M/+LKTVto2BfcWAJSH2lbqtscbA /mUA5+GDDz7wBPzHf/xHVSIGBgb8/q2cdBzjFE46ziiBCQ4fjmBY9Eo5OWfhvhjb1TgRTymwEDD2 9/d7PiRX7YUj6nCMU2DB0WRGH7l55/iNe0sEr6ukn2jmTDhjfPb2Vpx0PdE+lPFq7RV6KThBbm9f 5X33CmiFhUCBOnTY1dmt6mif6IfABJl5/VCH3tBPiiSR7Ed2GzZaPA0ODobV0baxH0pRXwFfKvtR wF00vsm2U/VZ5HCNePVq8+smpR/xFk0x7sRgoK/IVkJZ4Tij/hTuL3tb+qH94kAysCfwcH1KaT/g gJuwqK+ksh/1SckNsYAP/0DHhPu0zQMpr1+/nl28eDH75V/+ZVXbpzGw7xiwJMS+U7k12BjYvwz8 2q/9WjY8POxnIWZmZrYk4ujRo94Z5B3eYZmfn/ev1MLZSeF44cyMjo56CDzIKiyPHj3ywcKRI0fC 6mjbPFcDhxMOQicQR/TJkyd+XyoncHx83LczzwnvkieIGhsbSxL4E0yPjoz4AHdxcXED93Pzc74P SX8bdkb40u+CKILuxYXFDStEWJExPzfvHOOO7PDQUATJm085Njbu+Z+be75h54sXLz22IYfjUIKV EAjHPgiY6Ldh4ZYs+i59JUUh2UHgyJii2WzkgoE6dAcvKQptZmzJj28kJejH9FnwxC4a3wjc8rYM NpIhPFA4RSFo7O7q9u1fXq68ehe5jHXYzyG3UmLYjYEpCvyTjGDVXlh4TSXjG+NfitUHYKCvoJ+8 /Uhf6IfjYheuPfwhNwz+0Q91JExS2fKIG/Ppu1z3wiL9cB2sNknxb//2b57T3/7t3/bnCc9h28bA fmLAkhD7SdvWVmNgnzOA8/C7v/u73on5r//6ry3ZwCE9fvx4RqD/+PFj74gSyNy5c8cnH86dO7fl b8veAQ6W3t6+fTvTu9Jxuh48eJDRHv5SFJIup06d8hju3bu3vjqDbQIXOEmRmKGtJERIFPHKVYIV HFEcU/RD8HT27NkUlHgZ4EDm5OSkn1nmmQwkQ2ZmZn0wlyqw5HkP6Oe1u/ViamoqW3H9FSxT09PZ gtPP6dOn/KqNFMTggBNIPXny1D9/BRxvXCAzOTnhAkuH072OLkXgQlsJkrDn+/fv+35KHQEVfQfb SZXEI2FFv6SvYrtazo6uwMMr+lIssaf98AEvBFGMcSRCCDQZY8BJP0qpHwJIZGuVCDbNq5TRD7ae opDwOHnqpOPhjesbU04/zLivZNPOfl68fOH4Ou6fqZICCysu0A+JcpJl9Bn6i64/Kcc3+gJ4GOe5 /oDl+fPn3p7QTarrD/3x/Pnzngddf8CC/YAH+0mROEP/JEO4LqOfvH8ATq6FW61+w9Z/9rOfeRv7 /d///RTdyWQYAy3LQJsz4o3rmloWqgEzBowBY6BxBgjg//AP/9DPOv7DP/zDljMWOOY4OwQrOBQM lThjOICpggW1lgD72rVrfsaHIAFHhyAPB5HvqQoc4HTdunXLBy7gIPFw5swZzyffUxXphyCBwndm oMBSr35W5iYagk0Ad/fOXZ98aOMp++6qOjY+5mfmUuqHRihAIJhCHayAOH7saDaUKJh7SyQBwrQP Eqhz78zIenp7fJ/dySzugf7jb09d5xb2Q58l6Jb9MJtMoJdaPwSVPBmfPoNsJfdYoZXSfrBlAn3G OHFA8ogAqtosbp3U13Q4yQc4IXknDkgO7UQ/jdrys2fPswmXJCKZBxafnHABLqtYhK2WRrV19mdt hyq3vdRyfP4Y9INuCLIpjG+6/qTWjxJUJKzUV7Afrj9bBdv59pT1nQQVtkyfQTYrEk6fPu2TIfXo pww86AcbAof0g/1Uu/5wK+hf//VfZ3/+53+e/cmf/EkZMOwcxsCuZMDZa5slIXal6gy0MWAMNMLA P/3TP2UfffRR9jd/8zfZ7/3e7217KlZB4GikdnKKgIFFjmDR/lR1OMn8pXZCi9rXqH4aDVzeYuJh cistoh8eXMqrB5u/4JFnY7S1NWY/jSQh3uon87dmmP2EjFQCXMa2vTC+lWXLlfFtxdnPzpK8jSYh Qg01Or6F52p0m2C7FcaU3Xj9IZnzB3/wBz7x+Pd///d+IqFRfdjvjYHdyoC73jivwIoxYAwYA/uM AVZCsHzzpz/9qZ/B2K75BC2t4KCDsxUCKHDARys4o+KkNfRTeX0omJpduLy3in702s5mc4J8s5/N WqCftIb9tJp+dpaA2MxwYzWtdP1plTFlN15//vu//9vftvenf/qnloBozCTs13uEAUtC7BFFWjOM AWOgdga4d/Qv/uIv/BLgf/3Xf639h3akMWAMGAPGgDFgDBgDdTDArSz/8i//kv3whz/MfuM3fqOO X9qhxsDeZcCSEHtXt9YyY8AYqMLA//t//y97//33vWPAvchWjAFjwBgwBv5/e+cWI1d5JerdXDJx G86cGWMfJebi6OCB2AEFEhspcOTOAygYHZGQkJEJipGAh8PFT8f4PESynTfbPEFsKRossBRAAoaB BwyCh3FH8IDNIRLIDhaWsLnkwcbkjILbmXDx+b/dvYrf29Xd1V3dXbuqvhVVV9Xe/2X9397lsNZe a/0SkIAEZpoAhbAPHjxYrFu3rixsOdPjO54EupGATohuvGrqLAEJtE2AENe77767LLT41FNPlfUN 2h7UASQgAQlIQAISkMAYAR5yPProo8Uvf/nL4jvf+Y5cJCCBMQI6IbwVJCCBviXwjW98o9iyZUvx 2GOPFQcOHOhbDi5cAhKQgAQkIIGZJcDWqr/+9a+LpUuXFrfffvvMDu5oEuhyAjohuvwCqr4EJNAe ge9+97vFmjVrit/+9rfltnntjWZvCUhAAhKQgAQkUBS///3vi/379xe/+tWvivPOO08kEpBARsAt OjMYfpSABPqTwKefflq8+uqrxQ9+8INyP/j+pNC/q56pbf36l+Dsr3ymtuicfU2doZME6vJbnskt OjvJ07nbI7Bv375icHCwWL58eXsD2VsCPUYg7XAzoBOixy6qy5GABCQggakRqIvhMjWt+6u1Toj+ ut7TXW1dfss6IaZ7Be0nAQn0AwF8EKZj9MOVdo0SkIAEJCABCUiglwmc+qI+qzt1qj66qIkEJCCB GhLQCVHDi6JKEpCABCQgAQlIQAKtEzj1n5+23niWW576bKQo6uQUmeX1OrwEJCCBqRI4Z6odbC8B CUhAAhKQgAQk0KUEvvw8GchfdqnyTdROazn1+Uhx6rO/NjnZoUNJpy9HPi7KtIyBHvpP7bO/1iGg TisBCfQagR76l7HXLo3rkYAEJCABCUhAAjNEID2Z//LkJ0XxxWczNKDDTEjgyy+KUyf/X9FTiRln nV2cNXhBUQycPeHSPSkBCUhgMgKmY0xGyPMSkIAEJCABCUigywmc+ut/6IDo8mvYcfVLx8qfO66G CkhAAt1PQCdE919DVyABCUhAAhKQgAQmJHDq8xqlK0yoqSfrTODUF3+rs3rqJgEJdAkBnRBdcqFU UwISkIAEJCABCUhAAhKQgAQk0O0EdEJ0+xVUfwlIQAISkIAEJCABCUhAAhKQQJcQ0AnRJRdKNSUg AQlIQAISkIAEJCABCUhAAt1OQCdEt19B9ZeABCQgAQlIQAISkIAEJCABCXQJAZ0QXXKhVFMCEpCA BCQgAQlIQAISkIAEJNDtBHRCdPsVVH8JSEACEpCABCQgAQlIQAISkECXEDinS/RUTQlIQAISkIAE JCABCZQEToyMFCdGThbzB+el1+CUqbTbf8oT2kECEpCABBoEdEI0UPhBAhKQgAQkIAEJSCAIHPv4 eLH/nUPxtfG+8IJ/LJZcvHhaxn9jkDY/HH7/o2LzloeLW2++sbj1xzdOebR9b75d7Nj5eHHPnb8o hq67puy/7823imMff1KsvmFoyuO122H/O+8WixLXhRcsmHSoqbSddDAbSEACEugAAZ0QHYDulBKQ gAQkIAEJSKDuBHBAYKiPJxj/OAF6RZ56/sXiSHJuLL98aXFJcrLMthCNMfza3gKHCI6FUYdIcycE bXe/MlwcSNdksrazrbfjS0ACEmiXgE6IdgnaXwISkIAEJCABCfQwgTvW3FJ8/+oryhWOpBSIoyla YNeTzxZPP/diMX/evI5EDswG7nvvvL04cPDdOXFAbHv4keR8eKtcBpElEwkRHzgekMnaTjSO5yQg AQnUhYBOiLpcCfWQgAQkIAEJSEACNSQwmOouLMrSBJZcfGHxrfR6YOOW4ukUPdCJ9IXZwESKCa+5 kKMp1eWm64dK5w4pIBNFnBx+/8My4mTZ5ZeW6SITtZ0L3Z1DAhKQQLsEdEK0S9D+EpCABCQggTkm gOGH4UL4tiKBThDgifwF6UX6Qi6RNjD86t4UMXE8GfUXFitSFEU1baPajgKT0a5aF4HaFDt2PlFg jCO0W37Z0nzaCT8TRfDM8y+V0QQxT3UOBiC6g1oTGzfcX45H4csHU8TCiqtSFMhAUex+ebhcE2tf mXRYu+an5Zj52Kz3njtvm7S2w7bNGxo673n19cbnZh8e3b6lcXiyto2GfpCABCRQYwI6IWp8cVRN AhKQgAQ6R4AQaAz9XHgijAGCETR07Wgxu/z8XH3O88LHmxOD7YGNW6dduG+8cXe/sqdhjNEGo2v1 Das6ymM8XT0+uwRIzagKvxsMeYo9rlqwsjj8wUdl2gaOhHtSukNI3u5nl/2orHWAgU0diq3JQMdZ gNBv09hvccXVVxZLLlqcUiYOFY+9+WwMNeE7/ZmL8Yg84De857XX07h7z+iH3pH2wEnWx3d+S/Rb edWVZZ+9f3ireCE5JN5L7fktohdOFtZKigX6bt+2qWzrHwlIQAISOJOATogzmXhEAhKQgAQkUD7x xAjBwAjBKXEgGSVRRb/6dDfa1eEd3XGYNDMUp6sfERjUAaBwH44H5IWX9xQ7Hnm8GDlxsmfC8qfL p5/6RTRO7CwR9wKGPDUk8hQN7g8cDDjulqV7Z28y1Gl30w1DKZrglhIb43C/Mu5wahv9Y57qmI89 8Wwq1LhnUuREUCAbN6xrpFpQUHPbw/9SFoScdIDUAAdE7hhBN1JRcEBU9dr2UKr1kJwURIjMRXHL VvS3jQQkIIG6EdAJUbcroj4SkIAEJFAbAhgf1ZQHHBGbtjw0aoxfdmlpVNVG4UwRjL2ZfBrLE2Uc EEQ+RLg6061KhuV96zf1VG2ADKMfEwEiB4g+COE3QIQAToPcEYdTgGPhQIj2Q9etLMfA8cB9uTI5 9p569KE43Xin5kHxfFGQBhHCzhHNxiQaaTInBPcseuI0q9Z6WJGiGhi7FaF/RGbQHn1IRTmR1lNd K/PghHgvRU/ohGiFrm0kIIF+JKAToh+vumuWgAQkIIFpEwjDi+JwhI5jVOWC0XMkhWUjl6TQcQyY ZhLtTqQIAgyXPOIib0/uPMYSht9E7fI+fKYfRt+iZCxF/jtGGTsboBPn2R5wsvljXOan3+rrRyMg 4niZY58MujLE/fgnxcIFE1f6j36+dw8Bnup//PGfS4W5b3ASUJPhf6+7q2Gcc4x7hPuhWjgxnAqk K+TCb4B7kH5ItIs2HOfYsinUf4i+vHOvI8uSs3Cm5bzBwZke0vEkIAEJ9A0BnRB9c6ldqAQkIAEJ zBQBHBFVwTh78OGd5ZPX/NxNKW2BAnYhebt4uoqhxZibUkG8cBjQHqdB5MNHf9rNb8EAwgFBLjxP qgk/R55+7qXSWUAYPGkUufC0Os/Zz8/xGWfLxnEcKseS8wEZTNs1Kr1HgJSJSLvgXiXy5fAHH562 0DzthzSFqrC7Btt5hnBvNqIUkrOOqKNjx4+X9ReiTbzPn/9VvzjmuwQkIAEJdC8BnRDde+3UXAIS kIAEOkRgOIWnI4sWfuWMoLI+RlXkiGOs7U6GPjntg/MGG46AaJc7B2j3WOrPE2Ry10PCAcGYq1LO PIYeufWM2Y6Qk09KBZENPG0mP35P2s0gcvanMjYREKybon/hVJlKf9t2FwGu8ep0rbkHuW/DwYUT ASFqYX2KkJhIRgtQfvVbibbcR9yHITEm0TrTkbgfcwfJdMaxjwQkIAEJzCyBs2Z2OEeTgAQkIAEJ 9BYBthnMXzgLMJSISKAeAkL1/NKIT46CyBHHAMJA4x1DPWTJRReWjoow3jhOH/LHSe8IwVGAg4An 0JxnHOakX7u55tS5iDQRxsSoRIieaEXIxedJ9n3rN5dFKXFArL3tlla62qYHCHA/ct9wH+BsQ7g/ uS8PHHy3cWy8pUakxIrvXTlek/J43POtjNlsIKKKGKMatdGsrcckIAEJSGDuCOiEmDvWziQBCUhA Al1GACcAhnb+4ukvBvymLGIhCtyRe547LPhMsb3IbWf5GHC8SMvAecHTX14D6X+5hKG26tqV+eHy c7v56IsWLjhtTAxKJAzK0042+UJYPX2iH1sWHklrUfqDAIb9z1OaD/fLrif/tbHoHyanHMd27Pxd mUoUJ7i/700pHOwsg0SRyN0v/3s0KX8D1VoSnCQ6hzGfeX53oy2/HSKKJhP0pAAlzr08eqgsstpm NNFkc3fqPI4hdu9RJCABCdSZwDl1Vk7dJCABCUhAAp0kgBETWwiiB/+BT7TArTf/qGGAcxwnA9LM iCpPpD8jJ0+WT2UxoHa/MlyGsk9k9EcIedVhEON18p3UEF4Ia2e7wq0PP3LaNoad1M+5Z58A158I nzyNZ9S5drI0+HGisZMK9zjONhwPl6TvCH1fSL+lF14eLvtT4wSHXaQHRY0R2hL5w24T0ZboBpwI ee0U2o0nROjsT9EZ7OyCA/GruUadg+P168bjcGbrUv7dejTbWrgb16LOEpBAbxPQCdHb19fVSUAC EpBAGwTISY+CfAzDThPUaeCparMijXckg2e84oxxnCe4GG7UhFiVikFSsA+JQn3lly76QzQEaSk4 aIiGqO4W0kVLUdUKARwH3KffSu/NBAcdUUC5Mw2nAdE/b6TjR9ktJd0fOO3y3V8wkokkIjJi7x/e Lo3mtVf9pGyDs4DzuWzdvKEYTrVQoi1FVMt7roxKujRvesZnxqL/Gyk6IPr/r2tvK3/L/A7ztRF1 lO+kwe+f9UfkRj54tW2cW57WXhTjM4t2+ftknFtti2OG1KiIUMr7+VkCEpBAnQjohKjT1VAXCUhA AhKoNQEMbIwMwrsxoMLgLo2U14pye8rc2KouBmMNw4cQ8bwmRLUd32PMw0c+rMW2l4TUs53iTWmL zniiHXrjnEFyYzTO+d69BIhk4DWejHeeiIaoOTJe39J5lUXURLvxfhd59M1kbeN8vOOIaKV/7nCk L/3G06faNubi34T4dyGOTfY+Hsdm/SZqi77WZmlGzWMSkEDdCFgTom5XRH0kIAEJSKDWBGIbSwpU hoTRQTRAVQiRrtZLGDk5clozwssJR88lnr5WxySdo9o27zdbn6lYwa4GeZHNmItwd2QigzXa+i4B CUhAAhKQQH8TMBKiv6+/q5eABCQggSkS4Akuu0ngHCDHnDz4CKcmTYMCfGypSSj3gYOHyjb0ISSc J5URSUHxPrYzJP+dcSKKgHfa4dhYcfUVZbg7bYme4BzzRg2KKareVnP0YZ3k5qMH4fDUrSBfn5B8 w8DbwmtnCUhAAhKQQN8Q0AnRN5fahUpAAhKQwEwRIER7OBXlw+lAmHcetk2kwLZUpBHhOOkZeXFL IikoYElaBi/asAPAqdQeB0NeV4G2u+ZRQ2K0ACBj0nbhggWl8c/3uZT1999dFvjL9SnXnvLmxwtb n0v9nEsCEpCABCQggfoTGDiVpP5qqqEEJCABCUhgdgh8+Zc/zfjApGAQCUElfoz0ZkI0QVT5H69N 9JtK2+gzm++hz2RrnCkdzjr/mzM1VN+OMxv3ed/C7POF+3vs8xvA5UugTQIDiE6INinaXQISkIAE upqAxln9L59GT/vXyPu8fYaOMErA36N3ggQk0A4BfBAWpmyHoH0lIAEJSEACEpCABCQgAQlIQAIS aJmAToiWUdlQAhKQgAQkIAEJSEACM0eArW+pEcO7IgEJSKBfCOiE6Jcr7TolIAEJSEACEpBAiwT2 vvlWcd/6zeXOLdUu1DLh3K4nvtqmNtpQL4RzGNaTCTvJ8KJPvwo73VDotRM73vQrc9ctAQl0noBO iM5fAzWQgAQkIAEJSEACtSKwPG3JejQ5G9iCtSr73zlUnmMnmKpQlJV+8+c1L8iat1+Utq6tCjvE bN7ycLl1bX6OcTnebM683Wx+xrHy4NjON7M5j2NLQAIS6HUCbtHZ61fY9UlAAhKQgAQkIIEpEmDH loXJSXD4/Y/O6BmOCSIYDqQ0gmXJYRFyIDkokGWXXxqHxn3fuGHdGecOH/moaWoCc5GysOyyycc9 Y9AZOmDKxAyBdBgJSKDvCeiE6PtbQAASkIAEJNCPBHhajSy6YEE/Lt81t0BgxVVXFkQmHEmOiEsu XtzogaNh+WVLi/0H3y2dFKc5IQ6GE2LUMUHqxtGUcrDk4gtT2w+T0+JQsfLqK9J4o99xLhB1Ee2O HR+9Lw8fSREVx443+h35YNQZwnmcAURRLMzu3RMjI8W+N98u0xrmz59XrLjqitPOo3zej8/oMl7b xmLTB/RGz5H0Ylta+p6X3llDLvmYl1y0uFxXfn6yz6xh+LW9xYkTJ9O6Fxcrrr6y0SX4VNdNg9AP xpNt99sY0A8SkIAEOkhAJ0QH4Tu1BCQgAQlIoFMECG1Htm/bVL5jAO1+ebhhIJYH/dPXBDCEEYzr cELwGYN86H+sLI4mhwBREatvGCrb8QeDmH5hDD/93EtlCsXQddeUtQ9oQ4QFBvyuJ/+tHPupRx9K 5/YWTz//IqdL2TaW9rDp/6wr0zDiOO143XrzjcWtP76xPMyctM/rKjyW6lXcseaWhm6c455fniI0 0D+P8Hj6uReLTRvuP8OpEHOGnnyn7+g4S4uNqQ+Cg2ATKSRpjlyGrltZrF3z0waL/Fz1Mxx3Pfls OX6cg9PWzRvK/jhYHkvn83XTDn0e2Li1ZBq/5ejvuwQkIIG6EtAJUdcro14SkIAEJCCBOSRAnj9G IMbZ+nV3zeHMTlVXAkQoIIfHohD4jDGMLEuRECuu+qiMlMAQxukQT+SH0rmq7EuFLu+47ZaCCAGe 2FcFhwKvHY88Xjottj+4qVi4YLRmBE4KnB8Y/1UjHAdAOCxwCoxGVXySjv1LabTjEMkjNbjPb7p+ KDkQ1pU67355T9kOA79Zegh6hrOBIppI1dgPB8T6++8qoxfgwbj8nvi8/v67y34T/SEq4547f1Hq SsQFxSrpv+vJf03Hby9WJScO36mJEc4XxoMLwpoUCUhAAt1CwMKU3XKl1FMCEpCABCQwiwQw3u5N RlBu4MzidA7dBQR4Es8rDF1UJgUDw57jpFUgR1IkAhLRBc3qQeCAWJ0MZe6ziJIoO7X5h6gIIhB+ nhwY4TRBN6IgkHCaxDS0WZt0CR2I4qB96B7tWn2PnS2YL9InGJvfEVEXzE/djMkENvSnL/pEf9YX Th7SY1hrPt4bY4VDY+7J5vG8BCQggToQ0AlRh6ugDhKQgAQkIIEOE8D44WlrhOB3WB2nrwkBaj9g +B47nl4p6oD6EBjDCCkV3Ddh6B9IDgqkWaTDBWNRDWWDGfxzYKwGxbdTwUrqnMTrgmTIoxtOk1ww 8KtCXRQM/elIFOls5gS4cSw6oRUHRzO9LrloNGLk47E0D9I7EKI5QmAfTqE45rsEJCCBuhMwHaPu V0j9JCABCUhgzglgkLAV36r0H/2ERlMrAeMGQ4HcdkLCeTr8zPMvle8YOyvSU2GO58XyUJwQdXLK I1QdA231DauKoWuvOW1dhJqHMUGOOjoQhh5CgcBcD56ytppvThg34eGMGfPHuPEeerL7QERDMCdG zto1PzltDTxNviMdw9Db/crwaWMHhxiXd4zXHTufOI0BBhVPf5V6EyidUq8lw/ePXxnzEenAfc/5 velpPNEF7yUHRdzDc7WqKLB63/rNTaccTPf8bMrIyF/L4Zs5EWIL0jydZSq6lOxTh/fSvyHU5CCt BOaRksG/Qfym70jRHIoEJCCBbiKgE6Kbrpa6SkACEpDAnBDA8cB/4GOYUw1/5diTX4wtHATkb3M+ HBIYGREWnueVkwdPvjqGA04HxsKAIO+dp8s4LUIYjxeCo2BJahtCvjpOBIz/VdeuLPtGGHg+X7TP 3/O+OBiYl/mR3HDCmGH+/Fhsl4iDBAMIpwFrZV3kwdOWfuFQCT4cj2M4ICich+QMKBzIORwpSn0J 8ISfe4gICApRcm0j7QGtiYrgPPcFbfhNzLXw+1o7ln5RnZtznRJ+G8j8edPTgX+Hyv7ZGlalf0dw DpKSMZz+LUGoz6FIQAIS6CYCOiG66WqpqwQkIAEJzCkBnAZRnZ6JyR+nOB3Gel55n3Obt1A871AZ th4F9TDOMNo2pSJ4vCNEGdAWZ8by5BTIi+ZxPorr8RnB2YADolqQD93CGKmOMdqTsO13y74YkhTN CyESA8dCq8Lc+Q4IOBCYm6fMOR/SOR7YuKXY93/fbjghIqc9X1cweCFFmPzs5tWlk6ZVXWw3twS4 bzHk2WYTR0PugECTcFJwjyJE6MylEG3A7w49OuFwiN86dTGqW3bi8EOWXLJ4WkgOf/Bh2Y/fegh1 OPjtsWZeXI/4tyXa+C4BCUig7gSsCVH3K6R+EpCABCTQMQL8B35u2PAf+2GU5UY5CsbTyKPHjpf6 7k3RAhghzYyEiIDY89re09bG2FUjb3isTfUJc6RMVMfIB4xc/dXXr8oPl5EWuWFz2skmX1Z8b7QG QJyKcHwiG3I+hI/zPYwn2sfT4IHoPPZOBASOibx/pYlfa0KA603tA+7nFVeNFqMM1eI3QXQMEr+D OD9T73GfRHRAjBvzPfP87jjUeB+NFhr9PTYOzvCHqNNAhFNVSINCQsfq+fx71NOIY0QJ4WSo/puA w5FjjM31iPmjGYhAagAAEuVJREFUn+8SkIAEuoHAOd2gpDpKQAISkIAE6kSgFQOe0HSE9ImqYEiU xnp6ejqZkBJC2zBoqu3jaWv1ON+PpCfXSNWxUR5s408YhM2GqLK56Yah0oAlfQM9qBlBWkiz4oXN xvNY5wlQnDIcWs2ibkjJ4Ol8OCTa0bisg/BakSKFdpfGe0Q4RK0VUn6oj0C70fomQ2WKE1E1OLxI V8BRgdOEKJxqxFI7ukXUBVFMzI9u8BhKv/Hq/NRK4beJwxEukwm6IrBkHfCkf+zykffH+Rf/HrTi 4Mj7+lkCEpBAHQjohKjDVVAHCUhAAhLoOwIY6xElMNHiaYPRTx2KqlDVf3De16uHa/UdA4x0FML1 96edDEjlQEaP35/eF9RKX5U5k0BEvuBEauaAihSBJWO7OZw5QutHSOl5IRngGOW8tn97dE7mxaDH +N6x8/HyM04IjnN/4Rgg8iGMeY7PpAOCFTD/4fcfKXXg/sUJgdxz1+2p3stg6TiY7vz33PWL9BsZ bkl/orDggMOiFQdHqaR/JCABCdSIgE6IGl0MVZGABCQggd4hEMZaNXw8VshTzlYiFBgHh8Vvtm2M ri2/Dw6OOijCkdFyxxluiKHE7gkI66aoHvUyKFj5m22bmhq2M6yCw7VBAGM/36mlOhTRAOOdx7jm 1UxIx6kK9/v2dE9EUdiouUA7UpAwwE+MjKR7ZrDRlfuLObjHSGPg94KDLhfajKdjMz3yvvGZdXK/ MkfVeXZHmptX6F2dP8aovpNmFalWRDjwW43x49+Qah/mQFZ+7/TUmGo7v0tAAhKoKwFrQtT1yqiX BCQgAQl0NYEyrDytgLDwqvDEFlly0eQF6xgHwx3DZKoST6Ypmtcp2ZUiH3Ylh0MIxiBPuwmbx+AK gyrO+y4BCOD4aGbIY5hzvJmBzrHx+s0U1Zij2fwT6d3q/JONzzjsikG776dICEUCEpBANxLQCdGN V02dJSABCUig9gR4aspOAYRnx84BKI0zgVBqjIhqcctmi4oilmz1mTsi2PmCnTqiIGCzvvGEdXsK X8/7Mj+OjbkQilS+kFIxcgY8yT6QUjOQMDQ5NpoHP3Vny1yswzkk0EkCOOxwXu7Y+bvy35RqUdhO 6ubcEpCABKZKwHSMqRKzvQQkIAEJSKBFAvfceXuxKW3HSepB7GIRzoC1a25pKZ8bZ0bkwt+7fnOZ whERBEQV8OR3POE8efHMT1/axvycmwupMsD5QvQDa2BdoQc7AVAvYv8fDxXr1901F6o5hwS6hgBp XdTCQKhFEelNXbMAFZWABCSQETh7U5Lsux8lIAEJSEACfUXg1N/+0nS9Xzv33DKSYfE3/ttp59lq 8p/++5JiaXrlwnHyxPPifRjc7A5BVf3PPv+8+K9//1+K737n28Xda/+5bJf35zN9q+PGcaIamONv aRzG4/vdv/znckzajCeMh9HCehiAYnbMj/H/reSUiPkYmzx7ojfyNfMZR0jZf2ySvG01N55zOZ8q A4Zgt4Xbbv2fjVx4jsHm888+L7ccPGPMvzufJkobBMa7z9sY0q5zSCAip67/4bXFDT+8bg5nPnOq AX+PZ0LxiAQk0DKBzUkGTiVpuYcNJSABCUhAAj1G4Mu//KnHVtR7yznr/G/23qLmeEXe53MMvIen 8/fYwxfXpUlgDggMJLEmxByAdgoJSEACEpCABCQgAQlIQAISkIAEikInhHeBBCQgAQlIQAISkIAE ZpDA0VSAlmKrdRP04qVIQAIS6CQBnRCdpO/cEpCABCQgAQlIQAItE6Co6dPPsbtLfQ1pdp65LxWC 3fXkv7W8rrlqiF6btzw8V9M5jwQkIIGmBHRCNMXiQQlIQAISkIAEJCCBuhHY/fJwucUtu6koEpCA BCTQnQTcorM7r5taS0ACEpCABCQggb4jcOuPbyyWXLK43PGl7xbvgiUgAQn0CAGdED1yIV2GBCQg AQlIQAIS6HUCbC27+vqhXl+m65OABCTQ0wR0QvT05XVxEpCABCQgAQlIYHoEdr+yp9j35tvF2jU/ KesbUI/hxMjJYvnlS4t77rytWHjBgsbAu558tjj8/kdl2x07n0ifPyy2bd5QXHLx4rLNvjffKna/ Mlzsf+fdYv7gvBTJcEVx6803NsbY9vAj6fjX07i3N8bkA+NQW2HFVVcUq28YKva89nox/Ore4o41 tzTGph01Ip5+/qWk71uljksuvrAYum7lGQ6LHTsfT20/KTZuuJ9uDRlv3Keff7Gcj2KOzfRuDDDO hz2vvl6um3XQH53WrvnpGa1pN/za3nH5RIe8Hce4FquvX3VGZAhFMeG9++U9DR633vyjGMZ3CUhA Ah0lYE2IjuJ3cglIQAISkIAEJFBPAoePfFQaxVHIkAiEFVdfWR7blIob4pAIwQGBgwFnwsiYo+JU cao8jSHPcQTHw4qrriydG/kY8+fNK/Yk50I+Ju1xhDAuxjZy7Ngn5fdPs50ncEAwFg6IoWuvKefA 4H/siWeLHTt/V/aLP4zFqyrNxn0sOVYogkn0ReiNjg9s3NpSYUz0wTmz/LJLy/6DSacXqGmRxswF PjhHkHyenA/n8nZ33HZL2RbnCGwPVNbENWMenDGMueSixY1rwFiKBCQggU4SMBKik/SdWwISkIAE JCABCdScAEYsUQghGPc4B3jKTo2GXEonQHasjFBIxjBOhDz64MA7K0vHQYxBhMBoNMLrp81FAUqc ABFRkc8Vn4m8ILpha4q8WHLxaOQF50LPZWlu9JqK4AxBt6re6IlzgAKZa5MjYCLB6bBpw7pSf9rB 8L71m8p1BjciJHAWDF13TYoC+UVjONaBEyT4oE84FXKOOIUe2LildG6wToRoCZxC1TFD98YkfpCA BCTQIQJGQnQIvNNKQAISkIAEJCCBbiCw4ntXnqYmBjSRBjgNqhLGdRznyT/ys0oqAAYzzoUYg++M ue8Pb0XXMhUD58JNE9SAwMlBZAPRFbkDgkFCz+FklE9fRqM5ov8lKbIAJ0DulIlz1XccGKwxhPVd kL6zphCcOUh1jYxP+wMHR6M2+PzUow8lR8sDZfv4E2seOflVVEowxHmUSzgp8mN+loAEJNAJAkZC dIK6c0pAAhKQgAQkIIEuJRDG9JH0tH0yOfLBaBvqE1DzoCq5Qb4qRStglPPUnzlIfUB42j+evDem w/LLLz2jSehJVMBUhb4Y8aRA3JuiFxgfR8ey9B6pIVMdk/bnDQ6e1u3Yx38uv78w5ow47WT6cjRz WHAOh8sbyVFzNPUj7aWZ0Af9cwdIs3Yek4AEJNApAjohOkXeeSUgAQlIQAISkECXEqga05Mto5nD YqAYKBZlxS1XpmKVOCGIXCASgCf61WiC6jxhiJP60EzQ88jI1J0QjEUkxaIyWmNvWcMinCIcr0YZ NJu7lWPUdEAOpLSTqsxPuueOBBwikZJBnYnBi0bXXK1xAZPxeFTn8LsEJCCBThDQCdEJ6s4pAQlI QAISkIAEuphAXhhyomUMph0vkC0pjYCn8xMJ6QJEHOB8IOKAKImfZ/UlmvUNYzucEdU26DnZvNU+ +fdVqVYDLwRjHycJjgBqblZTT/J+rX4O/X+zbeOEXaK2RrXOA51wTuTCmB9XIijy836WgAQk0GkC 1oTo9BVwfglIQAISkIAEJNBFBDCIiWxoJS1hyUUXlitrVpeBbSSrsuyypcnYP1RGHnCO7xNJ6BB1 EPK2oWc+Rhj9ebtmn3E4sGNFHmXAXLGF6HRSPJrNs3xsfdXdLaptI+1kWYqAmEwYk5SWycacbBzP S0ACEpgtAjohZous40pAAhKQgAQkIIEeILD75X9vrALHAbs2IOy2MJlEgUWe1uMUCOEz20iy+0Mu UfCR9tRgyNMR8nbxmSgHtg7FcZFHBKBnY1vQH/8omqftMkedGsNZUU0cDVEgMxoOpA/sMsEWm7ne b6RtN5H58yeO6igbtfDnppR2gmxPDo98Hj5TiyL4nDcWRUIkRrRjjbG1Z+7QIa0F4TpFW77nfPiO 0C8fc/SofyUgAQnMLgHTMWaXr6NLQAISkIAEJCCBriZALYS9b75d1m9gS0mesq9Ihi6FJCcTnATr 77+rdAjcu35zGT1B/zCOqzs20J6UDJwKrTg5mJ+0CHaRIE0CxwF1JkJPajcsuXg0GoO2jIkDYvsj j5fbWpLGQV0Gdpk4RoMxQa8oTPnAxq3lGOjNuDhGZqomBGPdm7bmxAmR84l5oign+gxdyzame9OW nFtTislgqTf9YZYX+Mx1Z0zWD2+iQGibC5zZynT/Hw8V69fdlZ/yswQkIIFZIzBwKsmsje7AEpCA BCQggZoT+PIvf6q5hqp31vnfFEKbBKZzn+9IhjoRAtsf3FQc+OO7xd4/vF0asRjsRB/kgvGPITxe nQTOkTKB0YshvHDBP5bFJ6tGMWOSRkC7ZmPFOWojVKMkSPnYf/BQ6SRZlMb/fnKURLpGriu6EGFw 9PjoLhKrknGPHvuSo6U6brUta8f50kzvmANnBePTNpwIcW48Tq3Owxrz64AuHGPOiDqJuYIH34M3 x5CIOKHfM8l5g1Op6hAqGzb54++xCRQPSUACLRMYQHRCtMzLhhKQgAQk0IMEpmOc9SCGWi9Jo6f9 yzOd+zx3QmDEKhKAgL9H7wMJSKAdAvggrAnRDkH7SkACEpCABCQgAQlIQAISkIAEJNAyAZ0QLaOy oQQkIAEJSEACEpCABCQgAQlIQALtELAwZTv07CsBCUhAAhKQgAR6lAB1Aqi7MDjv9GKGPbpclyUB CUhAAnNEwJoQcwTaaSQgAQlIoJ4EppMrX8+V9K5W5qC3f229z9tn6AijBPw9eidIQALtELAmRDv0 7CsBCUhAAhKQgAQkIAEJSEACEpDAlAhYE2JKuGwsAQlIQAISkIAEJCABCUhAAhKQwHQJ6ISYLjn7 SUACEpCABCQgAQlIQAISkIAEJDAlAjohpoTLxhKQgAQkIAEJSEACEpCABCQgAQlMl4BOiOmSs58E JCABCUhAAhKQgAQkIAEJSEACUyKgE2JKuGwsAQlIQAI9R2BgoOeW1FMLGvA/VWbieg6c/bWZGMYx +pzAwLlf73MCLl8CEpgJAv4/+0xQdAwJSEACEuhaAgNfO69rde8HxQfOHeyHZc76Ggfm/UMxcNbZ sz6PE/QugYGzzy0G/u7ve3eBrkwCEpgzAgOnkszZbE4kAQlIQAISqCGBU//5H0Xx2Ujh/yXW5+Kk fcSLIjmIBr52fn2U6gVNvvhbL6zCNcw1AX6PZ50717M6nwQk0IME0v+/D+iE6MEL65IkIAEJSEAC EpCABCQgAQlIQAJ1I4APwnSMul0V9ZGABCQgAQlIQAISkIAEJCABCfQoAZ0QPXphXZYEJCABCUhA AhKQgAQkIAEJSKBuBHRC1O2KqI8EJCABCUhAAhKQgAQkIAEJSKBHCeiE6NEL67IkIAEJSEACEpCA BCQgAQlIQAJ1I6ATom5XRH0kIAEJSEACEpCABCQgAQlIQAI9SkAnRI9eWJclAQlIQAISkIAEJCAB CUhAAhKoGwGdEHW7IuojAQlIQAISkIAEJCABCUhAAhLoUQI6IXr0wrosCUhAAhKQgAQkIAEJSEAC EpBA3QjohKjbFVEfCUhAAhKQgAQkIAEJSEACEpBAjxLQCdGjF9ZlSUACEpCABCQgAQlIQAISkIAE 6kZAJ0Tdroj6SEACEpCABCQgAQlIQAISkIAEepSATogevbAuSwISkIAEJCABCUhAAhKQgAQkUDcC OiHqdkXURwISkIAEJCABCUhAAhKQgAQk0KMEdEL06IV1WRKQgAQkIAEJSEACEpCABCQggboR0AlR tyuiPhKQgAQkIAEJSEACEpCABCQggR4loBOiRy+sy5KABCQgAQlIQAISkIAEJCABCdSNgE6Iul0R 9ZGABCQgAQlIQAISkIAEJCABCfQoAZ0QPXphXZYEJCABCUhAAhKQgAQkIAEJSKBuBHRC1O2KqI8E JCABCUhAAhKQgAQkIAEJSKBHCeiE6NEL67IkIAEJSEACEpCABCQgAQlIQAJ1I6ATom5XRH0kIAEJ SEACEpCABCQgAQlIQAI9SkAnRI9eWJclAQlIQAISkIAEJCABCUhAAhKoGwGdEHW7IuojAQlIQAIS kIAEJCABCUhAAhLoUQI6IXr0wrosCUhAAhKQgAQkIAEJSEACEpBA3QjohKjbFVEfCUhAAhKQgAQk IAEJSEACEpBAjxLQCdGjF9ZlSUACEpCABCQgAQlIQAISkIAE6kZAJ0Tdroj6SEACEpCABCQgAQlI QAISkIAEepSATogevbAuSwISkIAEJCABCUhAAhKQgAQkUDcCOiHqdkXURwISkIAEJCABCUhAAhKQ gAQk0KMEdEL06IV1WRKQgAQkIAEJSEACEpCABCQggboROCcptLluSqmPBCQgAQlIQAISkIAEJCAB CUhAAr1H4P8Dr3vbLaDPrpAAAAAASUVORK5CYII= ) These matrices describe the full operation of an attention head when moving information from the residual stream at one position (which is attended to) to another: h\_i(x) = \sum\_j A\_{ij}W\_{net}x\_{j} (1)(d) Expand out the following matrices: Two token copy: W\_{\text{net}}^2 \cdot W\_{\text{net}}^1. One token copy: W\_{\text{net}}^2 \cdot \text{Id} ~+~ \text{Id} \cdot W\_{\text{net}}^1. ![](data:image/PNG;base64,iVBORw0KGgoAAAANSUhEUgAABCEAAAPHCAYAAAAWyFPvAAAMbWlDQ1BJQ0MgUHJvZmlsZQAASImV VwdYU8kWnluSkJDQAghICb0J0gkgJYQWQHoRRCUkgYQSY0JQsaOLCq5dRLGiqyKKbaXZsSuLYu+L BRVlXdTFhsqbkICu+8r3zvfNvX/OnPlPuTO59wCg+YErkeSjWgAUiAulCeHBjDFp6QxSJ1AHOKAC OjDi8mQSVlxcNIAyeP+7vLsBEMX9qpOC65/z/1V0+AIZDwAkA+IsvoxXAPFxAPB1PIm0EACiQm85 uVCiwLMh1pXCACFeqcA5SrxDgbOU+PCATVICG+LLAKhRuVxpDgAa96CeUcTLgTwanyF2EfNFYgA0 R0AcwBNy+RArYh9RUDBRgSshtoP2EohhPICZ9R1nzt/4s4b4udycIazMa0DUQkQyST536v9Zmv8t BfnyQR82cFCF0ogERf6whrfyJkYpMBXibnFWTKyi1hB/EPGVdQcApQjlEclKe9SYJ2PD+gF9iF34 3JAoiI0hDhPnx0Sr9FnZojAOxHC3oFNEhZwkiA0gXiCQhSaqbDZJJyaofKH12VI2S6U/x5UO+FX4 eiDPS2ap+N8IBRwVP6ZRLExKhZgCsVWRKCUGYg2InWV5iVEqm1HFQnbMoI1UnqCI3wriBIE4PFjJ jxVlS8MSVPZlBbLBfLFNQhEnRoX3FwqTIpT1wU7xuAPxw1ywywIxK3mQRyAbEz2YC18QEqrMHXsu ECcnqng+SAqDE5RrcYokP05lj1sI8sMVeguIPWRFiaq1eEoh3JxKfjxbUhiXpIwTL87lRsYp48GX gmjABiGAAeRwZIGJIBeI2robuuEv5UwY4AIpyAEC4KTSDK5IHZgRw2siKAZ/QCQAsqF1wQOzAlAE 9V+GtMqrE8gemC0aWJEHnkJcAKJAPvwtH1glHvKWAp5Ajegf3rlw8GC8+XAo5v+9flD7TcOCmmiV Rj7okaE5aEkMJYYQI4hhRHvcCA/A/fBoeA2Cww1n4j6DeXyzJzwltBMeEa4TOgi3J4hKpD9EORp0 QP4wVS2yvq8FbgM5PfFg3B+yQ2ZcHzcCTrgH9MPCA6FnT6hlq+JWVIXxA/ffMvjuaajsyC5klDyM HES2+3GlhoOG5xCLotbf10cZa9ZQvdlDMz/6Z39XfT68R/1oiS3ADmBnsRPYeeww1gAY2DGsEWvF jijw0O56MrC7Br0lDMSTB3lE//DHVflUVFLmUuvS5fJZOVcomFKoOHjsiZKpUlGOsJDBgm8HAYMj 5jmPYLi5uLkCoHjXKP++3sYPvEMQ/dZvurm/A+B/rL+//9A3XeQxAPZ5w+Pf9E1nxwRAWx2Ac008 ubRIqcMVFwL8l9CEJ80QmAJLYAfzcQNewA8EgVAQCWJBEkgD42GVhXCfS8FkMB3MAaWgHCwFq8Ba sBFsATvAbrAfNIDD4AQ4Ay6Cy+A6uAt3Tyd4CXrAO9CHIAgJoSF0xBAxQ6wRR8QNYSIBSCgSjSQg aUgmkoOIETkyHZmLlCPLkbXIZqQG2Yc0ISeQ80g7cht5iHQhb5BPKIZSUV3UBLVBR6JMlIVGoUno ODQHnYQWo/PQxWglWo3uQuvRE+hF9Dragb5EezGAqWP6mDnmhDExNhaLpWPZmBSbiZVhFVg1Voc1 w+d8FevAurGPOBGn4wzcCe7gCDwZ5+GT8Jn4InwtvgOvx0/hV/GHeA/+lUAjGBMcCb4EDmEMIYcw mVBKqCBsIxwknIZnqZPwjkgk6hNtid7wLKYRc4nTiIuI64l7iMeJ7cTHxF4SiWRIciT5k2JJXFIh qZS0hrSLdIx0hdRJ+qCmrmam5qYWppauJlYrUatQ26l2VO2K2jO1PrIW2ZrsS44l88lTyUvIW8nN 5EvkTnIfRZtiS/GnJFFyKXMolZQ6ymnKPcpbdXV1C3Uf9Xh1kfps9Ur1vern1B+qf6TqUB2obGoG VU5dTN1OPU69TX1Lo9FsaEG0dFohbTGthnaS9oD2QYOu4azB0eBrzNKo0qjXuKLxSpOsaa3J0hyv WaxZoXlA85JmtxZZy0aLrcXVmqlVpdWkdVOrV5uu7aodq12gvUh7p/Z57ec6JB0bnVAdvs48nS06 J3Ue0zG6JZ1N59Hn0rfST9M7dYm6troc3Vzdct3dum26PXo6eh56KXpT9Kr0juh16GP6Nvoc/Xz9 Jfr79W/ofxpmMow1TDBs4bC6YVeGvTcYbhBkIDAoM9hjcN3gkyHDMNQwz3CZYYPhfSPcyMEo3miy 0Qaj00bdw3WH+w3nDS8bvn/4HWPU2ME4wXia8RbjVuNeE1OTcBOJyRqTkybdpvqmQaa5pitNj5p2 mdHNAsxEZivNjpm9YOgxWIx8RiXjFKPH3Ng8wlxuvtm8zbzPwtYi2aLEYo/FfUuKJdMy23KlZYtl j5WZ1Wir6Va1VnesydZMa6H1auuz1u9tbG1SbebbNNg8tzWw5dgW29ba3rOj2QXaTbKrtrtmT7Rn 2ufZr7e/7IA6eDoIHaocLjmijl6OIsf1ju0jCCN8RohHVI+46UR1YjkVOdU6PXTWd452LnFucH41 0mpk+shlI8+O/Ori6ZLvstXlrquOa6RriWuz6xs3BzeeW5XbNXeae5j7LPdG99cejh4Cjw0etzzp nqM953u2eH7x8vaSetV5dXlbeWd6r/O+ydRlxjEXMc/5EHyCfWb5HPb56OvlW+i73/dPPye/PL+d fs9H2Y4SjNo66rG/hT/Xf7N/RwAjIDNgU0BHoHkgN7A68FGQZRA/aFvQM5Y9K5e1i/Uq2CVYGnww +D3blz2DfTwECwkPKQtpC9UJTQ5dG/ogzCIsJ6w2rCfcM3xa+PEIQkRUxLKImxwTDo9Tw+mJ9I6c EXkqihqVGLU26lG0Q7Q0unk0Ojpy9IrR92KsY8QxDbEglhO7IvZ+nG3cpLhD8cT4uPiq+KcJrgnT E84m0hMnJO5MfJcUnLQk6W6yXbI8uSVFMyUjpSblfWpI6vLUjjEjx8wYczHNKE2U1phOSk9J35be OzZ07KqxnRmeGaUZN8bZjpsy7vx4o/H5449M0JzAnXAgk5CZmrkz8zM3llvN7c3iZK3L6uGxeat5 L/lB/JX8LoG/YLngWbZ/9vLs5zn+OStyuoSBwgpht4gtWit6nRuRuzH3fV5s3va8/vzU/D0FagWZ BU1iHXGe+NRE04lTJrZLHCWlko5JvpNWTeqRRkm3yRDZOFljoS78qG+V28l/kj8sCiiqKvowOWXy gSnaU8RTWqc6TF049VlxWPEv0/BpvGkt082nz5n+cAZrxuaZyMysmS2zLGfNm9U5O3z2jjmUOXlz fitxKVle8tfc1LnN80zmzZ73+Kfwn2pLNUqlpTfn+83fuABfIFrQttB94ZqFX8v4ZRfKXcoryj8v 4i268LPrz5U/9y/OXty2xGvJhqXEpeKlN5YFLtuxXHt58fLHK0avqF/JWFm28q9VE1adr/Co2Lia slq+uqMyurJxjdWapWs+rxWuvV4VXLVnnfG6hever+evv7IhaEPdRpON5Rs/bRJturU5fHN9tU11 xRbilqItT7embD37C/OXmm1G28q3fdku3t6xI2HHqRrvmpqdxjuX1KK18tquXRm7Lu8O2d1Y51S3 eY/+nvK9YK9874t9mftu7I/a33KAeaDuV+tf1x2kHyyrR+qn1vc0CBs6GtMa25sim1qa/ZoPHnI+ tP2w+eGqI3pHlhylHJ13tP9Y8bHe45Lj3SdyTjxumdBy9+SYk9dOxZ9qOx11+tyZsDMnz7LOHjvn f+7wed/zTReYFxouel2sb/VsPfib528H27za6i95X2q87HO5uX1U+9ErgVdOXA25euYa59rF6zHX 228k37h1M+Nmxy3+ree382+/vlN0p+/u7HuEe2X3te5XPDB+UP27/e97Orw6jjwMedj6KPHR3ce8 xy+fyJ587pz3lPa04pnZs5rnbs8Pd4V1XX4x9kXnS8nLvu7SP7T/WPfK7tWvfwb92dozpqfztfR1 /5tFbw3fbv/L46+W3rjeB+8K3vW9L/tg+GHHR+bHs59SPz3rm/yZ9Lnyi/2X5q9RX+/1F/T3S7hS 7sCnAAYHmp0NwJvtANDSAKDDvo0yVtkLDgii7F8HEPhPWNkvDogXAHXw+z2+G37d3ARg71bYfkF+ TdirxtEASPIBqLv70FCJLNvdTclFhX0K4UF//1vYs5FWAPBlaX9/X3V//5ctMFjYOx4XK3tQhRBh z7Ap7ktWQRb4N6LsT7/L8cc7UETgAX68/wsztZC2JFMiHQAAADhlWElmTU0AKgAAAAgAAYdpAAQA AAABAAAAGgAAAAAAAqACAAQAAAABAAAEIaADAAQAAAABAAADxwAAAAAfIDXtAABAAElEQVR4Aey9 d6wl13XmW83YOeccGZpB6ZGSIBmiBYgeSvBz1IwtW08aWAP62ZYNGJL1zwPINh4M2BobeNZIHsCS LVnjgJGfR4YhyqD+MGlbT1YyJVJshs4558DMfue3ye9wd3Xdk+6pdere/jZw7qlbVaf2V7+99q61 VqUpl1qlcDEBEzABEzABEzABEzABEzABEzABEzCBGglMaZVraty+N20CJmACJmACJmACJmACJmAC JmACJmACbQJOQrRReMIETMAETMAETMAETMAETMAETMAETKBOAk5C1EnX2zYBEzABEzABEzABEzAB EzABEzABE2gTcBKijcITJmACJmACJmACJmACJmACJmACJmACdRJwEqJOut62CZiACZiACZiACZiA CZiACZiACZhAm4CTEG0UnjABEzABEzABEzABEzABEzABEzABE6iTgJMQddL1tk3ABEzABEzABEzA BEzABEzABEzABNoEnIRoo/CECZiACZiACZiACZiACZiACZiACZhAnQSchKiTrrdtAiZgAiZgAiZg AiZgAiZgAiZgAibQJuAkRBuFJ0zABEzABEzABEzABEzABEzABEzABOok4CREnXS9bRMwARMwARMw ARMwARMwARMwARMwgTYBJyHaKDxhAiZgAiZgAiZgAiZgAiZgAiZgAiZQJwEnIeqk622bgAmYgAmY gAmYgAmYgAmYgAmYgAm0CTgJ0UbhCRMwARMwARMwARMwARMwARMwARMwgToJOAlRJ11v2wRMwARM wARMwARMwARMwARMwARMoE3ASYg2Ck+YgAmYgAmYgAmYgAmYgAmYgAmYgAnUScBJiDrpetsmYAKT jsClS5eK3bt3Fzt27Jh0++YdMgETMAETMAETaBaBrVu3FidPnmyWKKsxgXEScBJinAD9cxMwgauL wLFjx4rf+73fK3bu3Hl17bj31gRMwARMwARMIJzAt771reKP/uiPiosXL4bX7QpNoC4CTkLURdbb NQETmHQEXn755eKLX/xicf311xfvfOc7J93+eYdMwARMwARMwASaReADH/hAwQmQr3/9680SZjUm MA4CTkKMA55/agImcHUR+Jd/+Zfie9/7XvHbv/3bxcyZMyt3/oUXXihefPHFgts2RlmoHx3PP/98 8eqrr45SSqr7pZdeKmDTFC1N4UJiq2laXnnllZHbC3YCF/iMusADLdhwE/o1/Qgtoy6waBoX2Iy6 jWgXxt6mHQeawAW7bcp4p2NSE8Y7HQfG0rJ06dLiwx/+cPHZz362ePLJJ0fd9V2/CQyFwHVD2Yo3 YgImYAKTnADPgfjTP/3T4oMf/GCxfv36K/b23LlzxcGDB5MDfO2116bvOXPmFAsWLCiuuy52qH3u ueeK/fv3JyeYqzZwPmfNmlUsXrw4XAuO+J49e1LQhBacrNmzZxeLFi1KV5RcAbLGGQQoBw4cSMkQ tdH06dOLJUuWFDfccEONNV+5abhgL1xei33QRtOmTUttNHXq1Ct/UOMcHOBDhw4VZ86caXOgrZYt W1ZEayH5cPz48XTWT/1mypQpxfLly4sZM2YUTEcV2uTo0aPpXuxrrrkm1U39CxcuLObNmxclI9UD F9oHm8F2YUO70afREs3l9OnTyWbQAhv6NX2a8S66oOXIkSOpD6GFMnfu3NRO+j9CE/Zy4cKFNMbA gz5EG8GED6wiS64Fe0ETxyTaSX0rSg/HpL179yYN1A0rjgNwiR57OQ5wfCQZouMjJxXgEq2FRAjH JNpKWhhzGXvLWt70pjcV73jHO4o/+ZM/KR588MFk31Ht53pMoA4C17YM+cE6NuxtmoAJmMBkIvCH f/iHBc7ur/7qr6ZgKN83HIjt27enWRs2bCjmz5+fppmHY0ECIKrg1HCmhKBk3bp1ybHC+d21a1e6 CgHnPKrg5D377LMpGSItN954Y9KCU4oDGhU8EQxs27YtBf1oIXhDC44xTinOeVTAQjCAbWA3q1ev LjjLheOJY0ygiTMcxYXgFi0E2xs3bkxcCPYJds+ePZuCucjgCQa0CU74ypUrU7ugA0cde4kMnghs eQAttrJq1aq2FthgL+Ugoc5+xaXY9CUSINgMAZwCO7Rgy1EFG0ULGtauXZuSICTVeE4O8yITVyTx GO+w2TVr1iQboT9jQ+hgflShPz/zzDPJRklUK/EAq+jjAGPv008/ncYR2ogxhcCfhDCFdooaY7AN tFC/xl76MeMO47KOlxHthG3oOKDxDjvBdtHJ8TGKCzyeeuqp9jGJZDj9mIQwJzXgkmthGeuQhMC2 7rjjjsuWR/BzHSYwLAJbWsW3YwyLprdjAiYwKQngKDz66KPFP/7jPxY/+ZM/mZy58o4SlOBM4ezh KOBgESwQvOgsR/k3dfyPVoI4ghO0cJafABJnBi0sO3/+fB1VV26TM9okbtDCmSa4EDARYGpZ5Q9r mElASfBEoIIWuOCIc4b98OHDSU8N1VZuEibsPxzggRYSVQS6LKOdogrOLk9dhwM8aCPshjaL1kIf IkiCBWzQQn8ioCO5RqCAjUcU6qMdCEpIEhFEogX7QRda0BtRSFoRmKh+kh9Mk4xgmkRJlBb2lzai XpJEcIEH7YX9KJiL4kKygf6DvRJMokW2DBcCy6hCG9EOBNrSwnGAz759+8IeKkgf4ZjEvmMjOg5w DEALdh35gEO0MM6QoCcpRHtx9Q79CmaMy1GFZCtJTexFWjTesIxPVOG4c+rUqdR30KBj0ooVK9KV YBwjyuX2228v7rvvvuLP/uzPiu985zvlxf7fBCYUASchJlRzWawJmEA0ARxu7sO88847Cx4OVS4E KzguBAX5WTfOYBDAsJyzmBGFs18nTpxIl/XjBOcF5xPnVJct58vqmCZwwsmCA8FJXnC4uOog0uGj DQiY8jZCk7jg8EUEuFx5gBa4kIDIC9oI7kgKRAWVss3yFTJKGtGGEVzggBbq4ixfXuScE6xg4xGF 4IC64EJgq6Iz2gQyJPsiCskggkZsNS9wIcGIDvRGFJKYBJT0YcY8FbTAiqsBorSornIbMbYQ5EaO vbQPYy8Bf3nspY3QwvKIQvKBvkQf5hYvFcYcbIixWf1ey+r6RgvjGTpgkxe40FZoYVysuzC2MJ5p PMnr0y1NUW2k4wDjPe2UFyUkOCZVcfn5n//5NCb9zd/8TWrL/LeeNoGJRMBJiInUWtZqAiYQTuCb 3/xmsbv1PIif+ZmfucJZQAxOC44CwQlOXl4IXphHEBERyOGUEzhVaWEehSACJ7TuQj04oDhZVVxw BKO4oAM9OLx88oI+CoFcRIBLHQSwtEdZC/9jM7RjRBsRGBEw0j55oA0P/U8yJOKsKX1IwWse3KKF Ai+0EADXXeirSkqp3+R1Mo91aMeIoqSUbDWvk3nYShQXEofs+1htxDL0RhTGD2yiqo2kDy5oqrtg u2NpoS/Rt2XfdWvR2Eu9jLN5ESuxy5fVMa1xVeNJXof00UZVwXa+7jCmGccYfxnvylzQwnzWoR3r Loy9tBP1jnUcQAu3j5TL5s2bi7e97W3pNqTHH3+8vNj/m8CEIXC5NzZhZFuoCZiACdRPgCDjb//2 b9PZ8/e+971XBNMowJGglB0J5snRwZHA6ai74PDhcONM8ckL/6MRLQTldRfqwbEUg7w+dKAHHRFa COrH4oIuNNI+VQ5frnsY09TDPotBvk3m8SGojEiIYC843LRFVTtpXkQSgn1WPTAoF82jLesuaJEt iEFep7TAL6IowaB68zoV3IldvmzY0/Rn2WUnLhFtxL7lY295vJM+2ihijNHYq3pz9mjjE5W0En/s pcxF9qJxKNdZxzT2Qn8aiwsa0cs6dRdxUXvk9aGP+YyHEX1JXNj/cr9GB/NYp8pvYNmHPvShdEz7 yle+0u6T+f542gQmAoErj/QTQbU1moAJmEAAgW9/+9vp3tp77733iktJqZ7AVmdNys4eyzWP9SKc LDksqhcNeWE+OiK0wEWBf66BaXRIozSX1xnm/wpC8nrz7TOfAGvUXNAkLrKrXOewp9VGeb15HZFa 1JdUZ64j1xfZRtRbDhDKWtBdd1EfqWKjeRH2kvcR1Zvvu+axXlO4RGkRfzEoc2E+TKLsl/rH0sKy vC35v67SyXZVJ1zQU3dpUhvpmDTWPstepLm8Hs/X4PkQ//zP/9x+KHZ5Hf9vAk0n4CRE01vI+kzA BEZCAKfoy1/+crrn+v3vf/+YGuRsVzl8+hHrjOVMaJ1hfMuR66QFJ1iah1HnWNvo1dmWkzrWdoYx vxctMOllvfHqURt1206EvXSzBdlRhJacv+rNGWleN+c9/82g07RRL+3Ujd+g9Zd/18kuxSWqjcRF 9Za18n/ellXLhzWvFy6sI83DqrdqO5205Ov3ul7+m36ntb9NbyP2SxojjgPi0oknthvRl7ppEZex tPB8jXe+853pKoi/+Iu/CLHxTty8zAQGIeAkxCDU/BsTMIFJT+C73/1usXXr1mLTpk3FLbfc0nV/ cV5GXeS4jFoH9U80LU1ov7zdIvTQRp3aKUJDvs+9aOm0Tr6t8Uz3U0cEo056IuofD8s6f9uJS531 Vm27Vy0R7dWLlggdcOpFC+tF6OlVC3qaUjpxufvuu9NDUL///e/7aoimNJh19EXASYi+cHllEzCB q4EA9/fy5GnKj/3Yj1U+hI1lODVV97qyjCIHott6r609/r+631f1Vm1R975WLRvmPD0ArWqbub5O 61X9dpB5vdQBl05tOUi9Vb/pRQu/U1tWbWNY86ijF8c8Qgs6eqmnV37jYYQO3YaR22p5m9iL1isv G+b/vexzL+zGqynvI524RI93VfslfWjuxcarttHPPPFXvWP9tpe2HOu3vc7vpEX6omy3kxb2R3oi uLDPYxXpiLLdqofM5tqkpxMXXnFKIoK3Bj300EP5zz1tAhOCgJMQE6KZLNIETCCSAE8O/973vpee A/H2t7+9Y9VyEqour5QjgWOj9TpubJwLuzl8aMwDiXFW1/Hncv6ruOQ/jOCiOtQeef1Mo5E2Er/y 8mH+L/7UWdbD/5onzcOsu7ytvA7Vm6+jtsvXy5cPc1r8cwb59qUvQguJBQUsqrdKi2w8X1bHtAKW Ki3RbSQuqjffX82L4iJbUL1jaYlIFGnsqGoj5vFBRxO0wAkd0pxzG/Z0J9tVXWiRXWleHd+yl05t FHWs7qSFfUejxsSxWMD2Xe96V3oD1be+9a32w3THWt/zTaBpBJyEaFqLWI8JmMDICfzbv/1bemXj xo0biwULFnTUo/ewV93rq3k4CxEOH++nx3HBKS87WprHq+siHD72GS1V97SihQ9M5KR2hDzOhdw/ ixaYUG9epAWnUI5hvnzY09pn2Ua+fWmhffSKwXz5sKdlC3Apt1POKkILgQj2SylryefRlnWXnH+V FrVdBBf2tROXSC3d2khaItoILqqHesvjndoteuwVA/SpSJ/0an5d37IX1ZvXIy4ah/JldUzTRxh7 x+LCfPRiW3UXccnHNtUpVrmNa1kd3xxrGGc05ud1oA893WwXre95z3uK2bNnF7w69wc/+EG+GU+b QOMJ1N/rG4/AAk3ABEzgDQI4Bd/85jfTjDe/+c3p9ZxvLL1yiiQFzgDOXdkR5mFbOBOsgyNWd5kx Y0Zy6NDCfuRFrxzEYYkI/NFCPTAYi8vChQtDuODwoYe2KHPRgw4JEJRQyrkNexpHeObMmcle0JMX tMFL7PJldUwTiMydOzcxKT8YTvZMG6K37oJDjhaKbDWvE31ogU1EwTYpeiVlXqdYzZs3L59d2/T8 +fPTGFPFhXk5u9pEvL7hRYsWpakqLeIidnVrgQs2TL3lMUbtxjrwqbtgu2ip4qLxWOzq1jJr1qyU UGV8KY8x0ie9dWthTOWjcTavj/EOfVFaGDtIilAvbZIX2ZDY5cvqmOaYxLhKveU20nEALd0SnfQ1 nlnF62q5GsLFBCYSASchJlJrWasJmEDtBA4dOlTs2rUr1fOOd7yja304EyQZcO7OnTvXXh+nmNs6 cICjghWCNBw6tMgJRxBaTpw4kQKZxYsXhwT++X5zz2pe4ITj1e0qk/w3452mLuo8e/bsZZs6efJk 4kEbRSSKqEP2UNaCI4lTitaIM4OAUCIIW80LWnDU0RLBhbrhgt2cOnXqsmQROi5cuJASEDqbmWut Y5oAgbrEQXXQr7DfOXPmtM/Ea1ld32ghgKIf5QELXJhHkIeeiEKyjuCI9uDZOSrokhaWRxTah/GO 9lBwTb0EcWghKRA1xtAGCiovXrx42e4fP348JdDU7y9bWMM/9CH2G1vNtcCFvsXYEsWFNmC/GdfK 4x1jDscmEkVRYwzHP+qs0gIftEQU2gDbpd8wxuQFbWhkOW3Zrdx5551pfW4hdTGBiUTASYiJ1FrW agImUDsB3ojBpY04Cb28FQNBq1atSk7A/v3728ETDhbO57Jly8LO3qIZLXzv2bOnffYJh5wkBI5n 1JlkuKxcuTI55nv37m1rwSk+cuRIcvZwsqIKD/EiaKGNFLCghf/REeV8sr/URdBI3Qrk+D58+HAK bNEaVdh3tFC3tBAw0GbYCvYUVQjk2HeccPoOhUAbW6bIttM/Nf/hDCRaCNoYDwgKCFJIUmI/kVqw 2yVLlqTA/+DBg0kLeg4cOJACTbSQDI0oGmPoO9gMOvjAhXFmzZo1Xc/eDksnAS5jDO0hLmyb8QUb Wr16dRgXgmj2nULfUbKI4wA2RPtF3Y5B4LpixYqkAS30IQoJV2wZLZHHgeXLl6exFy2MLRQSR9gM Y2HkcYDjsY4D0kKyBi0kS6Ku4oEBbYRNYLtKFpHco18zJveq5fbbb0/H/G3btqXjPNt2MYGJQGBK 6+Ax+vfKTQRS1mgCJnBVEHjggQeKr33ta8XmzZsL3r/da8GJwHngrAYOKQ4yzp7ONPe6nWGsh1NF gCst6EFL1FUQ+T4oiCQoIIBBC0zQ08tZnnxb453G6YQLwRI6OPyhBWeQ/yMLQQpa8isQcMhxkrGd yKLg+tixY20uBAYEeNFaaBMSEAS46OJ/khMEeAQP0QUmBG4KWAgaSE5wW1N0IZglYJEWEiUEeFFX QeT7i93SRow1FCVtos7251oYW+AiLVwRxvgSdbY/16LEJt+MKYx5tBF9O3qMIaCFi44D9GXGu8gk p9hwOwbJRDTBgQ/HIz4wiiz0n3379rWPA2jhVhm4jEILbUTfFhfGXo5JvR4f2Zf7778/jVO/9Vu/ VXz4wx+OxOm6TGAgAi17n+IkxEDo/CMTMIHJSIDA8Cd+4idSYPgLv/ALxSc+8Ym+dpOAiaCb4AkH gg+OxSiKtPCNY2Utr7VCzoW2wTEfZRthc9gLbaTPqO1l1FzYf/UjtGC70cGB2gB7kc2MWou4SBO2 OyouaMFudZYdHdEJK7WRtNCXYKM+Pap+3SQusl2+R92v4aLPqO1FXNAjLaOyF7ToODBIG3EF28c+ 9rHimWeeKe65557id3/3d8Ouusn7oKdNoB8CLVufEnu6pR91XtcETMAEggns3LkzJSBwBG699da+ a+d3UZdFdxNnLdWEmsZllIFbTqhJXNDVJC6w4ex6E0pTuMCC4K0pXNDCpwmlSVya1K+b1EZN4oKW 8fRrrhTbtGlTSkJwdR1XBkXd+tOE/mYNE5dAM0bsicvPyk3ABCYRAZ4HQeGyby6LdzEBEzABEzAB EzCBJhPguRAUbpHi+U8uJjARCDgJMRFayRpNwARqJ8BlmZxFoJCE4N5ZFxMwARMwARMwARNoMgGe YUXhmRt6kG+T9VqbCUDASQjbgQmYgAm0CHBvMw+IonBLRdQ73VOF/mMCJmACJmACJmACAxDgYZac POH5Ek899dQAW/BPTCCegJMQ8cxdowmYQAMJ8GR1JSF4qjr3WbqYgAmYgAmYgAmYQJMJ8HYavQ1m x44dTZZqbSbQJuAkRBuFJ0zABK5mAiQheI85ZePGjVczCu+7CZiACZiACZjABCHAA2J53Sll9+7d BW/McDGBphNwEqLpLWR9JmACIQSOHz+enipNZaN4j3rITroSEzABEzABEzCBSUWA1xjrOVYvvfRS +6rOSbWT3plJR8BJiEnXpN4hEzCBQQjwMCfup6ToYD7IdvwbEzABEzABEzABE4giwJUQuh2DqyA4 qeJiAk0n4CRE01vI+kzABEII5PdR6rLGkIpdiQmYgAmYgAmYgAkMSOC6665r347x4osvFufOnRtw S/6ZCcQRcBIijrVrMgETaDABJSF4wNPs2bMbrNTSTMAETMAETMAETOANAnqjF68bP3PmzBsLPGUC DSVwXUN1WZYJmIAJhBHgHkqdOZg1a1Z61dWgleMAUKZMmZI+g25nvL/TrSV8X3PN6PPNTeECV5jw GXUbNVEL7YS9wGaUxW1UTV9cWNqUft2kftQkLmhpAhuNvU2xl6ZxaUIbqV+PR8u8efNAm8qJEyc0 6W8TaCwBJyEa2zQWZgImEEWAyxd5OwaFV3NyNUS/5eWXX073YZ4+fTr9lO0sWbJkXAmNfjVofRya o0ePprMhr7zyStonbjGZPn26Vgn9RgtccIZhyxmbmTNnhmqgMrjgnJ06daqgvXivOvfRjurKl5Mn TxZ8SILBZf78+QXve48ucCEJRzvRF9CCQ8sHpzi6cE/zkSNH0hPer7/++mS3PCx2FEEU4wJcLly4 UHDJM0lKnhnDdHShbeCCFtqFPkS/hlF0QQtcsBvaZZRa6D/Hjh1rP1h4xowZicsg4/h4OTLewuXs 2bNpvOE4gO2OSgvPBuCsOLoY/xl7R3EcYIzBdrEXxl64jOqYVD4O0DZoGdUxieORjgMckxh3BzkO 5PqxQRcTaDqB0Z8eazoh6zMBE5j0BHCo+VB4wFO/AQYO3vbt24t9+/alAGXZsmUpsPzRj36UAt5I gDhY27ZtKw4cOFDMmTOnWLFiRXJAt27dmhz1SC0kHXbt2lXw0E+Ct+XLl6eABX2HDh2KlJICAl7B urv1+jIcYLjwRPGnn346aYFbZNm/f3+yGZzOlStXpkBSNhSpg/2mLeCA7aOFIIV2Q6POokZpwiHH Vl944YXURtgNAeYzzzzT7qNRWgj2H3/88ZQMwXZJEh0+fHgkWgjennzyyZTMI6glEULgAhclUKO4 MFY+9dRTaWxjrCOAI8n4xBNPhGth7MV2CXDRQTvRbtgQzCILXJ599tnUn0huwoaAm+NA9OXxOiYd PHgwJVnp1+jDhkhMRI53Og4w/tKf0UL9tFt0sEy9HAN27tyZkg5o4TiAPWNDkVywTR2TOA6sWrUq aUEbx+9+x948ueQHU0b2fNc1KIH4VP6gSv07EzABE6iJAAGPkhA4A/2eWcTRw8nctGlTClSQiUNA cICTwXTEmTAcKBwpgrabb765/ZYPzpDgDOPccIal3/0bFDuOEM4UXLgqhMJVB5y5JGGDLpzSiMIV ELTT2rVrU3BAneJCG5GYGOTs0yDaCY5IzOAAkwzhrLY4yF70pPNBtt/Pb7jqgLZg31evXp1+Shvh mJMUwVaiXlmL001QTV/BZjjDjhba6Qc/+EFqI9ovoqCF/SchCRfsg8L4QPBEO61fvz7k6gy0kBBi PLnrrrvaSVLO+BP4069vvfXWsKtWCOIYLzdv3lyggQIfAly0MPZgP3UXuGC758+fL26//fZ2H8J+ aCMSaXfccUcYF8Zekmhw0ViC/f7whz9MWmijiOMA3EmWMebRFhpLaCMSwLQf402UFrgw9rL/JPJ0 FQ9jD89ighH9KqLAhKRrfhyABfaMvZCIzW9rqFMTPgP2S/JBxwH6E+MexwemZUe96MiTECQoXUyg 6QR8JUTTW8j6TMAEaieAA0IigoIT0k+QTuCPU4MTlTsMbIOzlQScUQ4BiQYSEGUtODVchst+4pxG FLhQF9/lV57iiKIVhzCqkBCBQ95G/E+ADZfIM3LsNwEUbZLf7oDzS/CGPcEtohA0cdYULXmBE3wI fqMKZ9NJUHFGm7pVcMYJFLBt9VMtq+v74sWLqe9iq0pAUBeOPkkRAqsoLZzZ5/J+uORXadHP0Ucw w/KIQvvQlwgclYCgXrQQ7DLWRV2BAH/6Ev0m10JwzZgDl8ixF5vAPvIxhv5NApZECX0toug4QJ+h nVSwHfo5wX/U2Ms4h71QN/ah8Y5vxl7GHsbeqPGOsZWxRYkZsYELGtAaUaiL4yNalJihXrigDW79 csmTEIwZLibQdAJvHOWbrtT6TMAETKAmAjjWfCg4sLmj361KnBacYZIOeeCkbRFURjk2OP843lVa FEixHMev7oIThOONlvJZUd3yQlAZUZTwoF3LbSsusCMZUXchACAwqtLCPNjATfZYpx7sgDOU2C1B ZF5kQ+iNuNwfp1v2UNaCLvolOkgORBS4ECjkjj31EiSgBb30pYgCF2xTtprXyTzsW8+iyZfVMQ0X 7KaqjZgXGcixz9gD7aHgVvssVvQ1NNVdGOOxT7SUC/PoY1GBP+MH4xl9eKzxDi0RXHQcqOLCWMex gXaMOCbRPtRFW8AmL7Ih+jR9u+7C+I5t6liY1wcT9NGO/RyT8rGKcdvFBJpOwEmIpreQ9ZmACdRO AKdDTlCVU9BJAI4CpRxo5/NwCAgU6i7Ug2NZToZQL/Nw1HHE+nFsBtVMPXAtO8FlLRHBNoEKWtj/ crDC/7ChfdBcd1E9VfYifTCJOMvO/souyzYjLvCIOKtG+yjBUMVG+iLOsjMWyIlXvbldaJ705svq mNYYo3rzOjQvQkveRmP1a7RFtFFeDwzK/Vo2FDX2wn+ssVf9OoqL+qvqze1FXBhfIsY7jjWMMao3 14I+5qNXx998+bCnxYXtqt+oDnSgB63q+1pWxzf8qQsdZS3Uhx7ap59jdd4nI44fdXDxNq8uAk5C XF3t7b01AROoIIADhINN4UBe5TBV/CzN0sG+ypHQPJzTiGBbDkuV84lYOVkRDh/7280pR1MEFznb nbjARAE5uuoq7C91YRvoyQv/y2YiuGC7tFFeb1kP/0doof+hZ6w2iuSCHaiPqN6ci+ZF2Av15vab 62BaY5X6fnn5MP+njbTPYpBvX/OwF+yq7qJgsdyPqDfXorasU4/4qz3yutCCxigu0iIGuRbpy9sy Xz7sadlulRa1GzYVYS+duKiN0KH1hs0i3578BjHIlzGPzyDHJCUi1Dfy7XraBJpGwEmIprWI9ZiA CYQT4GAvRxUnrcphqhKFw9KL84TDp+1XbWdY86iHUuXYaD46tF5auaY/ve5vRIDbTQu8otuoG/YI LuxzJ/uVHSnw7KZ5PMvRIbtUvVXbi+LSixa4aL0qrcOaJ/vtxEXrDKvOqu2wr9rfTlrytqzazrDm aZ+rtGhernlY9VZtR1yqluXzIuy3ExdpieIiLaq3/K12iuRS1pD/D5dumvP1B53uVsegXJRkIpHS q00Oug/+nQmMl4CTEOMl6N+bgAlMeAI4BHIKdEZk2DvVKdgbdl2dtoeOCC291hHhKPWiJYpLr/vb 63qd2rqXZXJ2q9bVsl74Vf2+33m91BPFpVftvWjudVtjrdepDrWRuVTTi+LSqY1yZVF68jqrptEb oWWicYFVBJeqNqma1ys//TZ/1oWuQtEyf5tA0wg4CdG0FrEeEzCBcAIkIOR44NTLse8mhPV6uWqi 1/W61ddtuc6CdHJc0NuL5m51dVuuy0K7rSfN3dYbz/LcMRtrO1cjl27sZUfd1huLaT/z6SO92Ewv 6/RTb9W6vdpCVL/utM/RbaSxQ/VW8WsaF/TUXXrtI53aclgae9FCO/ay3ng1ddtf2VGEll7qwFZ6 Wa+pXHLtEbeVjJeDf391E3AS4upuf++9CZhAiwCOB0kILq/uNQARODlZcqY0n2/NY/taL18+7Olu daCHdRRIDLv+fHvUw36LQb4sn8eDQOsu3ZIQ6Il0yqkrZ5Dvv+ZHcFEbUb/qrdLSza7y3ww6rT6C jk5aurXloPXnv8v7iJKT+XLpy/nly4c9LVtQvfn2NS+ijQhwVI/qrdLS7xiab6Ofab1xoZMWNEeM d7LLsbQwXzbezz4Osm43LWwTJmrLQero9TedbFfbgIs0a14d36qjWxtpvTo0aJviov/L37KXfrVg 77pNLE9IlLfv/02gCQSchGhCK1iDCZjAyAlwwNanHzFyJqqCFc3D4evXmehHg9bNtZQdLbTwGWQf tf1+vhWg6TaX/LdoQwvOpzTny4c9nQcrZS78j0a4RLSR6mH/q7REcoE9etQeOXdxYV4El9wWqmwG LhS1Zfqnpj95gNZJSwSXfJ/FIN9t6YvoR71yiWgjGGifq/qSnmPCOISN112wBWxY7ZHXp37eJC4w iUhCdOJCu/FhHWyr7qL+SnuoTVQn7ca8qISIjo9VtosO5tNG/dqufsN+uJhA0wnU3+ubTsD6TMAE TKBFQA4IzmvZQekEaP78+clx4cFa5d/p4VALFiwIcbJmzJiRgjQcKpyYvOjSzJkzZ4YEcmjB6avi ojM1c+bMCeEybdq0YurUqSlAKAcJeiAay9Fcd1E91FtuI7TBBr0KsOrUQ1DEu+WxW3FQfeJEcDB7 9mzNru0b5xl7oOjJ8Xll6MNxnzVrVj67lmkcePosRf0mr0ispDdfVsf0vHnz0hhTpYV56I3Uwj6O pYVlYsd0nQUu2KfaI69L+uDSbyCXb6fXadWjevPf0afpT+iNKPQRuGiczesUK44DCsrz5cOe1lim evPtM/6hUezyZXVMiwvjHfXmBX3M17EiX1bHNGMZbVDVRhqPBzlWw5TxgE95H+vYD2/TBMZDwEmI 8dDzb03ABK56AgSuOC44MbkDiiPBu+MpUc6nAlweSKUgUg10/vz5NEmAgINSdyGIhg1OUfkBWbw+ DH0LFy6sW0Z7+3Pnzr2ijVh47ty5tA6OcAQXAiKcYeyj/Bo1gm9siKA/InBixxctWpTa6MKFC4mD /qBFgVOEFtjTRhTZqrTwTV8iaUJfiyjYA4EcXGgrFRx77BktEUkr6sUeqO/s2bOS0f6GFcGk2LUX 1DRB0pUACi55Eo1p7BktEUkrdo8EGh+0jDXeMcZE9Gv6NGOe+k2OH33YEuwiCjpoA+w0PyZRt2wo iouOA3ApB8XqW9gufOou1MHxD1spH5MYX7BhtETYC/2EJAN+Q1kL/QgttGFEoqhu7t6+CYxFoP5e P1bNnm8CJmACDSGQBxmDSFqzZk1yPg8cONAOWAgOjhw5kgJtnI2IQrC4evXqVBVa5Jjj7B09ejSd cYo6S4mIVatWpeBp37597bOVOMWHDh1KDlikluXLlyeHbu/evW1nGMd4//79KZiMTIgsXrw4BU9w 0BlCHPTDhw+nQAZuEY4wbUQbEMihRQELdkOb4QCvXLmS1UIKgRztcPz48XawhDOOLcNp2bJlYckZ kh1LliwpTp06lT4CcOzYsRT00kYEWBEFLdjMmTNnCupnvOLD+EISbcWKFamfRWghkGOM0fimOk+e PJnaDS1RiSISM9gn45u4oIc2O336dEGfj9ICF2yCfkzfwW4pBLcHDx5MyT4SWxGF4wBc0EK/lhbs h+MASfEoLYxj2AuJq927d7e1MPbChT5PIjSqYBNoYdxXUoSxBS0cpyO1aAyhbh0HdHxkTGb86bfk vkw+3e92vL4JRBCY0jLSN1L8ETW6DhMwARNoGIHHHnusuP/++5ODdN999xVbtmzp68wMTh7OHcES Z0cJ3jjjhAO8fv36sGBFWHHICbbRQqCkM3FoiXLK0cLhhWBgx44dKdAlaMApZz5aopIz4kJwAhec UK4a0dmvTZs2JS5RgT/7T/CIU06dtJPOCpLQigoQxIWAkjYikML5JUDg7BxadLuR1q37m7OABHG0 DfZB3yJ4wmFfunRp3dVftn0CAtoIPpyVJDnDNIEKQR68ogpBCrYLC11Jgw0RUNJOEVeraF8J3gji 6NvYLjbMeAcj+nUkF+wDLidOnEj2Age0YDsbNmwI5YIWEkMcB2gjxhnshTF43bp1acwRw7q/GWNI QJDYZKzjo+MA411UAo39hAtjL30JHRyDsF3aCnthzIkqOibt3LmzzYUxh/lr165N7RalhXroQ3v2 7Em2Ahe00NfhQn/qt9x7770FCUHKww8/HHb1Tb86vb4JtI4bU5yEsB2YgAlc9QTGm4QQQIIWHD2c LpwtnCsc9FEUAgUcYLTo0s9RaoELQRxaCFwiA5WcvwJJvgkSCBZGxQUNcKGt0IK98D2KosQIWggO aKNRaiEJQTJEiZHIoCnnDxd0oAc7oY1Ipo2qEKTo8m2CFrSMyn7RARcKWviMqmjspb0U6I6KC1rg wthL+9CXRlWkhbFGWkbFhbGlKccBaeGbY1KTjgPYy6BJRSchRtXTXG+/BFrj0JTReDv9KvX6JmAC JjABCBAojSpYKuMhgIy6T7xcd/l/tESf3S9r0P84d03SMsjZLu3LML8JTJqkBUd8lMGb2MKFoJZP E8qog/2cQZO4NGnstZbcSt6YbtJxoElaOCY1Zex9o7U8ZQL1E4i7lrD+fXENJmACJmACJmACJmAC JmACJmACJmACDSbgJESDG8fSTMAETMAETMAETMAETMAETMAETGAyEXASYjK1pvfFBEzABEzABEzA BEzABEzABEzABBpMwEmIBjeOpZmACZiACZiACZiACZiACZiACZjAZCLgJMRkak3viwmYgAmYgAmY gAmYgAmYwFVNgIfquphAkwk4CdHk1rE2EzCBMAI+YIehdkUmYAImYAImYAJDJpD7Mfn0kKvx5kxg KAT8is6hYPRGTMAEJjIBDta80513qV9zzTWFD94TuTWt3QRMwARMwASuPgLyY9hzXkPqYgJNJmAL bXLrWJsJmEAIAQ7czz33XPHyyy8XL730kpMQIdRdiQmYgAmYgAmYwLAI4MfwcTGBiUDASYiJ0ErW aAImUCsBrn6YMWNGuhLi2muvTVdFMM/FBEzABEzABEzABCYCAfyY559/PknlpIqLCTSZgL3sJreO tZmACZiACZiACZiACZiACZhAHwQuXbrUx9pe1QTiCfhKiHjmrtEETKBhBIZxsGYbFy9eLM6ePZv2 burUqcWsWbNGcl+mtJw/fz5d3SEt119//UjIw+XcuXPpCpMbb7wxXXXCd3SBC2eJ0MJZIjTMnDkz fUdrob4XXngh2Qu3ADVJyw033FBMnz69mDZt2khuTeLZLGfOnElthc3Chr40ime1oAV7wYa5xxom nG0cxZVS3DbG+MLl1rCgjfiM4t5vuFy4cKFgjOHqMY0xo+IiLfQr2oh+PQou1I+9oIcCl1FpKR8H ZLv071EUmPBh7B21FvpQfhxgfBkVFx2TdByADf16kEKbq+TTmudvE2gSASchmtQa1mICJjAyAuM5 YPPbAwcOFPv37y8WLVpUEDgdPXo0fW/YsCE8yD1y5EjSgvPLZ9++fSloQsugzs2gDXPy5Mli+/bt xZw5c1Ldx44dSwHUunXrkrZBtzvI79CyZ8+edsB0+PDh9AyQjRs3hmshyN6xY0eyDdgcOnQoaVmz Zk0xf/78QXZv4N/gjKOFYHLevHnFiRMnit27dxfYC/9HFoID6iZYWbJkSQpyYbVixYpi2bJloYkI Au1nn302JYvgQLCAtsWLFxe0U2RS5MUXXyz27t1bnDp1KtXPmEO/XrBgQdJC20UVkiG7du0q6E+w oKBt7ty5BX0pkgt106exWfoNYy+2zDiHlsjEK21CP6ZdFi5cmIJa/kcL4110kMsxifpnz56dEme0 GRrgQnIksjDuUz9jHUk87AWbXbt2bfjYe/z48WLnzp0psUnyAUYcuxnvOF5GFvqQWDDGoIUkDfYC KxcTmMwEfDvGZG5d75sJmEAIARwsHL7Vq1cnR4bvW2+9NZ2xJGjBaY8qBGzbtm1LAdwtt9xSrFy5 srj99tuTY0NANZ5kS7/7gJZnnnkmabnpppuKVatWFXfccUfazNNPP5009bvNQdfnjC1ccOw2b96c tNBGBClojHyYF3Wx/zi8aqPbbrst/c983dM76L728zuuxoALZ7BpG+zl5ptvTkEU9qIre/rZ5njW pU6SItgs9oLdEBwQNBBoRhX6LFyw4fXr16dAn+9NmzalQOHgwYOhfYnAlqQZNksChOANNiQ7GXsi CwEbgRztgg4+9CnmEVBFjjEKIGFC+2Az9ClsiKA3smj/0QIbjXenT59O9ht5HJAWEnf0Z/o1YwzJ LOw68nkB7D91ogWb1TEJLYx3JPuiCuMZSSqSRPQlaWH8ix57dRwgScV4pzYiSRN9TIri73pMICfg JEROw9MmYAImMAABggDOLOmsIJvgjNPSpUvT2ULOdkQUnDm0cPk6Z5FVOOPEmWQccwKZqEJdOLv5 GWzOksKF4JcrNqKK9pu6VXA8CRS41J5gJqoQTOOAw0VnsOFCm/FN4ioqkCPIJumBlrxgy2hAS1TB PgkS4JCfqeUMIY45Z7y5ZDmiwISrDriyibOlKiSxOONPII4NRxSCFWnJz9Ryhhst2C5JtohCIE0C hrrzK3YIpLCZSC0aQ9DBFSEqaMGGSNBg3xFFXBj38+MAt4Qw9tLnI48DjK1wwH5VOC4sX748McGe IgpjCPbCuAYHvik6JmHbaI0a79BCW+XHAfTAhbEl8jigRCb2Ii58MxZj2xyzorjAwMUEogk4CRFN 3PWZgAlMKgIEsAQAOJvle5C5txMnIsr5RAtON86mglvBJoijEORFODY4UZwBQ0f5MmT0wSrKEWa/ aQOueihfni0utGHEmUoc3bG4wIkPtyJEnR0kAMDxxVbzQhKAJA32EqEl7ycET3lBH3pIDEQF/gSN JNDQogABTTDBfkkiEUBFFOyFfa/iQlICnSRwIgr9CBvGXnIusiGWRQX+9BP6LbZR1gIrbAp2EYV+ wviLbWAjeVHfYp2IorGXsaR8TGK8g0vU2Ivd0k5VXGgj9EUdkxjH2G/14bwt1EZRtkufpS6N+bkW Hafo0xFjb163p00gksDlI2Vkza7LBEzABCYBAW7FoJSdPeYRgOMc42xEOBM45Dg3qhcNKuhDC+tE BHLUQ0CCk1Uu6MMRJEBAb90FJ5h9Zv/LAYJYyVmuWws6cC7VHnl90kdwS5Bbd6Ee2FAvHPICJ+YT sEQETwoQ0FDVl5hHkihCS54QGUsLOqMCFhIiaBpLC1witKABLZRyMo950qd1mFdnERf14bwu6cNe IsYYjWVikGuhL/HhFomIgi1gE3Ap92vpizoOMMYwtpZ1wEFc0BvRRrIFjbN5W6APPWiNSC4yvjP2 ikGuReMxy9HjYgKTlYCTEJO1Zb1fJmACIQTkJOA4lAsOBgWHMMLJUuCKlio9zENHREKEBARBy1g6 4MJy1qu7yKms0kLdtBNcItpI/NFS1sP/aIGL2rJONiREsE3VW65L9huhBR3U002L+ltZ6zD/xybV R6qCJ7VbhO2yX7JftUe+r5oXoQUmqkf15lrEJcJeqFfJ1Cotajf1t1xnHdPa5yotcOGjdeqoP99m L1xoy4jxrpMWccGmGPPqLrkW2arqpN2YFzX2ahxTvdLBNzqYTxtpHMqXe9oEJgsBJyEmS0t6P0zA BMIJ4LAQPFHKTk0+j/UiHD7VUaVFeqIcGzlPVVqYp/kKatBXV8nrUL15XcyjjaQ5XzbsadqIuqp0 5HXlmvP5w5yWvbDNKj2al683zPrzbfXaRyK40KfRQxGDXKvmqS3zZXVMi7/qzevQvCgu6iNjBdto i+pL2mcxqOICO43R+fJhT/fSRr3a+Hi1ddpfsYo6DnTiwn6iBy5qy/Hue6ffy3ar1hGXqDYSlyot zIvkMpYGzzeBugk4CVE3YW/fBExgUhPAaaHIianaWdbp5ABV/WaQedLS6bd5gNVpvfEu6+QI59vu 5ozl6w463SuXiDZqGpdObGTTEVxoW2lRvXl7a16Ell77COv12p75vvQ7LS5Vv4vkUlV/1Tz0dtJc 9ZtB5vXCvgltlO9bxHgnLrKNvH5N0z5RfUl1dvpuipbJxKVT+3dqCy8zgSgCTkJEkXY9JmACk44A B3mdEezkdOfr1QlBlyB30xLhnEhLt/3VPcrd1hvP8l7qoB3VluOpq9tve+USpaWTLciOIrSgoxMb aem0Tjf2vS5nfztx0XZYp5f1tP6g3532OZJLr/vb63qD8tDveuHSa1tqm4N+99pHOmketO7y73qp gzbqVXN5+/3832sdo+5H2qcoLr20EZp6XS/XXzWtef42gSYRuK5JYqzFBEzABEZBgLMw3HfNN2eq +nGI5CQoGMj15/N6CYTz3w4y3a0O9KC3V8dwEA36jbTkDLQsn6f1tKyO7251oAcm3dYbhjYemId9 5Qzy7Wq+HqyXLxv2NPsrW1e9VXVEaKFe9ND/qrRoXkQb5X1EZ5VzLtLCeuKXLx/2NPx5noDqzbev eVFcVE8nLjCJGGNkl2JQxYU24lN36UULGsSvTj2qoxOXyPGOfa3Sks8Xvzq5dKoj19dpvWHpq3pg c75t6VFb5ss6TfOGFj6UiD7YSYuXmUA3Ak5CdCPk5SZgApOeAA41T+gmCaEDeK87LWeiyinXPJyB CMdGdeDA8MkDJP5HT9Wr0nrd137WU4AmBvlvpY950pwvH/a02iivN6+DdmediGAFW6AeuKAnL7k+ ac6XD3sa9rKRqnaCC6VfR3gQnQqKYFClRfMiuLC/cuBVb75Pmpfzy5cPe1p9pGwv1CMtUW2kPqJ6 833VvCgusgXVW6VF41C+rI5p8a/Son4dxaWTFvVpuKgt6+ChbXZqI3HJ+5t+V8d3zoW6NfZRF+3G PI1DddSfb1N9WvXmy9DBfNpH41C+vNM0r9DVm3Iibv3ppMXLTKAbAd+O0Y2Ql5uACUx6AjhKy5Yt S5/Zs2dfERx2AjB37ty0WM5dvi7zcCh4N3uEw6d6VG+uRQ4J70OXY5gvH/Y074Bnn6seOIY+nCz0 yhkbdv359qgHB5Q6+eRFXGCid8Xny4c9TRIINlVthDbmo4X16i5Tp05N9ajevD7mYbuUmTNn5otq mcZWZs2albY9ls3gkGudWkS8vlGCkzlz5nTUwsIILdSjMaaKC/PQG9FGaOmFi/Syfp2FsZpS1ZfE Ci4RYwz1dBrv0BjFRXZJnWONd4wvEWMMYypjr8bZ3B407sBOCYJ8+bCnqYe+oiA/375YYSsRxwHq gT8Mym2EPniho982WrlyZduXyZMs+b562gSaQsBJiKa0hHWYgAmMjICCrUEE4PDhKHC5tBxfbYdb PHAwFi1apFm1fqMDPejAqcoL7xynzJ8/P8ThI8DFiUJH+dV0/M/8BQsW5BJrm1ZQiWNXdob1+kO4 RQQrONvUVcWFdoMNyyOccrhgD9i/OKgRZEM47v06wtpGP99oUZBWdTUS+qISReiGC5qoNx8f6M+8 6k/t2M8+DrqubLPcRmyPeSRn5s2bN+jm+/od9VAfrxjMubAR5kVqgQsJPRiMNd4tXLjwsjPefe1s HyuTnGH8wDbKXNCnvtbHJgdelX5C4rXqmKS+JfseuJIefygtcCkH28yj3ej3EUl62kcnGag7L7QR +mhHbLjuwv6ihXGWT174n+MU9g2/fkpue/l0P9vwuiYQRaD+nha1J67HBEzABMZBYNADNg7LkiVL kiOsyyCRgXN1/Pjx5AzqjN045PX0UxwWAnuCgVwLztWxY8eSk4xTHlFwuuHC94kTJy6rUv9HJWeo nLrUJhJDmx8+fDg5wJEJEerCbrCP3O5OnTqVNIqbdNb5DRe0qE1Ul/7nCqGoQjCCDVN3niwigUbw RODUr1M+qHaCW4IA2iRPop0/fz71LfpRlBbGD7TAJQ+eCJxOnz6dEhAkiyIKCSnsl/Hl3Llz7SrR hT6SFLCLKCQ5qe/s2bNp/FWdaKHdCMThFlEIcLEJbIW6VejfjL20T1SiCC30a/oMbFTQhha1oebX +U2wzXg21tjL2BN5HEALx0NuW8gLtktiMer4KL8BLrkW7IXjgrhw/HQxgclKwM+EmKwt6/0yARMI I8AlkARN27dvTwkAnBkcC+bddNNNBVcFRJXly5enQGXnzp0FwRuOOk4xjvnNN98ccrZf+4rDh1O1 Z8+exIXgRMELXKKCFfTg6OKE79u3L515IiggiIINWnQWXtrr/Cao3LhxY+KCjfA/WggoN2zYEHaF CPtIG6Blx44dxbZt25IWgluSM6tXr04BRJ0s8m0TsGzevDn1o61btxaLFy9ObYUWkiFr164NOauN Jvrwpk2biqeffjp90ELwghZsZd26dSFnTNFCQLJ+/frUPk8++WSxdOlSZheHDh1KgTY2ExmsoOXZ Z59NemgX6j5y5EgKbrElguCIQr1rWzaBvdJOaKHdjh49mpjRr2EXVVatWpVshH7EOAwHjgNqv0gu K1asuOI4QKBNgprjQMTVTeLO2MvxR8ckxhzGOh2TIo+P9GPG3L1796axBS0ae7HryOMASSn67oED B1KShoSZjkn0o6iTF2onf5tANIEpLQfx8idjRStwfSZgAiYwYgKPPfZYcf/99ycH8r777iu2bNnS t/PKUEpAy4dpnfmKdPaEkWAJJ48P0zh5nOFBS2Swgh5YcDYOBxgtaMApHYUW9KBFCSI0cFaXRM0o uOBwSgtn1TnTz9nbaC1wIZAjeCNRQ7CEvYxCC/bC5choIWAiMUFgEHWZNCzyQqBEG5HQI5hEC8EB wW50IXiiH2E3aEEHekatBXulfUaphXGXvo39wAU99O/oko+9aGFsYbyjjaL7NfYCExK/THNMYowZ BRdYoIN24uz/qI9J9CEdkxh7Ge9oq+gCF8YWrn6gjWgb2ojkyCD2cu+996bxiv14+OGH07ai98n1 mUAvBFr2PcVJiF5IeR0TMIFJTWAYSYhJDcg7ZwImYAImYAIm0GgCTkI0unksLiNADiLuOrWsYk+a gAmYgAmYgAmYgAmYgAmYgAmYgAlcfQSchLj62tx7bAImYAImYAImYAImYAImYAImYAIjIeAkxEiw u1ITMAETMAETMAETMAETMAETMAETuPoIOAlx9bW599gETMAETMAETMAETMAETMAETMAERkLASYiR YHelJmACJmACJmACJmACJmACJjB8ApGvpx2+em/xaiDgJMTV0MreRxMwgY4EWg/pbb8OK5/u+CMv NAETMAETMAETMIGGEMB/UcmnNc/fJtAkAvEvum7S3luLCZiACbQI8K5u3tHNx0kIm4QJmIAJmIAJ mMBEI/DKK68UL730UpJ9/fXXTzT51nuVEXAS4iprcO+uCZhANQEO3EpEkJTwWYRqTp5rAiZgAiZg AibQPAIvvPBCOwlBQsLFBJpMwEmIJreOtZmACYQQIOEwbdq0goM291E6CRGC3ZWYgAmYgAmYgAkM icD06dOL5557Lm2NkyouJtBkAk5CNLl1rM0ETCCEAEmHYZVXX301bWrUt3Von/huwgOqmsKFxoEJ n1G3URO10E7Yy6ivBHIbpWHkij/iwoKm9Osm9aMmcUFLE9ho7G2KvTSNSxPaSP16vFrYjosJTBQC TkJMlJayThMwgVoJjPfgzVmH48ePF6dPn046ubJiyZIlxdSpU2vVXbVx9uXo0aPFmTNn0tUdaFm8 eHHBWZJRFLTABWf4xhtvLBYtWlTMnDkzXApcTpw4UZw6dSrdekPbLFiwoJg9e3a4Fio8efJk+nAr EFzmz59fzJ07N1wLXM6dO5ds5sUXX0xa5s2bV/DBKY4unMk7cuRIOqPHfc3Y7dKlS0cSdD///POJ y4ULF4rrrruumDVrVrFw4cI0Hc2FtoELWmgX+hD9ehT3fqOFfo3dENyOUgv959ixY8XZs2dTk8yY MSNxoU9FF66mgwta6FeMvdjuqLRwEmX+rQAAQABJREFUTNJxgH7E2DuK4wAssF3shWPlKI9J5eMA bUM/GtUxieMRxwLsmGMS4+4wjgPsp4sJNJmAkxBNbh1rMwETmBAEcDy3b9+eHKx169aloACn+Ec/ +lGxYcOG5FRE7QiOx7Zt24rz588XK1asSM4eWrZu3VqsWbMmOaFRWkg67NmzJyVnVq5cWRAc4HCh D8d82bJlUVJSQLB3794UrFA3iQcSI08//XSxatWqpCcy4N6/f39x8ODBYvny5SlZhRZsiMQVeqIK 9nLo0KECPXDBGSfI3bVrV/rGhiLPoGIfO3fuTAEBdZOQOHz4cAqk6Es33HBDFJq0/08++WQxZ86c 1E4ET/v27UuJrE2bNoVqIXjDPkiEYB+MOdgPvDZu3Bia7CQB8dRTTyW7QAs2hBaCzM2bN4dqgQN9 mLZhfIMPWhjv4ELSKKrAZceOHclmV69enY4DJAE4DqAFO4oqOiZdvHgxjbOMvfQj7JljFMnXqPGO 48Du3bvTcQAuaIEL7cZxgTEnqmCraOGYiO2SeCABgD2jDS1RXNhnjkkkxqmX8RctjH8cBzg+Ro69 l7VBi9Olly+2Lht87crOy5b5HxMYi8A11xZTrp3GJWBjrXHZfCchLsPhf0zABEygfwI4vZxtIjDh bDaFs02cRcXJYDriTBgOFoEADtbNN9+cztiiBUcLJx3nhjMsUWdOcTQPHDiQuOBUUQj+OeNDMIeu qCABR492Wrt2bTv5IS60EWfmhnH2Ke1klz8ElCRncMAJtHF6xUH2QpAQUQjyaQv2HSecQhtde+21 KejFVnCOIwrByjPPPJP6Cn0JBxwttNMPfvCD1Ea0X0RBi4J+uGAfFM5UEjzRTuvXrw8JEtBCkojx 5K677mpfhUEw98QTT6R+feutt4YFTwRxBNwkHNBAgQ8BLmMMYw/2U3eBC7ZLwvX2229v9yHGWtqI RNodd9wRxoWxl6QQXDSWYL8//OEPkxbaKOI4AHcSDox5tIXGEtqIBDDtx3gTpQUujL3sP8dHxjv6 NGMPSRsYRV0xCBOSrowjSoLDAnvGXkhyciVCRMFnwH5Jhug4QH9i3OP4wLTsKEJPu45LrxSvXjxe FK/6wZZtJp7omcCla1pXx01vJRZb/bxbuabbCl5uAiZgAiYwNgGdScaJyh0Ggjcu2ybg5OxGRCHR QAKirAWnhstwcbRwTiMKXKiLbzjkBUcUrTiEUYWECBzyNuJ/Amy4cAl1VGG/CaBok/ysG84vwRtO MtwiCkETZ03Rkhc4wYfgN6pwNQgJKs4KUrcKzjiBArbN098jCmeQ6bvYqhIQ1EtCkQCKwCpKC1em cHk/XDjTr0I/Rx/BjG5F0LK6vmkf+hKBoxIQ1IUWgl3GOrhFFPjTl+g3uRaCa8YcuESOvdgE9pGP MfRvErAkSuhrEUXHAfoM7aSC7dDPCf6jxl7GOeyFuvOrL+DC2MvYw9gbNd4xtjK2KDEjNnBBA1oj CnVxfESLEjPUCxe0wS2SS77Pl15o9V8nIHIknu6HQMt2Lr3Y2zHgjaN8PxV4XRMwARMwgUQApwVn mKRDHjixEGeYoDLKscH5x/Gu0qJAiuU4fnUXAiccb7SUz4pytgmnlKAyoijhQZ188iIusCMZUXch ACAwqtLCPNjAjWCv7oIdcIYSuyWIzItsCL2cga+74HTLHspaqJu+hA6SAxEFLgQKBJV5IUhAC3rp SxEFLtimbDWvk3nYNwmciAIX7KaqjZgXGcixz9gD7UG75EWs6GtoqrswxmOfaCkX5tHHogJ/xg/G M/rwWOMdWiK46DhQxYWxjmMD7RhxTKJ9qIu2gE1eZEP0afp23YXxHdvUsTCvDyboox0jjkl53Wna CYgrkHhGnwRe7c1/cRKiT65e3QRMwARyAjgKlHKgnc8jkCNQqLtQD45lORlCvczDUccRi3BsqAdn ruwEl7VEBNsEKmhh/8vBCv/DhvZBc91F9VTZi/TBJOIsO/sruyzbjLjAg0Ci7kL7KMFQxUb6Is6y ExDRlyiqN99/zZPefFkd0xpjVG9eh+ZFaMnbaKx+jbaINsrrgUG5X8uGosZe+I819qpfR3FRf1W9 ub2IC+NLxHjHsYYxRvXmWtDHfPRGJCHEBQ3qN9KDDvSgVX1fy+r4hj91oaOshfrQQ/tEHKvr2D9v 0wR6IeAkRC+UvI4JmIAJjEFAwWKVI6F5OKcRwbYclirnE/lysiIcPva3m1OOpggucrY7cYGJAnJ0 1VXYX+rCNtCTF/6XzURwwXZpo7zesh7+j9BCgIuesdookgt2oD6ienMumhdhL9Sb22+ug2kFd+r7 5eXD/J820j6LQb59zcNesKu6i4LFcj+i3lyL2rJOPeKv9sjrQgsao7hIixjkWqQvb8t8+bCnZbtV WtRu2FSEvXTiojZCh9YbNot8e/IbxCBfxjw+UcekvG5Pm0AkASchImm7LhMwgUlFAIelF+cJhy/C EaYeSpVjo/no0Hpp5Zr+9Lq/EQFuNy3wim6jbtgjuLDPnexXdqTAs5vm8SxHh+xS9VZtL4pLL1rg ovWqtA5rnuy3ExetM6w6q7bDvmp/O2nJ27JqO8Oap32u0qJ5ueZh1Vu1HXGpWpbPi7DfTlykJYqL tKje8rfaKZJLWUP+P1y6ac7XH3S6Wx2RXAbdB//OBMZLwEmI8RL0703ABEygBwKdgr0efj60VdAR oaXXOnp13scDoBctUVx63d9e1xsPF34rZ7dqO1rWC7+q3/c7r5d6orj0qr0Xzb1ua6z1OtWhNjKX anpRXDq1Ua4sSk9eZ9U0eiO0TDQusIrgUtUmVfN65Vf126qrT6rW8zwTGBUBJyFGRd71moAJNIYA jryceX33Io51eznQ97peL3V2WkeX2nZyXNDbi+ZO9fSyrOqe8arfSXPVsmHNKz+ErGq7VyOXbuxl R93Wq+LZ7zz6SC8208s6/dZdXr9XW4jq1532ObqNNHao3jI7/m8aF/TUXXrtI53aclgae9FCO/ay 3ng1ddtf2VGEll7qwFZ6Wa+pXHJbz6fHq9e/N4E6CDgJUQdVb9METGDCEeDySC6v5sAtR7uXnZCT JWcq/43msU2tly8f9nS3OtDDOv3s36AaqYf9FoN8O/k8ng5ed+mWhEBPpFNOXTmDfP81P4KL2oj6 VW+Vlm52lf9m0Gn1EXR00tKtLQetP/9d3keqzopKX84v//2wp2ULqjffvuZFtBHBmepRvVVasO+I MUZvXOikBc0RWmSXY2lhvmw8Z1bHdDct1AkTtWUdGrTNTrardeAizZpXx7fq6NZGWq8ODdqmuOj/ 8rfspV8tjFf4MXwi2res2/+bQD8EnIToh5bXNQETmJQEOHDzoCg+HLyrnJSxdlzORFWwonk4fP06 E2PV12l+rqW8D2jhg1MedaYH57Lq3le0oYXl0txpv8a7LA9Wylz4H40wiWgj1cP+V2mJ5AJ79Kg9 cs7iwrwILrktVNkMXChqy/RPTX/yAK2Tlggu+T6LQb7b0hfRj3rlEtFGMNA+V/UlxnEKgVjEeIct dBrv0NIkLjCJCFI7caHd+LAOtlV3UX9lbCuPvfQj5tGGWq9OPbCnrirbRQfzaaN+bZcHgcqXYRsu JtBkAvX3+ibvvbWZgAmYQIsADtD06dOLGTNmpOmyg9IJ0vz585MzwYO1yr/jKds4AgsWLAhxstCP o4tDVXZA9MTvmTNnhjjDaMGZq+JCgIC+OXPmhHCZNm1aMXXq1MRFQZvaVA9EYzma6y6qh3rLbYQ2 2KBXAVaderAV7B67FQfVJ070jdmzZ2t2bd8429gDRU+OzytDH477rFmz8tm1TBMc0Gcp6jd5RWIl vfmyOqbnzZuXxpgqLcxDb6QW9nEsLSwTO6brLHDBPtUeeV3SB5d+A7l8O71Oqx7Vm/+OPk1/Qm9E oY/AReNsXqdYcRyICLY1lqneXAvjHxrFLl9Wx7S4MN5Rb17Qx3wdK/JldUwzltEGVW2k8XiQY7X8 GI5l5X2sYz+8TRMYDwEnIcZDz781ARO46glwsMdxwYnJHVAcCd4dT4lyPhXgcjZEQaQa6Pz582mS AIGgpe5CEA0bHE29pk118mo99C1cuFCzav+eO3fuFW1EpefOnUt14whHcCEgwhnGPvSKQe08wTc2 RNAfEThR76JFi1IbXbhwQTLSN1oUOEVogT1tRJGtpn9e/0NfImlCX4so2AOBHFxoKxUce+wZLRFJ K+rFHqjv7NmzktH+hhXBpNi1F9Q0QdKVAAoueRKNaewZLRFJK3aPgIsPWsYa7xhjIvo1fZoxT/0m x48+bAl2EQUdtAF2mh+TqFs2FMVFxwG4lINi9S1sFz51F+rg+IetlI9JjC/YMFoi7IV+QpIBv6Gs hX6EFtowIlFUN3dv3wTGIlB/rx+rZs83ARMwgYYQyIOMQSStWbMmOZ8HDhxoBywEB0eOHEmBNs5G RCFYXL16daoKLXLMcfaOHj2azjhFnaVExKpVq1LwtG/fvvbZSpziQ4cOJQcsUsvy5cuTQ7d37962 M4xjvH///hRMRiZEFi9enIInOOgMIQ764cOHUyADtwhHmDaiDQjk0KKABbuhzXCAV65cyWohhUCO djh+/Hg7WMIZx5bhtGzZsrDkDMmOJUuWFKdOnUofATh27FgKemkjAqyIghZs5syZMwX1M17xYXwh ibZixYrUzyK0EMgxxmh8U50nT55M7YaWqEQRiRnsk/FNXNBDm50+fbqgz0dpgQs2QT+m72C3FILb gwcPpmQfia2IwnEALmihX0sL9sNxgKR4lBbGMeyFxNXu3bvbWhh74UKfJxEaVbAJtDDuKynC2IIW jtORWjSGULeOAzo+MiYz/vRbcl8mn+53O17fBCIITGkZ6Rsp/ogaXYcJmIAJNIzAY489Vtx///3J QbrvvvuKLVu29HVmBicP545gibOjBG+cccIBXr9+fViwIqw45ATbaCFQ0pk4tEQ55Wjh8EIwsGPH jhToEjTglDMfLVHJGXEhOIELTihXjejs16ZNmxKXqMCf/Sd4xCmnTtpJZwVJaEUFCOJCQEkbEUjh /BIgcHYOLbrdSOvW/c1ZQII42gb7oG8RPOGwL126tO7qL9s+AQFtBB/OSpKcYZpAhSAPXlGFIAXb hYWupMGGCChpp4irVbSvBG8EcfRtbBcbZryDEf06kgv2AZcTJ04ke4EDWrCdDRs2hHJBC4khjgO0 EeMM9sIYvG7dujTmiGHd34wxJCBIbDLW8dFxgPEuKoHGfsKFsZe+hA6OQdgubYW9MOZEFR2Tdu7c 2ebCmMP8tWvXpnaL0kI99KE9e/YkW4ELWujrcKE/9VvuvffegoQg5eGHHx7o6ptLF08Ul155od+q r1j/u//+ePHdx54oPvhT9xWLFsZcBXSFiIbOeOgbjxRPPr29OHb8ZPH+e99T3POutyelF1rHvYe+ 8Wjx3X9/opgxfVrx0V/82WLN6hUN3YuxZU257sZiyrTXbmkca63WcWOKkxBj0fF8EzCBq4bAeJMQ AkXQgqOH04WzhXOFgz6KQqCAA4wWXfo5Si1wIYhDC4FLZKCS81cgyTdBAsHCqLigAS60FVqwF75H UZQYQQvBAW00Si0kIUiGKDESGTTl/OGCDvRgJ7QRybRRFYIUXb5N0IKWUdkvOuBCQQufURWNvbSX At1RcUELXBh7aR/60qiKtDDWSMuouDC2NOU4IC18c0xq0nEAexk0qVhnEuI3PrmlJzO+6y13FB/5 0M8WX/nq14uv/P3Xiwc+9fHitls29fTbq2ElmMCGJMPmFpd73nV3cddb70y7vuX3P9NKTmwr1q5e mRI3H/yp/5CmJxqXXpMQo/F2JhpN6zUBEzCBHggQKI0qWCrLI4CMuk+8XHf5f7REn90va9D/OHdN 0jLI2S7tyzC/CUyapAVHfJTBm9jChaCWTxPKqIP9nEGTuDRp7LWW3EremG7ScaBJWjgmNWXsfaO1 rpzafMvGy2YSLHMmnwCagFpl7ZqJd+Ze2vPvz33hL1vJxOeKT3z8Y/nscU8/9PAjKcHwB1s+dRm3 3Xv3pwTEPe9+e/Frv/JL466n0wYe+ea3i0f/9Tsjv9LCSYhOreRlJmACJmACJmACJmACJmACJnAV EygHxp/7/F8Wjxz/dvH+971nUl7pQJJl2IWkzYVWYmPzzZsuS0BQD/MpixbUf+vKsWMnU8Lj/OtX sqWKR/DHSYgRQHeVJmACJmACJmACJmACJmACJjDZCfCsg0e/+Z3W7TjPtW4vWNG+/aC8372ul//u 2PETxdFWcM8tH73+vtN6XJFAQoCrIKa3rvAgGTGz9b2mdYtEt8K6W1vPepgxY1rBbSmLFr7xXAS2 u3vvgdc2MeVS2i7/oJvffe+xx9OyY63n21TViWaeFUEiY9Gi+cVtN2+8bPuvbZhkxsVUj3Rsbq3H 7R0ULaMOyp7X9bA8v5olLQz44yREAGRXYQImYAImYAImYAImYAImYAJXE4FHW5f+/9fPPJECe+33 ba1bOx741G/q3/TNwxp5VoKuCGAmD7R8sPVMiTyYv+xHrX++8tV/LLi94AP33lN8rXWrQ17ueffd rVsbfjmflZ5TQT15yev50l//r3aCAC08p4FEAc+2GKuQCPn0Zz7/RpKhteIX/+rvig/+9H3pwZz8 Lt8uyQQ+1PvZTz+Y6tC2H2ndJsEnvy2D/ftSa3tlNr/TulUkT46g48GWXhIVeflA6+GXH/nFn0t1 cpuJyhf/+u/S5Kie2+EkhFrC3yZgAiZgAiZgAiZgAiZgAiZgAkMhQLDNcxUI5AmOP/2ZP20F+dtb VwxsSw9mpJJH/vXbKWjnLRE81JKz8rxdg4CZ4J7nJ3Qr32mtr2Ba9aRgvrVNHgBJoR4SEHk9j7bm fbZVD3WRGFGy4dc/+WD6DUmCboXAnysn8vq/8tWHUl1csUD9LEMX280TDGz7f/75HydtaOBtIiQv VEgscOsLCQs48M12Hvz9P04Jh//W0qerGKp0kJjhM33a9LRd6oZBEx4aGvd+KdH0twmYgAmYgAmY gAmYgAmYgAmYQC0ERvUWlPLOfLSVVCABQSGA/vlWkE0hEaFCQMwyJSCYzwMv3/++e16/taD78xl4 ZkVeD7+ltG+BaE2rnl/72C+1A/f3tIJy1kWPbk9IP+zxD4kNkgIkD/L6P/Khn0t1UOd4in7/YCtB AiMK39THlREkUSjSUebN/yQpuJqiacVXQjStRazHBExgJASacsAeyc67UhMwARMwARMwgQlNIPdj 8ulR7tTC0oMWddb+2InXbhngTD9BPLdo8MwCPioKukkk6GoGLSt/L170xvMXWKbf6hYGnslAPdyi cbRVZ14Wt56xQNnVWmfN6v7e7rH1mdcSJNRX3u7C1rw8CZLX2eu0btu4VFy6bPvav937XnvOxHdf f6bE2lUrr9j0a7e+XLpi/qhnOAkx6hZw/SZgAiMnwMGad7rzLvVrrrmmaMrBe+RgLMAETMAETMAE TGBCEJAfg1heQzoRCg+VpHAlwm98ckulZCUSKhf2OFPb0DMXevxZ19WOHT+V1uG2kWEXNOszFhvV efHi82myKonCw0CbWCaGhTaRnDWZgAlMGgIcuJ977rni5ZdfLl566SUnISZNy3pHTMAETMAETODq IIAfw2ciFm6/4I0SVUVvd6ha1u+8TvXodop+t8n65VeYDrKNsX7D/vMq1KqyOHsDR9XyJs9zEqLJ rWNtJmACIQS4+mHGjBnpSohrr702XRXBPBcTMAETMAETMAETmAgE8GOef/61M+KcVJkIRUH0xdZt GDw0sa6iergdZJj1LHr9dpPFrVsvut0y0u++oZUPt6h007xm1YrW1STbCm5zkSbVp9eOjifJom0N 89te9jBpelsmYAImYAImYAImYAImYAImMEICly417xkAVTh4tgHPg3jtwZD7L1uF4JtXdw6jUA+3 KvDWDZ5DkZfx1MMzJihfq9DJwyDLdeX19jL9ntbbPXiWBa86LZf8YZN3v/W1q0geevifLluNfeNW kS+9/jrOfCFv9Bhl8ZUQo6Tvuk3ABBpBYBgHa7ZBJv/s2bNpn6ZOnVrMmjVrJPdlSsv58+fT1R3S cv3114+EN1zOnTuXrjC58cYb01UnfEcXuHCWCC2cJULDzJkz03e0Fup74YUXkr1wC1CTtNxwww3F 9OnTi2nTpo3k1iSezXLmzJnUVtgsbOhLo3hWC1qwF2yYe6xhwtnGUVwpxW1jjC9cbg0L2ojPKO79 hsuFCxcKxhiuHtMYMyou0kK/oo3o16PgQv3YC3oocBmVlvJxQLZL/x5FgQkfxt5Ra6EP5ccBxpdR cdExSccB2NCvBym0uUo+rXlN/f7IL/5ssaX1mss/aAXLvJ6SM/kE3iQg+L7rLXe2HzQ5nn34z616 eI0ln7we3kBBQE7ArwdncmUDiRFeZ8kzFbiNo6pw9cNdrQQAD5D83Bf+R9oG23ryme3FQw8/UvzH 1lssfj575WbVNjrN+8C99xQ8dPKzrdd08pDL/611ywrbf+gbj6YrH2a07AVt6CCZ87WHH03PkVDy Qgw/2npNqoqeEcEynjnB77XfWifi20mICMquwwRMoPEExnPA5rcHDhwo9u/fXyxatKggcDp69Gj6 3rBhQ3iQe+TIkaQF55fPvn37UtCElkGdm0Eb8OTJk8X27duLOXPmpLqPHTuWAqh169YlbYNud5Df oWXPnj3tgOnw4cPpGSAbN24M10KQvWPHjmQbsDl06FDSsmbNmmL+/Nee1D3IPg7yG5xxtBBMzps3 rzhx4kSxe/fuAnvh/8hCcEDdBCtLlixJQS6sVqxYUSxbtiw0EUGg/eyzz6ZkERwIFtC2ePHignaK TIq8+OKLxd69e4tTp06l+hlz6NcLFixIWmi7qEIyZNeuXQX9CRYUtM2dO7egL0VyoW76NDZLv2Hs xZYZ59ASmXilTejHtMvChQtTUMv/aGG8iw5yOSZR/+zZs1PijDZDA1xIjkQWxn3qZ6wjiYe9YLNr 164NH3uPHz9e7Ny5MyU2ST7AiGM34x3Hy8hCHxILxhi0kKTBXmB1tRSeefDApz7eCuL/qvhcK9hW 4faBX/vULw8lAcE2CdQfbNXz2S/85RX1kAjJA3Fegbl77+fbr/UcKwnBdj/58f9SfPGv/i5drcCD Lylsi22MJwHBdriCg9dzfq6l+WutpAYfzf9kK7GQ6/pES8eXWjp4Xad0iG1+Kwa/IRFBkoXPA+lq lNdeo5o2HvRnSmvQfCNtFlSpqzEBEzCBJhF47LHHivvvvz+dqb/vvvuKLVu29HWmk4QDDtaqVauK 5cuXp10jaHjiiSeSU7Np06a+tjceNgRsjz/+eHLu0EMhmGIewcGb3vSmsCABLVu3bk1MVq9enerl kPOjH/0oBXZvfvObw85WcsaWekkS4WxSCKaeeuqpFFxu3rw5nZ1LC2r+Q6BNe+B0ogVnHC4ka3DW 3/rWt4YFCVyN8eSTT6Z2uPPON870EGyTzLr11ltTEFMzkvbmf/jDHybbQIsCJRx17OiWW25JwV17 5RonsA0SEKdPny6wDQI5CkmAp59+usCe6etRATeBNQEKfUaBEv0L+0UHeqIKQdPBgwcLxjWSIBSS Rti0tERxgQn9Btugb0sLNk1S5KabbkrzIv7Qd2knAuulS5emKgkov//976eAEi1RV4oQaGO/2MXK lSuTFvV1EjP066grRehD9F+OR2jBNjgm0dfpZ295y1vSGBjRRlxJRJ/BVtavX9+ukmM1fG6//fb2 uNNeWNMExwF8DxJW9CW46DjAOHPHHXf0fUy69957U3IQyQ8//PBACe1LF08Ul155oaa97r5Zzsxz C8P0VhCv5zh0/1X/a6ieRQsXXJZ8yLfUyzr5+kzz/AW0z2glH/OkRnm9Qf6Xnm5sel1PWofNecp1 NxZTpl3+ytTy/rbsfYqfCVGm4v9NwARMoE8CnHEiYNJZQX7OGSccUQIoPhEFxw4tXL7OWWQVglzO JHPGm7P/UYW6cMLzM9g4WnDB4SPIjSrabwUH1EtAgGPMpfYEM1GFs7YkqeCiM9hwoc34JgEQdX6A QJYz/GjJC7aMBrREFeyTIAEOSkBQN8kazp5yxptLliMKTAgECFY4W6rC2UmCW86cYsMRhWBFWpSA oF4SI2jBdkmyRRSCRhIQ1J1fscPZfmwmUovGEHQoGQIDtGBDJIex74giLoz7+XGAQJ+xlz4feRxg bIWDEjMw4LhAkggm2FNEYQzBXhjX4MA3RcckbButUeMdWmir/DiAHrgwtkQeB9DCfmMv4sI3YzG2 zTErigsMmlII3Dl7P+zAuLx/qqdToqCXdcrblfZO2y3/ptf/pacbm17Xk9Ze6x/2ek5CDJuot2cC JnBVESCAJQDA2SyfWeLeTpyIKOcTLTiYOJsKbtUYBHEUgrwIxwYnijNg6Chfhow+WEU5wuw3bcAZ wPLl2eJCG+Kc1l1wdMfiAic+nFUmoRRRCABwfLHVvJAEIEmDvURoyfsJwVNe0IceEgNRgT9BIwk0 tChAQBNMsF+SSARQEQV7Yd+ruJCUQCcJnIhCP8KGsZeci2yIZVGBP/2EfottlLXACpuCXUShnzD+ Yhvlqx3Ut1gnomjsZSwpH5MY7+ASNfZit7RTFRfaCH1RxyTGMfZbfThvC7VRlO3SZ6lLY36uRccp +nTE2JvX7WkTiCTgJEQkbddlAiYw6QhwCS6l7OwxjwAc5xhnI8KZwCHHuVG9aFBBH1pYJyKQox4C EpysckEfjiABAnrrLjjB7DP7Xw4QxErOct1a0IFzqfbI65M+gluC3LoL9cCGeuGQFzgxn4AlInhS gICGqr7EPJJEEVryhMhYWtAZFbCQEEHTWFrgEqEFDWihlJN5zJM+rcO8Oou4qA/ndUkf9hIxxmgs E4NcC32JD7dIRBRsAZuAS7lfS1/UcYAxhrG1rAMO4oLeiDaSLWiczdsCfehBa0RykfGdsVcMci0a j1mOHhcTmKwEnISYrC3r/TIBEwghICcBx6FccDAoOIQRTpYCV7RU6WEeOiISIiQgCFrG0gEXlrNe 3UVOZZUW6qad4BLRRuKPlrIe/kcLXNSWdbIhIYJtqt5yXbLfCC3ooJ5uWtTfylqH+T82qT5SFTyp 3SJsl/2S/ao98n3VvAgtMFE9qjfXIi4R9kK9SqZWaVG7qb/lOuuY1j5XaYELH61TR/35NnvhQltG jHedtIgLNsWYV3fJtchWVSftxryosVfjmOqVDr7RwXzaSONQvtzTJjBZCDgJMVla0vthAiYQTgCH heCJUnZq8nmsF+HwqY4qLdIT5djIearSwjzNV1CDvrpKXofqzetiHm0kzfmyYU/TRtRVpSOvK9ec zx/mtOyFbVbp0bx8vWHWn2+r1z4SwYU+jR6KGORaNU9tmS+rY1r8VW9eh+ZFcVEfGSvYRltUX9I+ i0EVF9hpjM6XD3u6lzbq1cbHq63T/opV1HGgExf2Ez1wUVuOd987/V62W7WOuES1kbhUaWFeJJex NHi+CdRNwEmIugl7+yZgApOaAE4LRU5M1c6yTicHqOo3g8yTlk6/zQOsTuuNd1knRzjfdjdnLF93 0OleuUS0UdO4dGIjm47gQttKi+rN21vzIrT02kdYr9f2zPel32lxqfpdJJeq+qvmobeT5qrfDDKv F/ZNaKN83yLGO3GRbeT1a5r2iepLqrPTd1O0TCYundq/U1t4mQlEEbguqiLXYwImYAKTjQAHeZ0R 7OR05+vVyUCXIHfTEuGcSEu3/dU9yt3WG8/yXuqgHdWW46mr22975RKlpZMtyI4itKADNmMFadLS K79u7dBpOfvbiYt+yzq9rKf1B/1mn8cK0iK59Lq/va43KA/9rpMtiEuvbaltDvrdax/ppHnQusu/ 66UO2qhXzeXt9/N/r3VE9aNu2qO49NJGaO11Pe1XzjGf1vJRfF9oPbB169PbU9VrV68oeCVmubAO r5XkrQ682jKiHG29BnTQ+kahtxcmTdU1lnZfCTEWGc83ARO4agjoYI3j2q/TKidBTm8OLZ/XSyCc /3aQ6W51oAe9vTqGg2jQb6QlZ6Bl+Tytp2V1fHerAz0w6bbeMLTxwDzsLWeQb1fz9WC9fNmwp9nf 3PbH2n6EFuoWfzHI9Wie1smXDXs67yM6q5zXIS2sJ3758mFPi7/qzbeveVFcVE8nLjCJGGN64UIb 8am79KIFDeJXpx7VIdvI69K8yPGO+lVvriWfL37l5cP8v1Mdub5O6w1LT9UDm/NtS4/aMl/WbZrf 8omw+25adu/dX/zGJ7cUn/7M59PnydeTEeXfffffn0jr8R1V0PVfP/OFgaobhd5ehO7eeyBxfOjh R3tZfeTr+EqIkTeBBZiACYyaAA41T6LmLCz3pvYTWMiZqHLKNQ+HL8KxUR1yQvL9YB56ql6VVgd/ BWhikNchfcyT5nz5sKfVRnm9eR2cZWadCKcNW6AeuKAnL7k+ac6XD3sa9rKRqnbS2fdBHOF+tSoo gkGVFs2L4ML+ooeievP90bycX7582NPqI2V7oR5piWoj9RHVm++r5kVxkS2o3iotGofyZXVMi3+V FvXrKC6dtKhPw0VtWQcPbbNTG4lL3t/0uzq+cy7UrbGPumg35mkcqqP+fJvq06o3X4YO5tM+Gofy 5Z2meesJH0pE+3bSwjKSD1zh8NEP/Wyx+eaNlVdBdNvGeJeTCEHDbbdsGu+m/PshE/CVEEMG6s2Z gAlMPAIc6GfOnFnMmjWrHSD2uhdz585Nq8q5y3/HPBwK3s0e4RCoHtWba9Fl7rwPXY5hvnzY07wD nn2ueuAY+nCy0CtnbNj159ujHhxQ6uSTF3GBid4Vny8f9jRJINhUtRHamI8W1qu7TJ06NdWjevP6 mIftUugbdRdshf5HGctm6Kdap049BCdz5szpqIWFEVqoR2NMFRclTSPaCC29cJFe1q+zzJ49O22+ qi+JFVwixhjq6TTeoTGKi+ySOsca7xhfIsYYxlTGXo2zuT1o3IGdEgT58mFPUw99W0F+vn2xwlYi jgPUA38YlNsIffBCR79txD7S/nyi3saSc8ynjx0/WfC56y13Fu9/3z3F2tUr0+0P+ToR01/66/9V bPn9z0RU5Tr6JOAkRJ/AvLoJmIAJ5AQ42OMocMCX46vlvFoPB2PRokWaVes3OtCDDpyqvHClB2X+ /PkhDh8BLk4UOsrOEP8zf8GCK+8NzTUPa1pBJY5d2RnW6w/hFhGs4GxTVxUX2g02LI9wyuGCPeD0 ioOYy4Zwavt1hLWNfr7RoiDtYuv+4HJBX1SiiLrhgibqVTKG+fRnXvWndmRe3UW2WW4j6mUeyZl5 8+bVLSNtn3qoj1cM5lxYyLxILXAhoQeDsca7hQsXXnbGuy5IJGcYP7CNMhf0qa/VVX++XfoJideq Y5L6luw7/10d09ICl3KwzTzajX4fkaSnfUhc0T7UnRfaCH20IzZcd2F/0cI4yycv/M9xCvuG36Cl bIeDbme8v5sxY9p4N+HfT1ICvh1jkjasd8sETKB3AuM5WOOwLFmypDhw4EBx5syZAqeXgnN1/Pjx 5AzqjF3vigZbE4eFwH7nzp1Ji5IfOFfHjh1LTrL0DVZD77/C6YYLl4aeOHGiWLZsWfvH/E+RvvaC Gieo69SpU6lNVq9enWqi3Q8fPpwc4MiECHUdOnQoadHZOQShD7uBG/wiClzQQpvk7aE2ytutbj0E I9gwda9cubKdiCGBRvCElvE45f3oJ7glCKBNCOaUiMGe1c+jtDB+oAUuBE/SQuB0+vTplIDAjiIK dWO/1Hvu3LkUSFEvutBHkgJ2EYUkJ/XRh2EBIwpaaDcCcc2rWw8BLmMrWqibIJ/CGMPYS/tEJYrQ Ql/et29fcfbs2Xa/xo7RojasmwnbJ9hmPKPPcDxcvHhxqlZjL8fPfNypWxNaduzYUZw8efIyO8V2 SSxGHR/lN9AeaNEVRnCBk7j0exzg9yr5tOYN4/vJp7cVf/v3/1jktzm85913F/e86+3tzX/pr/+u 0PMfWJ8rEXgo5Ud+8Wfb61RNXHzuYvGlv/5/i0f+9TvpFgqunvjgT/2H4q633nnF6g9945GCZx/o AZN3vfWO1rr3tW/5YDnPbkAnRVdDPPCpj1+2Le0P3zyoku185Bd/ruerNr7y918vHm3pRQd633/v ey5jQWU8MPKhbzx62XrSm4spryc9+X5p/WOt+j73hb9q7x/bu+3m6ltOtI9qs7F0attR39c+2CpR lbkeEzABE2giARzHf/iHf0gO46ZNm4of//Ef7ysIxNHlzMX+/a8d7AiYmMbpY3s4w1EFLQSy1I8m HHISJDjp0VrYbxzivXv3prM98Dh48GAKXjZs2NB2vCLYoAWHGBYw4WyTAoaNGze2z3xHaCEI4AML 2oV2QgvO6Jo1a5LTHqGDOmgfrlohESEtOMFwIllD4N+vIzyodhxvnHGCOIICnGimd+3alYKXtWvX hpylRD9aCP7RQbughYAOW2Y+9htx9hYt8KdfE/QfOXIkaWEaLgT82G/ElTNoobD/1H/06NH0P4mZ 3bt3J1u66aabwhJFcMFe0IL90kaMvbQRy2655ZYwLYDIjwNooT+RCCAJjL1EXOaPDtmLxtvycYA2 ijwmYaP0p/w4oGMSXLCnqKLjAO0CFxjpOLB+/fqwq/PYX+yB8Zc+ja3omMSYs27duoGOSV/+8pfT ttj+hz/84cFs7qXnWtmzy6+kZHsUAvv/579/KU2/6+63FW++49biyWe2peC6aOXN9dyF02fOpn3b tnN3sXjh/OLNt99arFi+JAXp6celPzxQ8buPPVFs27GntWTKa+svW1Lw+0e++Z3Wdi9/nsQXW0mO r3z162l7JD/mzpmdfv+NR75ZvKWlif+l4cDhI61x4bmUGFjU0iKNJA+Yz2+Yx20jp1pj/A+feLrY 3qr3nne/kVQpyW0F/q/p3b3vQJq+u/XbFS291EVCoqz3//q//6j4/7797ymZQj0vtWzvG//0zeJY q63zBEu+Hrew3NCyD/YfNujhfwoJiAdbiR10vPmOzUn7nn0HW+t+u2VHL6f90X5+998fL37vj/57 +t377nl3WgZXdJLk2LRhbVo2zD9Trmk99Pr6zsnoLa3iKyGGSd3bMgETuCoJEIzgwHCmi6CJy5I5 E8b/BJvRhUAW51xnKzkLt7YVwEVrwfEkiMXxw7EiWMDxIriN1kIbrFixIgUKBJVoQcPtt9+eNOG4 RxXq4owgLKSFs+oETZHBgfaXs3/US1BJoI1jDBfmRXJBD7Z62223JS0Et/QtgmzsGXuKLARPaKGN uBqD+gkOCJoig372GVvZvHlzux+hhX7O1SPRWmSr6tPYCFeujEILHG6++eY07mK7jL1Lly5N9hI9 xtAO2AfjPmMvAS5MONMf3UZwWdsa8zkOcRUCbNBF0B/NBS2rVq1K/YbjI2MvYwv2G60FW12+fHk6 DmC/jDFouOOOO1IfixxfqIvjI+MJiV8dk+jnjD3RY2+3fSfw/drDj7SuNJhfPPip30zf/Ob9997T usrgj1NS4J533Z2uRCBg5nkQJC3S1Qw/fV+3zaflm1vJgE9+/GPtdXfvfU/xOw/8QUHC4IHXHyz5 yL9+u3UFxCMpKP+1X/mly9blagcSFA+09BHc89n6zPak5YMVGnhg5R9s+VRL44q0Hdb5z7/+qRTc 65Wh7QoqJgjiqYtvytbW1RQkB7gCg32hwIxkwUdbV4HASuVzn//L1hUf307JEdb9TitZwHofaK2j K0bgCO/Xrrb4dvv3/A/f8ja/+Fd/l5irDr65AoOStxnb/Z0Hfr/4Wqt9ck1pxcA/1wXW5apMwARM YNISwGHA4dNluKPcUZy+pmiBCwEknyYUHL7IM29j7TNcmqIFjQS5BAWjLnAhyCWobUIhQCFQaEIh kOVycj6jLk3TQqAfeVn/WPybNPbSRk06DpCQ4TPq0qSxFy0kXvk0vXB7BYHvr3/sl9oJCDQTgBMM E3xzG0VVsN/rvt31ljsuW5UEBlcVpLpPnCwWLZhfPNq6MoDCLQp5Yd33tK6KIPGxpxXMr3k9sZCv U57magElILSMKxW4ouBi6+oUJRe0rPzN1Qr5OmtaGihHW5xU0EMioRzs39O6hYV6SDyQhLi7lTD5 n3/+x/pZ+3tza/+Lv+eWjtYVKq8XkhxV2+SWDOrLy/TpU1//941bdfjtJ7JkT75+5LSTEJG0XZcJ mIAJmIAJmIAJmIAJmIAJTCACBMuUtauuTBAr+NY6w9ytza3nHJCEOHqs9dyiVhKCZy8QRPMpFyUU drWeA9FLEqL8+/H+r4QECQwKiQMSN8z/3Bf+8rLNK6nALR154fkNJFr4HUXraR3mMw8uvZQPtBIl JC1+/ZNbXr86hGdHbEy3ZfTy+zrXcRKiTrretgmYgAmYgAmYgAmYgAmYgAlMYAI8P4EyveJtF+Xg u87dRMf0129/qLOeYWxbzNjW1lYipVwWL1xQzGhdhajC7SQkIdIVGqtWpP3kuRF6uKbW47vXt45w lcVnP/1gul2G53fwjAgKV2JUPfAyLQz64yREEGhXYwImYAImYAImYAImYAImYAITjYAC/2OvX5GQ 61eQPD0LqPPl45kuXwmAjjy4z7et+UqK5MtGMS1mXLXwyd/8WEcJPB+CBET5OQ/M4zYXFW3zwoU3 bs/QsrG+uWrk11q30VC4koK69FwJnmkxqhL7lKdR7aXrNQETMAETMAETMAETMAETMAET6JvA3a3n DVD06s18A7oN4+63Xf5Mh3ydQac5e0/hmQ8UXkNJYoKHQJYLb5GgaN3y8uj/SYaQANja2odyMqWs RVdK3PW2O8uLLvu/n21SJ7eB8OwJFfTw3A5uV6lqS60X8e0kRARl12ECJmACJmACJmACJmACJmAC AQR46OUwC5f1E8C+9uDH/e1N89YMzqqzrNfnFLR/XJp47ez8ifZc/uchk7zNQVc3cBsBhbdgULcK 63LVgN4oofl6MGO+rpZFfPNMhteSAf/jMr1o5c0fSqbAj/LQw//UlsU65WdJsJDXkrLNv/37h9rr Xmi9nvhLLSZ5gRnb+FLrrRn5/jPNVSOqM/9N5LRvx4ik7bpMwARMwARMwARMwARMwARMYMgE8sRD Pj2sanjN44Ot13F+shU889wCAmFuxSDY5bL+8Qa1PFiSByhyJQNBsh5Cmb8Jg2QItyyQhGBddHCL Aevy+3xd9vvu1tsueDAjAf+M1qtP/9unHxgWjp62w1sx4ESShKsd2DdxQ7se6sl6XLHwtYcfTbdf oJV90v4da70dRIUrGb772OPtdRe1ni1BYkEP5tR6fPPK009/5vNtVtTNunznr0PNfxM1fe2DrRJV mesxARMwgSYSOHz4cPHVr361ePnll4ubbrqpeO9739u4d3Q3kZs1mYAJmIAJmIAJNIPAn//5nxfn z58vXn311eJXfuVXCl4x3Hd5qfWsgUuvVP6MZAOvklzXCvaPtoLixa2z97zS8rf+z4+m6fKPbrj+ +lYQvbFYsWxJedEV/8+dM7v46Q+8L237pZYvxrMP3nX3W1vPMvjlgmV52bRhbbrigWs9Xmyty/Y/ 0Ari/8v/8Z/aV0xofYJ+dF5/w/UpSXJXS7/KbbduTEkB/a9vtkeyA/1jFTTd3tq3sjbWL2+XRAKv 2mR76EUPyZIPffB/b9fBMrHVPrHOT7WYsIzfsN8q7/vxd6d52t69/z97ZwJtV1Xf/x3I+DIPL/Pw QkYSCIgkqKgJDtgEK4pCBbRIobV/EFtbgbratUjW0tUCLm2hYFcLAm2BFhRBBRRaSTSoAUswkJB5 DiF5meeR/Pdnv/e72ffk3HPPHc6+9738flkv995z9tn7u797+v1+eziXXGw+d/ksFxbOcUwg4CPe gY39zK7de90z489qMn9py8yPT+KtxmeHMzqaDp0aEqOaY6XDCSuJofSmMqAMKAPtnIEFCxaYT33q U+b48ePm85//vLn33nvVCdHOy1yzpwwoA8qAMqAMtCcGxowZY3bv3u2ytGzZMtO/f4shWkoeTxzY bn0Qh0t5RMMqA3kMdOjYxXTollz37EqdDh3zntIfyoAyoAychgwwWzBy5EjnhOjTp4/BN5vFUsbT kFrNsjKgDCgDyoAyoAwEYGD06NFmx46WZftnnKHH/gWgXJOogAF1QlRAnj6qDCgD7YMBnA4M2NVw PrAMEsGJUUtHhixyk7zVuqTqhRd4gBMp61qWUT1ioZxoC8rLyRZTj/UFdPVgZFBfat3X+e2onngB Sz1wI31vvdSXeuOlHsqoWn0MZVwP5UwZqygDxRhQJ0QxhvS+MqAMnBYMoARUIpwnsW3bNrNr1y4X TTf7vuxBgwaZrl27VhJtWc+Sl61bt7plmWwxAcvAgQPtHsvkPXplJZbiIbDAC8owq04aGxtNjx49 UjxZ3SDwsn37drNz5053/gdlw3LVXr3y95tWN9XCsTFjxd/Ro0cdL/369TOsxAkt8LJ3715XZ44c OeKw9O3b1/BXC2fEwYMHzZYtWwyfnexeWOrt4MGDa6JcHzp0yPGyf/9+07FjR9OzZ08zYMAA9z10 OVE28AIWyoU2RLuGo9ACFto19Qajp5ZYaD/Nzc1mz549jobu3bs7XsraD18hkfS38AIW2hV9L3W3 VlgYk1ieDy7aEX1vLcYBuKDuUl8YK2s5JkXHAcqGdlSrMYnxSMYBxiT63WqMA+RTRRmoZwZ0rU49 l45iUwaUgTbBAAreypUrzYYNG5yBMmTIEGdYvvnmm87gDZkJFI8VK1aYTZs2md69e5thw4Y5BXTJ kiVOUQ+JBafDmjVrzLp165zxNnToUGewgG/z5s0hoTiDYP369Wbt2rVOAYaXM8880yxdutRhCa2w bdy40dUZlM7hw4c7Q1LqUEhiyDdlAQ+dO3d2WDBSKDcwyixqKEwo5NTVw4cPu7qL0Y+Byf5mDN+Q grG/aNEi5wyh7uIk4hDbWmDBeFu8eLFz5mHU4gjBcAELjpKQQjm89dZbrm+jr8OAw8n4xhtvBMdC 30vdxcAFB+VEuVGH4CykwMvy5ctde8K5CTcY3IwDsk8/FB4Zk95++23nZKWPAR91CMdEyP5OxgH6 X9ozWEifcsNhE1JIlzFg9erVzukAFsYB6jN1KCQv5FvGJMaBESNGOCxgY/wO3feGLAdNSxmAAV0J ofVAGVAGlIEKGUDRQ8kcN26cM1SIDkMO4wAlg+8hZsJQoFCkMNomTJjgDBWwMMODMoxywwxLqJlT lF2UKXhhVQjCqgNmLnHYgAulNISwAoJyampqcsYBaQovlBEzc9WYfUqTF4wjHDMowDhDmNUWHqS+ lHOgWJq0o2FYbUBZkHfORUEoIxRznCLUFYzeEILSjVFNW6HOMMMOFsrp9ddfd2VE+YUQsJB/Vj/A C/UDwVjAeKKczjrrrCCrM8CCQ4j+ZOrUqblVGMz4Y/jTrs8+++xgq1Yw4jBoJ02aZMCAwA8GLljo e6g/WQu8UHd5G8A555yTa0PUH8oIR9q5554bjBf6Xpxo8CJ9CfX397//vcNCGYUYB+AdZxl9HmUh fQllhAOY8qO/CYUFXuh7yT+OPFnFQ9+zatUq18ZpVyEETnC6+uMAXFCfqS84YlmJEELQGai/OB9k HKA90e8xPvBd6lEIPJqGMhCaAV0JEZpxTU8ZUAbaFQMY/ig1KFG+woDxxmwlBiczliEERwMOiCgW lBqW4aJooZyGEHghLT7hwRcUUbCiEIYSHCLw4JcRvzGw4SXkjBz5xoCiTPztDii/GG/UJ3gLIRhN zJqCxRd4gh+M31DCbDoOKma0SVsEZRxDgbrNCokQcuDAAdd2qavigCBdHIo4RTCsQmFhZp/l/fCC U0SEdg4+jBnZiiD3svqkfGhLGNfigCAtsGDs0teFWoEA/7Ql2o2PBeOaPgdeQva91Anqh9/H0L5x wOIooa2FEBkHaDOUkwh1h3aO8R+q76Wfo76QNvVD+js+6Xvpe+h7Q/V39K30LeKYEW7gBQxgDSGk xfgIFnHMkC68gA3eQvISIs+ahjIQZeDkKB+9o7+VAWVAGVAGijKA0oIyjNPBN5x4EGUYozKUYoPy j+Idh0UMKe6j+GUtGE4o3mCJzooy24RSilEZQsThQZr8+SK8wB3OiKwFAwDDKA4L1+AG3jD2shbq ATOU1FuMSF+kDoE3xHJ/lG6pD1Es4KItgQPnQAiBFwwFjEpfMBLAAl7aUgiBF+qm1FU/Ta5Rv+Us Gv9eFt/hhXoTV0ZcC2nIkWfqA+Uhxq3kWbiirYEpa6GPp36CJSpco42FMvzpP+jPaMOF+juwhOBF xoE4XujrGBsoxxBjEuVDWpQF3PgidYg2TdvOWujfqZsyFvrpwQn4KMcQY5Kftn4vjYHFS1eY+x98 1PCpUjoD6oQonTN9QhlQBpSBHAMoCkjU0PavYchhKGQtpINiGXWGkC7XUNRRxEIoNqSDMhdVgqNY QhjbGCpgIf9RY4XfcEP5gDlrkXTi6ovgg5MQs+zkV+pltM4IL/CBIZG1UD7iYIjjRvCFmGXHIKIt IZKun3+5Jnj9e1l8lz5G0vXTkGshsPhlVKhdgy1EGfnpwEG0XUsdCtX3wn+hvlfadShepL1Kun59 EV7oX0L0d4w19DGSro8FfFwHbwgnhPACBmk3ggcc4AGrtH25l8Un/JMWOKJYSA88lE+IsTqL/GUV 55w77zXX33y72X+gpX/OKp208TZv22Hmzl9g+BS59Y47zc23zq4JxuZt281Xbp3j/gRPPX+qE6Ke S0exKQPKQN0zIMZinCIh11BOQxjborDEKZ8QKUpWCIWP/BZTysEUghdRtpN4gRMxyMGVlZBf0qJu gMcXfkudCcELdZcy8tON4uF3CCwYuOApVEYheaEeSBuRdH1e5FqI+kK6fv31cfBdjDtp+9H71fxN GUmehQM/frlGfaFeZS1iLEbbEen6WKQss8Qj/Et5+GmBBYyheBEswoGPRfD5Zenfr/Z3qbtxWKTc qFMh6ksSL1JG4JBw1ebCj0/0BuHAv8c1/kKNSX7a9f69wa78amjoZh3W9eGEiONroN1OUyt5+PEf ma3WEcFfW5D8daltAbFiVAaUAWWgThhAYUmjPKHwhVCESQeJU2zkOjgknAuc0X9p8xvCwC2GRRS+ YuGqQVVa7kPwApak+iv1SAzPauS/UBzgEG4k3biwoXhJgwVeCBdnYMVhL/ea1MskXiRMuWmkeY68 puFFylIM3jRxlxNG8hzHi1zzMZeTRtpnhJdi4am/LMHPUpJ4kXRD8SJYJN3op5QTvMRt8YmGr+R3 MSzEDS9pwlWCg2eLpeHzUmla7en5W796Y91np1YYWZHx6muL6p4fH6A6IXw29LsyoAwoAxkxkGTs ZZRkbLTgCIElbRpplffYzKS8mAZLKF7S5jdtuJQUFAwmym5cALmXhr+450u9liadULykxZ4Gc9q4 CoVLSkPKSHmJZy8UL0ll5CMLhcdPM+47eENgaWu8wFUIXuLKJO5aWv7ins3SOcoZCEuWrnTJTpo4 1kyeOC4OgjsrYd2GTXbLzUHTNHKYmXrBlFPCrV2/0bAmcNTI4S488Xbv3s1Mfc+5pnFA/qoCwrIV Iy49MElaxTClwX4KUHtB8i344sLEYeS5gQP6uVUcr772htu+0djYz0yeMDaXx2jc0bzHpSXX2Ibx 5DPPmxkfvMjFTVxtQdQJ0RZKSTEqA8pApgygyIsyL59pEiRsmoE+bbg0aSaFkZnHJMUFvGkwJ6WT 5l7cnvG45wRz3L1qXYseQhYX7+nISzHupR4VCxfHZ6nXaCPUmWIrHdLWq1LT98OnrQuh2jV5LjRz GsPPnEgAAEAASURBVLqMpO+QdH3e5HtIXiTN6Kfgo+6CJ2tJ20ZC1N80WCjHNOEq5a1Yfv1yqjSt Ys+nyS91JU24YmkVu58VL35d978Xw5P2Psbu3fc+YNau33TykWeMNX6nmeuu/qzpbrdKiDz8+FPm uRfmyk/32WgN8dtuudE5HOQG8SEDrcPBN54ffuwp86WrrzCzLp3h7vPfI3a7AWGeeOie3LX99jyW b9/7YN6zxmK67NLpDpMELIQd58gdt381D7s8I59xaYBvsnXARCWKkfMiOMti6gXnOt788yPg46Yb rjU/eOZnefjj8h5Nx/+NAwK58vKZ7qBM/149f1cnRD2XjmJTBpSBYAyg5LO8moFbFO00iYsyIcqU /4xcI04J59+v9vdiaYAHg7yU/JWLESzkWzjw4/GvZb00mXSLOSHAE1IpJy2fgzhuQvAiZUT6cXjk WrF65eMv97u0EfZjS7p+XHKtWFn6z5T7nfxKG4mbFRUsPn/lppXmOeqCnN8RDe9jid6r9m+MM6kL kq6fhlyDO+HPv1/t7/LGBUnXj1+ugTkEFqmXkm4UC9eljvv3svheDAtpwomUZRYYJE7px+J4kTDw IpjlWhafkkYcFq5JGUm4LDBInMKL/I5+louF/kq2z2VRvrOtMY0RjeHMrDvy5NPPu1n4RnsWwpWf ntlyzRrFOCAu+/gM8zl7DecEWwVwTBDHXXNuy60A4AHinDxhnPnnu+9wzgjC8tYJwmPos0qikDxi w+CYEIcFKyVIG8O8oVtDHiacJ3HYf2DzcN01VxRKwsbV4iTw8zPPbn8AX1phBQQYp7fyJhhxUOAI kbyzkoJr4Ces79iJS4ttGHPnv2JuvvFay2m/uCB1e00PpqzbolFgyoAyEIoBBm6UfP4YwOOUlEJY RJmIM1bkGgpfSMWGdKN54Bp/KOWhZnpQLuNmcMEGFu4Lf4X4rcZ131iJ8sJvMMJJiDKSdOLKKDQv cA8eSdfnWnjhWghe/LoQV2fgC5GydD8y+s830JKwhODFz7Nw4Gdb8IVoR2l5CVFGcCB5jmtLviEW or+jLiT1d345+uWXxfc0vMBJFkZqND9JvFBu/BGGupW1SHulb+PPF9oR1yhDCeffr/Z3uCetuLoL Dq5TRqXWXQ4CFV2GOKopGPo4C3A+iAOC+HE84Ch47sW5LjmcADgmMIgx7MWIZivGzdZ50eIkmOfC yn+Eucka0ayGQAiLwY5gYBcSDHbug0dWTBAXmPic+/KC3KOL7TYPMJ2KfZzZ3/o2pFxg74s4NaL5 wUEwyzpZ0srU90xxGMHlY+T5W2/501zem6zDZfrFFzmetlm+k0S2YbA9hWfammTf6tsaI4pXGVAG TjsGUIAaGhrsPsTuThmKKihJhPTr188pEywjjz7HrC6KQH87QxBCyQI/BgAKVVQBkRO/e/ToEcSQ AwvKXBwvGAjg6927dxBeutkTtTn0DF7EaJMyleX/3Adz1iLpkG60jMAGN+AVQyJLPNQV6j31VniQ 9IQn6m2vXr3kcmafKNvUBwQlOirgQ3Hv2bNn9FbVf2Mc0GYRaTd+IsKV4PXvZfG9b9++ro+Jw8I1 8IbEQh4LYeGecMf3LAVeqJ9SHn5agg9eSjXk/HjSfpd0JF3/Odo07Qm8IYQ2Ai/Sz/ppCleMAyGM benLJF0fC/0fGIU7/14W34UX+jvS9QV8XJexwr+XxXf6MsogroykPy5nrBY9hrEsmsdK8yHnKEy/ eNopUbEV4+t2mwUiWypmxBjFk6yxjAH+ysJFeXHwxouoXNh6fsTiZYXPN2B1ATLJnq0gb4WQT86F wGmCEwHhTAZ+s8IA413kjttvcasj5Hf0E0cHMs06EaJSysoDzpGICvkmjmg8cITss1tNkqQlLy0r U5LC1es93Y5RryWjuJQBZaBNMMBgj+KCEoMCKrOAKBK8Ox4JpXyKgcv76FF6/Zmuffv2OSwYCBgt WQtGNNzs2LHDvWIQjkR4tR74BgwYIJcy/+zTp49pbm7OKyMShSsERTgELxhEKMPwAg+UmQjGN3Vo 0KBBQQwn0m1sbDS7d++2B4ftz3M2gEUMpxBGHNxTRuvWrTNSV4UXPmlLtC2/Hvn3q/2d+oAhBy+0 ZakbKPbMNoIlhNOKfOEEIr09e/ackk24wpiEuxCC05V+BV4wIuEI4Tv1GSwhnFakicHFH1gK9Xf0 MVJ2PJOV0Kbp86Td+G0GfPAEdyEEHJQB9ZT+xG8zUodC8SLjwM6dO51R7I9J0raou1KPsuSHNBj/ ZEzynTD0L9RhsISoL6SNk4H2SzlRj0VoR2ChDH2Mcr9WnxjwyMDGFgetj4PtBCLy+syoYS33B1ij u9gMP2ExxIlDnAjyvP8pmNi6UUgOWD6J66YbvtCydYTtI60rNVjBwTkKSYdAShqF8lMo3ayv4+xh FUgL/jB9S7XzpE6IajOq8SkDykCbYwAjoxIZNWqUWbFihdm0aZMZPXq0U2JQLrZs2eIMbZSNEILi O3LkSLN48WKHBVxcQ9nbunWrM7RDzVKS3xEjRri0N2zY4HhBoUIp3rx5s1PAQmIZOnSo2bVrl1m/ fr2ZMGGCM6QwGDZu3OiMyZAOkYEDBzqHiPAALxi377zzjjNk4C2EIkwZUQZvv/22KxO+YzRg0FFm 4Bo+vPBeXJ6vpmDIUQ7btm1zjhiUcJRxeMLJBxbfuKtm2tG4MNxwBoEFA0oMSBxZtCfaGVyFELBQ ZygT0pe6SpvGiQYv4vzMGg+GHHnHWUT/NmTIEJckhh1cUXd9ozdLPOSZvC9fvtzxMnjwYNduKC/a Om0+FBZ4Ie+rVq1y5QRHXMO4pX3h7MOxFUJoI8ILbaepqclhwdlIncEpHgoL/RhcUE/Xrl1rzjrr LIeFvhdeaPNwE0qoEzhi6PfHjRvnxgH6FrAwTofEQn2h3QgP/viIU4L+p1TxdRn/e6nxZBlenBRp 0iBs3CqJ6LNfsts+GryJDv++XMeJwJYPtmossQb8KwvfcEa8nKfQ1rYziOOlefuOvMMoWQmCyH3O wKhXUSdEvZaM4lIGlIGgDFQyYGOg4HzACYGRjTKBooMRxXV+hxKUujFjxjhjGywYShhNfKIAhjJu yS9YUPRQzHHSYDSglKMkgyWUEQcWZq3Hjh3reFm2bJlbgQAWZOLEicGMFdKDB3hBKV+6dKnDJrOC OEj81RGEz1KYmZQygheUXwwEZueou5RhKKFuNlmDCWN79erVzijACYHxRH3ByAwlYjyRPuWEsYBz Bucihnc5BkK52MGC8YSjCm4wsOmvMOxwToAnZLsm7xhuOCHgg7Tp78AybNiwILPawiWOM9LEkAMD fYv0vdSlkLzgHKKOMA7Qhmhb8IPRj8EZyoEGN8zokyaOTfoY+hT6GD5p7yGx0PdKf/fWW2+5vpa6 Sz9Iuw6JRcYk+hcZB1h5QFlRX/wVCVLHsvok/+PHj3cOPRxpOMzAQtuCr5DjQJo8utUOLxuzdt1G 09g/f+ad7Q1b7UoJziZg2wOybr33Bo3WBFjVwMqCuFdstgbJfRAn4TkjoZAIJhwN/lkP0fC83YK4 ZHUF5znwB5bb7rjTPGFXRhRyQsgqD1kREY27Vr85P4O/KC5x8kSv1wpnUrrqhEhiR+8pA8qAMpCC AWa8MJBwRqDoYbjwG4UmpBIsUJnNQfFFAQYLv5nlCY2F9MBx/vnnO15Q0DFUUErhLLSABccQvIAF XlBKa8ELOCZPnux4wbjEgKG++MuVQ/FD3ZgyZYozasGCUUAZ1QILijgKOA4iDDnqCcZBSIeV8E6a YAEHeKgnYMF4CC04MnEuYqTgIEKYYQZL6PorBhs4xJGHIyDUqgOfe6kfOGnEkScrIELzAhYcQjhG 4EXGAdpSaCHv8EC/Aha/760FL3DC6gvKSPreWowD5J1x4Lzzzsv1vbUaB6gTOIsYg+BFxgF4CemY SVs3xRjnAEoOjvSFN14g990928i5DxwKOcu+JtPf6iCHV/JKT18wnHE6+GHnvtxyIGU0rP8caSHE G3VCcJZDB3uPN2scOHDIfOXWOe4ATV7HKcLqCFZaiOEu1/1PnCA4L+LysyThvAo/jiy+c5ZFnPBm DbZqFLof90ytrqkTolbMa7rKgDLQ7hjAaKmFsRRHJIYCCk49CFhCLf8tll+Uu3rCgjOiHgTlvJ6w oIjXwniLlgW8MCNZL7OSGPq1MPajvPC7nnipp75XscTVFuOcmvXS99bbmFQvfa85Yd8ScvzUg4Ep UQx+3gaBwX//g//pVg5gvD/34jw3G88rIkXYAnD3vQ+413G6N1XYfmvthk25sxiiqw5YpYAjg7Cs sliybGUu7CT76s5CgmOEMxE4oPHmW2e7N2rgVOB5XoGJk+GuObe7T85/4A0ZPvZX7QGZrBjg1ZtJ QhryelEwGruDFwdE0ps7kuLTey0MqBNCa4IyoAwoA8qAMqAMKAPKgDKgDCgDpysDx+3bvA7tTMw9 Zy+wKgBHhBjg/OZ1mr5jgZUSt9q3ZWC43//Ao7k4uR53RgHOAt6m4YdlywZhuZckzilgA7BSAccH AibSuq71NZ9c+7p9DeYjjz1lcS/Iw46DQV7vSbg44T6OEueAac0P+Mg3eVQpj4EOdl9hZSeylZeu PqUMKAPKQN0wsHDhQvPlL3/ZLZ+dOXOmmTNnTk22C9QNIW0IyInDe82Jo/bNHzqUtaFSU6jKgDKg DCgD1Wbg+uuvt2fo7HHRPvTQ9zNd9cd2B1YdcC5BknBQIism2GqBcyAqrGBA2MqBpI3XBY78J892 t1sb49KS4GnDSXj/Uw5+LJZv/5nT7XuHjnabYLfkemFXGXbQlRCnW83Q/CoDyoAy0E4YOHFot3VA 7G8nudFsKAPKgDKgDCgDbYOBpAMj/RyUaqynjddPQ76nfTZtOInX/yw1P/6z+j2fgfAng+Wnr7+U AWVAGVAGlIGyGDhxrOXtFmU9rA8pA8qAMqAMKAPKgDKgDNSEAXVC1IR2TVQZUAaUAWWgYgZ0C0bF FGoEyoAyoAwoA+2FAd4H0SIcqquiDNQzA7odo55LR7EpA8pAMAZ0wA5GtSakDCgDyoAyoAwoA1Vm IN/v0DacEHIWRJWp0OjaAAPqhGgDhaQQlQFlIFsGcEDwTnfeX8673tUhkS3fGrsyoAwoA8qAMqAM VJ8BdBmkY8czqx+5xqgMVJEBdUJUkUyNShlQBtomAwzaBw8eNMeOHTNHjx5VJ0TbLEZFrQwoA8qA MqAMnLYMoMccOnTQWCXmtOVAM952GFAnRNspK0WqDCgDGTHA6ofu3bu7lRBnnnmmWxXBNRVlQBlQ BpQBZUAZUAbaAgMN9tWUhw8fcVCZVMlK9h84YJYsXele0cmbJpJeh5kVhmrESz7229eH6hsvqsFm 6XGoE6J0zvQJZUAZUAaUAWVAGVAGlAFlQBlQBuqGgRPeYc3+92oCnPvyAvPIY085451477j9FjN5 4rhqJhEsrudemGeefOb5Np2HYGRlkJA6ITIgVaNUBpSBtsVANQZr4jhgvep79uxxme/atavp2bOn 3ZcZvpsVLPv27XOrOwRLp06dalIw8LJ37163wqRLly5u1QmftZITJ941GzZsNFu3bjUXXnhhrWC4 Gav9+/e5LUDw0a1bN9OlS2eLJ/xS2iNHjhjqy1E7e9bZ1pNuXS2Wrl1qsjWJs1n2WixHDh02HTt1 NJ07dzE9ejTUhJd33z3ueDlksbBKqqvlpVu3LvbsmPD7rdk2tn//fnPw0CFzhl1uDZaGhq41wnLc 1ttjFs8Bm34HV186d+lk60v4FWT0d9RflqIj9He0p9qc7QOWo24sAFdLO+pcM17YXggvtClmyTt3 7uzOPXJEBf3vhDlqeaHuMkNPX9fV9S/h6wvZBgPjEp/S99Zm9WMLL6xeOHL0iMMCnrL1hhPZFiqr Bp58+nm3AuL/3XCtXUHQzzQO6J9tolWInRUPa9dvajN4q5DlNhFFeO24TdCiIJUBZeB0YwCFsVzh 2U2bNpmNGzeaxsZGg7GPgcvnmDFjnGJRbtzlPLdlyxaHpUePHtZ462EN7g1O8QQLimhI2bFjh1m5 cqXp3bu3S7u5udkZB6NHj3bYQmI5ZhXyTW+/bX7zm9+Y+fPnmyFDhtTMCbF37z6zYf16axR0Mj2s swpeMBiGDh1m+vTpHZIWs3/ffrN+w3pzpjWse/XuZXbt2mk27dtkRowcacutV1Ashw8fdnUXw6l/ //7mwJ4WB9bAQYPMQNu2QhqWGG5r165xBm7vXn2s0whsm8wAq3QPHTo0KJaj1sje/M47ZvfuPaZf v76G3mrz5s22rvQxw4YNC2pY4gzBibd7927Tv581QKzPjD4Gp+uoUaOC8kLlpO/dtWu3q6tnWqfv JltGXbt1NaNs/e1o++BQwjhAv/+OLae+ffu6tDdvfscZ2yNGjHDjQSgspMM40Ny8zfaz3Z1j5h2L BaceZYShG05OmB07dubGJMYgeMLoHzacpfxhx6SdO3e5+tq9e4NLe9u2bc5BPtLWl9DjI+35bTsm 4eDs1auXAcvx4++a4SOGm5527C5VTrieodSn0odfu36jad62w1z28Rlm2gVT0j9Y45A4IObcea+5 8vKZ5spPz6wxGk1eGFAnhDChn8qAMqAMlMkABiSKMEoMxgkyePBg88Ybb1gjZq0ZN25cMCMBw2DF ihWmqanJoPgiGNuLFi0yy5cvN+edd14wIwEsy5Ytc5zADQbkcKt0vvnmm2bp0qXm/PPPL3/Gx+Us /X/PPvusefrpp60B198alYetobDFltGQ9BFUMeQhOxu4evUq09sqnSOtQYAyPsga2evWrTNr1qw2 kyZNCmYkMIO8dt1a0/HMjmb8hPG5XOJQW7NmjRk7dkxAZ9EJlyYzyRMsFjGUdu3aZVatWuVWaGDc hRAMbcpj7979loOxzpAj3V7W0F69xvJljbmBjQMDtaUTzgFBP3P22WfnDCXqz6rVq50BI/1OCG7e scYthtyoplGmr3WCIP1suSxdttx0sk61oUNCOWhOOCMbPGdZp2a/fv0clj7W4bly5SrXJ4+y/WAo 2blzp3MMYVg3Dhjgkh1kHWdvLl5s1luHIxg7BDprCCxvv73Z9r1D3FgEGJx6K1esdGMSdRrDN4Ts 2bPXtaXBdhwabPs5xgGc9cuWLjOrV602kyedbc4IhIXVXjh/+/Tt48Zr8g+u5bburrZtibFa+p2s uTl8mHFgtXUk9rXjNeNj6ziw3o4Dq9eY8ePHOedR1jjKib+hoVs5j+kzykAeA+qEyKNDfygDyoAy UDoDOCBYAjxw4MDcwyx7xRGB8slqgAGtSmkuQAZfmLkFC0oURq0IyiazpTgEmKXDKRFCSIvlrqQn M9h8wgsOEWbqwBVCLrnkEvOBD3zALUf+1a9+Zb7/0MMhko1J44Sdtd3lVj0MsIq4LAGGlwH9B7h7 b9sybBrd5JTSmAiqeoltMjgihozKrxNg2bZtu5tVnjBxQlXTLBTZ/n0H3NYHDCffEGAVDcu3MaqY HQwxu33ErnpwM/3WcOve/aTC3dMa/r169TDMKvfp3ScPZ6F8VXqdrSB7WldA+DO1rHLCKYJzghUR /r1K0yz0/AnrnGEWm9n1Pt4qma62fPr372e22Zn3Pr37Ws6yn93GWUV6vXv1zFs9RF1htQorRXA6 9rT3sxZ42bKF1W+d7eqQFmcIabIyg76Y2e5d1ikbwonGFqJttk50s2OSnx5jEmPUOmvksm3Qv5cV P6wOoX6yXGaQTVvGAcakQYMHmTXWSb9t+3aLq9GFyQqHxLt16xbzrt2O54/V3OP3Grvqibotznt5 JqtP6su7755wdVW2McHPwAGNru9lVcTw4YyPdqlRHcj9Dz7qtjQAhXMhlixbaSZNGOtWFixeusL8 4JmfmakXnGtefc1OvtgVE1PtSomb7JYNpNmOJfc/+Ji7zpYODrOcdel0M+Pii9x9+Y80Dtj7Mz8+ 3cVHvBx6SbysYkCi8Vx5+R+4tCQO/5O0vn3vA7nzKwT3LBs/+ERIk7MhnnthrgsLvrh42dbx3Ivz 8sLF5UPi1c9kBtQJkcyP3lUGlAFlIJEB9pUyu8IsXHQfJ8owSlgoJwRYMJwwRqKzXLz9A0H5xAkg ymBi5iq4ydJ1jG1woPz6goEJV8zWhXJCwIkYaVFufGxZf8cpQxlwpkCUF2bXWZGAosOS3I4ds94v fcJstwYAdaFrl655WedsCvb677NnVuDcyp4z65zZvcth4KwDX8BHnaFuH7FbVkI4IXbauku+cS6K gQAmnEacUXH06C77KrxDQZwQOIoOW0dR/675e6/hpZtV0FmVwFkRUr997qr9HUOabU1RXlwdslyx pWjv3j1BnBBs2Tlg/xp7DMgrI8FC37t3394gTgjOMDls60N365wRx6JwT921UNw4EcLwP3zYloHF g/Mueg6QjEn0zWGwHHFnL7DtLMoLdehM254YPxsbKcNsjW3aM9sfwBHlpeX8mzNcO5Jyy/ITLLRr eInqDfRvHe24SZs+duz4KfezxJUUd9MIHCInnCOBsyBwQDSNbJlEYIsGDgOcDwhG/Ml7281td9zl ruN0YBXFKwsXmfsfeNS+YWOFdVR8wd3jP+LAIcAnYUmDsHPnv2Lj3uQcBLzJYpbdDkKaOBVwXNw1 cljBsymIo9mOc2AT3Hz6QhzOMRKJ95/tgZvy5g8cKbPtlg7wRfNxYP9B61SZ4Uep31MwoE6IFCRp EGVAGVAGCjHQMstjYhUFDDcUK4ynEIYcyhxGrqTrY0bRAQthcBCgAGYppINBEpcO+FAEUYTBG1XC ssRV67iPHG45tK5jx5a64eOBE5YlHzx4wHGXNS/MsKNQUS+iTgaWjfOHIbfPnhmR9dkQKNtyqGtc vsHHFgnO0sja2CbPu1sPmO3Y8dQzBZjdRjAiMPSyFpx1HKYaxwtOK+6BhSXuWQq8gAWJwyJ1iHaN ozNrIR3yTrpRA7ZTa7lJnxiHt5r4qLvHjtu+zJZHVDhrhbYNdyFm2fdZxwttxfUn1pHoC45OBIc1 K6CijlA/bDW+s+WA8YZVO1FpwXeGq7u0/06dsnW6UhdkHCZtX1rqkDGHLFaci3Hjlh++0u9ykCpO oSgW6Y/pmzmwMuu6mzYvGNk4B3AITJowLvZsBRwMd825PWe4EzeGOysS7rt7tnUUtBj/nMvwsH3D xnMvzm1xNnhv14gLO+fOe2zaK82MD16UW11B3MTHCgZWX8Q5AXAgkFYx3JNs+rfeciNROpF4581f kIuXdHB8+G8DIW6wcW+6xSYOC4lHP5MZyG+FyWH1rjKgDCgDykCEARQWJKoEc02UCxRCjO2sBcUG AUscHq6BA0Usa8EBgdFSCAfpc59wp5NgqMA/s+v55gHl1vLXwkuI+kIZvevo7xAxVrjIGxiQY8ey LyOpC4XqrrQljJqshZn+d1vbCDO1URFejgbghbQx4qgcwoGPR66FaEfUW+nHJF0fi6wYacFS/kG/ fpxJ3x0vNkAcFrkG3nftqqKsRfiPa0dco16H4oXVQkhL35vfywgvfllmyY2MSWfEvDUFfFx3ZWTH yKwlV0b0va19m6QpvLCtRsLJvSw+pe6CI4pFeGFbDX9tSdgy4RviGP8Y7jgPxAEh+ZEDIjHgfSFc NOwoe1AnMv3iaX5QFy8X1m7YlHe91B9sz/Bl0sSx7ifYRXC+8CrS6OtIp9sVGzhOXn1tkQTVz5QM tLhEUwbWYMqAMqAMKAMnGcBwwsGARBUJ/xrhULT8fe7uoSr/JwZCHBbBg/LJX9YiacRh4ZpcR+Fj Nuh0EXihPkR04Nbsn+QFQzhrOW4dImBBpDz8NC0a91PqlX+v2t/BkVRnJD0JI7+z+Dxu23RSu5bC w7jFiSPGdxZYiDMpz1JuSWGqhYsySuJF6nRLuBPWOZBvAFcLh8TD7HlBaU0avJwBkLUI/1IefnrS juAlxPJ6caDZVu3DyPvul2XejSr/kL5D6sYp0VuIYBH+TrlfxQtp00gbrhJokkbBEmrlRfirJK1a PitGfMtWjnwkOCtwNmz1DP38ECd/+Y6Nk1ez/7bfbvdCcKYgW93ZFo+67/IfDghE8irX9bM4A+qE KM6RhlAGlAFloCADKFBInPIpD4VSsgSLpBv3iVKeJlzcs6VcE2Ol2DNtXckqlr/ofQ4iSyOsmMha itaDVg05bVlWijeRm1YsorxXmlbR5xOKSQwHHBBgtrsBMpUk/sW4C8OLfQFga38Xn+EWZgiDISyz y/FhK78qq3jiYpIyanEUJRRm3MNlXEvkxYLxyynr5fWJ9aU1b9TbpHBlUBD7SLE0xEFzPMmhFBtz 6ReTnFHUF8ZwyjFEW0qsL2BpdSAV4690FsI+wZYSJOltGm3JeCc/S+y2kKhwToV/gHH0vv6OZ0Cd EPG86FVlQBlQBooygNIiinaSUuGHKxppBQFkT3YxLEkOkwqSz3tUsORdjPmRtUIek+TJS4kG1clg 1fyWmx0uYBcJJA6uzFpII01dSBOmUqykceaZZxQ2AFr5SluvKsHj8ttqxVqzu2BUbkuNWJcFQ1V+ gzYiS+yjsZ2sL6duG4mGrfQ3+U2uvy1cwV+IV1Em1QUpNdkKUWneiz2f46VAQCmnuO09BR4p+3Ia XsAbol0LFimPaKakfZ1h237WwtkchQR8LePmyTG9UNhqXM/pDQUiy/ESsx2swCPusl+mWa/QSsIh 98T5IM4IuS6fXJcDLOVaPX7KSgwOpLzumivqEWKbxJR9q2+TtChoZUAZOJ0YkIEbJQTlQH6n4SCn ZImW6T3kOwNCGNvF0gAPeEUB8qBW/atg8TmQRPxrEk7uhfz0cYRKt5M7INQe+Bhr3DpV2EHp1HqA XJa4/MMx47iQa3GHM2aBi7pQ0FhpbV9yKGQW6UuctBExWIQDucenXCu1r/DjKOV7ro+JeUiwhGhH 5PfM1oMX7XqqU9AIFtuBhu1jTkFCGbVchLtS+vOYqFJdkjaS48B7KncNR1vroabe7ap/lfpiWTgl bsHSUpaFjfJTHizzgtRLSdePhmvuuuVFwvn3q/09KQ3BZ6EEwRJ9O0c0r8LLybKMhij8W3gN4fAq jKLljrwG81X7houo8LYKtjLwVop6l0ZWOtjtI7yNI054o5VK6QyoE6J0zvQJZUAZaGcMsOSR12HJ Gx1KUVrldPG4ZZNyDYWvmNJRDUolDVFC/Di5Bh4UMfBkLaL8Cwd+ej4+wezfD/X9VBU9+5SZIWbG H16i6WM4wQ0SgpcWpbxlyv/UrRDUlxYsOCuyFtqce7NAaz2NppfjJZARJwcMCgc+nndbywiDvJS+ wo+jlO+uLtg0OTQvKtK+eKVf1sLsucxYS7p+msIL5RiMFwsgnpeW8yLo64L0d61t5ETMdiuuUX9D 8SKODmkzfhnJNWbJxdHm36/2d+nHYnlxZWcd4ylXZFWKTRxobOMRHiRO+Z3rh+RGRp8+L5K2JMVv +OLQzlLrLm894Y83LJXjwBAM1frEcJ96wbnuzRb+AZS87vLuex9wyXCYZdYiKxkKrcgolj7P81pQ nCaPPP7DvOA4JngFaVvaVpKXgRr+0O0YNSRfk1YGlIH6YICBnleIsReUgRsFO+3g36dPH7N+/frY ZeTEh0LRvXv3IAqBpCPp+uzK2QscAimOE/9+tb/zGkW4jDtpHHxwDF5Rxqqdfrr4om6AdE9VEgru u3bpYg4cPHSKIgwn7GHv1Kmz6WzfFZ+1cFAq76nnBPuTh9m1pNpiOFmj1zoHsn4lJilSVxq6N5h9 +1teMxvNO3UGw4k6k7VgiNAftLze8dQDQqUthcBCXnv27OleXyrp+vnnGngbAvDCTvWePXqaHdt3 mOM23ahQRkivXj2jtzL5TRk1b9vm+t7oAaFHj7bgYzl40gx4tYB1b+juxoy4s1yOu7ccvGv69u1T reQS4+lu+17abcvbFfLHMumPu3SxfUyXMH0M/MNBVE7YaxwC26NHmPGRukBbYUymr/WNdBkzGY+6 dusShVr133BC/wuOOCfEMduWutnXaJc6VtMnHTnS0mcdOXrE5iXbV3GnIeamG75gblt/p3ny6eft 6z0X2BUFDbbdbncG/ZeuvuKUN2GkibPUMLKS4dkX55pXF/I6z+nOqVBKPLzNA9zPvjDPvGJfCco5 EDgeOKySbRrRN3qUEvfpGladEKdryWu+lQFloCoMYCCgTGDIoeD5RvVBe7IySkZjY2NV0ioWCTjA w+u/UKp8JYuVHki/fv2CKOW8Zx2Hx969e095Hz1cgW/w4MHFspTp/ajyl2lirZGjfHbvaQ1cWx7w 4CuZGJTs/acMZTYzS0wo5L179zZbt241h+yrL3ECiICDMmoI5LQCS69evRwWZvKopyflhKvTzoFj 61UI6Wudi1u2WF7sK3ipJ+BDaM9Hjxx2bQvjKYRgbNOvyOuAT6ZpebH4cJjCXQghHZxB9DFRw/+I rUNg6d07DBYMLhx6hw7R37HKS1Z4nXAzwfDRt2+/XNllyQ9l1Nk6DzEAo7xQblSfXrathRDaCe32 yJFj7m0cnTuf5EXqEO0+xJkB1FvGAWagaTu+c/+o5QpHSU/rtPLHqqw4AgvldNA6gA8fPpJ3WCK8 UIf69u0RhBd4oP7u3r3b6Q2MlyKMA7y5qKFH3zx9Qu4nffpjmv896ZlS7g20b7Ng5cLk1ldYyrOc 68D10fYzKqwiuO/u2WaedUAsXrbSOR8mTxhrHQEzTjHcL7OrDeKkJb2ZzvD37+NYIt1i50qA4a45 tztHCGd+NvZvGV8KxSv5jMZ7041fMBdeMMW8ZQ+n3GqdsbzK8/9dfM0pr+30Mer3wgyoE6IwN3pH GVAGThMGKhmsUSYGDRpkNm3a5BSKAQMGONYw4LbZWToUjVAGAspn//79zerVqx0WcX6g/DU3NzuF RvBlXbQYbfDCbPL27dvNkCFDcknyGxF8uRsBvlDWB3jtlv3EeOFVmB2tchpK4KVP7z6meWuz2WUV UOqHGLgopCjljY0DcteyxoWxT93YuXNnnuEPFma+GwcOzBpCLn6cL6wCgRccVDKDjQGDQ29g48CS lfJc5CV+6WoNJxTcPXv2OGcRDj4EB8m+ffutsVK6gVAihFxw6kiDnWmHF4xcVq8ghw4eNnusk693 r95BVquQJjz06dPbOhf32S1sB5xRx3Ucart27nZ9XbeuYV65C5betKVtza4t9+jRotJiYOL8bOjW EGTlDPnHwO1jVzrQlvbssWXS6nDAIbFjxw5riDeYPoEcRbQbnC+b39ls6+reXLtmdQhYunSmDMOs ysC5wJi0d+9as2vXrhwW+mFWseAI6WexhpL+/QeY9RvW2zLaneeEAJvjrU/fIFDQG+CF8qCvpe9D 4GXnzh2Ol759ynCgZby4jxUFrAiICuc6FDvbYfoHLzL8JQmOiTiZNHGcNfjHnXIL50IcnlMC2gus VLjpxmvzbhWKt1A+eXiadULwp1I5A+qEqJxDjUEZUAbaAQOVOCKGDx9uZ5yOmZUrV7rZSpQZlAuu jR8/3vizHFlTNXToUDdLiSOC1Q/MQmFgMnM5YcKEYEYc+cQJAa/r1q1zvLCsH4ULAwFeQizzF77n z59vHnv8cXPE8gAXLEfeYJXRP/2zP3OrES6++GLzx3/8xxI8009m40aNGmU2bdxkl7UftYZcT7cN Ye/uvWbkqJHBDAQySf0AC1uKKCewMSuIMTV02FAzoHXGKFNCWiPHYBkzZozFss6sWrXK9LNK+jHL z7bmbc5hNWz4sGDOGbA0NY02a2w7WrNmjTMYWpx525zBQJv3Z3Wz5Id0RowYbtatXWdWrFiec95R RrShESNHBOOFfI4cOdKsXbPWrLV4Bg5sdGnjcO1s2xR1KZRTD+fd8OG2vzt00JbRascLfS9OTl5x OHrM6GBlBC84Wqkjwgs87LbGbUv5jQjKy+DBA61j6LDt4zY4Bx5jEOPAEeugGX3W6LwVWGDPUnB0 skqQPubAAcaklnEAZ9GYMWc5x1aW6ftx97f9GasMNm/ebI7avgVHFY7yfdapRhvrFWgVD5hwVNGW tmzZ4jAxDuy3Ds7d1vHZ1DTK9jM9fOipvmfsg0iFQQMpA2kZ6GAVRK2zadnScMqAMtAuGVi4cKH5 8pe/7BTImTNnmjlz5pSsvNKVouTxx3eMOWZLZQY1JHEowszs8Md3FFBWQIBFZt1D4YELZpMxDMAC BlZAVAPLu3vfTp0NDGuUTSd2Oaa8h90eAWann4wr7/wtAKmjLisgvOAk2rVrtzO04QOlFKdA6DIi A/CD4wxj4Ux7yF4/OwvXzc4yhcYCLzjvtu/Ybo1L2fbQ086w9yy5TZZVMJGHmOHHccYKCIxJZixp 2xi7oYXVVfQv+60hx+tV2VvNKquaYLFltNP2L/CCYEBRRjXBYnmhr2OLE425R/cerpz8rU4OZID/ 6OPo7/bs3eNWW3Xp0tU5sHBqhW5L1BdWrLAagjMZMLipL/Q1oYUVIWDZvXuPHQeOWQxd7fjYxzpD ajMmMRbQrll5xuorxuqQkwXCP7ywPYR2Td9b6Thw/fXXO46J/6GHvp9bkSPp6acyEIKBDh1tu+7W PzEp2x92CD+KJkLSm8qAMqAMtE0GUDAxYkMasoWYwliqFyzwgnEty5MLYc76OgpmLZTMQvmCF4xZ /upB4IZVNLUWeGFp++BBtT0vRHjAkK3FtiFJ3//EkMWZyL9aC2eWgKUeBF5Y2s5frYW+l60OobY7 JOUXXtg6w1+thW0XOEBCbU1Myi99DM5E2QKRFDbre/DCaqaQqwKzzpPGrwykZUBOq0kbXsMpA8qA MqAMKAPKgDKgDCgDyoAyoAwoA8qAMlAWA+qEKIs2fUgZUAaUAWVAGVAGlAFlQBlQBpQBZUAZUAZK ZUCdEKUypuGVAWVAGVAGlAFlQBlQBpQBZUAZUAaUAWWgLAbUCVEWbfqQMqAMKAPKgDKgDCgDyoAy oAwoA8pArRjYb1/hfP+Dj5q5Ly/IFEKodDLNRJ1Frk6IOisQhaMMKAPKgDKgDCgDyoAyoAwoA8rA 6cjA2vUbzVXXf9U8+fTzRbN/wDoh5s5fYJYsXVk0bCUBQqVTCca29qw6IdpaiSleZUAZUAaUAcdA hzM7KxPKgDKgDCgDyoAyYBngzR8ivHmjrQoGf+OAfvYVwAfbahZS4Z5z573mkcefShW2PQbSV3S2 x1LVPCkDykBJDDBwy+DNK9bke0mRaODgDHTo2tucOLDNmBMngqetCSoDyoAyoAwoA/XEgK+7nHHG SYdEPWFMg2XSxHHmvrtnpwnapsMsXrqiTeOvFLw6ISplUJ9XBpSBNs/Au+++az3uB8yxY8fM8ePH 1QnRVkr0jE7mjO4DjTl2yPoh3m0rqBWnMqAMKAPKgDJQdQa27jxgtm/fa5dEGNOxWx/ToUu39Gkc O2hOHD+WGB6jeaBdodDQ0M3Me/kV09i/n5l6wZTcM/utHsX1/fsPmqaRw/Lu5QLZL2y3WLJsZS4c YRsH9M8FIZ616ze5tPzrBADDug2b3LOTJo61YU4+JxE0b9tutm7bYTEMN90tVpG08SZhl7iKfYJT tohE4yP/nDGBgImwPSzOURavj5H7ry58o2yeievV194wzZaL7t27manvOTePZ+IHC+4q0hbM0bA+ 59G8EEe5ok6IcpnT55QBZaDdMMDqh27dujknBDMJJ+zMuj+j0G4y2h4z0uFMYzp1d4Noe8ye5kkZ UAaUAWVAGUjDwBHT2fBn7OJAPhs690zzWEsYwh7ebU4c2R/7DIYs2wemXnCuM64xoifbFQvihHju xbnuDAcxromELRWzb78lz/B98pnnTznrgXBfuvqKXFw4IEjrystnmis/PTOH55RnnzEOTy5A65cn n/6ZO6jyvm/PznNCxMWLof7tex90Brg4LMhDHPZoOnG//fj8+358d9/7gHMMcF8wweUdliv5PeOD 05wDIY7nhx/L38Lhxy1p+uVBvoiH5+B51qUzJJgBC4IzB2eDiITdf/DgKeU12Tp/7rj9qxK07E91 QpRNnT6oDCgD7YkBnA44IxB1QrSnktW8KAPKgDKgDCgD7Z8BdBhfjyk1xx262C2ORw8kbnFkZv2y j88wZ7tVCP1cEhwMidHKLPk/W+MUo/fV1xa5t1bMts4E2VqBkcthkzgubrrhWheOa9+2hjBvuPhn a4iLIyCKnTR4FoObZzHamcUXIzoaPu1vzmQAg+/weO6FueZhex1MpRrbz704Lxcfxj75icYnfHD4 pjgfonjnzn/F8XyhdfoIJ8LzjIsvMtddc4W7Pg/uI1hfsdxTHj7POJFm33mPCzv9gxfl4iRd7k17 zxTz9VtudNel7IgXvnGOgJNwd9/7bzZ/K60jaoVh20wl0nZPLakk1/qsMqAMKAMeAzgdqiUHrdeY P7Z41FLI0+HDh902k1pjgQfBwnaXWsuRI0ccL/WCZf/+/W4VTq15OXr0qF3iWh9Y2BoFFjBVs32W wzH1hO1a1OFaY6Et078cOsQWpOr1W+XwItvY6gELXNRT3wsn9TIOgIX6W+txgDKiDdGu66Hvrbcx qRq8+H2C/72U9t3hjOQDn6daYxUDeJp1JLDdAXnWroLAUL71lj/NGbcYwLOsswLDFYMVke0JMy6e lguHcYsDgLjYRlFI2OaBzLZODp5BeIY0K5GmEcPd6gB/xQXOg1Ejh7lVCaXGLXkUBwTP8x3Hgdsn kzLCGdZRAM/kVXhmJQhOAXFAEBUOBXgWxwDX2NoBp6x6EAcGz+E8QnAy+EIYP07KjnJGrrIrUYRv 4vicjRchvUpFV0JUyqA+rwwoA+2CgXIHbMn87t27zfr1691PZiKIr3///qaxsdF07Bi2q0X5Xb16 tTNsSRuFb8CAAWbgwIHBsWDwr1y50hmTYMG4hJdBgwaZTp06CX1BPlHG165d6xRhsKCU9+7d2wwe PNh07pyseFUbILyAhbKSMurVq5fD0rVr12onlxgfhv6mTZvMjh07TJcuXRwvYBgxYoQJjYUy2bx5 s9myZYsrE9oR7WnUqFF2T6vdduOd/p6YqSrcJG2wNDc3mzPPtNt+rIBlyJAhpm/fvlVIIX0U8EL5 bNiwwdUXcNCuhw8f7rCE5mX79u1m48aNDgtpgw9e6GdCy86dOx0W0pW+l36XP36HEurL3r17zbp1 6xwftGvalvAidSgUnn379pk1a9Y4LPS1YIGTWowD9HOrVq1ydVb6O+oKf6H7XsYBeMEJIVhoz7Uc B3BAwANtumfPnmbYsGHBeUlbLzkvwBeW+a+z2ydYon/C/tvqORKaRg1zQdliwKw5YZ60WygwpjlX QoxbjHT+koRVD6y0wBD2RYxs/1op3yVdtlHgMCE/iD2uPPe9lPgmTRjrVkI88vgPnXNAHAg33Xht KdG4cyD8B8g/+OAQrPyJCCfCM1zLKgXKg+eQZttvxwllERWJc4A998MX4bt5e0uc/r1Sv4fVjEtF p+GVAWVAGWgDDKDsoWBhJJ111lnOYNmzZ49ZunSpwdjEgAolKFiLFi1yxvX48eMdFvAtX77czUCN G1fZ8rlS8oHiCQcYthMmTHBKFfiWLVtmcNqcffbZwYwEFE44wCgRLOBbsWKFgR+4QiENISiacIDh NnbsWHceCbxQh3bt2mWmTJkSjBecQnDATOnkyZNdGXENLGCEF85LCSUYB1u3bnVl1KdPH+e0whGw ePFiM2nSJKegh8KCkc3f6NGjnfEmTgmcapRbSEfEO++84wwnsODAo+7AE2XHNQzLUILRT7o4QDCw cULgIKF9wRGGbiiRfhYDEjy0bxwk1CPa2dChQ0NBcQ6It956yxnW9PnwgnH55ptvupUrI0eODOZE o08DC22IdOnbcJBQd8FE/Q3luKKfpf02NDQYxh8cImBgbKD86I9DYZHxh3rC+AMWrsEVnAm+EJWG +kndoGwmTpzoxkmwUHdpS/AS2lFfTr4xjhFmxr9y65zYKMSwxzBmhn7uywvcmQ8YtBjp0+35By0r BWIfdxeJo8nWoWoLxvyTz/zMsK1BcFaSBk4N4uFMBrZUYMzjOJhuV0KI06Wc+AVbGp6J3z8Topz0 QjwTRuMKkRNNQxlQBpSBGjHALDJGATPHojQww45hgAGFscIsd9aCAQAWlBsUT8FC2hgGzKKirDPT EkIwklDsUHhlRh1FFF5YBYDhEmrmVLBgyAoWnEbwgWLO/VAGC/nG2YDiCQYEXjDoUMwxfCm/EIIx ABbqrjgbmLHFoENBfvvtt82YMWNCQHEOO9oLdaJfv5bZF2YHwYIRzgwzhkOIGWVmjcGCEUfbkTSp L5QfWGhHIRxXOIWon9QVqaPgob5s27bNtWv6GGnvWRcWeae+woXkn/KCK1aD9ejRI1eXssRCP0db kToi+ae8cJSAk1VXOEFDCG0FY5r6KrxI30v9haMQfS/jAOkh1BfJP2MSbYsxIuQ4QDvCuD7nnHNy WOABDIwDONVCOfRkHMDhKrxIXaa+0J5oVyGE1V44f3E2SN8rWJYsWRIUSyX5lZnxyRPGOWdCXFyy GoB7bHvAUOcchres4+KVhfbsiAceNb+zWwS+XuHWiri0i13jTAicBThHcIbI2zY4GBOMpQp8fMlu o7jM5pFtKDgNiMelYfNOOpVIy1aJc2OjEJ7l7IiWLTHTc84PrnPORb1Iu3BC0Gng+WZQZsAJ1ZnV SyGGwIEhwR+DhooyoAycZIAZdhQbFHAUCBGUURRQlEH6qBBOCBQa+kJwiIIleOgbMRDAg5GQ9cwT hhNKFjNOUcWb3/TX8BbSCYGxIsqezwuzuvAms7pyL4tPDCfqg9QPPw2MTMoNLBgPYsj4Yar9HQMB LNRfX8BC+tQXVvdkXV8wnKgPfIoDQvBQV6gzOEtob347kzDV/qQMSAuDkvRFMHRpPxicOHAw7LIW nB6khaPIF9oW+g7ORcJgzGUtlAG6AMvFxegnTTiij4M3wkTbWRa4mNknrej2AuoqvFCf+IvylhUW HB/U02jfS9uiL6TdR/vCLLBQb9lCRHmIw5V0hBfqC7yEwgIv4PCxgAdeqMNgpR3xPUuhb6Hd0q/R hn0hffgBSwgnBOMjaTEmiSNa8ICN9kSbpm77/Y+EqafP3Cs0O5wwnGOQJLJVA0OfMyX44ywCnBCs jkg67JAVBbLqIimNUu6xugDnAOcf+GdClBJHNCx5xBEBXs5r4I/tELfdcac7oJLzG8RxE3026bc4 GNDzivEs52dwJoRsq0iKu1b3sm3xGeeKmYqf//zn5u677zb33Xef+/zmN79pXnjhBTejknHyp0X0 cLxw4ULzne98x8ybN++0yLNmUhkohQEUBcRXyOV5lB0UGxRlFKCshcGJ7R8oLaTrixi0hKFdZy0Y TaQTxwvKJn8sxQ3BC5ywJBhOogqd4ENxJ0zWQjoYT6QbLSPBBw6U1KyF8mFbjKTrpyc8scKHssxa qAdgQaRM/DSpvzhw4C5rAQvGNCLtxk9TroXAQroYcZRDHC9c4x6OgawFXpL6O8EH3hBCfaFOUB7R tiRY4AV+shaw0GalbvjpcU36O/96Vt+pl2ChDUcNe+FFwmSFQeJlBQT9WRwvYAMjZUQ5Zi2kQ58X 199JGdE/hxgfpS8j/9G6y2/wwBt46l0wqNluwIy/HEApmNnqwLYAke/ZWXi2bEQPoOSsB0S2HEh4 /xNHAfejaSStVmhuzj8DYcmy/MMUD9j4kAMHT56vwG/wicOE32kFfLffcZfbauI/gyNgkjvL4VQd wz/bwX8m+h2eOSwTnqP8RXluaGg5Vyp61sOSZaWv7IjiqObvNrsSgsHwqaeeMi+99JL5y7/8S+dR ff311833vvc9wyezO1/84hdP6YCrSV57juu3v/2teeKJJ9wMC4YCfyG85u2ZU81b+2RAlP+oskdu xZBDqeGPWY8sBaWFvhEscYoN11FqUFCzxoLyCRbhwM+34IMTHATRmUM/bDW+Y0QLL3HxgREscJP1 LDv5lTxHywhewIKCCpbozGEc9kquUV8w0CTdaFxSdjiuojOH0bCV/ibPpAMn4ImKjyV6r9q/aR+U EVLIeOIe/IWQNH0M3GUtlJHkWcrDT1PKLQQW0k3DC/0QbTvrPiaJF+o0f6GcVoKF8oj2MVKfpe+V 3345VvN70jggbZ36EsJRJLxIefj5lPpM2yecOGv8MNX8ThoyJkm7kfjBBx7CgKctyHV2xp3tC7wu kxUFo0YMc7P/OCA4LBEHAoa4vMWB13YSrtEeerhuwyZ3UCX3ZbY/Ls9sb5hnV0uQBvHguFhrn+X1 l1GZYbdVsLLiPuv0cCsc7BwQBjirHnwhTXGg3P/gf5pJdksJBy4SpzhE+Ey7coFwnP1AvomP7zg6 wMmrTdke4cdF+qySIHyT5UwOk/Qx+t+vtzzDnc8fz3PQJ+mQHvGzNYb0WH0Bbw32XCdWR4jDRpwv fty1+N5mnRA4Gp599llz9dVXu/3GdKR/8Ad/4LZj3Hzzzebpp592S/A+8pGP1ILXNp8m2y5mzZrl luGyAuJf/uVf2nyeNAPKQLUZQIkQJSGq7JGWXJNwWRv+gqVQPsFDGIyJrIV0yLdw4KfHNbkuBrl/ v9rf/dkkSddPg2sowSEUYbgvxIuPCSMhayENsCCFeOFeiPoC98XqL1jShCFcJUJ+pS4U4yVNWVaC hWelLhTDUmk6xZ4nr1IXkrDAXb3wIliK5a3S+1IvC/HCdeFPDN5K0yz0fDEsPJe2vRVKI+31pLor cQgv8jurzyQslI+UkfCXFQ7iFSxJacBLCCxJGNLew3lw15zbnYPg4ceeyj3GQYx33H5LbksARvjN N1xrnrBGM1swRAiHIwOjvJBwj7hwQmB0IxjcPBc950AOwCScpMPzhOUMCF9uuuELLk4cFPwRJ04O jHscGevswZvFnAN+fJwHQRw86zs9yONNNu++XGXPh8BRAmeOKxsmScBx15zbHF7JF+GFP9JF5HBM 8i/lQRi4Jz3yVg/SZp0Qss/6W9/6ljuoqqmpyfH53ve+1+0t4z77fGfMmBE7q1IP5NczBvgUThcs WFDPUBWbMlAXDMQpnwIslJKVZDgJllBKuWCRdAt9hlCy0mAhjBhYhbBW43oaLKQTihfqZiGROh0C CxiEG0nXxyXX0ijv/nPlfAdHEi8Sp4QTbHK92p/CS1y8knaIMoIT4UXSjcNEGNpS1rPsaXjxHUpx WKt1LQmLnwbllLUTQrAklRFhJJyPr9rfi/WpgjFEu5a6m5RHqbtJYapxLwQvwm018BIHRvwTD91T MDruYyCzcoDtAmwFkEMe/YfknAS2OzAjz5kSYjxLOAzmuLRwdtx392x3NoQff9wZCayAwBiPYmFl gC9R3D4e/7WaxfLvx0na/HGGBTi723O6onkkPFxcaB0zYJSzNQrlXeIXDorxLPmX7R5SFqTpC3zG ieQheq8Yvmj4pN9t1gnB6cccwkQj85cy8Z3DbtiTyB49ZtmyXtKaRLDeUwaUgdODgTQKTtZM+H1h 1mkVi7/aClCx9JLup+ElFN5Q6STxIffgJQlP6DoNnkLKuWBJwiv5qvQzTX2RNASX/M7iMynPkn5S mGphCpFGKViT8AgvpcRXSdgkLJXEW86z9YQlbVsKgTlEGmnLKwQvtcovxnZ36ywoJmIUFwsXdz9p 24YfPi0WniklrJ9G0vc0OMtNN81zLkzr6ogknLW612adEOeff747LBGvMqe+i+BN5fRbPM2cjpz1 8mdJVz+VAWWgbTOA8cMfymvawZtwSbN9oggXC1ct5gSLpBsXL31jWgUo7vm015L21Pr4QvTRSVgk P3Ai/Mm1LD7BkqZ+pcFcKT7yWy9YpI1IGyyUtxC80EZkttqvq1Ew++AVAABAAElEQVRModoSbaTY SocQdZc2koYXylLCRTmr5u80fUeoMhL+k+oLeQ9Rf5OwCD54CVFGkl9JN1r+cl3CRe9X87fwEhen 4JB+KC5MNa8Vy6/gKRYuion+U/qKNH179Hn9rQyEZKBkJwQNg0pO5WZAkkrOdf5Y3kXH5l/nmh+W DEp4wklYybjE49+T8BIPnUncqzh/+ctfugO9pkyZYt7//vcHUbYFdzU+6TzIt+STOCXvwiP35Lpc 8zkkvNyXsO6C/U/i4nf0GZ9vCa+fysDpwACHZ+G8pP3JCf1p8y2KDW0xKtIWaVsSLhqmmr8lDWnn fhsnHTByQFu0X6gmBokLLKQfxwthhJtSlSyJv5RPSUN4iT7LmAbeEEo53PMHL8KB4PHxCWa5l8Wn lBFxR7FwTcpO6hXXshJpI7TFOBF8IXihHkgbEQ58THLN13X8+9X+jrHN4X3CgR+/YAlZRqQv6RbC Eu17/HDV+i55TsLil2W10o2LR+ploTLiurT9uOereU14icMi18Ai4aqZdjQuSUPS9e/LtXrhBTzU WylLH2u1v4sDTTiIxi9YSh2T1q1bl9NhGNdUlIF6ZqBkJ8TSpUvNm2++mWuoNBAaC4MACjy/zzvv PDN69GjnDHjttdfMpk2bXGfnd0YSni0VF1xwQe5Eck6D5QwC3qdMx0SHwB9p0KDe9773ufd2x5HK ORDPPPOMWxlxxRVXmIkTJ8YFq9trnOD+k5/8JMcVfAm38MUqj0mTJplzzjnH5WHJkiVm2bJljid4 l0EfngjPtUsuuSTvrRa/+tWvHLdEAL8IafDHu7SnTp3qrsX9RxgVZaA9MsCWraFDh7o+hveFU9el PRXLr7w1IG7Ap08krm72ZGLp/4rFV8l93uwAbtp/tL2Cj2s4IUJgIR36IDiIimBBEQuh8MEL/R35 L8QLOEJs3SPPpCOnxvvcUG5wQ/lkfZo/6YKDtCgj/vw0wcIfQv3NWigfykleoxdNj/GPut29e/fo rar/Bgvp8FaoQvWXRLN+k4pkDCy85jdu37zgC4EFXqQuJGEJ9SYtqQtwQLv2+2zBF6q/gxfSl/KQ suNT+mMZK/x7WXyXuiD9bBwv9Hd+e88CB3HSxxQaB6S/Ay9hshbhhboCN36aUoe4FmIcoN9lLACH 9LOSf/CBR/pnuZ7mc+TIkbnX6PrlnuZZDaMMhGagJCcEjWLRokXmRz/6kWlubs552wDN4NSvXz8z fvx4p8zjhGAQwKEwf/58Z/jK65QIT+fHuQ44IHBaiHCGA4Y4z/kzIjRYwk+ePFmC5n3y6iN5pSSv 7OStGDzTlgQPJq8dJS+8i5vOCaFTZPBi5ceXv/zlPCcEbwHhdaQoSyIMLmxRwalw9tln55wQdGyv vvqq+elPf3rKq6KGDRtmrrzyykQnhMSvn8pAe2OAwZr+gs9SB27aJf2fKDH+8/SBKBj0jf71rPhD aUHRIl3auy/yhohevXoF2aYGDvoiHMtwAEcicEX/Rp8eghfKlj4UDkgXXCLCC3hldkruZfFJGhhP OJ2jyie/KTvu+xizwEGcjC2cobR161Z3fpIYddyT+gxeUd65npWABQcgWKRM/LTgBU58jP79an+n XTO2FsJCvQVvCKGMmGSJw8I14S4ElgEDBpj169fHYqGM4CVulWoW2OCFti39nd+XCFf0vX7fkwUO 4qQugEVe7eqnQ1uibftbh/371f5OG6HdSj8LLhHhRcYtuZ7VJ2MSf4V4oT+GO+pw1oKjCCykCTd+ mlKHcKD5fGWFSfoy9H3w+EJdAR+8gLcUId4Q+EvBpGGVgUIMnOyZCoXwrtNgP/rRj5r3vOc97uBH XoWJ0PFjHLP9gQYsHS2Ohj/6oz9yr3p86aWXzKOPPuqWFDbZNy9cddVVzvmAQux7h+kkbrrpJvOJ T3zCfO9733OrKDj/4ZprrjFDhgwx48ad+voSOjecFszy88rOj33sY0EUOI+aqnzFafDNb37TKWEP PPCAW3HC4Ann8Ag3eDlFPvzhDxu4eeutt8x3v/vdnGMBJwzbUeB14MCBEtyVEzx+8IMfNI888oh5 5ZVXnKPiuuuuc86j4cOLHySTi0y/KAPtiIGowV5K1hjwaWes3kKhoE9DUCxwJqIMoiyHEBQWlG5W n7F8W7Cg1OA4pg9n9VkIoe/CYNmwYYMbL2RcIG0ODkb8a+5Chv+BZe3atc55Lood5Q4vYAWLb8Rk BYU0wIKxTf3AASzCbDfKJ86ZEIYT6ZLWli1bHBbfeMSxTR1mXAjBC1iou7Qn2hLjvXDAhASTGOAL MXsLFtoO9YTtWZSJKPY4jwQLY3IIYSzHsNy+fbsZNWpUDgu6D3WG+6EcIjik6M+oHzgYhQOw0K7B wl8IAQt1Bg6oI+Iso7zAQvnR1kIIaVEGYGEc8FeD0L7A5revLDFhiNKu6duor1I34IU6RF0O1fdK WjIO+BzQ/yGUUag+htfP40SjXfv6sWAJxYuMj9QXsPhtBiyMTfAifWDa+uLrMv73tM9rOGUgJAMl OSFEeZJOnUGRDo6Kzow7WwV8oSNEkeCPzlAGcQYNjOQJEyb4wd13FPYxY8a4zprOYNq0aeav/uqv CioepM3KiRdffNHMnj3bYQDnxo0bXZqk43s7T0mwji4waPGHkSAdIZyxqsRfLSKQCccf+aVjZeBD CFtoKwplAf84IOD6a1/7mlNqJM5Cn/CsHVohdvR6e2CgkvqNIYlhwmqmsWPHuj4HZQ/lAiMu1Owt 5cC2EhTw1atXm7POOssZCeDgGv2EGA0hygwsjBEoffRlKOL0U3CDkhzKOUNe6StR+FCGMRjAAjZm mOlv6RdDCQYu+SdtsPAbgw7jG07AGkoYcxgXcIpgIPAbg5K6DC7KMJQwJuFoZ/zmjBZ4YIaS+gNP OOpLVcrLxU59Jb01a9a49Gnj9BFydgwTIqGwwAv9yMqVKx0emYygLuMoop2H0nPQN0ifraCkj1ME gRfqMHpgKEcR/MML7Zr62mQnuOABox8nLLpTSF7gQtoOaVNu9HX0M3AmTpJy62Ta5ygj2i2OIhyv jEno44wB4KG9hxyTGHcYg2jHlAdjEOMA/R1tnD4nlOB4oDzoY+hT+KOu0P8xBoh9EwIPWHA40Hbg BB7AwiQC40PIMSlEfus5DV4XilTyxo56zl+9YivJCRHNhDghuE6jLiai5LMcDI9sIWFQ/fWvf+1u swIgaUD7v//7P8O5E9dff33OAcHzHFBJh3vuuecWSqZur9MpMngh5IVOKUngVbglXLHwKFXw9oUv fCFvZUVSGtzz0ygWVu8rA6cTAyiXbBVDmUA5RwmkDeNoxajkdyghXc6NAQsKKKsguIbR5M9ChcBD ujhEUarAQh+CMoyCDpaQvGBU4ixniT0GC7ygEIMFhS8kFjhg6yLGEuWEMQc+lFL+Qhm31AHSarLG GzNxKOaMOVzDUAELuEIKxhNYqDMYKZQRugaGNvUplFAfxGlHneE8LIRZZfSKkM480sVRRv6pL5wH BT54on2FNOKEA/o7nxfKCF78GV3CZi2SLliWL1/u+hiu0ffKCoCsMUj8jAO0a7DgMKK/Q3/FMYND L2Qfgx5Jf4ejU7BQZ+EltHELB2Ch7qJ/yjjAhCP1OiQvYMEpAy+ChT6G/gUsofteygMHCP0dfS9j A87P0OOA1OHT9XPOnfe6rN939+zTlYKa5Lsi7YIOnsaD+GcSRHOCd/x3v/tdbg8hv2VfWjQsv+kY /vd//9fMnDkzb6lqNOzrr79u/vEf/9GdK4HD4YUXXnCKJQ4RjOwLL7wwuAIVxVjObxQNBhAERbmY g+eNN95wnbukJSsi5Lf/SSf3+OOPO17ZPhOy8/dx6HdloL0xQLvFmKsHQZGRGcpa40HBYxZQZm9r iQcFkxltfwtErfCIoY+xX2sBizhAao2FMQndIrQBWSjf9YQFZwNGSz0Ihj7GXD0Ixj8GbT0Ihj4G bT0IBjfjQD2MBTIm1QMWHKusoOGv1sL4yBhQD+NAGi6a7YqBrdt2mKaRdoVnQ5jtaMVw1SOmYpj1 fgsDFTkhfO97kqGMs2DevHm5g2nYu8eStTjBSGarAEsweVMDnWic4PF+6KGHnOeb5WX/8z//kwvG swwEl156ae5aW/qCEoYTgo6SFSOyciRuRgoe//u//ztv9YN/AGg03ziDOEPiG9/4RuJyPBwZLJ0j bWZd8FwzW/eb3/zGefSZIcJrrKIMKAPKgDKgDCgDyoAyoAwoA+2bgbnzXzFPPvO8ueP2W8zkiaee 0VeL3Ncjplrw0BbTrMgJwfIyEfZ7xQlnQXBmAx5HllayVYC/QishWMbG+Q6f+cxnEmfOcEIQNm7v LJ5FvK0h97LG5b2SayxpFCcEThtWj/hOH+LGMfD973/fOXRw2PDmC6RQWbBv8t///d/dwZ0cLpok a+3S6dmzZ7vln8ySMftC+d1zzz3OMXLjjTe6w0OT4tB7yoAyoAwoA8qAMqAMKAPKgDKgDCgDyoDP QNWcEHFbAFiR8Pvf/95t2fiTP/kTd1AaDohC2zHYevDYY4+5Jc28wSFJPvnJTxr+2qvgcGD5Gg4I HDZxTghWNPz4xz82OARwGogTIq4sWFHBFhckDW/s7fzhD3/YXunVfCkDyoAyoAwoA8qAMqAMKAPK QBEGFi9dYX7wzM/sVoyWAxwfefwpux2jwdx047Xm1dcW2b833PfG/icPeGbFxJKlK83Xb7kxb+sG zzbbLR1cF5H4167faPYfOOi2e8y6dLqZcfFFEuSUT3kmDpPgYKvG/Q8+ZkqJVxLimUce/5FpHNDX XHf1Z3N5kHS5j7A15aYbrrHhTq4Ov/9B3gZ50D73mbz0WT0SDUsccDXPrjIhL2xzmXrBuebKy2fm xUm49iYnX9xeRs78mfm4MyE4p+GZZ54xH/rQh9y5DXLOAVsIMKyZyfdl4cKF7jWbn/70p4MfnOPj qIfv7LeU7RfihPBxwffTTz/tXtH5kY98JO+QrLiyYOUIW2JwQIQ8/dfHrN+VAWVAGVAGlAFlQBlQ BpQBZaDtMIBhPGnCWPv2iBYnA4Y3vxvs1nf+MMwXv7UiL0Nz5y9w19e1Guty89kX5lr7T34Z58Tg YEiMepwOGN/77YT1/Q886ozzkyHzvyVhIiQOiNvuuCsvXnuMv4v3kceTJ1nBAiacAldePivngMDh ItdnfXyGw0sY0vHz6fiwnHAdIezUC6Y4PnBQ+PKwdco8+fTz1uHQz+V96numGLaYzLbpk4f2LBWt hPCdEHHnEPziF79wr0665JJL3BkHcmoyzgdWRMjp5BDMjD9GNQ4LTl0+3YUzLVgJgbCiJLp9hbMd Fi1aZG6//XbngPBf/RQtC1ZBvPzyyy4+tmGEPGH8dC9HzX9pDHBitZwJU9qTLaf8f+pTnwr2SrZS 8Wl4ZUAZUAaUAWVAGVAGQjBQzYPncTrw9+TTxhrSK830i6flzoTAuDbWsF6ybKWZ8cGWlQsY4ax2 QFglMan1/AiuI9Pee677ZNUDRjgG+Ozbv+o+uTHr0hnW2L/HGeeTrbNDnncPtf6XhIkgGPHEzxsv iB+58tMzrRPiP82zL8wzGPtx8WL4333vA/aVut3yMHFdsN415/acYwKsN9862927w+ZBhLS/dPUV Li9yjTzBX/P2HYbVGoR5zjplwHLrV2+UYA4vqyNwRoC5vUpFKyH8MyGis++8o56zIK644orc2Q7+ 6+HYMiArIXBGsGKCgw+/8pWv5FYAtFXSyVfSK0jT5AsHj6yEwEGD00aEszB++tOfmg984APutUc4 K8TBQ5hoWfC2EZwQvG2EAyVVlIF6ZYBVUpw9wlkyhf64zx8H0sof79rme6EDb9Pkt5oDdpr0NIwy oAwoA8qAMqAMKAPVYsDXY/zv1Yo/Lh5WJEyeONataJD7815e4AxpVgDMtd9FcEggkya0HGrJygKc FZfZcOIo4D5xYsAj8oz7kfI/cYLgFPHj5fHrrvmsiwVHRFRwNOC8YCuF7xQhnHMeWKxXWacA+ESI n/xz3xeu46DwRfK9tTl/hcOBgydtPMLzHId/ilPHj6M9fe9YSWYKOSE4v4ADEDkYctasWbkk/Fdu 8TYN3oSBoY3z4Ze//KUzkv0Z/dyDbegL+WKVAm+W4N3QrDzgYMdSBaeCrITwt2PwndeP8vnRj340 9ypPtrqQjrzSEycI3LKK4pFHHjHnnHOO4ZWcKspAPTPAaxyvv/76ohCrPbgSH22HPol2VO34i2ZI AygDyoAyoAwoA8qAMlABA6LHEIVMZFYQXepHmcnHCF+3fpMZNXKY+875B9Ps2QbPvTjXng2xwq06 WLdhk11RMSznGFhrwyM8E5VRI4e7S7wStFSRVRhsGYkKDgQcBGs3bIzeat0CscPcbM+6iDovBOtL bptJvsNBMMoKh1MiLnABLDhq4IjVFPCFo2KSdWrUy9tHCkCvyuWKnBD+7Lu/BeDNN980y5YtM1/9 6lfzXrHpOy1kOwaz/L/+9a/tspcGN7NflVzVMBIOi/zud79rNm3aZMaPH2/+/u//PrcSpBRYbMeQ DoTZXRw7CLO9zz77rOHgTv8d3bzKlG0W8InjQQ6yZMvGG2+8Yb71rW8lvpKzFGwaVhnIkoFaOAAY uGkzOO9oP7XAkCWnGrcyoAwoA8qAMqAMtG8G0GPEXgiZU4xmhBUInLuAE4AtG2x3wNBmNQNOBe6z 6kGEFQdID2sDRkVWGxxotX+i95N+S7xJYaL3xHFBuk/YMxoutI4VwUBYiXPbtp12Ne7OvMc7mA72 vIyTB1Pm3Szy40vXXGFGWyfM3JdfcdsvWKFBuqyG4HyM9iwVOSHEUEZxxwnBNgSWRT///PNm2rRp bhWAT57vhJCVEKyC+PnPf26uvfZaM2TIED94m/tO/rds2eL++M62CZwRzO6WKhxMKSshfCcE3GIk ffzjH89z8OCAwBGBE0LKA8Pq0UcfNZdddplblVEqBg2vDJwuDLD6gTbHSghe8UvbKWcF0+nCl+ZT GVAGlAFlQBlQBuqLAfQY7AAEWyCUcD4DKwdeXbjIJYkRLTP5rJJ4xV6/0M7yI/LJd1ltwOGO0dUQ nJdQrnCeAyKOg2g8XJe0/Xs33XCte4azH+5/8D/Nrbf8ae62xHmrfatHFGsuUJlfptttI/whOGp+ YM+D4LDK7nZCOrqlo8wk6vKxipwQGL5U+N27d7vKzifbKjiD4G//9m9zWwUk5/36tRwMwm+cEDQQ XgM5YsQIc/HFF+dm/iV8rT5lOZM4AdLiYPaUQzUvvPBCs3TpUsNrLs8///y0j+eFY5WJrISgQ2Hl CCtMeM0mr+QcOHBgXni2Y+CEQOCVcyGee+45Vza8HrVeDCqcM7INJy8D+kMZsAywaueee+5x/UOp hOA84Fn/7JlS49DwyoAyoAwoA8qAMqAMtHUG0LdDCs4GzoLYZw18d1hla+IzPjjNnQvx3IstM/zi nOA2Zyk8+Yyxb7DYlPcM9zgvAuH5UkXSwCkSNeIx8nFwTG3d7iFx45SQMxh4OwYrEzg0Up5nGwny isVVLSeEe73pwjfs6pDpbqUI8YOdVRVsz2CLi6TPvfYmFTkhMNLFCQExb731lvnZz35mpk+fHjv7 758JwcoJTsFfsWKF+drXvua2Y9QDuRj8v/rVrwwHZ3KoZqnS1NRk/uEf/sFwyj/fMYzKEXgVJwjn P3BIH2dBjBs3zvEbjVNWQnCdlROvv/6645fXnYozI/pM6N84dziw9NVXX3UrOfRVoaFLoP7To818 4xvfcI6qUtHiBPT7mFKer8ZgTRw4AHHGImwxk7NaSsFSjbBgoR/AeQkmsPBXq20mrN6izxcsrKKr hWMUXsACL3yCgzKqVR8pWOjjwUK/XwteqHOUDbyABYc29aVWvDBWsKQYPIyDcMMYV4v6G8UCL2Cq BRbqr0xKkD5YKKtaYIEX+hh0JfScWmKBF+otk1t8px3RrmvBC21JsMBRLbHAhT8OgIX6Uq5eSt4q EXiRbY/UF9p1rcqI/o4xScYBsNSKF8oIXvik3vIn+n+pfFPmIv53uVatz7gVBnL+w37rULjK20bA NgxWRmBw46jwhe0aLY6I592nvK2CAyJ5OwSOATnM0X8u7ruPiefksEjfkUC88orMpK0OHF652L7t gxURxEMecKw0Wkyc39DYaB0W9nWiCK8TfeTxH9lvJ8xNN3zBXUv7H7zwKlNwfd2uuuA3IitK5JWo aeNra+EqckKgoNCRiPDGBhSFD3/4w7GvgfRf6cnp9g899JA726CeXsnJKg4OcqSzLscJARc8O2bM GKGlrE94pCNC6CR/+9vfOiXxtttui+0oCS+dFqsgXnjhBbe9hVUZ9SJ0sLwFhe03rBJRJ0S9lEz9 4KCP8PuJkMgqGbB5lq1XGzdutINTo2uL/KZN0hfIKqVQ+WFbGFhYUcXf22+/7YxbsPh9dgg8bNFb uXKlcxCRNtvUUDxHjx6d91afUFjWrVvn+lbq2datW13/yvk6/hlHIbDgrFq1apWrGzjPBMuoUaOM v2owBBYMSbCghLOSiDLjGvUl9MoilPG1a9c6g3LQoEFm586dzrE3bNgwN6aFNFpYtbd8+XJnVMKD /GYlIuUUEgvjJwdewwfp0+dQZv3793dYQhpQGNdMKFBPZFUmWPr06ePOqgrJC+2D+sJ5WbQb+lza OH0N7Vr0ohDtiDKhf2ObMfoNehnfwUJ/x++QwhgEHrZCo5eCBQzwIvplKDxMpKFf09eBBWzU2SY7 8RC678X+oP6KvsF4SdumvwuNhTZEu5a+Fyw4SKgv5U6qZFmmsnKBV1gyW+9vTRBnA6sMxJkAFgxr VhEwqy+v5vQxYrjPtq+u5K0UbOtAWInAc7zyMm7bhP98IUwY9bfdcadzJDxrHQfdbTvE2Acfb95I ipe0ydsci+kum1d5JSdvzADr/Q88aldJzHMwiBO5rvVtHu5Hyv/gCWcIDpev2JUP5B985B987XkV BBRV7ITwGyyvgfybv/kb16nE8e8rM3SMDFif//zngw4Scbj8aywHZxXDZz7zGf9yTb4LtwxsHPR5 ww03uO0ecWAwcmRQ2bVrl+OUtwygxNWLMIPDag6UlloZmvXCheJoXwygYKFUcf4LbwVCeB0u/QkK MiuYQs1wY9yywgzljq1uCOftcEgtBtV5550XzHgCC30XnMANxsnw4cPd1jK2rLFdLdRsO7Ne8IKT CGUTwZhiBR8YeZsRs2EhBEOb/DMmggUFFCMbZw3XL7jgglx/njUeZifhhXLAOSxCvaW+nH322c6I ketZf4rRP2XKlBwHKOpLlixxBlQo5zV1A16ow9QNOdOKMqOMMG6p16EMbgxIdBPajOgGYKH+Unbl nD1Vblni4MSQo1/DCYLQruhjMHKlrZcbfynPoUuCZ+LEiQ4Dz+KMWLx4sTN6OSA8lMAJBiV9r7wO nb4XvQejFyyhxgHBQlnQ5yKUEbxQr2nXofpedFLSZDwCC20GXn7/+9+7tsRb5EI50Zikw2EGF2ed dZbjBUyM1bRr3iQnurS7meF/Mg7Qp9GW4EXGAcYk+uNQY1LabGI032zPTXjFbiHAUJdzEnie3xjU 3bvbVX32uy8Y6BxOGbeqAWMbI/93dqWExDv1Pec6Azwajx+nfC+EiWfvu3u2mcfbLOyqBoz7yfZt GRj2UQeEf1imxItDgDMi2CqyzjoFSIfncETwtg8fa/RVoHHxES8OE2Nm5h1ieaV95ScHe/7O8rN1 +w6XBltQptuVFmnyL3jb4mdFTggGYTyaIsy6c2BioU5WBisJzzaM6DW5F/eJMR53ngBeQzqwNMoA cbCyAOzR8MRDB00nxdsnai3iBQUnA8all15a0GFDpymzrXyCf+rUqQXLIpo3FC7+fB7himuUZ5Sr 6PP8ljgIGzegMFPBQPyJT3zCOaDi4tBrykBbZAAHBG1QZgXJA8o4iihKKUZUCOOJ/hEs9AG+A5L2 iHKDYoMhE+oQYNKiXyU96UP4hBeMTWZ9wBVCwIKIccB3+jYUY5RhjBlRSrmXpdAXMrMNL9JXwgtl hvGAA2DChAk5zrLEgpGNgxgl2BfqMpyBBYdACGH1BeMvhpNvCGBso2sww824GGJ2G05k1YHvNCd9 JlAwfNFffJxZcYSxAhYMJ3FAkBaOEbBQdzG8/XtZYWGcZ2UVafsrdpjtp84IFp+zrLDgQKMPAYev S4KFtoTjhk/RpbLCQbzCC/2+Pw5g6NPHUXdDjgPwAg/UGRHGBRxnODulPsm9rD7RI6kv9G/wIOOA jEkyDvhjRFZYiBcslJU/DnAdXsBC/WUVQggBC/xQX4QXPuGCvpe/JuvQknshMKVJwz9EMRq+0Mw9 Br2scog+w2+M7aR4457xryU9m3RP4iiEmy0Y/vkWhMcRUSzOQvHhyOAvKpwDIedYRO+1598VOyFk 0GMg/PM///OcQhVHGoMSSh8dAEbyhz70obhgsdfoMF977TWnYL/3ve91Xm+UALyrKG3gYLYC5SXO uyuzgyjofKdjRuliWRodMp0PM2V4SBGMZZbH4l2X2cRYYBlelJkXPmfOnJnzZsclycAnyhB5+uxn P5tzSsSF969huPzmN79xgyQe4YsuusjgucZggQPKDK7ggnSigjINb3DGIMvAh8cfLy4K5cKFC93S WmZsSAtFk3M3GATqabtINF/6WxlIwwD71uk7UISjfQ+zGCgZoZRPsNC/0QbFuJU8iMOYNknby1qx wUCgHwFHtN9AGYYr+vVQTgjKAOM1asAKL5QhYxP9XZaCE7wQL/DEn7w9KlqfssCFsUJdiM64MZ7A BfWFfjtan6qNRdoJ8VJ/fQEfeChD6lW0DP2w1fqOowgHGlj8tgIn1F/GPZwDMu5WK924eKgvOEV8 xyLhwIXuA1bGVdHH4uKo1jXKQPbR+7zwnTrEPfqgEE4I2gntNmrAgoVyo07BXQgnBO2E/hd9LdqH SNsiTAhntPS9OO+ifQj9HbyEckJQbykn2kyUF8oIfKHGJPox8i1t2G8TUkbU3RBC30Ja0uf7aco4 RZsGc7QM/bD6XRloywxUpG2hlEjD/chHPuIM+iQyaFh0OjgAWDEhzyY9wz0GNYxWjFiM5e9+97vm xz/+sXsVKA4EOjneBPFP//RPbnYiGh8e6P/4j/8wTz75pFueh9KAYfxf//VfzoHBTCVnLvziF79w hjQKOsvncHrg4KiVMFgwmLJUDadNUkdEnujkkUsuueSUWa2kPLD6Y/78+c6RwNsFfvKTn5inn37a La1E0VqwYIH59re/7WYLo/GwDJ2yuP/++50jh/s4eh544AHnxWVwYZaRcnv22WedkkR5ESeOHxVl oK0zQBtA4tonfSRtGGUDZSJrQSFHuZF0/fTABxbCoKRmLaRD3x11QJAu+FAEMRDAm7WgBJNn8h9V hIUrUZazxgIOlEspDz89wYdxS9+btZAO3JAuPPgCT1zHYKEfz1rEQCCduLbENZxEIbD4DpFCWMAZ ymDByQCmQljgJQQWMIAFiXMECT4J4wJm+J/wIm3YT0rwUV9C9DHSlwkHPhbaEn+h9EnqAnUCXqLt WvCFGgfoY+hbozjgR3gBb4gykrog/axfRuADD1jBnLXQv9P3Cgd+etIfcx88KspAe2WgopUQGL5N dqkQDoirrroqFUcY1Myo80pOGl8aYTYeRwGOC9LEwKUjvfrqq108DIxc/853vmNmzZrlMEm8GAgY 0Mz2cKgjy0pRitlf+q//+q/OafG+973PGfovvfSSe6sE6Xzxi190nSbx1ko4+IrVHddcc41bKpaE A4eOrGLgQM203MLdo48+6vLLDMfcuXPNv/3bvzmu3v/+9zunEasg/uIv/sJw5gdbPEQYOB5++GHz 4osvmltuucV87GMfc44lBre//uu/dhyz2uW6665zA8y8efOco4oVM6xYSeuEkvT0UxmoRwZESUBx iIq0QxRCcQ5Ew1TztxiuYInDwzVwhHCI4ICgfymEg3xzn3CiGFeTCz8uUSrjsBCOcoKXEIqw8E+e o3j4DRZ4oSxx2mcpOETEWJG66qcn16Re+feq/R0cpCMcROMXLNLeover+Zs6KW0kzniSciNcCJH6 Kxz4acq1EFjgRNKRdH0swkuI+kK64kyNwyLl5rc3H2u1v0ue47DAC38SptppR+NLwwtlCTcyeRWN o1q/k7D4vNDnZS0+FqmrkiblxjXph7LWT6Ufk3QFB5/g4DplJP2Qf1+/KwPthYGKnBDM1F9++eXm k5/8ZN7ZEEnk3HHHHc4JkLaB0zHh7RYjm5UPLDEjTZwHKHIM0AyMhPU7Fjr8e++91x2C9rnPfc4Z 9DRolmPxCkuWLmK4o+gRlr1gCI4StpfUWlj9wCE5YPHzFYeLAffP/uzP3KBcyjJItrTAHYejwS3L CdnKgfNABic6bjpmXxi8fvnLX7pVEKy8wKkE/8wG/PrXv3b1AR6ZjWApJKtYKCf4Zq+zxO3Hqd+V gbbGAHVe2kZcG5VrhAuh8JEGIulG+eR6KMVGlKc4LFyT6/Q/aceDaH7S/hbDifCSrv8s1yhHwezf q/Z3yig6VsWl4WOOu1+Na1JfiKsQL9zzw/E7C5E2UizuELxQF8CDFOMlTVkWy1Ox+8J/EpZQvEgb KWRskxc4IZw4Aorlr9z7kuckXuBO+uhy00nzXJoykjqetdM1Kb/CFeUjZZkmf+WGSeKFOMEDL5Rl 1nphUn6FFymjcvOb9jnhpVB4n5dCYfS6MtDWGajICUHmSzF4CV+qcU9DxGgdbQ+KoZNi6wD7Ipml lw4LpwLX2ebB4TIinHbLac10yDT4p556ysWBU4Nnb7zxxtz+PIxnwnOeQnTfpcQX+pOBKu3+QXjC mVLqzBkH4vzd3/2dWxbG6hCeZ2WLcEue2ZbCihDKQQS+nnjiCee8oVyef/55xzGrKejEWRmDs0iE 7S3EwetY/bjlvn4qA22VAeo7IkpMXD4Ik6QAxT1TzjXBkvSsb2Alhav0XpIi7MddTBnzw5b7PS0v Icqo3nhJ4kbqdAheKFvBIun65S3XQmBJ20YIx1/Wxrbw4vMh30PyImkW+wRvEuZiz6e9n6YtSRml jbPccGnzS38XygkhdSMuT+AN1Zbi0o9eqxcs7YmXpPKP8q+/lYFaMFCxEyIEaIxW/jjHgdl0DiHy DVycEBjQvEHCP62ZswkwlnFMcLgk2zCYcWNLAVtC/IGAvWIcrIhzo1RHSQgOskpDuF1rT0BnJQTb JHwO4YUDKuHNd0KwYoIzHYRbVpWwMobtLjhyfOcUnTqvPoL/UKfPZ8WXxqsM+AwwyMuMYJIS6ofz n6/2dzGGimEJoZwIlmJ59PvhYmHLvZ8mDcpRyrLcdNI8l5aXUFiS6oLUoxBYwAE3hZxSgiUtf2nK olAY8pvEizxHmDThJHy5n+S5kJEWkpe0+U0brlw+5LmkuiC8pC1LibPcz7RtJAlzuWlHn0uTBmWU FnM0/lJ+p00jVDsqhj0UL2nKCKxpw0m+fB7973JfP5WBemKgTTghhDCMYYxfZtP9jo3VDpwbwRkR rBzAKcEJxRzqwiw9b2ngnIQkwQHBc5x/wPYBBjDONmA7QS3PhUjCXM17HJjEm0Z4C4d/kjTcsmUD Bw9nVMApnMipvTgV/vAP/zDxDSKUDa/KgkecPwj74dgCI28AqWZeNC5loFQGZLCm3dO3yO808YiS IEqv/4x/LY0h7D9bzvdiaYAHvH7/WU46aZ4RLD4H8px/TcLJvSw+i6UBHjgpFq4a2Og/qV8+B368 cp1wWQv5lbou6calGQIL6YIHJ0QcFrkWooz8NhI32y5YCCf8xfFWrWvwz3gp6frxyrVQvEg6SbzA SYg+RuqlcBDHC2XEX9aSBgsYhL8s8UgaSbyE7O/IaxwW/7rwlyUvSWn4+JLCVQtf3IHNftyCR8rS v1fsuzxb7Xq//8BB88jjT9lXTI41M/4/e/cac9tV3od+OQaM78bbG3zB29v4Ar6QEHMSoCmQCurK RIKqSSqVnDatTo+QGjXqhzZRPwWkfkl7Pp2KfMpRWlVNmpOeBkepiShIAbciQDDF+Ipt7L2NL3jb Bl+wIYXkrN+0/6/Hnsx1e981x1rv3uOR3nfONeeY4/mP/7g9zzPHnPNn3rEIxtrP33nPfZPP/o8v Tt77Mz+9Lz5n+Vv/z3+cHL70ksmsz3WunaB9mOG+CUIIJghCMFD63yz3DgKPFbhTL3Dw0Y9+dOLd ExxcHXho+T9n2ioJjx9YVWElhfRWAhiAOM5/+Id/2H1polwBsA/reCnIAhA4FmzIKgZGxhe/+MUu OHPjjTd2n0f7z//5P0/e9773dWnwhNv+IOm6BB082qLerKjw2U4rJtTh7/3e73Wf5pNvk8bAphnQ Zo0J2qZ+sIpjEWNiyCjPMQZfDcMmOhgh/spy+A2PPlvDQYiDFg7KOg4+x4K5PL/u/dRRqbfU4S6z NOs22kod2cc9PXiBp5QSXzCX59e9j/u0kaF6yt33/hi/bhzyi1OEgyEsOVaDF+VNH4nessw5VvJX nl/3fvpIv73QEyy16ih9JHrLsuZYLV7SFqJ3CEvGofLcGPvhfwhL+nUtXuZhSZ/GS+pyDD6S57w6 Ci9lf8t1Y2xLXujO2EeXenMs49AY+ss806ejtzwHh+PqJ+NQeX7ePh/IH1l3/b4wDUL86X//Qpf3 JoIQx56cvjx/qv/aN1/ZYdj2f7Be95ardoIQfgtM/JN//EujBnFu+W9/OrnlU5+dS8/BC86f/Mav /9O5aWqcXO7zFDWQLNDh0Qp/ggYc5VIs9fdZTYEIX2pwt/3AgQPd1jfoH3jggc7BcI3O7S7+pz/9 6YnPUebzeo8//ngXjPAoghcx3nzzzZ3D7NGCk0E86iIY48WRGRytOvFVEXx4TMU7MxzDkWCCF2da JeEvA6nzX/rSlyYf//jHO95x5x0cjgvm4F8+gh5vf/vbTwZqWxn3AQMmem1aAM7ErT0vK3l8K8Zd eZ1j2rxHldZtEJR6sh890ZvjtgIsxKNVMQy7AyP9M1Yrs6BOX+DDMbwZb/pp1vmbHgYonf26DS84 GfsFmcokCISboTqCzXFYhoLn6+REXlan0RO9Zf6OabtE3xhbtJUEwGe1Gf00acbEwznJisBZWOiv gYWejDFDWByDt0YdwbIML8Er/ZiSlZRDfSlc4aXGGEPPvPEOxlq8pF3SOWu80+9rjDHGVGNvxtmy PWTcwV0CBOX5de/To68Y1/q8hCttpcY8QE/G3j4W+PAFx6p1pIzq35/VU01eYeBjv/lvu5Ucrxyp u5fvp2U7lvaDB87vVqtYsTL098STL31meSz9q+S7b1ZC+HKFT0K6C9+fUDizAhSf+MQnOofZ5zWJ l1n+0i/9Unf8t37rtzoHmzMsSqhj/4N/8A+69xdIy6H+/d///clnPvOZ7u6/zvuLv/iLVRwH+jct AjW+kCFoEzEAWr3w1a9+dfLHf/zHXTABTwI8Bkk8/87v/E73wk/Xe1+HFQ8GeZ9K/Wt/7a91WVm5 Im/Bh//0n/5TF/jxtZJaE3LK07aNgTEYMNkbT4wZDN9yfNInGBhemltD4IBH4JBRVQY+rPQggog1 DD4OrjFE0Bc3ZeDDb/gEj2tInErjE+OuxKKOCN7KuhsLF+7pggUPpZGp/ThmbKxRR3jRHsyfeGDA RmBRR46VGHN+3VtYlPvRRx/tgtb9/OGrFSiiGy8C7PSa7+Aj+rP+lXrsDo78L20zbbVU55jgTPme rPL8uvfpoc8jlSUv9DhWEwteBPRwoK2WfSbjnUd0U3fr5qLMT3DG+KFt9HmBL32tvGasff1E4NVY oh+X/ZcNTLTvGrwEC7tb39E+Ipmn9Ptyrsr5dW/Vj8AVDujWdiLqCD71WGLM+XVvlRcWL3NXR6X4 bZ7Svsu5qkyzzL522OQVBjzOsUl5719/x8Tf2PJTN/z4xN+QWI3h72f/+k8Pna5+bN8EIaxI+JVf +ZWddwqUTP2zf/bPupdKGnCtgrAighhI/tbf+lvdex6888DjF15QaVLyfoMy2vlzP/dz3UsTGcyc bPrKAarUdyLu/+qv/moXpCmdAoOkT7D61KbHU3yK03kTlz/v5vi1X/u1zoB1XgADdzgujVnvjfiX //Jfdi+nNKi+973v7b5AUmPSORHrqpVp/QzsZbI2znjsiCMnUJov2jCKjTuMwdyxWz/y43NksOiD 3nEDS4IfjCurvhhhwXf8lev/ZYzAC+PTaigvFI74TYIvx8fc0uW9P+rEY3dEvVsFZyzCWw3BC10C 67AYKx0j8Gk3eMuxsTHhBRZ1UtZH6qist7GxcEa0YbqtyotTyaHkOMCyF6N8Ffzmf/OVOimDRdpz +nktLMYPWPDCeYpTyXFi1wgMlHPuKuVcNS3d2i+97KWMbXDBB0st24kNR58+jAscEVjUm7E3x1Yt 56rpM7bCQjcnnxhjjL3qB9YaAou+7LFYwc70a+0YltRhDSzGVuOZPmO8i32esdf8GXw18MDiphnn v2yn2q7xptb8GLtBfcCSFUZ4wVN4WXUecH2k3M+xdW2/Ox2Pv3Tb1yYekTh48PzJddNHJA5e8KNz aJnuzDNPn1w2fUeCRxSGpEw7L8/+tQ8d/ebE+ypeP33EYAhDzruODsGIs86YYjn0xp2sSt1w/tRP vnUwr50LejvHpqsLvvSVr01X3b/YrT4YKiMdDx195DicwSY9XHfdc3+XsxUMycPxIw8/0uV9+NAl MwMMPUiDP//g5k9Oy3X+9L0a4wdDBgH0Du6bIISBwV14HbMvHGODmA7Xd2wNKu7me+GkyKLzQ3mY rKyoYAA6P5Smr/dE+v0zP/MzHTf9AQ/vDA7c4bI8j0tGIf5Fbv3u848j1whMSHsycnsitZMTuSx7 mbA5TPqIL8y4G6ivMCwcM/7UfLmtL9YwwgUiOG8MdUaxYx6JYqDWEgYfXo8cOdLxwuhjjHJe8FIa gWNjMkcwwhnmxitOARy4gYUDXEs4bgLdeNFG/IaFY2fVWK2AiPKqA1gY5h6Tg4VDx5kSrFGHtcT8 ce2113b9yHuaOCzqChbzx+HDh4+bg8bEpQ+zGzzu6Q8WwTxYtBUrLWvZCfQI5qsfnyPPzQLBI7aL NlPOzWPyIm9YfB0LHvVCt0c3ObfaUq0xht7D0zahvaojWNSbmyI4069r1RFe2DnaCF6Mw3gwD6T+ avLiplB/HuBom5/MAwlkwT22GHtLLMYcY13mpJrzo35szD169Gg3tsCSsVe7rjkPCErpu25g8D0E zDIn6UcJ8I1dP6vmz+H+td/4110AItdybD86fcdAGQTgYHsEQoDgzKnTb0t+7sb3Tn757/18Lu22 3mPwB5/45E4aB13zDz/8d+Y6zHTA8pL+Xz0uz/z4N//2t3ewCgLA9LPTFQn/5P/4pS6JPMo0Dv67 3/0vgziTZ7nl2MO+Ize/VMad3y/vCNp074SY6v3Zv/5SwObf/94fTjwe8fppAOe4lRrTPP7h3/s7 k+9Ox7bj8p7mJRDxG7/+qx0/fR3zfn/xtts7Hn5l+k6KbZF9E4Qw2cybZIec35Jk1y6aAKQxgZ2M Mq/ci7hzfpk7QvN0nIyctzKfOAwYfxgwjApOLUPPnTC/axp7YdRKL3dWcreSw314aqzXxsL45hhw lBjADCxBEc5tbSy4YZgz9DgGsMBw/fXXd5iMY7WELsYwLoLFGGolH65qi2AzvZw3d07NlXhxrCYv yq2tWmUHi1UH+haDvNYy6ZJ7Dgos6khAT3sWfOAc1J7PtBUBmvQjWPRzTlNtLGmrwaKNCMRuAgse ONXGXW3X2CtIo73UHmPUg/Zh3Df2CnrihBNeu47wcng65puHBH5xE6e3Ni+wCNDoN+rJ2Gts0X5r Y9FWBYjMA9qvMQYGX9HTx2qL+REvVj9kTtLPjT21x95ly37n9G79z/3Nn538wt++qXOEb/nUn07+ 3fSrGR+dOvcf/zcf7bIRqODYnzENJPzzf/qPu7v6Vk38wSdumfzX6UsTr33zVTt39AUBOP2c63/9 T//PLqAg7Ud/8/+efPy3/2P3IsoyuBGcCXIkAGE7JMH0d//Rr3Y4yhcywikoAafjVh+UOM84/YzJ L07LOUtcL0hAt6CG6+ESbFhW6LtuysfvfPw3Oz4FI/6vKXc4FYjp4/rT6RdCPjt9pGLVr2588r99 tsO5Lasg8HNyetzLtoyWrjHQGGgMLMkAg4HBl2W4S142SjJG37ZgwQuHwN82CIPP36YFL9uCBRcM cE7BpgUvnFxO7TYIB4WjsA3CkbUypebqlFnl3jYsHH1/m5ZtGnvV0TbNAwIy/jYt2zT2wiLw6m+/ yE/95I9Pfnm6QiHCGbbC4E//xxemjxPcN30c4aXHLXxB47ru5Ygv/eao/+Lf/sA03Re7Rxfy3gKr Aci/eDkAYV/aX5k69QIbvvRQ6nO+DB58dLoqYFYAQtp5YhWDFRoJIEjbBRT+8f8+eWj6CIQVGson GDAkrifl9Yenj3lYqfCPfuXXhy4ZPOaLGRGBDIECun/xQzd1gQ3n4IIFf7CtIgIjghtWgGyT/Oiz DduErmFpDDQGGgONgcZAY6Ax0BhoDDQGGgONgY0z4J0Jffnfbnhrd0gwgli5YAWBgMRL70J4yQk+ 7pGDLuWkc4453v1AgmutAujf8T82XcEiOEH2EoBw/YNTvAIMCYg4FhFEEaAQWJkluR7+UmYFLco0 8/Zz/WWHLjku2ZnTFTK7EQENIqixTdJWQmxTbTQsjYHGQGOgMdAYaAw0BhoDjYHGQGNgnzAQpznv fQDb3XerHIYCDylWmT7Hym3fuXfuDz7xJ12Sl3S+8iLO8rpl949MgxAeAxkSj2iQeRhdP4RxKL9N HbNq5E//+xe7VRD9QM+mMEVvWwkRJtq2MdAYaAw0BhoDjYHGQGOgMdAYaAw0BnbNQB6X8NJFqxm8 7+D//Z3pOx5efmfErjOeXij48HenqywEB1Z598KQTnnNCzIMXVMee+n6lz57Wx7fpv08MrJtqyBw 1IIQ29RSGpbGQGOgMdAYaAw0BhoDjYHGQGOgMbBPGPByRfLS6oRJ9/lOzn3elZDj/eI47k+wYkis ojgyXVFRCmf6F6Z/H5i+HNNLMv/97/1/5emV9q2CeGGKcygQcde9Lz2GMW/1wAXT9zQo+9D1KwEZ KXFWQQw97jKSypWybUGIlehqiRsDjYHGQGOgMdAYaAw0BhoDjYHGwPYy4KWXY4jHLPpO9y3TLy+Q vFsh5xOcCI4vfeX27O5svYRRuv67F3xS0pcr+i9hzGMSPt8piOBrG/1rdzJfsOMrHbD6wkcpnHcB DgGIeY9b/PT0JZ2u97WKUnC0DfJfX66XD/zN924DnB/B0N4J8SOUtAONgcZAY6Ax0BhoDDQGGgON gcZAY2D/MFAGHsr9dZbAyyd/7Td+86VPV05fyfBZX2uYOt2+vJBVA76K8Qc3T9/fMP16xAsvvjg5 Y/r1JSsLvJugLz93489O8/jC5OPTz1p6meXBA+dP097fBQbkJ1AwS3xRAxbXfnT62MfQpzxzrRUX cHpJ4+FLL+lemumllwIjcHrhZQIi3e9pYORfTD8vOk9e+lrFF7rPaR576unuc6LdVzV6QY15eYx1 TnDkS9NADg4THBrS5ZEWK05+bVrWy6Zf9qgpbSVETbabrsZAY2ArGTBZZ8L2ibXsbyXYBqox0Bho DDQGGgONgcZAj4HSdmHLjCEeg7huGhj4rd/+j907GTiwnNxf/nuvfLbTly3iwP+73/0vL6e7f/LP p46uYED5skpO8r/+2K9PXj/9ooY8rX74g098crrK4Y0Lv37hWo98WEmx6P0QHuPgmMPjM5cEFoEM X8IQIKE7+cA/z3nP9flCx3+dBh7+zb/97S54Qpe8NykCEHjx/ox5YtWHdHt7xec8DbPPnfJXU5l9 up1pDDQGGgMnPgO33Xbb5Jd/+ZcnP/jBDyY///M/P/lX/+pfnfiFbiVsDDQGGgONgcZAY+CEYeA9 73nP5NixY1152DWnT1cgrCp/9cJTk7/64fcXXsah58BafTDP4bb6wCMUggyLZNk8F+Uz6/y8/HNu Wax9HSmnz2jO46N/3Tb8VvZ1Yj7lVadNTjl9fn1PA2anvGobCt8wNAYaA42BTTLgjoHJWhDCnQSx 2fKOwiaxNd2NgcZAY6Ax0BhoDDQGFjFw2mmn7QQevv/97+/sL7puN+c5rWcusXzfioZlZdk8l82v n25e/vPO9fMZ+r1KOYeu3+QxZd+EtCDEJlhvOhsDjYGtY0DQIcsXWxBi66qnAWoMNAYaA42BxkBj YA4DbJjSjpmTtJ1qDGycgXEeGNp4sRqAxkBjoDGwPAPtqbTluWopGwONgcZAY6Ax0BjYPgZKW6bc 3z6kDVFjYDJpQYjWChoDjYHGwJSBdU3YHunwt678dls59P/whz+c/K//9b82jkUZGpbhmvzLv/zL ramjYLHdtGi/2q52s2kpsWxDv84Ys2le6FdH2zLehZdN1xFetg1LmwfUyvFibFFP2zDebdPYezxL 7VdjYDwG2uMY43Hbcm4MNAZOIgZeeOGFyaOPPjp5cfo5Ko92nHrqqZM3vOENk9e97nXV3y/hWVBY nnvuuU43PK9//esnF1xwwc5SzVpVw9B76KGHOl7ohOX888+fXHjhhdV5+Yu/+IvJ448/PnnmmWe6 4lu2Cot6yhLWmrw89thjk+985zs77yA577zzOl5e9aq6U7M68jIzf+rH35lnnjl54xvfOKmNhTGO E+03zqTnnC+77LKJbU2hX1vRZjhxeHn1q189ueSSSyZnnXVWTSido/Td73538s1vfrNznLRXYwws Z599dlUseHn++ed3sOAFnksvvbQ6FgWHRR1l7IXF+FJ77MWLsffIkSMTY036kjrSt2sLPvQj7Sbj G14OHDjQYauJR//By/e+970dteYkWLTjmgLLI488Mnn22Wc7XtSTuREe+zUFlieeeGLy5JNP7tSR dnvRRRdV56VmuZuuxgAG6lo6jfPGQGOgMXACMsCwuu+++zoD5sorr+wcFQ7dvffeO3nTm97UObm1 iu3Ozj333NNhOXz4cPdiqqeffnry4IMPdgYyR6GWMLDwwjDHwxnTt0Zz6mDB2eWXX14LSndH/YEH HugM8iuuuKJzIgVp4IPl0KFD1Yw+Tj8s9Kujc889tzOIBWscu+aaa6rxwnE6evRo58S9+c1vnpxz zjmdM/f1r3+9qzdc1QxEcCbhufjiiztHUhvy+84775xcd911VQMRHAPtg0Og/ao3jpT+de2113bt uVZFCcyoE84SLOrt4Ycfntx9991de6kZiOD008uxhoVwdu+4447JW9/61qoBGo72XXfd1QU/jL3a KgdTvV111VVdkLFWHRlHjPkw6DcCVtoQfH4LdtYSYy4sgg/GWS9e/ta3vrUzHutftRxufRgWcxMs ApzmpG984xsTwXtjYC2BxdirDV999dVdHxaM0LeCpRYv+vD999+/ozfzQOZHfSvBo1r8ND2NgZoM tMcxarLddDUGGgMnJAOcAcYNo4GjzfjkuBw8eLBzoNyJqiFxTBhYjF4OJSyMX3d5OFDO1RJG71NP PdXxwsCChRPlbhyn5dvf/nYtKJ2TTZ86cqcJFqsgBGU4LZyFWsKhxI2VBu4EclqCBV/aUy1hgHP8 3a2FARYOrfZTGwsnhWNCv6DQa17zms5h4SxwqjgP2ngN0Z/1F+1WG4GFI4cXd245D/DWEMEPbUKb 5WhbEfLa1762a8vqqyYW5VVHsGR1CjzqS71x/q0CqCEJCnEa8aJ+4ILF2MexrIVFea1S0U5hKecB fVyAkZNbQ/QRY1oCvepFOxF46boHUQAAQABJREFUyDxQC4vywmLs1Y/1J1jgMC9p11mZVoMbY50A iHlAG4HFuGf8szLNuFxLYDHv9OcB7ReWmnNSrTI3PY2BkoEWhCjZaPuNgcZAY2BFBhjCjIU4TLmc YczIYQRz5moIA5ixxxjvL11naDFO8w3xsfFYVp/l/XgohVHsDk8tXuhmeHJQOAelCEjgBZYaDi5e YEn7KLHAph2pQ+2qhsCi3P06creSs60Oa/CirMHCaSsFDo8/CJho4zWELk5cHJXo1IaCxfkaYnUM p1EfLgUv2i8c8NYQKw/g0T7KMQYWXAm4CrLVEFg4sP06MrZYpSGQpC/VEHWQsVeAqBR1VBOL4Ji+ pI7MBRFjDizGllpjr3JrD9pKf+zVnmGqNfYax4xn2oc5qBTBEVhqtRdY1JHxRD2VYnzJ/Fhr7C31 t/3GQC0GWhCiFtNNT2OgMXBCMiAAwehjTDBiSuFQMiYYGzXEKgfOCiz0lhKHgQNR4w4uLJyEIV7c feK01OJFIIizghN6S4mRjrcaDi5nhVEeDkossDnOkasRhOAgaL940VZLSb3hpMZdU8GZ3PnrY4FL +4W31koed0Q5AOk3JTfwOVfL8c8Y03duw4v+XOtusru3yp5+U/KCK+dqOXL06NtDdZRj0sA0tnCk YRlquxmPa413xviMvf3xLm3IGFRjjDF2+BvixVjnT9uthSXvx6C3FPjM38aXGvNj5iRto19HfjsO a61AZ8lF228M1GLgeCu1ltZCjwnC0r6aS6AK9SfNrsHMUkE81zIQ9kougxPu/S57LQcOavIQvLZN FjMQp6hvSLgyxzhyNfhksDC4GVP9gIjfnE1pGEBjCz0cy76xRy8c8OClhuOv/8ziBR71pH5qGHwM XPzT2a8jvPiTpgYWOuiCI20VHxFYSI0gBCeE40TmYUl/6xKO9C91tAhLDV5g4FSS1Ef34+V/4aoG FnWUdhm9JZbgC97y3Bj7aQvab78vZdyBt+bYO8RL8NUKWsVeUB+zeKk19tKj3cziBUZtt1YQQjtM fZRtEj7H9f2MQ+X5de+HF+VPv4kOOByr1Xajt20bA7UZOD4UuKR2Axxn1pIhz3XtVj7zmc/sBB8s Ebvpppt2m9XWXifAQjx/tin5sz/7sy7QU+oXZX3LW97SvUSqPL5N+17+9bWvfW3yvve9r+oLndbN wV7L8elPf7qD9KEPfWjd0AbzC953vvOdG223g+C27CDHNgZu39gDNcekY9zod2NKggvR29flOGOv hsGnvHH8h3AEI/5y17Kfbl2/S16it8zbMQGTGrwoL11DOGDK8bSrEue69+lQRyR6Sx0xjtXl2JK+ NISD7hyvgYUOdUTCQffj5X/Bor3MauNl+r3ul+23n1fw1WgvOAn/4aDEk2PS1eQlHAxhUUepy/L8 uvfDfzgo83fMHxzwDDnkZfq97i/CIv9g2auuRdfPa7u5FpZtqSPtNm08+MbYLgq+ay97xTLUFsco S8uzMbBbBnYdhODYcqx3G4TI6gd5eCP3fpcsC+y//RhPxIutxnZAhjjEsz/P3t1www1dEkGkz3/+ 81Xvrg9hyzETpmWKglr9Z+OSpm3Xz0B4twxRELDJ7hhgKJChCT/H9mpMLIsshlz0Dl0Xx2no3DqP BctQniW+GgafMi8SdbRMukX5LDq/bbws035r1FHJf9k+wmeOxcHK8TG26mhePUVn+lKw5fi6t/Pa ZXTPS7MuPOoovETvUN6py6xGGEqzjmPzyhx8y9blXvHMw1LmLd3YQYhV6qjENsb+Il5ST/r10CM+ 68QUXublqe3WGO8WYQkve8GSPOaVt51rDGySgY09juHRAOJzThyg/e4E3XbbbRMrO/ribrK/TQQg YPH9YfKOd7yjCxgJGu02cNRlNMI/ARzcZdXICCpalgMMhHefzjrZxWSdCdtdteyvwgvjpS9Dx/pp 1vl7N7jXqX/ZvEpeyv1lr1813TK81MCxCHeJodxfdN1uz+Ml3AzpGzq2W13LXBcsQ2lrYpmHo4+t Bq55eKI/2z6+E/n3PF5S7lq8LIMFphp4lsFSA4fyLoNlm3iBZZtk1Xoq+S73t6lMDUtjIAzsaiVE Lt7LNnczTvQ735t8DKOsn6GXApXn235j4GRmwESf5+PduVl28pZu3l2tGBCL0q2L+9x5jN6hfHcb ZBnKa96xYBlKU+KrEaBdRod6nFeXQ+XYzbFlsMi3BhZ1tExbn1eXu+Fg6Bo46Mny7aE0jtXAgvss 8S/bah+TNEnXP7fO39rMojuiNXhR1rTLebzUHu+GuA4+eJdp40N5rHIs/EfvrGuX7f+zrl/meOpo KG3wlW18KN26ji3iJXi2hZdabXeRTb5bXjzmkXda1OB0Xe2k5XNyMrC2IIR3RNxxxx3dd4C9dddd bcv+BRm8eyCPXCRd3hKc1QMeFchqCNe5O+ub4IIVjvuOrnzKTmX1gbu5rrUvb++VoPPWW2/t7vjr 6J5xl6drBQXe/va3dysEvG/ANcRjFFYs9IMiVmzAAm+wKEuCC/L2DeSyPPS85z3v6fL1OAajKr+7 g9N/VijQn3zpl69yRlynHJdffnl3CA7lLcuRtP2tdDgJrjwWgqt5gx8dJV8eI7FaJbzgUV5wpk7p znFYw43jqe/rr7/+R97rkPLF6NRm8NLPw3lcwTVUB/SUbQE+5dcW0qZSj2V99zGFs75++Q/VIyww lbhwpQxemNWv86FySF/WOV2zJG1GGbSBft3kutRF2hYO1Enqq8+7/PTDfrmVbxHv0XkibBlLDBB/ DIBlDVd1QYaWV8aQkFfSjcnVIoMPRv1/npG6Lnzhc4iXUkcNXqIj9VHqtw8jByv89c+v8zfu/dHZ x+N3js0bp9eFJ7zIL3rLvFN3Zbry/Dr39RH8M6KHsORYDV5SR8oXvWVZcyxtvDw3xr4ycyyit9RR s47KIET0DmHB37LjZ3n9qvtpl4uw1AgUZewYqiPH/MFRA8s8XoIPjmBelfdV0qe/Ru/QtbBoM2NL eBnC4pi/WnP1PCx4CJZV6wiPyZut1/8s6tgct/wbA6swsLYgBMeHM8P50ok5SITz9+Uvf7n7ZjNH 24DkcQDpODl5NCADlXy8iM82QQf7nCF5vf/979/pYJxGOj/3uc91uvI+Bvk6zgmDhQOmUz744INd QEF+8uL8cQLl4ze95cv/6PQnX+nkKw8OKbyu55z3y5OyAAVHX5KvdyAoI5GvcgiQxFkcKgedKQc9 cA2Jc3AZhJSPI4qDElv/OvUkCKG8OAtPuPFySHkor7rjeAenfKRJWcsgRI4L8AwJjPTAmBedJuCR 9F/4whe6c1dccUVXHnmqA+lS57PagjzCt/dihC8ccrzhCt7wnTYZ/bYpW3lMe6EXhpRDHc7ieagc 0i/z4k3Y6INVGQQkEuhzfQSXt9xyS9fu8QWLtOoW1tRD8DqWttLnHW/qZB7v0bvft8aJ0qGIAbBM ufLJs3mGsHpY1ZhYRnc/TV7wCEu/DDkGSy2DD69DzwTD4g+OeWNSv3y7/Z1njXESvckrx+DAzdii zNoCHH0JFrylLvtp1vlbmeGBpV9PwUJfjTriiCizMWzorn/wpb+tk4d+XmW7nIelRh3BpsyzPmMY XmrWEUzRaz+SNp3+luNjbaOHXu1Vv4kEX62xF//zxjv4gjcYx9qmXQ7xkvZctvGxcMh3ES/qCd4a wZnwkrGt1AmH4xmHxuRE3uYAutRR+k10wgEP7la1G6TPNfJp0hjYZgbWFoRIIQ3C7kDHmOM0c544 fZw1jqw/jhFjI05hruf4OF46487JL3fp+9fEMUsericGHA5anCvpPvnJT3YOcz9/Ti2njlMGH4kT Vjp58rj55ps7LMom2OJvVnm6jIp/sCXfkif5coqdsx/+XNovB+f/j/7ojzq8fS6iSpmdo0+ZXBMe wk/S2io7fumOo+q4tDjjLCdAo9zScoyD0/UGTM56eRwvs144KT2M0rieYzxUHsfLVQUPP/xwt0LE CpQEIWAl8Jd5lHyrx+AN31ZQqL8cfymXxf/TVvp8+a29K1tf+uWQh3Y3VI7+tYJJAnDyiCRohI8E /ejGv7pKfUuftK4PR+Fdey85S/6O4z3cSC//ZfAmj/2y3ctkfeDAga794l0+pSGs3hgT0pTHx+JF nXNY6GTYlEZW3sZ97rnnDrbPdWOCRT/Ine2y/AxhGC+44IIqvMBhHFJHfYMvvMBbw0mg4+yzz548 9dRTHQcl7ziBMdyV58bY17fPOeecLmDfv2umjrTncDeG/jJPTpExx82D1El5Hr5aWOjVNp988slB LOqI6Es1RADd+JvPY5Y68RLuyuNj7R88eLCbA+jtS44Z72qI9mI+orc/9oYr3JXj4Fi4YHGDJByU ejI3aFM1xPiib2ecjVNKd/qWfq/djC3uxLNh02dKfcY7f9pUibFMs879jKvmANwYTyKpI/NEbJ6c G2ObscwKp/6cFHznnXfecRiXwVHaMuX+MtfupDl16hr+8Ps7P9tOY2BlBk5ZbmxZ+4spLXEvO3Cc piHHt18ogzfHjGNZ3mWXjpOk0zrfF07VkNBt0InYz+/+NcHJYY98+MMfnnzwgx/Mz26bPKwG2I0E f58n+SpzOCjz7pcDD9Ivw2mZz7x9kyfpO6P0WKZPV1YDxOHl/BLnEujwO8eVxfFw69xupN8WGBVk aLLv47dig3gxZ9kuS76Dt0u45L/w1cfWr6syu37aBFCGylFeZx/ePo9ZRcMIIx6DUhfatvSlpP8I Bi4r6r3PmWuXwbusjm1Kt9sJW3/kjDDufO88Ir/nnnuuM/QYEzWEsQcLA7ysJ1j0RYGAWo4/oxIW uvvjpd+Mz/TlGtxwEhiafSz6DV5gLQMlY2GiQ3tI+yj1aD8wwlrDcaKbAzALC2O4Jha84EdbhSkC h7FNAKfGSgh6OXL6tj6srUb0c1iMsbWWOnOM6Hr22Wd/hBfHMgYF45hb5VYHeIlDSx+OHHMOdzUE J5xpfbp0crUXK0f0IWNMjX4Niz/jbgIg4cAYYy7Vl2qIsVdf4uCWWEpeas4Dyo2XvKsgHKQ9O1+j jugw3uGhPw8ES632om2qI8GQPha/YXS+RnAm9ZHtKa85e3LKjy3nROaatm0MhAFt55TTlpsD1r4S IiB2szVQk6GB2iTLCeP8mWxK52gVXfIhy17v7gOd5SQXA2QVvUnLuCJxQHPc1gqDLJsvjw/tpxxD 53ZzDC559p1XeaU+lJsYGKXFDYc3wQlOcfgqj3NmNyVpU0POTuogaVbBGC7CzSrXritt6iptigFI hjClbpN2XRhaPi8xoI3feeedkyNHjnSPWDEwtCt94+KLL65mlNN72WWXdU6cgNNVV13V9VUG+bFj x7oABKO9lhw6dKhz2GAxPmiHHO3HHnuscw6G+uVY2NSD9n/06NFuVYSADSx+G9Nq3aVUPrrUhwAo h41zzUB/9NFHO0fGXFBLjBf+BDPhgsUc+9BDD3X72lMt4cSpJ/0GP+Z8Rrp+RWCpcfeWLu0DFrrh ueiii7oAgDpSV9pzLQeBY0+/92SxR9I+tB9t+Morr1zaplG2vYgxxvwuwI+Lw4cPd9np0/qXYLs2 VEPwr03cfvvtHS9wcTTVlzEPNvVYQ+g1D+BF3zH2aqvmAW3ZzRvBpBpCr7HXylpY0latMnr88cc7 +7NWoEh5tVc8aL/aB/ubvWLMMQfUnAcuueSSrm0Y93EAi0ANLOaB/s2eMesLL1bE0Q2L8Y9tqV8L itfEclw5T5l+IeyMg5NTfuD9PD847lT70RiYx8Ap07YzedUZk+lAPC/ZzrmtCkIE1aIAgYjqojTJ ay9bA7g/hrPJLTr34sglWh8Hci/41n1tytfPNxNnHG98MFpN9CSP2rjeoJlHNQQkpI2z38+35u+h suVYecd4WUyu2YY6xG/wp22lXP2ySLuXttvPr/1+hQFt4brrrusMvv/5P//njqPECOY8MFBrCaPb yhf90suCGeruUsLBAKspnCervu6///5uLIXFHR5jAgePQ1NLtH9Y8CJgFF7cGePI1MSij1577bUd lrvuuqubWzjbDE/OA2y1RLk5KYzyu+++u2u72gujWPudNZ6MhY8jxxjnbHNstRd8/PiP/3h3fCy9 /Xz1WY4j3Ry3zHf4wNdQsLefxzp/6790c1g8JgKfuuPs1nr8IeWhjzMpCGG8g8OfY/pTTdFOr7nm mq69fOUrX9npx/o0zmpKsAhcsR3VkfYrSMQ2qjkP6EN44dB+9atf7fq1OtKmE8SqxU3mASuBBYy0 Y+OdOcCcVJMX86N5QEAkWDI/4gVHtUSwCBZ1ZOwNL+YB7bfmPPAjZeZMvvr0ST3L5UcQtAMnAQP1 LJ0lyGQkkjhT/UtyvIbzx+E2iTCWy3dCwJRHKvr4lvkd7PLPfq4rnfwcq7kNv32dcVxFiSMG66za YJzlPRL4wpuIuyDExiK5LwM1qBNly/7Lp7qIs/20uxxfZiswE6N0mfRjpRGA6BvD8+qx3+bGwnUy 5otbRrglygxPxo42V9PACu8MUI6SOzyWtTNm4NkEFhxcffXVO1gYXozSTWHxotXU0SZ5oVtwWx9m BONFHdU0gtNe6GT0Gq+1Xb/V0SawaBfmEfMNRwUGvOBnE3LhhRd2d2qNq7DBsinnwEoVji5eiLmr P6/V4ihYMt/gZVNY3Ek3J8NivNtkHcFhHshjEDiBZxNi1RssxpgE8zaFxXhiHgiWTfJCt+CdecDY qz9vak7Shw9Pg72ZkzY5D2yijTadJzcD9UJ+S/DMcNYhh5x8Djqnj3FSQxIQGHKiM+nuBkccxqEy ioaS0tnfjY7dXCOoYHIYcqxzrFwyl0i6FyuS1Au+OGPuNgpeJN1uMK3jmuAa4jvHcucmwYjU/Tz9 qaNwU6bdS/so81m0L8hD0qbCtTuIfbEUUv2Gj/759ns9DDBmtH+Ogva0CUc7JWHMwMIo3pSBNYTF Uu1N82Ku2RZe8AGL7Sac/tSROgkvtpvGos3iBZZNBSBwk8ADLPrTpgIQqSdOJBz+NuX0l1jw4m/T WIy3Ge9g2eQYo42El005/akjWNKvNz0PlFg2zYsxBS/m6m2akzY9D6TdtG1joAYDWxWEMIkkautu eoRT5647sVSzhpjMCIcuTmlwcOb6kgk4afvn8xt+k5NHFsq0HEpOsXPujtWW6BRUKHHBiQPnwwls 9v0JNHBsy3Oc4Tj4yzi9uTYrLtZZdnxrV57TLMtV8p0XbZqQpC3rHBZtsbzWsVzT50vaofbhmr0I /WWf8Pvzn/98l6U+Q2BXTwIjZVrt9rOf/WyXpnxxZ3gfA2+nrP1rDDQGGgONgcZAY6Ax0BhoDDQG GgM9BrbqcQzYOElxuOKUW9rPkXIuznKvHGv/yUGjjzPnk5x+c9ZEbzl78JTC2ea83nLLLd35fM6y TGPfte9+97u7T0zK13Xy5YALQPQf/ehfP9Zvd9M9UsGpDi4v41QXVkCUzmswCDYIUuQOfI5z0B2X ZxzdnBvaSqPs+KObU20p+ToE3zi99dZbd8o1i29plVPASz3ChAN3VJwrxYqPsn0oq7RpH2XadezL 3+oSwZTgSp8oOb7hhhu6tqTd6j9w56WqPktbph2T93WUueXRGGgMNAYaA42BxkBjoDHQGGgMnHgM nPrRqeymWJwbDnT5pnXLiByz7QunjSNVit/S94VTy0Hi0HleyzOZb3vb2wYDEMmjv1zT7z6+6Mk1 +Z1tH7/r4VBWjp+X6LzrXe/qHpfwm3MevZ6PlDbL8OKYc2Cd8xehJ2X03KC83LH/6Z/+6W7ZdNLZ 7qYc5fWej/SSG2UJ1ln54kWQB2a8uw4uwYm+Ey4P5fDnWeLyPB6I42X76A7O+IcPeuWHq1yHG9jL /GUxxMu8el2W77IeYVHn6gUe9V3WI1x0Sgdf2kdWgZSBlFXK0acILwl6WE4p+ACLoEI/KIcXzzqq A9f5rf/81E/91I8Ei+hZxLuylTLEe3l+v+578dwf//Efd88T4+9v/I2/0fG3X8vTcDcGGgONgcZA Y6AxcHIx8B/+w3/Y+RTq3//7f7+zT08uBlpp9wsDH5vKKVMn9ZWPce8X5A1nY2ALGBAMyGM4gWPl SFaSbGpVS7C07fIMeKv6Rz7yke7lXTfddNNkOjZu9Jn45ZG3lI2BxkBjoDHQGGgMNAYmkxtvvLF7 MTwuPv3pT2/kHXOtHhoDyzAwvVHqGyxNGgONgVUZEGzwyEb/3Qt5UacVJE32FwPeHm4FkLisVSRN GgONgcZAY6Ax0BhoDOwXBmLHsGWaHbNfau3kxbl174Q4eauilXw/MeBxB48yCEL481tggnhXRP8x if1UtpMRq/eEPPLII91KCF8SadIYaAw0BhoDjYHGQGNgPzHw0EMPTZ577rkOskBEk8bANjPQghDb XDsN21Yz4J0M3vvg06p5NMO7Gry/ocn+YsBjNerNpO2TXW01xP6qv4a2MdAYaAw0BhoDJzsDl1xy yeSpp57qaGgrIU721rD95W9BiO2vo4ZwixmwAqJ8AeUWQ23Q5jDwYz/2Y91LYQUhvFy2SWOgMdAY aAw0BhoDjYH9xICXl/tr0hjYDwy0d0Lsh1pqGBsDjYFRGWjv5x2V3pZ5Y6Ax0BhoDDQGGgMjM1Da MuX+yGpb9o2BXTHQghBL0ma5/Te/+c0lU7dkjYHGwH5jYF0TtneDPP/8892jHZvkQHlefPHF7vlQ L6vatLzwwgsdLz/4wQ82DWXi08iem90GLN///ve3Bgtenn322Yl3pGxa1I06ws+6+uZuyxQs+Nk0 Fqu1jC/606axGFfU0TZgwQVejL+bHu9ggcPfpp/Lh8U8gJtt4CVz0qZ5MRYEy6bnAXW0TfPAbsfJ dl1jYFUGdrXu+Bvf+MbkiSeemHgmvv+JwlUBrDO9IIE/Lwa0TH5dYiLxqZuzzjqrexnhuvK99957 J9/61re6ierNb35ze5nhuoht+TQGNsDAk08+OTly5MjkNa95zeTUU0/t+vXBgwcnntGsPU4+88wz kwceeKD7zCg8nITXve51k0svvbTDV5Mexu99993Xval7G7B4cReD7/TTT++MUHPF4cOHqy9hZQDf f//9XSDktNNO6wIjltHCcsYZZ9Ssoo6Ho0ePdg4lPjjbHku64oorqmPhEJjHjx071s3jAiKOwXLe eedVfeM7pw0vnrFWN3EW9KPa7/7BAXsBN+qoj6Xm8990P/7445NHH310ou0SY4yxzl9tgQUv6ggP 2q+x18ujPWpXS/Dy9NNPTx588MFuzDfusx8vvPDCycUXX9zNC7Ww0AOL9kuMvbB4b9WhQ4eqz0mC VV//+te7cSXzACzai981BRZ+jECINqPtnnvuuR0vac+18JgHtBdtNnMSTL6w5neTxsCJzMCughAC EDowZ7+2cT2vMr797W93uHTedQYhfHbRxPae97ynU+9FhLfeemsXhBE82I3kqwoG39rGzG7w7uYa 9fHJT36yMwTC3W7yadc0BradAW2dIaEvc1CMFwy+u+++u7v7xLGsZQzTe+edd04uuuiiDouACAPn rrvu6rbXXnttNUeOsXfPPfd0ARAcwGJVGV4ch6XWOzgYmgK/XjzqPS7BAgfj2FheywAVBMEBfTgw D+BFsEbd/cRP/EQ1w5xe5edwv+1tb+t4sa894+uqq67qAvC1+iAsAld4ibPNycTXNddc07WlWljY OYIh2oYACIHloWkgSx+/4IILqvUlX+/haIcDDq/giGCjfY5uLQkHAkPhwFeF9CW8GHtsa4jgLw7Y fcZf/dp4rC8RDnctoVdgUcCBc40DY+9Xv/rVbnURjLXmAfWh/wYLXswN+he+rrzyympjr9VV+q+2 oj6M+cZAY5054vrrr6/GSzhwU1H7xYtgxB133NFxc/XVV1ebB4y9t99+++Scc87pOOBPCTZqQ/qS vi4g0aQxcKIyUC9EvE8ZZIS4+/C+971vJ+DCcNxr5Fa+BsGbbrqpC27s5086ulP1mc98ZvLlL3/5 uFrGkTL6a9IYOJEZYJQzLjkCtoxPDhSjy/jB0KohnJHHHntsBwtjDxZ3VGBzZ4xBWEs4CIxwTkmw GBfcpcQJPLUEFmPVG97whuOwwGbliPO1hD7GMN3mE3WEF79hdIe5lmQJOwcudcQw117cpauJhWPC sbZqx7yBl/Qr2NwAqLWMm4MQLO6SwuEPT1aq1MSiTegrAmjuHocX44txRnACd7XEGEPvgQMHOk7g 4UipN8ESfb6GZHWI+jCmpP3CARteamFR3qF5wNgLm5t3NecBwbOMteFFn9JmtGtB2VpiDlRXAvQc be2Fc20sNhYK3tQSvPTnJPzAYm50vpbgxTijfagrvODH2Gtcrjkn1Spz09MYKBloQYiSjYF9KxYs 6TPhRhghv/ALv9DdHcmxVbcMUHmW+a6ax7akN4gaTEXeS1G2D37wg5MbbrihPNz2GwMnFAMcNcYC Z9JfJIa5wEAtY4JhycBk4DFqSinv5MI0tnCcGN4c2v44xxhm+NU2+Bh4/SWuHBbCyavBCydauTm0 fV5g04bUIaO9hggypK2W+jh26kgd1niWHPfmEcLpLwVX2gzDXH+rIepIG+Zca8MRnODGHA5PDdE2 6Usfjk715pgARC1HLljUR8mLfUESDl4tLPhnd9CrXiJpz/qa9ltDOLAcam3DOFOK9qwP4a6GqAPt N+NJqVN7gSV9rTw3xr62qY6G5iS8aDfmxxpjDB0CaOnDZXn1c+NMrToyvqsj4702U0rGXrzVmgdK /W2/MVCLgVdG7V1qNNkkmsc5F00ckjIdo49jPyQcWkvFbBlo8usbarnO5EJ30uV4fysvdy1M4iYH +mfhLK+1WsE1fSdafgYqk3CwxSCRL1zOS9fnJOnoYeBkIujjKfOYxZdrORpwhDOPyASfOyYk58p8 Sk7CX8rSXfTyvzId7kSyy3QwKBNJeWCiiwRjfncHp/8W5StduJpXjuQ3b7sMl64vMc3ipMTEAFI+ 143BbcpU4hqqg6Rr280wwGEkpREcJAwsBjFjggHEyBlTtEnthRFDbylxGDhx+moZMCnTrWtfQASW fjBE/nioyYsyM4YZ5f06SL0x3KXrG4Tr4iP50KGe6O3Xkd/wSYO7YMu1697iBBZ6+7rCk+CANIz0 MYXDyIkjfSfOMfikcTfZnDemKHOckSEs6Uuw9AMDY+CCBaYhLHgxtsDiDuqYAkPGu1lY6IfXYwBj S8ZVHPT7UsYdbVe7SZ2NhSkO45Aex/QndrBHMsYWbSFlTj+OztQbXowx+Z3z695m7B3qs7D5g5ez nTpbN4bkp8z0mPv69aQNwWIeMC6OPT/inj05NCdpy/CZQ83VsDVpDJyIDOypZVslwFEvxWMF73zn O8tDk7z/oDzIyXv/+99/nENbpjMY6XxEfuXjCjqvF0WW0XbpS+c4unRyaW1L8YynF2vOE84r6U/s 9Hr8gMPvj4QLv+2XUupKOueTD9wf+tCHukuU7XOf+9xOcCL5DPEFg4GdYZDy0S/g4j0WAj1xkst8 vJ+BjlzjHP48clIGCzwT/Od//ufdREW/9Lfddttx5YYhkvKkDUjvfH4nnSWSn//857t8c6yfr+Pa Fr5cr0y4icDzgQ98YOEE6hERz0WWMsRlWdYybVnHjpeY+m0f3+oat32s/baOm367HOIAp9KV+Q2l KzG3/boMZFlr39iDIoYOw0cdjm3YMKCMB7D0jXK/HWdgxRAbkyl6YAkHpa7g4zwZ58d+7pUhHF5K HNmHEScwjx2EKNtCv47wAgteYGGgjinaS4JjdPcldYe/sYMQ6oeetNNZWGAeW1JH9ISDUme4qoGF 3mXGmBpYOLbaJdkGXmLDpD46YC//Cz541Wd+l2nWuR/+h/Ro0/5Sj+vUO5RXsERvmSb4cGLsHTsI AcuseSB9PWlKnGPsG1/IEC/akOPaODybnqvVE7yl7TcGJy3PxsAmGdhTEIKTzrlytz8OJ8eMIxZn lhPpz4vAOHQGPE4h55CjzPElce7KdPL0YkNOV54lk9a1nLMyreul60scuBInPDBw4GGdJXQoxyqD NGe21MUJp4sjLS8BFX+/+7u/263GSPmDQdkEDsqyhS9l6TveIrucXwGVfhRZZP69733vTv3IWwAA p0P54y94cC9QgCM6E4TgYOMvKzw+/OEPd3V/8803D5Yn5cpWvvKQbx+bfJWhXycwv+td7+ryN2lK p77T1pJ3f5t6LssqmPGFL3yhc+wT+CnLqu7UU3DKQ9nVXyn9tq9McKo7LzvS1omXC8nDX4Jz8laX JryyrfS5db1jJsOSq6F00japzwDjitFC1GdfcqxM10+zzt8My3kSI4vjObYESzgo9TmW44yssYMQ CWjDEL19PDhJXZbn1r1Ph/YwhKPUVcP4VEewkCE8OVaDl2X5T7squVr3vvKmj4SDUkeOLVuX5bW7 2U+Zo7fMI8eSpjy37n1tZRlepPE3FBxYJ6b0kXAwlHdZl0Pn13UsfWQIS47hTz2NfWc7bSF6yzLm mPoJ5vL8uveXqaPwMrbjvwwvsNTgJVjm8R1e5qWZdy51PS9NO9cY2CQDP3rrYwU07373uztn1CUc NW+4JZz3CEeRw+mRhjjznMz+XXrnOG7OJZ08L58uXWNAcrYJB24oT06itH2RnlPJaSbypMfjD6Vh 2r/Ob+XoO/ZD6cpjfU7iUJeclOnLfY6tsilLny+YU/byGtxa2YDPlDHn6c4x5c7Kj6H6wBEHOoIb OpXHtcQ2dZxVIkm/7JYzTvo8cdBxnfNlfrhQPu0CBkEFsojToXYimIUXXKa8gmGkXAlCj98wDQW3 3vGOdxzHbXjRrlJ3adPyiC567NPvmrJ+BFqIF3sRafxJU6ZTj/Q412R7GJg34dcybOgh87AwPpNu TPboWUaWMcaWyWdemmWwSLNMunl6ljm3rI5avMxrC2lHNYxy3IWb6C35zLE4NeW5de/DMY+X6Fs2 XdLvdhtehq4PLzXqCCfhJXqHMEkzD/PQNbs5Nk9H8G1DHZVlq9Wv6QwHpf7s42Uef0m31+0iHcFY g5e03XllkqZGX1qkY7e85DplLPfnlbmdawxsioE9rYRY5KBzVDlKnMa+w8TpdefY24Q5hvlDhLQJ OsTgSMAgz2kOBRyG8MS55pBy3DiXcTDnkR680q4ifQwJqCyTRxzVoe9sC0xw0PuO9zx8s3QPPcPa x403f6Ssj70uJ1TngiDJO7zQr06t+lDG8ny/jLPKlbxsy7ZXHrePS20hOvCetlGmhUnwQ9vBQR9H mTYR/ORZnuvjtVqG9AMJ0tGZdpDfOIMhq4EEJLJipdTT9nfPQCZrBkiWZa6a2yIDJzpWzXeV9Lnz OA9LDRwwL6snmFcp56ppswR53nXwLot5Xj6Lzi2rY9l0i/TNO4+XeXrSjualmZf/que0hVnGeU0s q4wBwbVqWVdJvy28aAe12sIy/MwbO2rUS4lxHpYyXQ3+GpaS8Vf2t4mXZeYkyHfTXtL2l9XxCkNt rzFQl4E9BSEWQU0gwZJ0f0OSIINzHFDfzC2P9a/JuXkOYXmNVQIeicjdbk4i58/d8GXzKPMbcz+B D056X4I1afrnV/ndDzjMupYjL/ARp3hWulWOCyYNOerySLn7QYhV8k/aPLfaDwA4j8s+n/md67PN cW05+zm32236hUdYhiSY1ZNgw6233tq1X20Yd9ruUNBkKK92bDkG3K3Rt9yNMcYsO/FLN29pbYyB RemWQ7k4VdrOrJTwMEyWNcZm5bPM8WAJB+U15bGkK8+ve3+RDnhwUsNog0V7KDkYKu8izEPXrHpM 212mrc9r46vqnJU+fUQQYh43NXgp+8gQlhwr080q1zqOK7NxKXqH8qxRR2UfGcKSY+qyVl/CRfQO 8VKrjsL/EJbyWI32uwwWvGxDHYWbmrwMtZPgyDg0lGadxxaVN3iWtdeDjX0ZG7NG/UZv2zYGdsPA qEGIAHLnOcvJcyxbzhTJew+smigDBJzgoSX6uX7RluPoM5EcW860pe4CIu4sl8vvF+VT4/yiQakG hujgkHlvgS9T5D0J8AlMOD6GZLXLNvEwRjnlaWLx2MaQlJOOoIP2i3ft11YwQrtWL+sKjAzhOJmO Ma4FwThAJm5BiWUd9bTXoWWnMSRK431MXmN00Ouv72A6Js2yZdsLVoYw/eGgn1eOh7/++XX+jg46 o7fMX93BG/7Kc+veD/9DWMpj5TiwbgzJD5a0kaH2m2NxanLdGFttkp5ZL6kLltTlGBiSZ+rI76GV GWlDaeO5bqyttmAVYvSWesJLrTpKH9kGXtIWFvFSa7xTL6mPso7Sr8t2VZ5f937awhCWHKs1J82r o/ACS406StsNByXvJZbwV55f937G9+gt83cMxt3wwo7JTTg29dgvWi5xt/3GwKoMjBqEyJ1tnU0g Yp4ICkg3yzHLtcnTSxcXiTsHOqF8OXP+BDniSAtuWCkxJHHu4hgPpVn3sZRtaCVAyhtc69bdzy8r V37yJ39yZgCpf80yv+FXviHJ8fAwlGbZY/msZ1bOlNc55rEeeuDRPmat9tBWSPIr89ntPr305fGK Wfmk/cJYvhdCXxGIELhb1F9m5d2Or4+BtNcho9zKCgaFrxzUMGwYHAwtBgy9pQSLR4e0+bGFHliG nvXFFXxw1OAF/7DEuCvLHiyM5bG/RkEvPeppyKlUb/DgZOyXdcJChzowz/XbLyz+SI15RzCEHnfx hsZt7Uias88+u8M05j/Gv3797LPPzmy/9Ncy8JU5n4Dslztc1aqjlHmoX+fYueee24c5yu+UmV59 OwE1ysKLPh1HeBQQL2easTcclLrSl8b+wkx0po7o7fMSfPp9HiPNdWNs6TCepT5KHfAZd/S1GvOA 9pI2Qm+CEjDhBZ5aYy/duIGD3r7A4/xe5mp136QxsM0M7OnFlIsKxnHSgThN/ccI/HY8Eme/P1DF CUw67zNInv20fR2cWkve41Anj6zK6F+f89kasOIY59iY27zngnPZl5Qhafrna/2e5awvq18wSl33 y6juZr0vYtm8y3TqzsSmjfXr2SM/HtHJkjWc0t8vW45ZrbNOI8Z7JggcfSn7in3tt89VAiLpM/08 2u/VGdjLZC24yWmJEVNqV0cMjAMHDuwYP+X5de8zPhnd9PbL5A4z4SDUMvj0G/2v7+DC59jBgwer 8KK8HLmhOspdI+PFXgy+ZeuSDvWk/H1e/MYNLKWBvGzeq6ajw5yqjYaH5IErbYghXCM4A0uc17TV YLHFC+5qYKFPnyV9XnACC2emllOJF/z0sQRfyV0HesR/F1xwQZf7EJbwYkysIbEH6Z013kkTx3NM TPSYB/r2Bp2OwZc2NSYOeRtfBBiHxt70rcxbY2PRX40hQ7wYY4x52jfuxhac+DPe0V2KNkTMEzXm xwSjh+YknDgOy6pzUtkPyv2yrG2/MbAtDIze631FQOe2hJ9TxdHjUPt9xx137AxMcU59flA6aXxK Mo5hBggd0lcb/L7lllu6tNJzKm1LMSkw5qx48IUDeXF0pSWLHHrOIkd0aPAs9axrX9l8+UDgA8bg dddb2aziyOMr69I5Kx936Qnd6iIc59EYvEQ4/LDDLV0/cJR0th61USc+F5o6cU0e8chnLMtrdrtv FUfaiXr3Ry+nXnsLl3iFybsXlK9so8q1TkzKot0xAODAb1nPfsOZdGm/6iDpfDqVJJjW/Wj/9szA bidsDsBFF13U3dnOiiVgGBLHjh3rnKYEjvYMckEG2ivHnqH5zDPP7KRmdOmXjCtYawjD/+KLL+4c gCeffPI4lVYi4RvWWqK/MOzycmN6YfByZHUY52psPHiBxVb7KNsdbDAam2oY5coKC12wlAKLNjz0 ouQy3Tr3OWmMc1jojlg1Imhs3ORE1BBjr7kND6XDDYsVEtpLrYCIYAdHDS+xhXCgn8OX8zV44eCq J+NLAvn0wvXUU091QS3OUw3BPyzGXfUSCRZYa/Vrjrbx7MUXX+zwBIv+/dhjj3WBgVpBCH1Iv2an lfOAscU8kHkiGMfcGluNZ+zoch7I2FuOh2PiSN7mJPOhtlpKxr9ac5Ix98ILL+zG+z4WPOEFllrz QMlF228M1GJg1McxFIIz63OMX/nKVzqHKwUzKHHuDJaEc2ri4Jhxtog0Ag4cQ5NMHOOkdZzT1k/b HZj+M9B674N0nDh/xHHO/qJHROiDh5NMZw1J2XxBIcESePHgr5aYwOiDI5+oZAB6D4GAQRmEgEla Dr76wKvrh2RWnahrn6icdd1QXouOqT9tTLBLcIuES4GHCENTO4FdW4nA5HEH59cpMHgMiC5tyx9x XLtMWwtXcKUOynSL2m+XaftXhQGOGiNYf2F0cfYZe46pJwZqLWFkcZKOHDnSObgMdY7Kc889Nzl0 6FDXzmph0Z/pPXr0aMcLLLAxuvRPjl4tYdCZR7wXSB3RzUBn8Ak45y58DTycRm0GD9oGp41DB4sg US1DWFnxQCdHSUAGNo6UOoPDOFhL9BtBWm03+jkv9uHSfhnnNQSWw4cPT+6///5OP4eB4wQLu+Wy yy6r5iBor9ooLA888EDXdmDRljkpOKvprKgHthEuYpfBwrkz3uGuhiizetCPH3rooQ4Lrh599NHO 6b3qqquq8qJP6zvmAW1HO9HHBbGuuOKKamOvPmIeyNirLgTvzEmOwVYrmEd35gH9mpgH3LRSb+rP 71piTGO/utmjfug29pojccbOrSWCUsZePg8cbE31YxyujaVWmZuexkDJwCnTiazaQ0M6vkADoyfB hxKMfQaHAWFemlwjrfzIIkcx+XLsFqVN/rZWY8QZLI/X2DdI0+1vFl81cSziLRwvU3dwJ/2qdbKb Mi/T9mpjWkXfKm19N/yc7NcIkn7kIx/pjOibbrpp8rGPfWxl49VYxHgwfjECGeIMHs5tLccp9ai9 MGwYNIZ4hjksDKzaWNzR5uAKPhBYGF+w2K8pnCS8CEbghRMDi7ultXnJ6hRY7MNi9Z562gQvnCVB ELzgIlhqOZRpB/RrK+rJnVu8MNAZ6+aK2sJRygoEvLi7zqmq6TilzPozR9IKCFhg0HZrrTwIDltz Kiycblg4tdqu+b+2CPYm6Es3LPp1zcBiyowPWPBD2G7aiyAanmqK4Acs5iT9Sv+xKq/WyryyrOak zI+wGFfUESy1eTGuGF/0bbr96UewGG9qivnR2OsPL8b+jL27wXLjjTd2ARVl+NSnPrWRuq7JX9O1 fxmY9rtTqgYh9iNVBnB3/t0pr3lHaD9y1TA3BvYrA+sIQuzXsjfcjYHGQGOgMdAYaAzsfwZaEGL/ 1+HJUgIxiLohv33IrAi25fGWxCeyvQ+L0SA3BhoDjYHGQGOgMdAYaAw0BhoDjYHGQGNg4wy0IMQS VeA5fasgBCKaNAYaA42BxkBjoDHQGGgMNAYaA42BxkBjoDGwOwbqvD1od9i26iovOPSOhiaNgcZA Y6Ax0BhoDDQGGgONgcZAY6Ax0BhoDOyOgbYSYgXear41dwVYLWljoDHQGGgMNAYaA42BxkBjoDHQ GOgYmD5y35hoDGw1Ay0IsdXV08A1BhoDjYHGQGOgMdAYaAw0BhoDjYH5DJSBh3J//lXtbGNgMwzs OQjhEQXf222PKmymAqPVSzN9Amk/CbybxB398zjzHpCvfe1r85K0cycAAybrTNg+i5X9E6BorQiN gcZAY6Ax0BhoDJwEDJS2y24+8XkSUNSKuEUM7DoIwYH7zGc+M/nkJz85+dznPje57bbbtqhYJxcU jvzNN988+fKXv7yvCv7www93uG03IfjC27yvnvgyyr333tsCEZuooIo6//Iv/3LiW/Pagu92lxN5 RRhNVWOgMdAYaAw0BhoDjYFdMRA7hi3z6le/eld5tIsaA7UY2PWLKb/xjW9MvvWtb03e8pa3TC6/ /PJaePesx4qNv/iLv+i+drHnzNaUwTZiWlPR9n023gPy1re+tQvw+Fyrr6Q0OfEYcMfg9NNPn/zg Bz/oAhB/9Vd/1QIRJ141txI1BhoDjYHGQGPghGXgtNNO62wZBfz+97+/s3/CFrgVbF8zsOuVEB7B IBw0jtp+eWmjFRtWcGyTbCOmbeJn01ishhB8uOOOOzYNpekfkQGrH7J8URCiSWOgMdAYaAw0BhoD jYH9wgAbJn/NjtkvtXby4tx1ECKUteU+YaJtT2QG3vjGN3Yrf6z+aXLiMdAm6xOvTluJGgONgcZA Y6AxcDIxUNoy5f7JxEEr6/5hYOXHMThh7gg//fTTXSmzquCd73zn5Mwzz5x4kV8ed7jnnnu6Z6w/ 9KEPdedc4Pn6HBfAuPTSS7vVFK6NRIfHPDzXJL13UFht4a70m970psF83v72t898BkoejzzyyHG4 6X/Pe96zg9l+Kd518ZrXvGaibKUo83nnnTehL+LxlAcffLBzVOV7/vnnd9eV5UrabOdhSponnnii ex8BvnHgjjxecLBIcG3FSj+94/B6DGSoDtTfrbfeOvH4Ac6lVydJ2+dZHUnj3Q72lRnOfrpZeJXL yx/LerbChuNfinTSwJ7n3Ybaj2uc1xbDW9l2yjz7+67LuzXe9a537bQnHDqujtsjGX3WTozf65qw PdJBTj311I0+0qE83nXh71WvetVGseDDuzYaFkwcLzjBzTbUUbBou1kVdDzaer+0X30JDng2KSUW eDb5zhhYtBeizWxazMv42IbxLrxsGos62ZZ5IFi0m20YY7ZtHsBLVg5ssi9t09i7SR6a7pOLgZVn ME455/T555/vnGL7xHHCWTUpcX45fmedddbOOU4cZ5UTx4nm8Enn733ve9/OIx2Oy4eTLA/vnUha jqX0nHN5cIw5hhxT+2VgoAP08j+OcR93MLsujm2CBt/5znc6PS4vnWl6YSsdZA60P+XiPKf8Xnqo XLOc1nmY6E2+OLziiit2yooDOgRkZkmuDddJVx6HFa/Krkwf+MAHOh3y9psDz7jAM+7CM9yujXz6 05/u8lFPuEye6u+mm25KsplbmJQx9YxjAaCSO3nSAxs8Z5xxxkQdwd7XI+0tt9zSYQ9v2gzenCux l6BKHXQrS0Rb0Rbl06QxMMSAF0I9+uijkxdffHHHKNf/tJvaTotnQWF57rnnOt3068MXXHBBdeeS 0fnQQw91vOANFkHaCy+8sDovAqyPP/745JlnnumqkPEJi3qyX1Pw8thjj3XjGEMYL4LbeKntXMJy 7Nix7g8Of8Z581xtLIxxY7v2ixfiOefLLrus29asI/q1FW3G3IMX88Ill1zSzVk1seDFHGUOKoMz sJx99tk1oXT1wgYMFrzoP24K1Mai4LCoo4y9sOhHtcde7cXYe+TIke5mXPqSOtK3aws+9CPtJuMb Xg4cOFB97NV/8PK9731vhwZzEiy1g4ywuCn57LPPdryoJ3MjPPZrCixs3ieffHKnjrTbiy66qDov NcvddDUGMLByEELn8KfTzHLoGHoc99JJ5iwKQHAgy5UF8vCFDe9F4PiVwgAq09LJETXxlasr6Pmj P/qjzkmeFYQwOfobwu04bPINZk54xB1+uEmOG6wIB5gTDWu5kkI+ymVFAaylQ9tdOP03DxNe4pxz 5HO9fP/kT/6kOwdTjidPW8EC1zKuS0zK7ni/DtSnAJFVBqWDbjCmO4GZ8IyrpMvqB79zLFgS3JgV hEk6XJY4lV0Ax4qbXKtuHNceUhe5vq8nQZqyjcBmBYvyJ1iS623lnWAKLDjpi2PqW9pw0k/Tfp+c DDCs7rvvvs6AufLKK7t+yanLmJd2XIMdDoq+rP8ePny4ezGVgKJxgYFs3KklDCy80KvfCh5y6mDB mRVGtYSj/cADD3T9V3BS4FOQBj5YDh06VM3oCxb61dG5557bGcSCNY5dc801tWjpHMqjR492Tpwx /pxzzumcua9//etdveGqZiCCMwnPxRdf3DmS2pDfd9555+S6666rGojgGGgfHALtV71xpPSva6+9 tmvPtSpKYEadcJZg4fCaf+++++6uvdR0/jn99HKsMx9zds3Z5lp9q5ZwtO+6664u+GHs1VY5mOrt qquu6uygWliMI8Z8GHIDRBuCz++a84AxFxbBB+OsFy+zXzMe61+1HG59GBZzEyzsJ3MS203w3hhY S2AxD2jDV199ddeHBSP0rWCpxYs+fP/99+/ozTyQ+VHfSvCoFj9NT2OgJgOj3PoxwDBmStGpSN9R ldagZHCMg5/r4ujndyLJBvLSCXSXWhBA8IODuKrIVx4czIjB0QBADwc4wpE3wcZJNbCSfnnDAUwG 21UlOq+//vrjAg1wOiZfGPvius9//vMdxv7d/KTvY/W7X375KmOfZ8fojhjQSXnMb/X8/ve/fykD QN2VQqc61h5SnzB++MMf3jF4kj5tJOlsXSfPErv0Jhz59usDdisvXOsRjD6e6Ep+/bLmfNuevAxw BvQFYwZHW3CQ43Lw4MHOgUr7HJuhOCYMLEYvhxIW7V5f4UA5V0v0xaeeeqrjhYEFCyfK3ThOSznm jo2Jc0ufOjKOwSJQKyjDaeEs1BIOZcYpdwI5LcGCL+2pljDAceNuLQywcGi1n9pYOCnmKfoFhcxL xl1jN6eK86CN1xD9WX/RbrURWDhyeHHnlvMAbw0R/NAmtFmOtpUhr33ta7u2rL5qYlFedQRLVqfA o77UGye31hyZoBCnES/qBy5YjH0cy1pY8ML+0k5hKecBfVyAkZNbQ/QRY1oCvepFOxF4yDxQC4vy wmLs1Y/1J1jgMC9p11mZVoMbYx0b0DygjcBi3DP+WZlmXK4lsJh32JzlPKD9wlJzTqpV5qanMVAy MEoQolSQfYZvDIocyzYO/SJD3fVjiHxhiOMPhwHToGSgEniIs22AiuMLSzAbTPuSoInrV5UY5+Gm vD66kibnYBGAIO9+97u7yTjnbJNe4MRqgfKvPF9es2jf4CkoI8+sNMARwxFPDILdSDguHSZ14A4U 3HRFX5l/6mOIN4Zk+YhHrrMKBzfKssyd2ZpGTTC27fYywBBmLPTHN4YxI0d74czVEAawtswY5xiU wtBinFqhUUMsH6crPJQ6GcXu8NTihW6Gp/GIc1CKsQIvsNRwcPECyxAvsGlH6lC7qiGwKLe2Woox nLOtDmvwQnewMMhLgcM8I2CijdcQujhxcVSiUxsKFudriNUxnEZ9uBS8aL9wwFtDrDyAR/soxxhY cGUOFmSrIbBwYPt1ZGxhf7EZYveMjUcdZOwVICpFHdXEIjimL6kjc0HEmAOLsaXW2Kvc2oO20h97 tWeYao29xjHjmfZhDipFcASWWu0FFnVkPFFPpRhfMj/WGntL/W2/MVCLgWpBCAPRIoc0DmStwpd6 OKCcBQNQIqHu1nH4HTdYJJhQOqqlk1zmZ7+coPvnFv2Oozsv8JI0yQtugyjx2EFfkl66/l8Mq/41 i37D98EPfnDnsRl6BQc8TpGgzqI8ljmvbXjPg4CBfUEKfybUUubVR5mu3MeFcqjfWhNQqb/t728G BCAYffpQ+l9KpF0xJowfNUT756zAQm8pGY84EDXu4MLCSRjixd0nTkstXox9nBWc0FtKjHS81XBw OSuM8nBQYoHNcWNcjSCEeVn7xUt/rkm94aTGXVPBmdz562PBkfYL727G+JLjZffNCxyA9JvyOvic q+X4Z4zpO7cwwac/17qb7O6tsqfflLzA4lyteZQefXuojnJMGpjGFo40LENtN+NxrfHOGJ+xtz/e pQ0Zg2qMMcYOf0O8GOv8abu1sBhbjXf0lgKf+dv4UmN+zJykbfTryG/HYa0V6Cy5aPuNgVoMHN8L R9SqQzEghiTHs3JgKM3YxwQhvBeBI8p5FnyAmaNrcLKczG+SlQj2OcEGCmXIecfJXgzaoQH7paCk Pw4AAEAASURBVFwnOzyKlpbitzv9AgGWS1qdUD56IU84vSth3WLFiD/5W17n2dD+yyVX0WmALkWZ 8Ox9DeoqopxlsKN/pyjp5m09OiJP7/CAOS/onHdNO9cYCANxivqGhPM5ZiwYGiOSx7q2DBYGN2Oq HxDxm/Eljf7VN8LWhSH50MOxjDOQ47ZwwCMNbobSlOn3um/smMWLvNWT+oEnRvpedc66noEbZ6Vf R3jxJw0s8+aBWfmvcpwOuuhMWy2vd5xwIvp368p069jnhHCcyDws+ptHesaU1NEiLDWCMzBwKknq o/vx8r9wVQOLOtJmSPS+DKPbBF/wlufG2M/Yqx/1+1LGN3j17Rp9yRgzxEvw1QpaGe+I+pjFi3HX X391wrrriQ7tZhYvMGq70qTO1o0h+aWPpD5y3BY+x/V941B/pUSZdh374UX502+SLxyOpe0OBfyS tm0bA/uZgeNvlY1Ykqw0cHehLzm2Gweyn9dufzOw/HFq4SlXO9h3XICidIDpyp34oQi3wAXJowXd jyX/5Rp6+xJd/aCNa5Thhhtu6JaLCqqUdyScZ/yG736+u/mdRztM8kQgRjDCeyvIbpdl5rq0iUyq ff6jt1M2/ZcJNatWctxWua3S6JcfXvXopab03H777eVlO/vhMph2TrSdk5YBRmfaYN/YQ0qOSce4 GVsSvIvevj7HGXv+xhbljePf1wVHMIa/fpp1/i55id4yf8cETGrworx0DeGAKcdr8EKHOir1dj9e /hfjuEbbTV9K+UscJb4aWOhQRyQcdD9e/heM2kv4K8+ve79sv/28g69Ge8FJ+A8HJZ4ck64mL+Fg CIs6Sl2W59e9H/7DQZm/Y/7gqDXG0D8Li3O1sMxru3AQWLaljrTbtPGX0I3zXxBinqi7vWIZqv95 Otu5xkBtBqoFITh6xPP8cSj9dofbnWznx77TQl9WK2TCcCzCwY2jWa528C4BAynnte8E56WO/XJ5 EacAgtUJZV7RVW6HMNHjWk5+yRd8ggvOhdMyL/si/t4JYevOfsqa9H2s8vf4BF2rCl6Us++4h8cE Beblm1UOSeM3ruENN2kbJUaBBu+IKEWZXef6Mq0ypty4GxJ1iXfXldcmrTxcG0w53rYnNwMxtocm /BzbqzGxLMMx5KJ36LpajlOwDGEo8dUw+JYx/NXRMumGyrPKsW3jZZn2W6OOSv7L9hFucyzzWY6P sVVH8+opOmv1pXntMrzMSxO8e92qo/ASvUN5lnU5dH5dx+aVOfiWrcu9YpqHpcx72XTlNavu75c6 Uq7UU61+vYhLbbfGeJc6moUnvOwFS/KYpaMdbwxsmoFqj2PkTjMHmsPLMbeUjmPnznL/qxljEQOH oMcnPvGJzkkvH00QbOB8whanFw4rCDi3HO5+QMFxKw+8qyDlko4TzmHtf3Z0qFxDmOQrkOATn0P5 OjdP5IlTfHtZZT476ROXnPGhPPsBlnn55xzHHZ94s1VmKzVMKPLD6SKBFR5b3MXZL9uEfUEH5RHw wE/ajq2/iPpIWkEKabU1A7Lyl3Wba7J1Pp9AhT9p5a9OE8hJ+rY9MRjQNjJhu6uW/VVKx3jpy9Cx fpp1/t4N7nXqXzavkpdyf9nrV023DC81cCzCXWIo9xddt9vzeAk3Q/qGju1W1zLXwTJL56zjy+S7 appwssx1NXDNwxP92S6D+URJM4+XlLEWL8tggakGnmWw1MChvMtg2SZeYNkmWbWeSr7L/W0qU8PS GAgDuw5CXD59RCGPDCQz27e85S3lz+P2yzvNnEIOvTz6jl0caNu+cEbjGJbnfMnCcQ7nPMn1HOZ+ WlicHyrXO97xjpnPxcKvLBxx5YIDPz7ltcxd81mYlF8Qw119eGflS4c8+nzh27k46a5fBmvyk74v /XrH4U033dSthvAIhfILPNAzxGOZX+pZWg5/rvdbOyq5gwUXOBaoINJ5JMWxsuwweVlmHqGRXv79 PIfajGsFePCd1Rh02SfK3+TEY8BEn+fj3aladvKWbuhZ1zAUA2JRuqTf6zbP1EbvUH67DbIM5TXv WLAMpSnxlf18KO06ji2jQz3Oq8t14JDHMlikq4FFHS3T1ufVJazrEDjoyfg+K88aWHCfJf5lW+1j kibp+ufW+VubWXRHtAYvypp2OY+X2uPdENfBB+8ybXwoj1WOhf/onXXtsv1/1vXLHE8dDaUNvrKN D6Vb17FFvATPtvBSq+32/Y8+37vlxWMeebdODU77uNvvxsAqDOw6CMEBHBKO7zzhTLpTPU84lKVT WablbA8Jx3eZu+6uhX0W/nn5D+nNsWXKlbRD21mY5DvrXPIxmM3CPVTORVhXzQ+OIT3BN2tb1vMy 18/CPa/s8/Kd1WZKXMHuERGBjP5KmJxv2/3PAGOJAeKPAbCs4ZqJfmh5ZQwJeSXdmEwtMvhg1L/n Ganrwhc+h3gpddTgJTpSH6V++zBysMJf//w6f+PeH519PH7n2CIjdR2Ywou8orfMN3VXpivPr3Nf H8E/I3oIS47V4CV1pHzRW5Y1x9LGy3Nj7CszxyJ6Sx0166gMQkTvEBb8LTt+ltevup92uQhLjUBR xo6hOnLMHxw1sMzjJfjgCOZVeV8lffpr9A5dC4s2M7aElyEsjvmrNVfPw4KHYFm1jvCYvAV0l3kk emzeW/6NgVkMVHsnxCwA7XhjYNsZEIDweIkXVzY5MRmIA5QJf8hImVXyfE1hniHMKEjes/JZx/F8 ZQKWfhlyDJZaBh9eh56BhsUfHDFS11H+WXnk7eI4obeUHFM/Md7K8+veV2a6+jjoCRa8pS7Xrb/M D/fw0Nuvp2CRvkYdcURS5qG7/sGX/laWY937ZbuchyV4162/n1/KHA7K8zlWu46it8SSNp3+Vp4b Yz966NVeSwm+WmMv/ueNd/AFb4lzjP20yyFe0p7LNj4GhuS5iBf1BG+N4Ex4Kce24ITD8XIcyrkx tuYAutRR+k30wAEP7la1G6TPn3yaNAa2mYEWhNjm2mnYNs6AF4wKQlhtYSVGkxOTgb1M1gcOHOiM CYGqfj7uRDAmpGGgji3aKIeFzr5hk7dxn3vuuVWcSlgYUUO8MIRh9KnFGrzA4X0183iBt4aTQIfP v4WDsk3Ah69wV54bY5+Dds4553Rtpf8YBHzac7gbQ3+ZJ6coKyDTVsvz8NXCQm8+AzqERR0RfamG eG+WfpLPY5Y68VJyV54bY//gwYNdtv324mCOGe9qiPai7PT2x95whbsaDi4snL9wUJY/Y2DaVHlu jH3ji749NMakPev3uBtb3Inn/KfPlPqMd/48Vruqs13ms+x+xlVzI25KSR1JUyMYnbEMjv5cHXzB W+JctF/2g3J/0XXtfGNgEwy0IMQmWG869w0DDAorIBY9ZrRvCtSAzmRgtxM2Y4IzwrjLd8gpkd9z zz3XGXr9z+nOBLHHE4w9WBjgpTEMixercmRqOf6MSljo9q6YUvxmfNb83C0ngaHZx+JdNHiBtUZA hA7tIe2j5EX7gRHWGo4T3ZzKWVgYwzWx4AU/2ipMETi820gAJ6sCcm6sLUdO39aHtdWIfg4LB6HW UmcBNLqeffbZH+HFsYxBwTjmVrnVAV7i0NKHI8ecw10NwQlnWp8unVzt5Zlnnun6UAI4Y+OBxZ9x NwGQ6DTGcGz1pRpi7NWXPMJTYil5qTkPKDde8q6CcJD27Hytsdd4h4f+PBAstdqL8V0dCUL0sfgN o/M1gjOpj7ZtDNRmoAUhajPe9O0rBgQfWgBiX1XZRsB6YSmn6ciRIzt3NRieXmh60UUXVTPKGTaX XXZZZ3xbxZNABIP82LFjXQCC0V5LDh061K1AKLFwtB977LEuAFEzCHHxxRd3TsLRo0d3nCdY/Gbs 1bpLiXu6BD0efvjhHcOcgf7oo492GL1/ppZwAPw98sgjO1g4dA899FDn9GtPtYQTp544BNorYaRr PwSWGndv6RLQg0VARD/WvzkG6khd6fO1HASOvXGEc+KlybD403604cOHD1e5e4sXY4x3LdGLi2DR p3EFS40VRbDgX5vgaIcXx9WXMQ+WLMF3fEzhROfF1fpOAlfmAW3Z+6QEk2qIPmLs1V5hyV1/Xy57 /PHHOyy1AkXKazzThh944IGdYJGAlTHHHFBzHvBCcu3TuJ/AlfYDi3lg0QvV11l/eDHm0a0/EQFO /dr8UBPLOsvV8moMLMvAKdMJ5JXbDcte1dI1BhoDjYETiIGvfOUrk4985COd0eZrLx/72MdWvhPN eGDwuTsYR8ldF85Djbs8ZXUwaDhusDDUGcQcXwZYrTvswcPQu//++zsjCxaGseXaHLxaTlyJBS+c ufDC8EzgJulqbONcc7izdJrhyXmosRy4LKM6YZRzUrRd7YWTwomrjYVJwonk4MICm7rytSkGe23h XHPcIvjQjwRuasuTTz7ZOSw4Maboy7DUevyhLC8s6khbgcPfhRde2K2sKdPV2NduBSH0qYxvnH5j b20xtghGG/fUkbqCg0MZbLUwGVs4tJxsfYl+Dn/NIGfKCoMvlgngZbzLnJT5MmnH3qobARFzNiza sD6El9p15EaBOspqGW3YPGBO2s38eOONN3bjOA4/9alPVQ3wjF1vLf8Ti4Hp+HhKC0KcWHXaStMY aAzsgoF1BCGoZUBw/Bme7sAxcBiimxCGFcOPU8eYgWcbsDA43RXbJJbU0TbwwghVV3hRR7WN4LRN 7YSDoO3CoI42iUUdxanES21HpeRFHcWphGU3zkHy2+s2vMjHYxjGmE0JLHgheNkkltSRdpw62tQY o90aewlO4NmUwIIb/XrT412JZdO8GHO1X1u8bMuctNd5oAUhNtXTmt5VGZiOz6fs+hOdqypr6RsD jYHGwInOAGNmkw5KyS9jxvPb2yDbhmUTd9SH6gEvtZauD+kvj3HYtoUXWDgF2yCwcCI36UiWPDQs JRuv7AvI+NsGMQfUevRiUXm3aU7aJizG3m0Z77YJy6L21M43BtbJQHsnxDrZbHk1BhoDjYHGQGOg MdAYaAw0BhoDjYHGQGOgMTCTgRaEmElNO9EYaAw0BhoDjYHGQGOgMdAYaAw0BhoDjYHGwDoZaEGI dbLZ8moMNAYaA42BxkBjoDHQGGgMNAYaA42BxkBjYCYDLQgxk5p2ojHQGGgMNAYaA42BxkBjoDHQ GGgM7C8GNvVi4f3FUkO7SQZaEGKT7DfdjYHGwFYw4OVz/ki5vxXgGojGQGOgMdAYaAw0BhoDCxiI HSNZub/gsna6MbARBtrXMTZCe1PaGGgMbBMDPuvmM2Y+IWa/Td7bVDsNS2OgMdAYaAw0BhoDixjw yVyfHiWb/GTuIpztfGMAA20lRGsHjYHGQGNgyoDggz/fU7dt0hhoDDQGGgONgcZAY2C/MPDDH/5w x5ZxU6VJY2CbGWgrIba5dhq2xkBjoAoDVj6cdtppE98x9xxlWw1RhfampDHQGGgMNAYaA42BNTHw 2te+duKPCEg0aQxsMwNtJcQ2107D1hhoDFRhoK18qEJzU9IYaAw0BhoDjYHGwEgMlLZMuT+SupZt Y2BPDLSVEHuir13cGGgMnCgMrGvC/u53v9utpDj99NMnp5566sboUZ7vfe973XsuzjzzzG6Fx8bA TBW/8MIL3aMu7tJYcbJJwYtnZ9XRprF4ftf7SLYBC15gUUevec1rNllFXbt98cUXOxywbPI9LZY1 w+IZayumNonF3U1YrNjSZjaJxaNrxjvj3KaxGO9gwQcsm/wyACzGO6IvbcM8oN2cccYZG+clcxIs m+RF3ehH+vamx17txbi7LfMAbpo0BmowsFlLsEYJm47GQGOgMVCBgSeffHJy5MiRzmliXDGIDx48 OLnkkkuqvyDqmWeemTzwwAOdwcmBYxC/7nWvm1x66aXVncvnn39+ct9993XOwTZgeeihh7oXdzE8 GaECNIcPH95ZwlqhqXQq6L7//vs7I5hjyzjnsMDCQK8psBw9enTy3HPPdXzAIjhzxRVXVMfCKfjm N785OXbsWIeFYe4YLOedd15Vp5uTjZennnqqqxvOgqCRfvT617++ZhV1HHzrW9/quNFm+1hqBiPo fvzxxyePPvpoF5RBhDHGWOevtsCizeg/eNB+jb1vfOMbqzrdeHn66acnDz74YDfmC1qZBy688MLJ xRdfXN3phkX7JcZeWM4///zJoUOHqs9Jxpavf/3r3biSeQAW7cXvmgLLN77xje5xBW1G2z333HM7 XozFNcXYq71os5mTYLr88su73zWxNF2NgdoMtCBEbcabvsZAY+CEY+Db3/52Z0hwTDgoDGEG3913 393d/edY1rozR++dd945ueiiizosAiIMnLvuuqvbXnvttdUcOcbePffc0wVAcACLFQh4cRyWWisR GJr33nvv5Oyzz5685S1v2cECB+P4zW9+845DNXYD5cjigMGLA0Y4XgRr1N1P/MRPVDPM6VV+Dvfb 3va2jhf7DGN8XXXVVZOzzjprbEp28odF4AovcbY5mfi65pprura0k3jkHY6KYIi2IQBCYHloGsjS xy+44IJqfemRRx7pHO1wwOEVHBFstM/RrSXhQGAoHHznO9/p+jRejD22NUTwFwecNuOvMcZ4rC8R DnctoVdgUcCBc40DY+9Xv/rV7i43jLXmAfWh/wZLAuP6F76uvPLKamPvs88+2/VfbUV9GPONgcY6 c8T1119fjRfzIw6MadovXqwSueOOO7rjV199dbV5wNh7++23T84555yOA0ErAVdtyLykrwtINGkM nKgMtHdCnKg128rVGGgMVGOAUc645AjYMj45UIwudy8ZWjWEM/LYY4/tYGHsweIOC2zujDEIawkH gRHOKQkWDre7lDiBp5bA4q76G97whuOwwGbliPO1hD7GMN1Z3o8Xv2F0h7mWcPhh4cCljhjm2ou7 dDWxcEw41lbtcBK03fQr2B5++OFqL1vjIASLu6Rw+MOTlSo1sWgT+ooAmrvH4cX4YpyxCgB3tcQY Q++BAwc6TuDhSKk3wRJ9voZw2Iyv6sOYkvYLB2x4qYVFeYfmAWMvbE888UTVeUDwLGNteNGntBnt WlC2lqgjdSVAz9HWXjjXxmJjoeBNLcFLf07CDyzmRudrCV6MM9qHusILfoy9xuWac1KtMjc9jYGS gRaEKNlo+42BxkBjYEUGOGqMBc5kuZQzhrnAQC1jgmHJwGTgMWpKKe/kwjS2cJwY3hxaDkspjGGG X22Dj4HHKSiFw0I4eTV4cddNuTm0fV5g04bUIaO9hggypK2W+jh26kgdWhkxtuCeUU44/aXgSpth mOtvNUQdacOca204ghPcCNzAU0O0TfrSh6NTvTkmAFHLkQsW9VHyYl+QhINXCwv+3fGnV71E0p71 Ne23hnBgOdTahnGmFO1ZH8JdDVEH2m/Gk1Kn9gJL+lp5box9bVMdDc1JeNFuzI81xhg6BNDSh8vy 6ufGmVp1ZHxXR8Z7baaUjL14qzUPlPrbfmOgFgMtCFGL6aanMdAYOCEZ4DCS0ghOQRlYDGLGRA0j i1Huzkr0BoetY4QTx7kaWwREYBnihbFXkxdlZgzTSXcpwcdwr+Hg0qGe6IWnlOCDBXdjC05goTc8 RGd4EhyQZmzhMHLiSN+Jcww+aWqsKlLmOCNDWNKXamBRdlhgGsKCF2NLDSwwZLybhSV4bceWjKtD fSlBWG1Xuxlb4jCmbZT6HNOfagVntAVljt4SS+oNLzXGmIy9Q7zgxB+8NZxtZabHeNfHow3BYuw1 Lo4tuBdYDAelvuAzh9aYq0vdbb8xUJOB462xmpqbrsZAY6AxcAIwkGWtjIm+xNBh+NQw+BhQHAVY GDKl+O04A6uGwUcPLOGgxBJ8nKcaRhZDOLyUOLIPI05qGJ9pC+EgGGwdgwUvNbBoL3SlbZRY7Kfu agRn1A89i7DAPLakjkoOSp3qidTAQs8yY0wNLBzbtMu0DfgitXnhxJHoDQ7b4IO3xngX/qO3xKJN +0s9lufG2A+W6C11BB9Oaoy9sMyaB9LXk6bEOcZ+xrEhXrQhx7Xx8DcGhuSZMkdvjtvCoZ7UTw27 odTd9hsDNRn4Uau5pvamqzHQGGgM7GMGGFe5y8Zw6EuOlen6adb5e5GxHSOL4zm2BEs4KPU5luM1 jKzS2I7ePh6cpC7Lc+vep0N7GMJR6qrBizqChQzhybEavCzLf9pVydW695U3fSQclDpyLHVZnhtj P2WO3lJHjiVNeW7d+9rKMrxIk3TrxlDmlz4SDspz2S/rMsfG2KaPDGHJMfzVqKfoiN6yvDmmfoK5 PL/u/WXqaJt4gaUGL6mjeXzX4mUehnauMTAmAy0IMSa7Le/GQGPgpGEgxt1QgWsZNvSQeVgYn0k3 hHVdx5Z1QpYxxvaKaRks0iyTrgYWOmrxMq8tpB3VMMqVOfxHr2ORHItTk+NjbOGYx0t0Lpsu6Xe7 DS9D14eXGnWEk/ASvUOYpJmHeeia3RybpyP4tqGOyrLV6td0hoNSf/bxMo+/pNvrdpGOYKzBS9ru vDJJU6MvLdKxDl6GVgjNK3s71xiozUALQtRmvOlrDDQGtpaBGCkxAFYBmmtnXbObPGflNet4jI55 WGrggG9ZPcE8q0zrOJ4lyPPygndZzPPyWXRuWR3Lplukb955vMzTk3Y0L828/Fc9N68t1MQCx7Jl Dq5Vy7pK+m3hBSfL8rJK+Xabdhledpv3qtfNw1LmVYO/hqVk/JX9beJlmTkJ8t20F2OSv91c+wpb ba8xMD4Dr7xOeHxdTUNjoDHQGNhKBjyf6XNu7k6s8glLk7wXWs2SOCiL0s26ftXjeenYrOvgYfws a4zNymeZ48ESDsprymNJV55f9/4iHfDgZFnDcC/4YNEeSg6G8luEeeiaVY8NvdBvKI95bXwo/W6O pY/og/O4qcFL2UeGsORYmW43ZV72GmW2AiR6h66rUUdlHxnCkmPqslZfwkX0DvFSq47C/xCW8liN 9rsMFrxsQx2Fm5q8DLWT4Mg4NJRmnccWlTd48oLVZXUfOXKkexG29ItWWyybZ0vXGBiLgdnW81ga W76NgcZAY2DLGPCZLN9PtyTUZ80YAIyRZSTGxNCy0xgSpfG+TJ67TROjkt6hMjhW0yjHYTjolynH w1///Dp/Rwed0Vvmr+4Y7uGvPLfu/fA/hKU8tqrxuRucsKSdD7XfHItTsxsdy16jj9CTF7b1rwuW 1GX//Dp/p47kOWTIpw3BG/7Wqb+fl7bgpYbRW54PL7XqKH1kG3hJW1jEi7Y1toT/1EepL/26bFfl +XXvz8MSfLXmpHl1FF5gqVFHabvhoOS9xBL+yvPr3s/4Hr1l/o7BuBteDh48uDOP1RibStxtvzGw KgPjj8yrImrpGwONgcZAZQZM9oIPQ994XwTlrLPO6pIMGeWCGgwKedcwbOBnaDFg6C0lWARcYgCV 59e9Tw8sQ8/64go+OGrwgn9YYtyVZQ0WxrJ0Yws96in1UepTb/Dg5LWvfW15apR9OtQBXvrtFxZ/ 5MwzzxxFf5kpgzl6ht77gC9pzj777PKyUfaNB+nXs9ovxeqxhqTMQ1jCVbgbEw/+U+YhLDl27rnn jgljJ++Umd7+eBde9Ok4wjsXjrCTsTcclCrSl1KP5bkx9lNH9PZ5CT793hg9ttBhPEt9lPrgM+7o azXmAe0ljnl/vMMLPOaJGmMvPbiBg96+wOP8qnO1Mqr/tIF+vu13Y2CbGGhBiG2qjYalMdAY2AgD fUNtFRCve93rujsWMWLKa32VgYFx4MCBHeOnPL/ufYYHo5vefpny2TEOQi2Dj/HP+OwbfPA55q5N jMJ1c1Hmp7wcgKE6yicHGcKrGnyljmX36VBPyt/nxW/cwMJIHVvoOO+887o2Gh6iE1faEEO4RnAG ljivaavBYosX3NXAQp8+S/q84AQW7facc87p0oz9Dy/46WMJvpK7sbFYMRa9fV3hxZhYQ7RdbYLe WeOdNDXGGHoEr4acbcfgS5samxvjC0d6aOxN38q8NTYW/dUYMsSLMcaYp33jbmzBiT9zMt2laEPE PFFjfiyD0f0gBE7gg2XVOansB+V+Wda23xjYFgbG7/XbUtKGozHQGGgMzGFgtxM2B+Ciiy7qlkt/ 5zvf2dHAkDh27FjnNJ1//vk7x8fcYbBw7BmazzzzzI4qRs4TTzzRGVew1hCG/8UXX9w5AE8++eRx Kp9++unOKIe1lrz+9a/vDDu6I+r88ccf7xy8OFc5N9YWL7DYah9lu4ON8fmGN7yhilGujLBwAGAp BRZt+JJLLikPj7rPSWOcw0J3xKMIzz///ITjVOMuJb0CQe4q4qF0/mHx3hjtpVZARLCDo4aXOEsw 6ufw5bxjYwsHVz0ZX9RJBK6nnnqqC2rVuuOPf1iMu+olEiyw1urXHG3j2YsvvrjzTD48+vdjjz3W BR5rBSH0If36u9/97nHzgLHFPJB5InyNuTU/Gs8EIcp5IGNvOR6OiSN5m5PMh9pqKRn/as1JxtwL L7ywG+/7WPCEF1hqBGdKHtp+Y6AmA+2dEDXZbroaA42BE5IBjhoj+MEHH+wcWndSGHuOvelNb+ru BNUqOCOLk+QFVQw9hjpH5bnnnpscOnRo5Tsre8HNEKb36NGjHS+wwMbouvTSS3eWvO9Fx7LXMug4 K4888kiHhZP5/7N3ZrGyXFf5r0s83Xmep3MH2xlslIBIFEFE4A9BYQiDIobwQCIBFihCAeURhPKU B4SCBDxACEKgRPAAQYoEyBGIRCEEYeLg6frO8zxPvrbjxP/72+Q73qdu9XiqVlWf/rZ0TndXV9f6 6rfW3rXWrupqCikSvm3bts2ehR92e/NZj6KRmIEDxQtFGwUdWpgkikqE2Qc4YJNCiQkZtFFI4TN0 UEBENfrNrl27UuzKPsULz9FF/JKcRzS0zMzMFIcOHUr2KRjoT2ihyNu5c2dYgUAhR4yi5fDhwyl2 0EIsU6TALLJYwQ/79+9PLOjHNLRQ3DHeRZxJxib7jB/ox8eOHUtjCqzOnDmTit6HH344lAt9mr7D cYDYIU7o40xi7dmzJ2zspY9wHNDYCysm7zgmsQxtUZN52NZxgGMSjePA1atXk9/wH6+jGmMakzPc iBr/YJuxl2MkzKKu4mF/mZRi7D1//nzSwaQn/mEcjtYSxd92TCAnsOjugWzuF4fzd/3cBEzABKaA wNNPP1088cQTKYl+//vfX3ziE58YOXnl7BvJAwkNSSCJOAkPZzCjCie5isKNxIaEhiGexBwtJFjR WjijTYGrXx1BC8kXWnge2SiS4MJkBFwoYtDC2dJoLro6BS08RwuXdOOnNrhQLDEJAhdYSEtUQak4 wD6xgp84cwsXCgWS9VEvTdY25/NIgasrEODC2XWKqsjCSfrpzxSSXAGBFjQQu1FXHkgHjxRyaKHo RgtFLbHLpFZ0Y7JXk77YRgv9mrE3usEDLfCh6aoEJtHgFNmY/EALxyT6Ff2Hq/KirszL95Vjko6P aGFcwUdoiebCuML4Qt/GNn/0I7Qw3kQ2jo+MvfzBhbFfY+84Wt73vvelCRX24cknn2zF15H8bGty Cdztd4s8CTG5/rNyEzCBmgjUMQlRkxRvxgRMwARMwARMwARGJuBJiJGR+QMtEWAOInbKr6UdtVkT MAETMAETMAETMAETMAETMAETMIH2CXgSon0fWIEJmIAJmIAJmIAJmIAJmIAJmIAJTAUBT0JMhZu9 kyZgAiZgAiZgAiZgAiZgAiZgAibQPgFPQrTvAyswARMwARMwARMwARMwARMwgVoI3P3KfS3b8UZM oCkCnoRoiqy3awImYAImYAImYAImYAImYAIBBPKJh/x5gGmbMIGRCXgSYmRk/oAJmMBCI8DBWgds fhZLzxfafnp/TMAETMAETMAEFiaBPHcZ5yc+FyYV71VXCdzXVWHWZQImYAJRBL797W8X/NY8vx/O 73bnB/IoDbZjAiZgAiZgAiZgAuMSII+5fft2+vj9998/7mb8ORMIIeBJiBDMNmICJtBlApwxWLx4 cZqEYALi9ddf90RElx1mbSZgAiZgAiZgAnMIPPjggymXYeErr7wy+3zOSn5hAh0h4EmIjjjCMkzA BNolwOSDLl/0JES7vrB1EzABEzABEzCB0QiQw+R5zGif9tomEEvA94SI5W1rJmACHSTApIObCZiA CZiACZiACUwqgTyXyZ9P6v5Y98Im4EmIhe1f750JmMCQBOo6YHNfCf7q2t6Q8u9ZDfvc3+Kb3/xm 61oQZy33uCgt4H4kXfGRtPDYdiN+4ULctN1yLV3o1xpj2uaCfXzUlfFOXNr2EVy6pqUrY0zXjgP4 qQvjXZfGXuLXzQQiCPjrGBGUbcMETGDBE+CGUGfOnCnu3LmT7ifxpje9qdi4cWOxevXq8PtL8F1Q tNy8eTPZ5qsmGzZsKNatWzd7qWaUQ0g6jx07lrhgEy1r1qwpNm3aFM7l1VdfLc6dO1dcv3497T6X raIFP+kS1kguZ8+eLa5duzZ7D5JVq1YlLvfdF3toxkcXL15Mf/iHv6VLlxbbtm0rorWQjMOE+FUx yfecd+7cWfAY2bBPrBAzFHFw4WZvW7duLZYtWxYpJRVK3HDu1KlTqcAlXhlj0LJ8+fJQLXC5devW rBa4oGf79u3hWthxtOAjjb1oYXyJHnvhwth7/PjxgrFGfQkf0bejGzzoR8SNxje4rF27Nnzspf/A 5eWXX57FwDEJLcRxZEPL6dOnixs3biQu+IljI3p4HtnQcuHCheLSpUuzPiJuN2/eHM4lcr9tywQg EJvpmLkJmIAJLEACJFYHDx5MCczevXtToUJRt3///mL37t2pyI3abc7svPjii0nLzMxMujHVlStX iqNHj6YEmUIhqpFgwYXEHA5LlixJRR1aYLZr164oKemM+uHDh1NCvmfPnlREMkmDPrTs2LEjLOmj 6EcL9vHRypUrU0LMZA3L3vKWt4RxoXA6ceJEKuIeffTRYsWKFamYO3DgQPIbrCInIigm0bNly5ZU SBJDvH7++eeLt73tbaETERQGxAcFAfGL3yik6F9vfetbUzxHOYqJGXxCsYQW/Hby5Mli3759KV4i JyIo+rFLYY0WGsXuc889Vzz++OOhEzQU2i+88EKa/GDsJVYpMPHbww8/nCYZo3zEOMKYjwb6DRNW xBD6eM1kZ1RjzEULkw+Ms9x4+fz587PjMf0rquCmD6OFYxNamODkmHTkyJH0q1SMgVENLYy9xPAj jzyS+jCTEfQtTiSgJYoLffjQoUOzdnUc0PGRvqXJoyg+tmMCkQT8dYxI2rZlAiawIAlQDJDckDRQ aJN8UrisX78+FVD6yaymd16FCQkWSS8FJVpIfjnLQwHFe1GNpPfy5cuJCwkWWiiiOBtH0XL16tUo KanIxh4+4kwTWrgKgkkZihaKhahGQQkbrjTgTCBFi7TAi3iKaiTgFP6crUUDWihoiZ9oLRQpFCbY Z1LogQceSAULxQJFFcUDMR7R6M/0F+KWGEELhRxcOHNL8YDeiMbkBzFBzFJoc0XIQw89lGIZf0Vq YX/xEVp0dQp68Bd+o/jnKoCIpkkhika44B90oYWxj8IySgv7y1UqxCla8uMAfZwJRorciEYfYUzT RC9+IU6YeNBxIEoL+4sWxl76Mf0JLejguERc68q0CDaMdUyAcBwgRtDCuMf4x5VpjMtRDS0cd8rH AeIXLZHHpKh9th0TyAl4EiKn4ecmYAImMCIBEmGSBRVM+jiJMUkOSTDFXEQjASbZIxkvX7pOokVy yhUaEY3L6rElDrlNkmLO8ERxwTaJJwUKxUHemJCAC1oiCly4oKWKC9qII3xIXEU0tLDfxGreOFtJ sY0PI7hgW1oo2vKGDr7+wIQJMR7RsEURp0JFNokhaeH9iMbVMRSN9OG8wYX4RQd6IxpXHqCH+MjH GLTAiglXJtkiGlooYMs+YmzhKg0mkuhLEQ0faOxlgihv+ChSC5Nj9CV8xLFAjTEHLYwtUWMv+008 ECvlsZd4RlPU2Ms4xnhGfHAMyhuTI2iJihe04CPGE/yUN8YXHR+jxt7cvp+bQBQBT0JEkbYdEzCB BUmACQiSPpIJkpi8UVCSTJBsRDSucqBYQQt286aCgQIi4gwuWigSqrhw9omiJYoLE0EUKzDBbt6U pMMtosClWCEpF4NcC9pYTiEXMQlBgUD8woVYzZv8BpOIs6ZMzujMX1kLuohf9EZdycMZUQoA9Zuc Dfp4L6rw1xhTLm7Fhf4cdTaZs7fsu/pNzgVWvBdVyGGHvl3lIy1jHTQ13Sik0VIVuxqPo8Y7xniN veXxTjHEGBQxxjB28FfFhbGOP2I3Sovuj4HdvKGP4zfjS8TxUcckYqPsI16zHK1RE505Cz83gSgC c7PUKKu2YwImYAILhICKonIiwe5pGYUcBVTTjYSFhJtkqjwhwmuKTdYhAWq6YYfCspzsYRcd6IFL ROFPMteLC3rwE/6JSPhIcOGPzbKP4MIf60RowQa20KFYhYcaWmgRkxAUIRROtH5a1N/Sig39k48G aYngggaKSpr8kV58559YRWjBR4pL2c21SJ/05u818VyxQPyW+5LGHfRGjr1VXKQvatKK8Y6GP3px iRp7sUPc9OKCRmI3ahICLvIHz9XQx3L6vsYhvdfEo7iw/+o3soMOlkXFruz60QSiCXgSIpq47ZmA CSwYAhS2SnDLyR47qWWsR3LTdNPkguyW7bGcZC8i4WN/VfhX6ZBG8SuvU+frnIvs5ttnGRMmEVzY X2xV6UCTlkdwwQY+yu2mF9/5p+Q4InbVl7T/uY5cX4QWbOAjmhikF9/5J43Ei/jl79f9PI/f8ral LyJeYCL+YpDr0TLWi+QiBlVa8JF8mb9f93PxF4N8+yzjDx1RYwz2e2nhvSgt/WIXHTS0dMVHxK1i /P/UNfN/0OQ7vpuvlir/N7M33qoJjEfAkxDjcfOnTMAETCARULJddcDXsvkmE8OiViInu1Wfiyqc pKVKQ64vIuEbJvHHR8OsV7U/oyzrGpdh4jfCRzn/PD7EVstU7Gl5E4/4qJ+fZDOqL/WLS3Hpt470 zvcRH4mL7FZtM/dl1ft1Leu3z9I3rC/nq6mflnzbw66Xf2bU55PiI/ZLforq14NYErsR45181EuP uMxHi7bRy4aXm0DbBDwJ0bYHbN8ETKB1AhysdcDmrJqejyKM5KXcqpaV16nz9Ti667Q/7LZyLvnz YT8/6nrDcInQMUh3riF/Puhz474PF7Gpsle1bFxbw3xOWqrWjdTST0dZW4SufnpkX49lfQv5dT8u 2u8oLsNoQVOEnmG0ROhgf4fR0iUuaOlSG9VPOe/8eZf2yVpMQATm3plFS/1oAiZgAlNEgAO9vh/P maphD96sV/VdV6FTAjFoPa0/30d9D1p2q7Y37iRL1bb6LZOWqnVyfdyAq+k2jA382M+XdWkcRgu2 IrTgo2FivZ8v6+KCDuzo8u1e243QAntd4p/HalkT62i98nt1viZmBp0RjeDCviou+3GJHu+qWEsf eoeJ8aptjLJM/GW312eH7f+9Pj/Mcvmoal3py2O8ar26lg3iIj1d4RIVu1U36syZj8uFr3nonhYR THPNfm4CoxLwlRCjEvP6JmACC5IAyRIHbZIQJQDD7KgO9FWXV2o7bFPrDbPNcdcZlPChkXX6Janj 2i5/DjvsdxWXfN0ILrIhf+T2eY5GtIpf+f06X8OeP2yW9fBaywYlqXVoEhe2Jbv5duW7fL38/Tqf i3/OIN++9EVwkY+wL7tVWhTj+XtNPNc+V2mJ9FE+CSG7+f5qGfzwZ9NNcSm7uT0tQ0vERJHGjiof sYw/dERo6cdF+tAhzTm3up/3i13ZQgt+arqJixjk9uQj4lbr5e/X/Vw2qrRgi+UaE0exDUe2zd+g Cd1Rtut1TaAJAp6EaIKqt2kCJjBRBHSwV1LWKzGo2in95JmS3nwdLSMh0Lbz9+t+rp+lw255H7QM LVEJH1yrvgONFv7QoSS1bhb59vRzgjDBbt60DP/ApunGPmOrrAO70gI3+bJJPbBHD3bLfpIW7Ef4 iEJE+1x11l/61N+a5JLHZT8t0tukFratfRaD3J6WRftIdnMtimn1t/y9Jp7LDnaJ17xJX9TYC/9+ 4x36pDfX2cRzxWUVF8VzHuNNaNA2B3HBT+iNmJwRl3xsk050sDwfh/ReE48cA7CFj9RvZAcd6IHd qHkD6+uP7biZQJcJeBKiy96xNhMwgRAC8zlYr127NiUT3FirvB3ORJBMsA4JatNt6dKlqWDBZjmx 0d24V65cGVJUooUkqooLiTAa161bF8IFHcuWLUs2e3FBb0SRgI3ly5eny+thkDdew0vs8veaeE6B tmLFihQr5bNm+Ih4Frsm7OfbpChavXp1WqRYzd9HX5QW7BKbtCotuokefSmirVmzJvUT/TxmbhMu Obv8vSaer1+/Pm22HC8s1DLGu4hGvLDv2C2PvWIFu4gCFy0Uf2KQ77/GQMVU/l4Tzxlf6NsaZ3Mb imf6PeyabkuWLEmTDOozuT3GO/5WrVo1crGdb2fY5xpXOQZoMkaflY9YJ2IyWmMZOsrHJOmTXmkc 5jHvB/nzYT7rdUwgmoAnIaKJ254JmEAnCYx7wCaZoBghueP3ztXY3s2bN1OiR5IV0TjTgxYS8DwZ RsvVq1dTIRNV+JOQowXbt27dmrP7vCb5pECIahQJJJplLdeuXUtc0BoxUYQN4kHxke8/8YNGtEYU TtimqOylhWQ4Ugtc4EOsokkNHbdv306TRLoqQO819UghR9+mDxOravRztFAgUGBFNCbQsHXjxo17 uLBMY1CEFvYbH8BFBS12YcQy3oNdRIMJxTR9Oi9yiZfr16+nPqQJnKb1oIU/xl1NgMgmYwyFLX0p ojH20pe4N0CuJecSeRxgv+GiexWIgeKZ96PGXsY7OJSPA9ISFS+M7/iISYiyFl6jkffxpZsJLFQC noRYqJ71fpmACYQR2LVrVyoOjh8/PntWg8Tz/PnzxebNm8OSchKbnTt3puT76NGjsxMRJOQXL15M Z3dJ2qPajh070hUIuRYK7bNnz6YJiMhJiC1btqQi4cSJE7PFE1p4TbIXdZYS9thi0uPkyZOziTkJ +pkzZ5LGbdu2RbkoFUYUAadPn57VQkF37NixVPQTT1GNIg4/URAQrzSSdOKHhpaIs7fYYkIPLUyI 0I+ZFKEwwEf4ij4fVSBQ2DOOUJycOnUqaUEP8UMMz8zMhJy9hQtjzO7du5NdWKCDP/o0rNAScUUR WuBPTFBoiwvL8RdjHlp0CT7Lm2wU0cQEjb6jiSuOA8Tyxo0b01jYpAZtmz7C2Eu8okVn/a9cuVKc O3cuaYmaKEIT4xkxfPjw4dnJIiasGHM4BkQeB7Zu3Zrik3FfE1fED1o4DmzYsEEYG3+EC2MetunH NCY46dccHyK1NL6zNmACFQQW3T14vHG6oWIFLzIBEzCBhU7g6aefLp544omUtL3//e8vPvGJT4x8 JprkgYSPs4MqlDjrQvEQcZYn9xEJDYUbWkjUSYgpfEnAos6wSw+J3qFDh1KShRYSYy7XpsCLKuJy LXChmBMXEk9N3Gi9iEcV1xTcunSaxJPiIeJy4Hwf8QlJOUUKsUu8UKRQxEVrISWhiKTARQva8NWe PXtSwp7rjnhOcU3hpgYP+lHUWW3Z5fHSpUupYIEJYwp9GS1RX38oa8FHxAo6+Nu0aVO6siZfL+I5 ccskBH1K4xtFP2NvdGNsYTKacQ8f4St0UFBKW5QmxhYKWops+hL2KfgjJzm1r2g4cuRImsDTeKdj ko6XWrfpR3zDhAjHbLQQw/QhuET7iCtE8JGuliGGOQ5wTBrn+Pi+970vjeMwfPLJJ0MneJr2m7e/ sAjcHR8XeRJiYfnUe2MCJjAGgTomITBLAkHhT+LJGTgSHBLRNhqJFYkfRR3JDHq6oIWEk7NibWqR j7rAhSQUX8EFH0UnwYpN4oQz/MQuGvBRm1rwkYpKuEQXKjkXfKSiEi3jFAfa3nwfxYXt8DUMxpi2 GlrgQoNLm1rkI+JYPmprjCFuGXtpMEFPWw0tsKFftz3e5Vra5sKYS/zyCJeuHJPmexzwJERbPc12 RyVwd3xe5C8bjUrN65uACZhADwIkM20WKLkskhm+v92F1jUtXALbhQaXqEvXB+0vBVtXuKCFoqAL DS0UkW0WkjkHa8lpvPGcCRn+utA4BnAfjy60Lh2TuqSFsbcr412XtHQhZq1hegj4nhDT42vvqQmY gAmYgAmYgAmYgAmYgAmYgAm0SsCTEK3it3ETMAETMAETMAETMAETMAETMAETmB4CnoSYHl97T03A BEzABEzABEzABEzABEzABEygVQKehGgVv42bgAmYgAmYgAmYgAmYgAmYQH0E2rqxcH174C0tdAKe hFjoHvb+mYAJDCTAzef4o+XPB37QK5iACZiACZiACZhABwgoj0FK/rwD0izBBO4h4F/HuAeJF5iA CUwbAX7WjZ8x4yfEeO6D97RFgPfXBEzABEzABCabAD+Zy0+P0tr8ydzJpmj1UQR8JUQUadsxARPo NAEmH/jj99R5dDMBEzABEzABEzCBSSHwrW99K+Uv5DCcVHEzgS4T8JUQXfaOtZmACYQQ4MqHBx98 sOB3zPkeJQdwXw0Rgt5GTMAETMAETMAEaiDw0EMPFfzRmJBwM4EuE/CVEF32jrWZgAmEEPCVDyGY bcQETMAETMAETKAhAnkukz9vyJw3awLzIuBJiHnh84dNwAQWCgEfsBeKJ70fJmACJmACJmACJmAC XSbgr2N02TvWZgImMDEEuPTx+vXrxbVr19LXOZYuXVqsWbOmeOCBB8L3gQmVq1evFjdu3EjfC0XL 2rVrW9HCzsOEP76jumTJkmLVqlXpMRoMXPARf9zAi8tW4bJ48eJoKcnezZs3E5eXX3458Vi9enVr XO7cuVNcunQp3aCVryatXLmyWL58eStfS+LGapcvXy5u376dvialvtTGV6S4Ye2VK1eKW7dupRu9 wYT4bePn74hZacG+tPA1sugGF/o0Y8yb3vSmYtmyZQXx24YWxl6Nd/RxuDD2tqUFLowxaNHY28ZN Arm/kbTwnLEXLvTvNhqxy5in40BbWvALcUvMoKXt44COSfQpjkWML8SNmwksdAK+EmKhe9j7ZwIm 0DgBkuCDBw8Whw8fLlasWFGsW7cuFS3PPPNMSnQaF5AZIMFCy7Fjx1JCs2HDhuKll14q0HLx4sVs zeafkvgePXo06SHxRQuF1P79+4uzZ882LyCzAJcTJ04Uhw4dSkUTWtD37LPPJi28H9lOnTpV7Nu3 LxX5mzZtSlpeeOGF4uTJk5EyUqGEL5577rlkFy4Ub8QQGmEU2SgM0EKxsnHjxlQwoePFF19MEySR WpgE+cY3vpEKyvXr16dCmxjCbxQMkQ0ecLlw4UKaOGOcwW9wYRIrsrHvMDh//nwa6yia0MUYE62F sVf9Bh34iWKXfg2zyAaXAwcOFMePH589DhBDcKHQjGw6JqGFwpZ+zUQaWphsjBzvGEOOHDmSxl4d B4iT559/PsVNJBf2m2MjxyC0MMbAir5FPEdyYb/xD3kDk4ocB9CCttOnT4ePvZF+sC0TgIAnIRwH JmACJjBPAiTgJJkzMzMpKecs8s6dO9OZU5KJyLtUcwaZZGrr1q0pwaJY2bVrV0q4SHgii0oKyjNn zhRbtmxJCRZa4EJSTGEZWbDgH4o2Er1t27alIoFHChcKS4qFqMakEDaZrMJPnLndvn17Ki6ZhIgs nvABMcqZ7B07diQumzdvTrGDFgq6qEYBQLFCjO7ZsydpoUggfimc6GdRDQ0UKxQHxAl9mqtmdu/e nWKFWIoqWLCDPeJm7969KWalhcIyeuKKvkvBTbzQf7gCAn8xzkVPXDG+cFabGOHMusY7irloLsQo Yx5c6NvEDLqIJfp75HGACWeOBYwvjHmMMWjhrD/HAfhENcYQ/AQX+jNaOFYy2ckkNRPTUQ0tHB9h wjiHFo5JjH/0d/pTVMMW/YU+hJ/QwHGA/kS8cHWamwksZAKehFjI3vW+mYAJNE6AAoFkV5d0yiCX 35JwcTls1BUIJHMUK1x2S5GiRiFFkkNSQ8EZ0eCiszkkfPml9Loi4ty5cxFSkg1soaHMhQKTgook OapRTFMElLlwFhdfUSREFbgk5cQNWvKGj9BCYh7VKOAotJm0yi+lp7CkoCO2+apGREMHeigm80uj KVrQEjmJRrGCFnyUf3UIXcQMxW/UmXZihb4kn8gXjH/EDO8xKRDRGM/oS/Rp9KhxhpsCkyKcv4jG BANjCGMvMaOmSSz8g58iGlrwA7FKQatGn4ILsR01occEDP0W2/l4x1jM2KtjVtR4p3GeY3Pe4IJW uEU0HR/hQB/W8ZFHOHF84PgZxSVin23DBMoEPAlRJuLXJmACJjACASYZKIyYdCDhzBvJMN+Xjjqb TLHCWfQqLSpeeD/iLBgFAnrQAoO8cZ8MklIKq4hGcoktbPKXN3HhSggS4qYbVx700oI2eMEuQgtc KEaIW2I1b4oh9DJJ03Qj2VbBSEFbbsQM/Qw2EY1iBE2KD9mkSIAV7IjviKaJoiouLKM/RxX+mkAr xwscWAazqH7NWEY8EBsq4uQPsWJ8jijk8BF9pYoL+uhjaIloTDIQm4wn5bFX8RzFBS2Mrb24oI8J mohjEuMYtvAF41ve0EcM0Y8i4oWJInyAjvIxCSYsh13EcSDn4OcmEElgbsYcadm2TMAETGABEFBi WU722DUtIwmLuBSXpIVkrjwZghaWkWSxTkRRiR0KtXKClWshUY5IsrADf/a/XKzwGjYwiShw2V/s KDbgoSYtFDMRZ/zRgS1aWY+0kJBHfD2EuFVRX9YifWiJKLbRoq/nVGlR/4rgwr5TOLHvVVpYxnti x/pNNfqz+JeLOGxKn8bEpnRouyqksVvu1xp3osZe+OMHxYY08qixN4oLcSktZT3yEeOz+n6ute7n jGOMebKbbx+fsRx2EZMQ4oKGKi7oQStsmm6w55jD/pe1YJvl6Ig4DjS9r96+CfQi4EmIXmS83ARM wASGIKCCviqRUGJMQhgxCSEt2JXtfBdYho6IhI9kTolwrkHPpY/1mm5K5PpxgUmEj9hfbBEvYqD9 57XiKIIL8YKPcrvSwqP0RWihwEUPNmU31xLJJe8jslulJSJesKtisYqLirsoH2mf+3FBC3HVdOvH RfpyXzapR2Ov7Oa2WIbvorgM0oI2+lvEcSAfe3MmPFc846OIeOnHBS38oSOiL0mLGORspAX/RPgo t+3nJhBJwJMQkbRtywRMYEERIGEhmevVlGBEJXzSIrtlXSwnqdF65ffrfN0vecr1RSR8Kpx67R96 YBLBZVgbgzT32pdRluOjfsm//NTPl6PY67cuOoaxExEvaBEXMajSPohf1WfGWSYu/bRExEveR/pp gd2wcT4OD31G+1ylRctgF6FFNmRXGsuPEfHbT4v0sY7WK2us8/Wg2JUe+bJO2+Vt9dtf6WAdaS5/ vs7Xg/ZXejRZMY7tqgmxcbbjz5hAUwQ8CdEUWW/XBExg4ggMU3iMu1P9EqBxtznO5/ICa5zPD/sZ sRy0fpe4RGgZ1kZEIjzIN0qEh9U8aHuD3lfMyG7V+hFahu0jrBelp4oFy8QqQkcvDVXL5cuq9+pa NoyNYX05X03DaMFGhJ+G0cI6EWPMsPsboWUYLvgoQgt2hmnDas63BXM+p7Ehf8/PTaBLBObeoatL yqzFBEzABIIIcFkvv3DBwXuU7+1ykB/mbMOw6813d3Vpdr/EBb3DaJ6vFn0ne9B2pHnQevN5fxgt 08gF9v0SVcVRhI/QgZ8GnSkexpfziRU+O2wsRPVr9rlXYRQPPb6BAABAAElEQVTtI40dslvFOpJL lX2WSd+wvuy1nWGXD9tHIuJ3GC1wGWa9Yfe/13qD9ld+itAyjA1id5j1eu3vsMub4sIvKymHGXS1 xbBavZ4JNEXAkxBNkfV2TcAEJoYAdy/nZ9VI9PmJNRIjkpFhmhIWJVP5Z7SMbQ1KOvLPjft8kA30 sI4KiXHtDPM57LDfYpB/Jl8G+6Zb1U30cpvoiUzKe3FBk9gM0pzrH/e5fJTbzbclLYrx/L26n8ME O1x+LLu5DS2L4qI+UnUmV1rQi+6mG/vMd+tlN7enZYP6fv6ZcZ+zv4oF2c23pWWwE7/8/bqfa+yQ 3Xz7WobeCC3iL7tlLVqu9fL3634uG7KZb1/LYCJf5u/X/Vz9VXZ7bV/r9Xq/juWDuKCR/qz16rDZ axuK3V7vi9eoXPi5Wvk1Iu576fdyExiGgL+OMQwlr2MCJrCgCZB0LF26NE1A8Lyq8OgFQD89VvUZ LSMZGDWZ6GWv33IlNthVEqP1WcYfCYqSFL3XxCM2SOjEILeBNpbzvjTn79f9XD7CbpkLr5l8ivIR dthn9r9Ki7hIc90s8u2hQ7FQ9hPatCzCR3ksyG6uVcsiuOAjFSKyW6Ulggt29XOT5XjhPemLGF/y PlJ1ZYa0RPiIfRd/7JbZ6Cxw1HiHFmK4iou0dYGL9MElImb6ccFv/LEOsdV0U7zgD/lENuHCsnwc 0ntNPDK+YKsqdtHBctbR+DysBmKMEyn8aQwb9rNezwSiCTTf66P3yPZMwARMYEQCeaJKMkICMGxb tWpVWlVJb/45Li0noVi5cmVIkqXEQwlVrkX6li1bFpJ8MqlDYll1eb30oSUi+Vy8eHFKdKt8Ky4U eqzXdJMd7FYlwmgkkYwoELCDHuJdHLT/So5JlJcvX67FjT3SB2Wn6mZs6CNWiKumG/tMn6X10sJ7 0svzJtuKFSvS5qu0sCzKR4gQl6p+rWUaE5tkwraxw76XY5f3pAUfjVrI8flRG3aIzyotLKNfR3Fh XIVL1XinGOJYETHGML5gp4oL4x8axW5U5qOuLy6ym38efSzXmJi/18RzJgg43lT5CB3owUeaOBlW Q8551M8Oa8PrmUBdBDwJURdJb8cETGBiCZAQ6KwBSQFJwLCNBIpkgeROP0fGZ9kGv1HP4/r164fd 3LzWI6mhYOEeF3kywkZv3LiRts3XTiKScpI5tMDzzp07c/aL3z9H38aNG+csb/LFmjVrUmGS+wh7 4kKBEMGFOKOQo8gv/x492vhDi+KxSSZsGx+g5datW3NMEUP4Dm4RWiiauJSYJp/kguhLxHfEJAR2 0UJRCZd8PCBuiWfim74f0VavXp0mi65fv36PuZs3b6YiT+zuWaHmBYwfxANciBs1xTMFJ3ojGkUl 8YAWYjVvcKER38RW040+TUwQG2UtxC6xBLuIRuGPHrRo0kF21bc2bNgQwgUm+KnqmKQYYoyJGHux oa9dlo9J+IgYRktEvNBP8BH+gU3eFEMcB0adSNBxn32AvZsJdJmAJyG67B1rMwETCCFAckKSSCMR yYuOYQRs3749FY8XLlyY/SyJBa9JyEnCIhr7sXXr1qTh3Llzs8kwWs6fP590RBUr7O+WLVtSInTm zJnZSRGSJLRRUEZq2bRpUyqe0KIiAS2nTp1KBV5UgQAXbJEgEh9KGok7fEQcbtu2LSQRRgvxiS/w ibQQ/3CSFtaLaBT16Ll69erspAha4EKyjg8jihX2lUKO+OQmb3nxf+XKlYICl9getUAYl6H6Cjqw r/Hp0qVLacJm8+bNYQUHMcEYQ9F28eLF2V1CG3rwEXojGn0IexSzORcKbWKICYgoLeJCH6Lv0J9p TCrymuKWojOi0UeICbTkxyQ48ZrJ4SgtFMOMZ/DRzZ9hIE70s8ixl5hgEi0/PuIrHZOiThjAgH6E FsY3jb0cm86ePZv6MxNFozZtJ2psGlWf1zeBnIBvTJnT8HMTMIGpJEAioOKGJEAJ5LAwSKIoklRU si0SYa6Q2LlzZ8hlr9LK2RNskviSkJCMcMadhH3Hjh1hxS160LJr167ixIkTSQsFAVrgTWIamShR 4O7evTslwocPH07Fia5EePjhh5OvxLDpRzjs3bu34E7mBw4cSGfUdabwkUceCSuc2E98sGfPnuLo 0aPFoUOH0pllCicKFnwXdbYfLRQs0oIeCiX64uXLl1NMR53VRgtF08zMTHHkyJHEhgkJtNCvKXwp 8qIaXJjopD8fO3YsTYIwEUGhTdGEHtaJakzAECMUbjqLCxcY0a81lkboISbox4wxTIxo7CVu8V8k F4pGjgNMzvDI2W4mrBgHo7ngC86oo4WvpjD+w4f+Tr+O9JHGXiZ84cJxkfFFxySOB1ENX3AcYOyl bzMJQvwQJ8QL2qKajgNMznBMwjb9CkYcH9A2SmN80gRl5LF1FI1e1wRyAovuBuzw1x3nn/RzEzAB E1ggBEhInnjiiXQm713velfxyU9+Mp0tGnX3SCA4c0oyQELBGafIJDjXixYKAwoXkhmSr7a0kFRx plRaKC4p8tpoJORoUWKOlsiEPN9neKAFPiTkxEtkQp5rIWYpatGiS4V5bKORlhC7FFH4hiuJos5o l/cXLRRMFHDELAVVW1rQJi08l5a2+jXFG0U2jPBR1BVf7Hu5lbXwNY22uDAxQ/zSp4iVqKsOykx4 nR8HunBMIl4Yezkmdek4wPGxzWMS8aLjAP16nLGX8fI973lPCgMmxP7pn/6pKiS8zAQ6QeDu+Lwo bvqxE7tsESZgAiZwL4H8SgiSNpKkcRqFJGfmutDQEnlpab995qxMV7SQ3EVe/tuPC3HHGcsuNBLw rmiheKRAabN4k0/QQlHAXxda28V+zoCilr8utC5pocAe9Sx2Uwy7dBxAC39daF06DqCljrGXiVK1 qPuyyJ4fTWAcAu2cihpHqT9jAiZgAg0RIDHS5YucxRp3EqIhed6sCZiACZiACZiACfQkwJUQalxV 52YCXSfgSYiue8j6TMAEGifAJIQusZ7PlRCNC7UBEzABEzABEzABEygR4CtJany9xM0Euk7AkxBd 95D1mYAJNE6AS3l1+SxnE3wlROPIbcAETMAETMAETKAmAtwvRq2Or3doW340gaYIeBKiKbLergmY wMQQ4OZ3OnPA1zG4GsLNBEzABEzABEzABCaBADc5VlM+o9d+NIEuEvAkRBe9Yk0mYALhBHTmQHfl DxdggyZgAiZgAiZgAiYwBoFLly6lT3Ez3a1bt46xBX/EBGIJeBIilretmYAJdJSAfr2Bn1W7fPly R1ValgmYgAmYgAmYgAm8QYC8RZMQfLW0K78A9YZCPzOBewl4EuJeJl5iAiYwhQR27949u9fnzp2b fe4nJmACJmACJmACJtBVAq+99lpx4cKFJI9f+tq0aVNXpVqXCcwS8CTELAo/MQETmGYCe/bsKbg3 BO38+fPTjML7bgImYAImYAImMCEEuJm2roTg17709dIJkW+ZU0rAkxBT6njvtgmYwFwCK1euLDZv 3pwWnjx5cu6bfmUCJmACJmACJmACHSTw6quvFhcvXkzK+CoGv/jlZgJdJ+BJiK57yPpMwARCCHD2 YMeOHcnWqVOnim9961tj2eWySP5ef/31sT5f14ewzz5whqRtLeyTtVR7lu/ydsVH0sJj242Yhcu4 /bBO/bmWtvsS9jXG1LmP424LH3VlvBOXtn0Ey65p6coY07XjAH7qwng337GX/dCVEPlXS8ft1/6c CUQQuC/CiG2YgAmYQNcJMAnBwfurX/1q8dJLLxVXr14d6eZOfObMmTPFnTt3Cu5OzVc7Nm7cWKxe vTq9jtx/fmIULTdv3ky20bNhw4a0P9/1XbFzzySdx44dS1xggJY1a9ak76zyPLJxtoj7feinzGCB FvzUBpezZ88W165dS5NEsOBn1fgu7333xR6a8RFn0fhDB39Lly4ttm3bFq6FZBwmxK+KSfrmzp07 Cx4jG/aJFWKGIg4u999/f7rz/LJlyyKlpELp9u3bBROkFBzEK2MMd8Ffvnx5qBa43Lp1a1YLXNCz ffv2cC3sOFrwkcZetNCPosdeuDD2Hj9+vGCsUV/CR238ZCI86EfEjcY3uHCpPtoiG/0HLvwEthrH JLToa5Ba3vQjWk6fPl3wS1hwgQVXD6CnDS7cy4EJBPmIuOWqzFG4sC/80R5++OGmEXr7JlALgdhM pxbJ3ogJmIAJ1E+Awo9Ch0aidOXKlaEnIVj/4MGDKYHZu3dvKlQo6Pbv358mNihyoxoFyosvvpi0 zMzMFIsXL077cvTo0ZQgUyhENZI9uJCYM8HDJaIUdWiB2a5du6KkpDPqhw8fTgk59/+giGSSBn1o 4SqYUZK++Qin6EcL9vERXwUigWSyhmVvectb5rP5kT5L4XTixIlUxD366KPFihUrUjF34MCB5DdY RU6KUEyiZ8uWLamQJIZ4/fzzzxdve9vbQiciKAyIDwoC4he/UUjRv9761reGXvLMxAw+oVhCC37j a2P79u1L8RI5EUHRj10Ka7TQKHafe+654vHHH099a6QgnMfKFNovvPBCmvxg7CVWKTDxG8UYk4xR jXGEMR8N9BsmrIgh9PE68jjAmIsWClvGWY4D3OtI4zH9K6rgpg+jhWMTWpjg5Ph65MiRNOHPGBjV 0MLYSww/8sgjqQ8z9tK3OJGAligu9OFDhw7N2tVxQMdH+pYmJgbxYSxge2jXFZ2DPuP3TaBtArGn xNreW9s3ARMwgT4ESGJpJLacpR62kQCQ3JA0UGiTfFK48LOfFFCciYpoKkxIsEh6KSjRQvLLWR4K KN6LaiS9/NwpXEiw0EIRxdk4ihauNolqFLfYQwtnmtBCgcKkDEWLLmWN0ENBCRuuNOBMIEWLtMAr 8p4kJOCw4WwtGtBCQUv8RGuhSKEwwT6JNHd5p2ChWKCoonggxiMa/Zn+QtwSI2ihkIMLk1UUD+iN aEx+EBPELGMUV4TwM3zEMv6K1ML+4iO06OoU9OAv/EaRy1UAEU2TQhRecME/6EILYx+FZZQW9per VIhTtOTHAfo4E4wUuRGNPsKYpole/EKcMPGg40CUFvYXLYy99GP6E1rQwXGJuNaVaRFsGOuYAKHv ECNoYdxj/OOYH3lTarRw3CkfB4hftIxyTHrqqacSPmINtm4mMAkEPAkxCV6yRhMwgRACJAMktEpu hzHKuiQLKpj0GbZDkkMSTDEX0UiASfZIxsuXrpNokZzq5lVN6+Gyel3eD4e8kRRzhieKC7ZJPClQ yjfsYkICLmiJKHDhghbFR84FbcQRPiSuIhpa2O+yjyj+KbbxYQQX9lVaSKTzhg6uXGHChBiPaNii iFOhIpvEkLTwfkTj6hiKRvpw3uBC/KIDvRGNCVr0EB/5GIMWWDHhyiRbREMLBWzZR4wtXKXBRBJ9 KaLhA429TBDlDR9FamFyjL6EjzgWqDHmoIWxJWrsZb+JB2KlPPYSz2iKGnsZxxjPiA+OQXljcgQt UfGCFnzEeIKf8sb4ouPjMGMvjJmIpM3cvZLDkxA5TT/vMgFPQnTZO9ZmAiYQSoAEjYM4B/5nnnlm KNtMQJD0kUyQxOSNgpJkgmQjonGVA8UKWrCbNxUMFBARZ3DRQpFQxYWzTxQtUVyYCKJYgQl286Yk HW4RBS7FCkm5GORa0MZyCrmISQiSV+IXLsRq3uQ3mEScNWVyRmf+ylrQRfyiN+pKHs6IMg6o3+Rs 0Md7UYW/xphycSsu9Oeos8mcvWXf1W9yLrDivahCDjv07SofaRnroKnpRiGNlqrY1XgcNd4xxmvs LY93iiHGoIgxhrGDvyoujHX8EbtRWnR/DOzmDX0cvxlfIo6POiYRG2Uf8ZrlaB1mopPY474StO/9 3u9t5f4jOUs/N4FhCczNUof9lNczARMwgQVK4Md+7MdSMvK///u/QyWvKorKiQR4tIxCjgKq6UbC QsJNMlWeEOE1xSbrkAA13bBDYVlO9rCLDvTAJaLwJ5nrxQU9+An/DJPwsf58Ggku/LFZ9hFc+GOd CC3YwBY6FKv5vqGFFjEJQRFC4UTrp0X9La3Y0D/5aJCWCC5ooKikyR/pxXf+iVWEFnykuJTdXIv0 SW/+XhPPFQvEb7kvadxBb+TYW8VF+qImrRjvaPijF5eosRc7xE0vLmgkdqMmIeAif/BcDX0sp+9r HNJ7TTyKC/uvfiM76GDZsLHLJAR/tB/+4R++x+farh9NoGsEPAnRNY9YjwmYQKsE3vGOd6RLVjlT xHeK+zUKWyW45WSPz2kZ65HcNN00uSC7ZXssJ9mLSPjYXxX+VTqkUfzK69T5Ouciu/n2WcaESQQX 9hdbVTrQpOURXLCBj3K76cV3/ik5johd9SXtf64j1xehBRv4iCYG6cV3/kkj8SJ++ft1P8/jt7xt 6YuIF5iIvxjkerSM9SK5iEGVFnwkX+bv1/1c/MUg3z7L+ENH1BiD/V5aeC9KS7/YRQcNLV3xEXGr GP8/dc38HzT5ju+G1cKNLJnI0Q10m1HsrZpA/QQ8CVE/U2/RBExgggnwCwEczEkAnn766YF7omS7 X8I3bDIx0NiAFZTIVWnRR6MKJ2mR3fwx1xeR8A2T+OOjYdbL92Oc513jMkz8Rvgo55/HhxhrmYo9 LW/iER/185NsRvWlfnEpLv3Wkd75PuIjcZHdqm3mvqx6v65l/fZZ+ob15Xw19dOSb3vY9fLPjPp8 UnzEfslPUf16EEtiN2K8k4966RGXYbQ8++yzKV957LHHZnn22q6Xm0CXCHgSokvesBYTMIHWCXBT qPe+971Jx3/913/NXukwSBjJS7lVLSuvU+drJS51brOJbeVc8udN2GKbw3CJ0DFo/3IN+fNBnxv3 fbiITZW9qmXj2hrmc9JStW6kln46ytoidPXTI/t6LOtbyK/7cdF+R3EZRguaIvQMoyVCB/s7jJYu cUFLl9ogP3FFxde//vV0z413vetdXZJuLSYwkIAnIQYi8gomYALTRuAHfuAH0s3OuMyx3w3fSLCq vusqXkogBq2n9ef7qO9By27V9rh0edjEsOrzwy6Tlqr1c33cgKvpNowN/NjPl3VpHEYLtiK04KNh YqGfL+vigo5h7Ayzznw1wV6X+OexWt4u62i98nt1vh4mZiK4sK+Ky35cose7KtbSh95hYrxqG6Ms E3/Z7fXZYXzZ67PDLpePqtaXvjzGq9ara9kgLtLTFS5RsVt1o86c+bBcuFqTn/PkFzbe/va355vw cxPoPAFPQnTeRRZoAiYQTYCfuOK33s+cOVPs37+/r3klT1WXVyqRILHRen03Ns83ByV8aGSdfknq PCXMfhw77HcVl9mV7j6J4CIb8kdun+doRKv4ld+v8zXs+cNmWQ+vtWxQklqHJnFhW7Kbb1e+y9fL 36/zufjnDPLtS18EF/kI+7JbpUUxnr/XxHPtc5WWSB/lkxCym++vlsEPfzbdFJeym9vTMrRETBRp 7KjyEcv4Q0eEln5cpA8d0pxzq/t5v9iVLbTgp6abuIhBbk8+Im61Xv5+3c9lo0oLtliuMbGXbb7a 86UvfSndwPKd73xnsWPHjl6rerkJdJKAJyE66RaLMgETaJMAvzvPT12RyH7hC1/o+x1R/eSZkt5c t5aRcEQkfPpZOuyWkxstQ0tUwkcSVfUdaLTwhw4lqTm3up/r5wRhgt28aRn+UWKYv1/3c/YZW2Ud 2JEWuMmXddvPtwd79GC37CdpYf0IH1GIaJ+rvgctfepv+X7U/TyPy35apLdu++XtaZ/FIH9fy6J9 JLu5FsW0+lv+XhPPZQe7xGvepC9q7IV/v/EOfdKb62ziueKyioviOY/xJjRom4O44Cf0RkzOiEs+ tkknOliej0N6r4lHjgHYwkfqN7KDDvTArl/ewM2zv/rVr6b1fvInfzKEoTT60QTqIOBJiDooehsm YAILigDJ5Ic+9KH0e9tPPfVU36sh1q5dmw7+3FiL5CFv3BmcZIJ12GbTjUsyKViwWU5sdDfulStX hhSVaCGJquJCIozGdevWhXBBB/f66McFvRFFAjaWL1+eJrbQkzdew0vs8veaeE6BxoQbsaK72MsO PiKexU7Lm3qkKFq9enXavGI1t4W+KC3YJTZpVVp0Ez36UkRbs2ZN6if6eczcJlxydvl7TTxfv359 2mw5XlioZYx3EY14Yd+xWx57xQp2EQUuWigYxSDff42Biqn8vSaeM77QtzXO5jYUz/R72DXdlixZ kiYZ1Gdye4x3/K1atapvsZ1/Zj7PNa4y3mkyRtuTj1gnYjJaYxk6ysdq6ZNeaSw/ckNKrtZ8/PHH ize/+c3lt/3aBDpPwJMQnXeRBZqACbRBgITxp3/6pwt+2/3LX/5yTwkkExQjJHf8TJYaSfHNmzdT okeSFdE404MWEvA8GUbL1atXUyETVfiTkKMF27du3Zqz+7wm+aRAiGoUCSSaZS2cTWKCCK0RE0XY IB4UH/n+Ez9oRGtE4YRtispeWkiGI7XABT7EKprU0HH79u00SaSrAvReU48UcvRt+jCxqkY/RwsF AgVWRGMCDVuMRWUuLNMYFKGF/cYHcFFBi10YsYz3YBfRYEIxTZ/Oi1zihXv50Ic0gdO0HrTwx7ir CRDZZIyhsKUvRTTGXvrSnTt35mjJuUQeB9hvuKAnb4pn3o8aexnv4FA+DkhLVLwQm/iISYiyFl6j kfd7XQkBT66CgNuP/uiPhsVW7j8/N4H5EvAkxHwJ+vMmYAILlsAHPvCBVGz84z/+45wkt7zDu3bt SsXB8ePHZ89qkHieP38+/dxnVFJOYrNz586UfHNTTU1EkJBfvHgxnd0laY9qfEeVAirXQqHNjbRI 9iInIbZs2ZKKhBMnTswWT2jhNcle1FlK2GOLSY+TJ0/OJuYk6JzVopDZtm1blItS8koRcPr06Vkt FHTHjh1LRT/xFNXYd/xEQUC80kjSiR8aWiLO3mKLCT20MCFCP6b4pzDAR/iKPt+rQODzdTYKe342 mOLk1KlTSQt6iB9ieGZmJuTsLfvEGLN79+5kFxbo4I8+DSu0RFxRhBb4ExMU/eLCcvzFmIcWXYLP 8iYbxSAxQaPvaOKK4wCxvHHjxjQWNqlB26aPMPYSr2jRWf8rV64U586dS1qijkloYjwjhg8fPjx7 HGXCijEn+jiwdevWFJ+M+5q4In7QwnGA+0FFNbgw5mGbfkxjgpN+zfGhnxb62//8z/8U/KT4//t/ /y9Ksu2YQK0EFt09eLxxuqHWTXtjJmACJjD5BH7v936v+Od//ufiU5/6VPGe97yn5w6RPJDwcXZQ hRJnXSgeIs7y5MJIaCjc0EKiTkJM4UsCRhER2Uj0Dh06lJIstJAYc7k2BV5UEaf9RQtcKObEhcRT EzdaL+JRxTUFty6dJvGkeIi4HDjfR3xCUk6RQuwSLxQpFHHRWkhJKCIpcNGCNny1Z8+elLDnuiOe k+xTuKnBg34UdVZbdnm8dOlSKlhgwphCX0ZL1NcfylrwEbGCDv42bdqUrqzJ14t4TtwyCUGf0vhG 0c/YG90YW5iMZqzBR/gKHRSU0halibGFgpYim76EfYr+yElO7Ssajhw5kibwNN7pmKTjpdZt+hHf MCHCMRstxDB9CC7RPuJEAT7S1TLEMMcBjkn9jo+f+9znij/90z8tPv7xjxc/8zM/E55jNO0jb3/h E7g7Pi7yJMTC97P30ARMYB4EvvGNbxRMRHCjSg74nNnv1UggKPxJPDkDR4JDItpGI7Ei8aOoI5lB Txe0kHByVqxNLfJRF7iQhOIruOCj6CRYsUmccIaf2EUDPmpTCz5SUQmX6EIl54KPVFSipV9xoM81 9SgubJ+vYTDGtNXQAhcaXNrUIh8Rx/JRW2MMccvYS4MJetpqaIEN/brt8S7X0jYXxlzil0e4dOWY NMxxgAmLD3/4w2lM/Iu/+ItWJkTbimfbXTgE7o7Pi+5bOLvjPTEBEzCB+gk89thjxfd///enqyE+ +MEPFrzu1Uhm2ixQcl0kM3x/uwuta1q4BLYLDS5Rl64P2l8Ktq5wQQtFQRcaWigi2ywkcw7WktN4 4zkTMvx1oXEM6DdZHamxS8ekLmlh7O3KeDeqli9+8Yvpqqhf+7Vf8wREZGeyrdoJxF6XW7t8b9AE TMAEmiVA4vTRj340JZV/8id/ks6eNGvRWzcBEzABEzABEzCBuQT4us9nPvOZ4p3vfGfxcz/3c3Pf 9CsTmDACnoSYMIdZrgmYQDwBzmr99m//drqfwL/+67/GC7BFEzABEzABEzCBqSXA11n++q//On3F 8mMf+1jYr9BMLXDveOMEPAnROGIbMAETWAgE3vve96a7UJMEXLhwYSHskvfBBEzABEzABExgAgjs 37+/ePLJJ4uPfOQjrdxcdAIQWeKEEfAkxIQ5zHJNwATaIcDXMn71V381fe/4s5/97OwN2dpRY6sm YAImYAImYALTQIAbwH7+859PP8n5C7/wC63dNHgaWHsf4wh4EiKOtS2ZgAlMOAF+3uyJJ54o+ErG gQMHJnxvLN8ETMAETMAETKDrBL70pS+lKzB/4zd+o7Vfluo6I+ubPAKehJg8n1mxCZhAiwT4qc7f /M3fLK5fv96iCps2ARMwARMwAROYBgJXrlwpfvEXf7GYmZmZht31Pk4JgUV3f1P59SnZV++mCZiA CZiACZiACZiACZiACZiACZhASwTu/vz1Il8J0RJ8mzUBEzABEzABEzABEzABEzABEzCBaSPgSYhp 87j31wRMwARMwARMwARMwARMwARMwARaIuBJiJbA26wJmIAJmIAJmIAJmIAJmIAJmIAJTBsBT0JM m8e9vyZgAiZgAiZgAiZgAiZgAiZgAibQEgFPQrQE3mZNwARMwARMwARMwARMwARMwARMYNoIeBJi 2jzu/TUBEzABEzABEzABEzABEzABEzCBlgh4EqIl8DZrAiZgAiZgAiZgAiZgAiZgAiZgAtNGwJMQ 0+Zx768JmIAJmIAJmIAJmIAJmIAJmIAJtETAkxAtgbdZEzABEzABEzABEzABEzABEzABE5g2Ap6E mDaPe39NwARMwARMwARMwARMwARMwARMoCUCnoRoCbzNmoAJmIAJmIAJmIAJmIAJmIAJmMC0EfAk xLR53PtrAiZgAiZgAiZgAiZgAiZgAiZgAi0R8CRES+Bt1gRMwARMwARMwARMwARMwARMwASmjYAn IabN495fEzABEzABEzABEzABEzABEzABE2iJgCchWgJvsyZgAiZgAiZgAiZgAiZgAiZgAiYwbQQ8 CTFtHvf+moAJmIAJmIAJmIAJmIAJmIAJmEBLBDwJ0RJ4mzUBEzABEzABEzABEzABEzABEzCBaSPg SYhp87j31wRMwARMwARMwARMwARMwARMwARaIuBJiJbA26wJmIAJmIAJmIAJmIAJmIAJmIAJTBsB T0JMm8e9vyZgAiZgAiZgAiZgAiZgAiZgAibQEgFPQrQE3mZNwARMwARMwARMwARMwARMwARMYNoI eBJi2jzu/TUBEzABEzABEzABEzABEzABEzCBlgh4EqIl8DZrAiZgAiZgAiZgAiZgAiZgAiZgAtNG wJMQ0+Zx768JmIAJmIAJmIAJmIAJmIAJmIAJtETAkxAtgbdZEzABEzABEzABEzABEzABEzABE5g2 Ap6EmDaPe39NwARMwARMwARMwARMwARMwARMoCUCnoRoCbzNmoAJmIAJmIAJmIAJmIAJmIAJmMC0 EfAkxLR53PtrAiZgAiZgAiZgAiZgAiZgAiZgAi0R8CRES+Bt1gRMwARMwARMwARMwARMwARMwASm jYAnIabN495fEzABEzABEzABEzABEzABEzABE2iJgCchWgJvsyZgAiZgAiZgAiZgAiZgAiZgAiYw bQQ8CTFtHvf+moAJmIAJmIAJmIAJmIAJmIAJmEBLBDwJ0RJ4mzUBEzABEzABEzABEzABEzABEzCB aSPgSYhp87j31wRMwARMwARMwARMwARMwARMwARaIuBJiJbA26wJmIAJmIAJmIAJmIAJmIAJmIAJ TBsBT0JMm8e9vyZgAmMR+Na3vlV85StfKU6cODHW5/0hEzABEzABEzABExiGwLlz54ovfelLBbmH mwksRAKehFiIXvU+mYAJ1E7gqaeeKv7wD/+wWLJkSe3b9gZNwARMwARMwARMQAReeuml4m/+5m+K L3zhC1rkRxNYUAQ8CbGg3OmdMQETaILA6dOniz//8z8vfvmXf7lYt25dEya8TRMwARMwARMwARNI BHbv3l38yq/8SvGXf/mXxcWLF03FBBYcAU9CLDiXeodMwATqJvC5z30ubfLHf/zH6960t2cCJmAC JmACJmAC9xB45zvfWXzf931f8cd//MfFrVu37nnfC0xgkgl4EmKSvWftJmACjRL49re/XTz55JPp XhC/9Vu/1fOrGKz38ssvF9/85jeL119/vVFNgzaO/VdeeaV49dVXB60a8j5ausIFHfgJf7Xd0AKb rmjpCpfXXnutMz7qopYufD8cDcQLfNpu0mIucz0BD8aXLvgoPz7OVRn/Ci06JsVbn2txGC0PPvhg 8ZGPfKT4t3/7t+Lzn//83A34lQlMOIH7Jly/5ZuACZhAYwTOnz9f/NVf/VXxrne9q3jLW95yjx0K fs5OnDx5Mt08atGiRcV3fdd3FTt27CiWLVt2z/pNL0DLmTNnUpIlLVu3bi1WrFhR8DqqwYXvs8KF Yltatm/fXixfvjxKxqwdkk643Lx5M2lBz4YNG9JXa/BXZKM4OHbsWHHnzp1kFi1r1qwpNm3aFOoj jDNRxc3Prl+/nrTAAi0bN25McZwWBv2Dy6lTp5KPiB+4rF69Omm5777YVAUuXP585cqVtPdvetOb ipUrV7aiBS6MQ5cuXZqND/rzli1bivvvvz/IO/9nBr9IC0vw0dKlS4tt27YV0T6Cy+XLl4sLFy6k iV+0MeYyxrTBBf/wR9FPP3rooYeSjxYvXhzqI7jQnxnvYIIWClmOA9FasH/t2rWkhYIbLcQJPmrj 3kpoOXv27OyxGi3ELlqI5agGF45FjHf4i4aWnTt3VnLBdz/0Qz9U/MM//EPxjne8o3jssceipNqO CTRKYNHdztDuabtGd88bNwETMIHxCFC4/v7v/37xH//xH8WnP/3p4s1vfvM9GyLpPHz4cEo2KSJp JOkU329961tT4XLPhxpaQMF04MCBVFxToFA4kXDxx3dL169f35DlezdLsnfo0KE0+TEzM5OSTwoG WD388MOhWpgEefbZZ1MivmvXrvRIkn7w4MFUVJL4RSWgnD3GLgn5nj17UqFy+/btYv/+/SlWHnnk kTAtFEvYpeDGJxQoaCGGVq1alfRFcSERxy4TM3ChmGQSa9++fSkpf9vb3hbGBS3ELjHMxCNFNlxY RmH56KOPpr51b9Q3s4R4uXHjRurDTD7gL7jQv7/7u787jAsxe/z48TS+EadooW8RQ2hhfIws/hlL mCgiXpg4g8uRI0fSBOzjjz8epoUUmoKfX01ifOGeQXA5evRoulqE2GUSIKpR2OInfMQkHv2c18Qw xyRiOKpxLyX0cDzavHlzKrg5NnLcfPvb3x6qheMgPmG8ZwKaBheOm/RpJhmjGhNnTEQz2QsbGvqI I3xE3yo3+tnHPvaxNGnC/amixuayDr82gboI3I3hRbGngOpS7u2YgAmYQMMEnnnmmeJrX/tawX0g 9u7dW2mNxPOBBx5ICRYJOH8kWyR6JMQkxhGNwonkjmKAs0zYRwtneSjoSNgjL8slmaJoIeGDD2d5 SLhIriK5wJ4kmOKWyRDOeMGIooVElPd0FUCEn0i+KW7Rgl/gQvJLzOi9CB3YYLKMfcdHaIEL/iEp 5uoIdEY1JqjYf7jAAy1cMcMVReggaY9qFCVXr15NtuEhLvQrtMAtquEffMEEJwUlWpgsovDWmdQo Ldhj34kPtBC7aMFnFLj0paiGPWKCMYXJ1ZwLfZ2xMOr8GhOL2EMHY0rORRMAjIURDS1wYXxDD1o4 FuAjjhEcB6K4sM9wYRKPM/loYTKG8YZjE1rQFNE4DlPkM6bQl3Ss1vEpkgv8mZyBh67aQQ9jHY9M dlblDUyU/MiP/Ejx9NNPF//yL/8Sgc02TKBxAp6EaByxDZiACUwaAZKjv//7v09J08///M+nhKG8 D3z1gbO1JDYkFGoqWkiUowpcbKGFM9i5Fi5/pajjzByFVURDB/tNYVs+A4i+SC0UARS4FEvlM4Cc sSQh5P2IxJyYoqDkDFb5TBcxhK+iim24cBaZpJciIW/igtYILhQr4kJxmzc4MYmF1qhJNIoV+nD5 zKj6Vhtc1q5dm2NJk2nENBMmVQXLnJVreoEPiAc45Gdh1c+juLA79BP6Uzle6OPEM31aX3eqafd7 bgYujGn4iD6shhb6NZNrUT7SfjMJkTfGYSZgmUiKurkh+w0XtORcOD4RQxwjOG5FNPabeNAEp2wy /uEjdERx4TjMMZLYzbmgCS6812sC+AMf+ECa0OHeEFG5hVj50QSaIPDGiNnE1r1NEzABE5hAAlzu /NRTTxXf8z3fky6DrtoFJeXlQpt1KZxI2El+IhoJCUUaSVW5aRlJVkRRSZJFgSC7uR4t65Vk5evW 8ZyEjrOD2M0LJ7YtLSSg6G26YYdipEoLiTmFb1RiCRPigSS4nAgTuzT08pWkphtaKBBgUPYR2liO 1oizyRRN9Fns4o+8KV7wIXHVdIMLX8Oo8hGc0IOOCB9p4hC7+CNvil3Gn4hCDjuMMVU+QhfxG+Uj xg2NZYoPsZE+1omIF+xS+Ff5iPfwE+wijkn0VSZEaGUu0geXiHhBgyaay1p4T/EcwYXjL7ELH42z aFBDC+sw9lY1vlbJr2Xw9UK+zuFmApNOwJMQk+5B6zcBE6idwN/+7d+mwuhnf/Zn7ymMMEYCpTNt 5WKF97UsKvlUAkXiS5KXN5bRSGyaLuRIoODCoxiUtaBPevP3mnjOPqMFm2UuvIYNxV7EmUrswF9J b76/8hsFZURRSVz24oIufEfhieamG3b46xUvsOF99bcm9WADHyk2yrbQGFVUss/EAjZ7xS5FZcQY AxdsoaOXn2AV0a819qJDY1vuJ5bjw4h4yftrr36Ntl5FZa67jucU9f1ilz4fES+KXfapH5cILWgg LntxUQxpTGT9ploel736ETp7xQufISeB72c+85kQXzbFwts1AQh4EsJxYAImYAIZAc5sffnLX043 GeO711WNZIJkmEbSUG5axjokfk03khKa7Ob2tIwiomktcMHOIC2sx1/TTZMLYlC2x3J8JF+W36/z tfhXaWGZlsuXddoubyv3kezm67Asj/H8vbqf9+OCLemT5rrt59uTDWzKbv4+y+hDWi9/r+7nGjuq dGBLy6PiRWOH7Ob7q2Xqb/l7dT+HC7GJTdnNbWhZZJ/O7Vc9j9AiLlX2WSYuEbGLf2RHRX6uK1qL +MtuLy2K8fz9Op/DRf21nxbprbLNDWC5kTD3hvjv//7vqlW8zAQmhoAnISbGVRZqAibQNAESBK6C 4KzIu9/97vR91iqbJCskFLSqZEKfYZ1+CYXWm+9jPxvSR1LYdJLFfkiL7Jb3jeXo0Hrl9+t8LRu9 tGCLdSK5DNo/JamD1pvP+4rdftuI8tEwWtCpoqaf5vm+p3jptR3FUYSWQVwitQzbR7rARb4b5Eut N59H+oj8JH/k29OyCC75/spurkXP8/W0rO7HnEvVtqWvC1ykZdgYr9qfUZbNN16Y1PmJn/iJdKXP F77whZBj1yj753VNYBQCnoQYhZbXNQETWNAE+FWHL37xi+kSUn6XWwlK13e6n86IAjvn009Lvl6E rqqzcLmGyOfDconU1M9WhH/62S+/F6FnkI/QEKGjvO9tv4bLIDZtayzbnzY/5f7pt+/93isznM/r XE95O1EasJvr6Ge333tl/fN5LT1V9rRMj73scK8qbsDKL3jxC11uJjCpBDwJMames24TMIHaCXz9 619PPyvGJY/8fFavxncz9Z3OfgkDRbDW67WtOpZXfe+2vF3WUQJUfq+u12xfWnpxYXm+Xl22q7aj G5H10sJn8E/TXLAjLTwvt1xfv/XKnxv39TA28hgf184wnxtGC9tRXA2zzXHXGcZGVOwO0qKYGZbf uEz4HFqG6SNRWrQvYqDXPGrZIH75Z8Z9Th/RRKfs5tvSsggtg2xEahl27BikOWc57vNB43vOZZgY H1cHnxt27BjEha+Jkp9wI+P//M//nI35+WjzZ02gDQKehGiDum2agAl0jgDJyFe+8pV0eS1nGvjp uV6NZILkhqbLK/N1tWxQApR/Zj7PlfzLbr4tJVmsE5lkyW6uRfrgouQ9f7/u57oDeZUWbKEHLfJl 3fbz7SkWxCB/L38uX+bL6n6uJLcfl2ET5vlq68cFfdIYwUU2crv5/uE7uGi9/L26nysue2lpgwv7 KLv5/mpZBBfFpWzmOvLnEX2afqRxrJ8e9bdcX93Ph+USoSWffK8a78QqQou4wFt2c/ZahhbWbbLl WnpxQc8gLvwi1w/+4A+mrxLyK14RN2Ftkou3Pb0EPAkxvb73npuACWQE+Bmv559/Pi3hZ7D6JSQk WfwOPK3qe61apnXSig3+4zfgaVXfa9V3gElclDA3JQVm7DOPYpDbkj7pzd9r4vnixYuTFhI+JZuy o2VwiShYmBDpxQUt/JF8auJEOpt4ZJ/RAhPs5k1aKCgjikr2mT/FaZUW3kdz0y3nUtYjLvShCC2K hbIOGOA3lhO3EVqwgS3slvu1tKArol9r7GX/y7GLBvRpHOJ1ky3vI/24RPiI/eQS/dwf+b7DK4pL 3l/LXNCkmI48PvbjgiYdt3jeVFPssv35ciFHob9xc0r9TGxTur1dE2iKgCchmiLr7ZqACUwUgYMH D6bLG0kUHnvssYHa169fn9ap+klF3SV+1apVA7dTxwpr1qxJEwzc1JBkK2+5FpLQphv7TBJa9fOO 0rdu3bqmZaTtk1jyR8JXLljkN654iSj8KRCwIwY5AJaRmMMlwkcwQU9VIad4IcFlEqfphhbsVPkI n8Fm5cqVIT4ibrEFl3KRwGv6Fnr7XSVVFy/sLF++PGlR0aZtwwU/4UPWa7rBhX4tf+T25DdiG3ZN t1xL2UfY5qww8RThIyZmVq9enXa5PN4phtC7YsWKprGk7a9du7Zyoog3Ge/ErmkxjGEck2Q3t6cY gl0kFzT0Gnt5L+JYDRcdH3X8wbYa+sg/hulHGzZsKLZs2ZJ+zpNfynAzgUkk4EmISfSaNZuACdRO gMsaSQK2bt06VAKrIo3f9M6TYZJPfpc8KhEGhAoj7ObJDQkfvx1P8jlMYlMHVO03OsqXifIdVs4K UlxFNBI6kj605EUCxSRXvvA+yXJE4Z8XLHDIG34jbigiIhr7jS1s3rhxY47JK1euJB5MskVwwYYm 9PBJ3ohd+qQm2fL3mnrORBD9psyFs430840bN4ZxUWxevXp1zu6qn+PDiAk0jMOFGIYDfNTwERMi aImKF7hwFUI5Xhhv6OeMLxGTMzBgv9FS5sKYAxu0RvmIop7xtRwvMOEXn3ifiauIxrgLF8a6/PjI c/oWPuJYEdGYkCIe0MJ4ooYWjt/oiOIiH8Eg58IxiWXkFcMcq+G7c+fOtCvf+MY35uyX9s+PJtB1 Ap6E6LqHrM8ETCCEwNe+9rVk5+1vf/tQ9ijktm3blpI7flWDJIK/s2fPpmUkCFGX4TLJgBYS3/Pn zycd7MS5c+dSIrxjx46QS+uxSSHCGRoSrJwLBQNJ4ObNm8O44CNu4MXj8ePHU7GERnSghwIi6mwc dvERCTF3NNcVBxQH+IxihcQyosFj06ZNKTE/derU7MQVWniNDvRENSYhSLzhookrHk+ePJmSck1S ROjBFgUJtqWFIg4uLCdmohpaONOObWKWRh/nNUUVPowo/LGrmNCYwjL6OD6jyO53I1/WrbMRK3C5 dOlSwaQZDS5HjhxJWuhnUVwoYPHDxYsXZ7XAhfihRWqhsCdmGNvEhQkjfMQEEhPsUVyICbgwOcP4 RmPSEy30J46PTFJENPoKxySKfNhwnKahhXF4ZmYmjAvHao6BxCu5Ag0fcayEC/EyzKQVcadJiGef ffaeCbm0Yf8zgY4TWHS3M869drfjgi3PBEzABOomcOzYseKDH/xgStR+93d/t/ipn/qpoU2QZJ0+ fXr2TAQTDyR7kcWtxJJkoYUEh0YyQ8KjS4a1XsQjZwEplkisaCRfcKGYiUqEtZ8qItGEbf44o80l rdFaKFCYEMFXTAZgn7PM6KFQiGycFVSBiw7SAbREFivaXwoUuHCWX42Cn0KG2IlscKHYpmBRfNCH KGSitVCg0KdVVOIj+hA+iirixB4tFE4U/zRpgcswhZO2U8cjWihuKf6VxjJJxGRI1ORvvh/iQhzT mBCAS9TZfmmBC3GLHp7T4EJxG60F2/jnwoULs8dHNBC7EV+Xwb4aMcIVInBh4oF+jRZ8hK+iWzlv oP9wsmAULv/+7/9efPzjH0/HjT/6oz8q3v3ud0fvhu2ZwNgE7vbBRZ6EGBufP2gCJrBQCHz2s58t PvWpT6UC7JOf/GTxjne8Y6RdI9mjuCSx4S+6UMnFooOEiz90SFO+TtRzccEeBXebXOAhNmih4IdN G62rWhS7bXKhiCNuuuIjaVFfaiNesEnsogXfELvwaaPlsdt2vORceI6P2uQiNm37qEtc0KLjAFzU r1neRtMxAF+pT6OrjSYu2KZPjzoJzcmTX/qlX0oTPB/96EeLD3/4w23shm2awFgE7va7RbGnF8aS 6Q+ZgAmYQHMESEZefPHFZIDLfDk7PmojsYo+E9hLY5uFfllTl7iQaEafOS7z0GtrEYm5j3DpSvx2 yUdQMpe5saJXXeJCzPg4IM+88dil40BX4gU68+XC1ZZcPXH48OFi3759bwD3MxOYEALtTKVPCBzL NAETWPgEuFSfy0VpXLbK5ehuJmACJmACJmACJtBVAnydZObu/Sxo3A9FXwVKC/zPBCaAgCchJsBJ lmgCJtAcASYh9B1n7p/QlTNZze2xt2wCJmACJmACJjDJBMhVdNNe3WNikvfH2qePgCchps/n3mMT MIGMAN8R1ZUQu3fvbu0+AZkkPzUBEzABEzABEzCBngT4OgdXbvLIzXS5sa+bCUwSAU9CTJK3rNUE TKB2Aty9Wz/Fp5+8qt2IN2gCJmACJmACJmACNRHgHiT8ghD3uSCH8SRETWC9mTACnoQIQ21DJmAC XSRw4MCBWVl8HcPNBEzABEzABEzABLpOgJ925msZ3A/i0KFDXZdrfSYwh4AnIebg8AsTMIFpI8AN nWj8PNbq1aunbfe9vyZgAiZgAiZgAhNIgK9jPPjgg0n52bNnJ3APLHmaCXgSYpq97303gSknwM9z 6hLGZcuWdeYnHKfcLd59EzABEzABEzCBAQSYhOAECu3q1asD1vbbJtAtAp6E6JY/rMYETCCQAHeU vnLlSrLIb27ff//9gdZtygRMwARMwARMwATGI0DewgkUmnKZ8bbkT5lAPAFPQsQzt0UTMIGOEHjp pZcKfqKTxoHcP8/ZEcdYhgmYgAmYgAmYwEACK1euTOvcvHkz3Rti4Ae8ggl0hIAnITriCMswAROI J8AkxKuvvpoML168eOyf5+RrHd/+9rcLHttu0oKeLjRxaZsN9sWmK1zEpm094tK2j+BgLdXR0DUu 3AjP8TLXV13zEeNLF44D5jI3TvSqLi5Lly5Nm6RPXr9+XZv3owl0nsB9nVdogSZgAibQEIH8Sghu 7sTvbY/amMQ4d+5ccfv27fTdTBIC7ljNz2ZFN34r/Pz588WtW7fShApa1q9fP3vjqkg9JERw4ewM PyXGJA+/PtLGV15I9i5cuJASNHShZcOGDcWSJUsikczaQgtfBaJAIO7wkS6pnV0p4AlcLl++nL5L /NprrxUPPfRQ+sk3LvFto126dClxIY7hwved29ACF5J59NC/4bJmzZqCM47EcnQjVvATWuCyatWq pCdaB1zoz/ys8SuvvJKuHIMLN/Rtgwta8BFXszGu4B/6UhuNMRct/FQiWohb4necY8p89aOBsRcf cb8AtMClLS3EC8dHjomMc4y9uo/BfPd1lM/Dg+Mjx31YLF++PHFp41hdzhs4FnF8HIdLfhzj2EKf dDOBSSAwesY9CXtljSZgAiYwBAGSNQoeGl/FGDUBIAl+4YUXUrK3ZcuWNPnAsmeeeSYlOkNIqG0V kmC0UDwxCYIeki2W3bhxozY7w2wIrtileCKx4o8iFy7RZ2qYdNi/f39KPinetm/fngqm559/PiXq FFZRjVg7fPhwcerUqVQwbdu2LcXdwYMHi9OnT0fJSHaYADl27FjSQjIOFwrcF198MS2L5IKPuEHs 0aNH08TQjh070iNc9Os1UXDYb7Twc3ck99KC31geyYV9Ji5gwPiEFnyFFthENvabwpa+RJGNFvoT TNAXfcYdLfp5ZbQwAUG/In4Za6IaXJggYrxjIoY+DZczZ84kVpFa2GcK/ueeey7tPseAtWvXpglY WOlYF8UGLviDghsuaGGiBn3RWrhfwrPPPpsmrDgecYzkRo779u2b/UpmFBdyBNndunVrmghhkgYu HLNHbboSgs/5vhCj0vP6bRLwJESb9G3bBEygVQIc8EmQaCT5o54pOnnyZEqmSLBIgvmbmZlJCTkF HsVVRKMAQAv7g33OTFKs7Ny5MyXGaIksnkjASbTQAhPOxPGcFskFeyTlJMMk5JyB40wcBTeaKKDG SfrY7jiNpJefUSPxJAlGC8/xFcUmzKIaySqFHDo2bdqUtPDI2Vu0kBRHNfabAlITViTVaOGMHrwi J66YsCNm6EPEjM7cEjvEdWSSzxl++guxyhijK5soujmjS1xHNSYWGWPQQv+BC7FC/MKLM7BRjTPa aIEHY5y48JwiFz1R450m0Li6Sn0ZLmghronfKC0U9thjMh37+Io+xHPiGmZRWrDD+ErT5Bla9uzZ k64WOXHiRJgWJoIYX2iw0LEaXcRSpBa44Af6NsdEfMRYs3v37lkuo05cEXtq9FM3E5gUAp6EmBRP WacJmEDtBJiA0BkZLrkeZRKCz1FUkkTkSQDboaikWIkqKkmkSDJJZnItnNlmGUVc1NUQJFBwoTDg Tw22nAmDSRQXJmcoSPAJZyfV0MLlyfifoiWikXxS9PNIkZI3knO4RRWVcGG/4VDmQvEPF7RGNQpY NGE7v6wfH1FQUXDDLaLBBS35Zf2KFx4p8qIa+40WJmdyLvRpJk3RMmrBMq52+jS24JJrIX7QoiJv 3O2P8jnGVsY8tOATNfoRYx5cNK7rvaYeudqLCTsm8/LL+jkGcCVNpBbGeK6Iox/lXDhGMQYyFsIt osGEP8Z84kMNHcQv/SxqApjCHC5lLUyk4Sf6WRQXtOAnYjXnwtVFcBknb8iP+brRtnj70QS6TOCN 0bvLKq3NBEzABBogQKKqqxXyA/kwpjgrSoFAUlVubIuiKeoMLokwhSMJeF4goItlNNaJKORILkmE 8u+pJgF3/4kV60Q0Ej58QIKXJ+XYlr+ZEFEMNKmJhJxEGC3lr/2QjFLAUCRENLgw4YHNvHDCtrjA jZhquqGF2KzSwjLY8H5EUUkcMPmCXfUb7T+viSF8GJHoy0fECjGTN3SgBy0RxRNcGO+wqz4sPYpn 9BLjTTcmQpi0QktexMku4w7+gU3TjfGfSQZaebwjhmCDf6KKbbQwxldxIV7oQ/Slphtc9PUyjSey qdhlbInwEXY1yQmDfsdHaWzqEd+gBT+UuWATffS1UfMGYk2NfuhmApNCwJMQk+Ip6zQBE6idQF5k lRP9fsZIspRY5gmAPkOi73hsJgAAQABJREFURYtIyrGjqxywW06yVPCiBd1NN7iQbMlubk/6pDd/ r4nnJLlokd3chpZRsORxkK9T53PswL9XvOC3aC3YLMcLr2FDsReR0GKHpLwqXqSPQi6i2JaPZDf3 v7jgw4h+Lf6K01wLz1kON41D5ffrfA0X9NCwmze4yHcR/VpjLzrKWqSPQi4idhk3FJe9tKApIl6w w4QqOuQPlqmxnLEwIl6IFWKGVqVFyyK55HEqJrk+Hbfy9+p+nvMXg9wGy9A5Kpc89sQ9366fm0BX Ccw9mnRVpXWZgAmYQAMElECy6arisJdJEuFeSTmfUYLBOiQeTTclHiQw5aZlFCxNa4GLCnrZzfUo WYriIv9WaUEXy9FC0dJ0E38xyO2hQxoVV/n7dT/PfSS7uQ2WwSSKC3HTi4uWw6/plseL7OY24UIf itBCHHSJC/vN/ldxEaMoLrDppUPLI/oR+4uPaLIrFjyqb0VoQUc/O9IS4SPGDWnRsbCKi8ah/L26 n8NFdsQgt6FlcIk8PtYZLzljjWH5Pvq5CXSVgCchuuoZ6zIBE2icQH7Azg/kgwyT2Cj5VBJT9Zl8 var361qmYrGfFtZpOsli+4O4oJH1pLkuBlXbkY1eXFjOOtJctY26lknLoO1FFQn9dMAFJsNq7ret Qe8Nyz6CyyAtil0VWIP2bT7vD6OF7Udp6Td2qH9F+0h2c85aFsVFfpLdtrTkfbWflny9XGudzweN 79IXoUX+Yf9kN99XLUNLvxjPPzPu85yL7Obb0rJRYzc/gaITEvl2/dwEukrgjS8SdVWhdZmACZhA QwRImpWkKAGo0xRJR9OJDXr72dB+aT/r3L+qbfXTkq8fpSe3WfW8Cz7KdU0bl2H3d9j1cpZ1P6cv dS1eIgo5OGoc6cc0wkeTNr7AK5pLP0YRWvrFCO9JX4QW2crtVumL0JLbzXVpuZaNqoWrKvSZ/MSK tutHE+gqAU9CdNUz1mUCJtA4Ab7DrJsBjvJ9Zg76unJCiUOVWBJ3rVf1fl3L+t3PQvo4WzJMITEf TWxfZ2Vkt2p7+XpV79e1TFz6acGXVZfG1qVB2xEXvc4fc3391ss/M5/nw9iI4iIfDdqfiH40iAt+ iurTw2iB2bD8BvHt9/4gLfpsl7QMq1nax3kcto90IXa1f1FcsDOoII7QMsiGxt5B64nffB4ZO4bp I6Nq4eoH5TGj3tRyPvvjz5rAfAl4EmK+BP15EzCBiSXAz8vx02o0fjJr2EbyqcRSSUz+WS1jHRKP ppvuhi67uT0tk978vbqf55MLspvbYBl/wybv+WfHeT4omUMLSSF6mm5ogU8Vl9z2MElqvv44zwfF gnw0aL1xbJc/oz7Si4uWK8bLn6/zdb6/sptvn2V5jOfv1f1cfaRKB7a0PCpeNI7Jbr6/WhahRWMv NmW3Ssugvp9/ZtznxIvGjn5aorj022fpy2N83P0e9Dn5iPVkN/+MlvXTm68/n+fqr1z1KLv59rQs 92X+fp3P0SL+sptvX8tGjRd+alR5TP6Ty/m2/dwEukig+cyri3ttTSZgAiZwl4AO+sDInw+CQ5Kl BKrqcmgtiyic0Co7XJJZ3g9dpsk6SoAG7d+475NkkUDxKLv5tqSPnyKLaPo5QdnNbbKMP5g0zQW7 xAtcFBu5FnyGFt6XL/P3634u/tgtxwuv0QiTUZPhcXSKP/tfbtFcxF52cz3iohjP32viufjDBdt5 k76oeNHYIbtVWlimuMrfr/u5+MOlKmaIXa1Tt+3y9ujTGjuqvscvfRH9CG3w78UlUgtM+h0fI7XA hfitil3ey7UQN022vL9WHQfQgs5R44XPuJnAJBLwJMQkes2aTcAEaiEwn4P36tWrkwbdeTsXpGUr VqzIFzf2HC0kOFU3hpOWlStXpnUaE/GdDWOHJLTqBlnSsnbt2qZlpO3zW+wk5iR8SjZlWFqWLl0a Ujxhh+Sy6owcBQz6YKczq9LZxCNMmKCBSzkZVgzxPpqbbthBTxUXtMFm+fLlYZMzS5YsSUzKRaVe U9BE9Gvxr4pdxi14Ed+s13TDP3CR3dye9BG3EWdhGVvoJ9jlr9zo1+glZppuxAJaaBpPZBNtxAxc IuIFu1zNxziiWJUWHomXSB9pn8tfyVAMRXLheCO7ORNx4TEidrXPPJa5SAvHcbFj2aiN/XQzgUkh 4EmISfGUdZqACXSKAAkfCTG/6Z0f+EkC+Y16Cs6IRBgoFAgk3mgpJ+ZooUUkWdhRwcgkRLnwR4sS MdZtuuEDCmm0lLncunUrmdcETtNaKFjQApPyBA1c0Ldu3bqmZaTtw4V4oDApF083b95M65C4N31m EEMqKulDL730UrKtfyTq6EMr60U0fEARV/YR2vAR8RKhJeeiWNX+4zf0UaxETEJgd/369Sl2GWPy Jh/BRWe/8/frfg4X4oHYLMcLfmMZTJigiWj6Gl8vH2k8jNDChAg+KGshbhlj4BJ1TNJYhpb8+IgW fIR/IiY54U4/gQt282OSjtXEVBQXYhct5bwBnejjOBGlBZtuJtAmAU9CtEnftk3ABFolkCdHowqh mN6+fXtK+C5cuDD78StXrhTXrl1L70UlWUxAbNu2raCA5AZV2q9Lly4V3KgKnVHFClzQQnFy+vTp WS5og9OGDRvCkk+0zMzMJB4nTpyYnYggATx79mwqZpQszwpt8Ala8MPx48fTBACmKLLPnDlT8L1e Cr2IJh8xGYEWnTnFZydPnkxaNm3aFCEl2SA+6StHjx6dnRRB07Fjx1JSvmXLlpAJEcSw3xRI2NYE DVyIH3ihJapt3bo1FVD4SJOJcIEThdPOnTvDuNBPmGg4depUKpZgQBGHNuJpx44dUVhSP6GYwycq uMUFEfQzNEU0CtzNmzcX586dK3RzY7gwvhA/+ChicoZ9hQljCLbFBS30aSau4ELcRDT6EFwuX75c cByicVyCE8ckuHDcimiMLfRrdFy9enXWJOMuhf+uXbvCuLDPcCFWlDfAheM2XBhfOBa4mcA0EPCN KafBy95HEzCBRgiQTHBmh2JbCSjFNsUD70WcSdaOUSRQLCkZJtkkqaFwoMiLbJxFpyhAC0keWuBD IhhZOLHPJKB79+5NBcuBAwdmrxghSd69e3dY8okWrlhBy+HDh4uDBw8mLfChSEFLVLGCFvZfXPbv 35+0UbgQs4888kgoF4p7tMAELfiMiSIKyUcffXTk70izf+M2igR8ceTIkTlaKOYoVqIm89CPL/bs 2ZO0HDp0KHGhsOUPH3HWNKrRh5loIHb5I5bRwRUZ9Gl8FtXggk0mIeDCmWPGG8behx9+OPRMsiZg 4AAXJiXgQr+euVv08zqqwYUJYFjAhWKW5/QleHEsiGzEC/2GYp9jEc8ZY9CiK0ii9HBMhgUTMkyM 0IgXjkm6qWOUFmyKC8dF/MaJC7hETnJG7a/tmEAvAovuzsD5C0S96Hi5CZjAgibwZ3/2Z8WnP/3p tI+//uu/XvA3TiMBJemkUdzx11ZDC2dNGdrRQdFEktNGY1IELmhBB4VLW03JOJNGFL0k6G1xKWuh gIs6c1vmDw8KAx6ZBKGga4sLGiiY4IMWfNQWFxVMnEFWvLSlhf5DwQQXNMAlcsIqjxm04CMKbWt5 gwxcGHcZf+HCWBc5SfSGkv97Rp8mdunLbWopc+GYFHUFRJkJr+UjnotLG+MdXIgV/mg6VqcXI/77 u7/7u+IP/uAP0qc+9KEPFb/zO78z4ha8ugnEE7jb7xb5Soh47rZoAiawwAhQYEeeIe2Hr0taSDbb TDhzThRtuolcvryN513SwhnurnBBS+SZ436+p5DsihaKpC5pYRKkCw0uXdJCUctfF5q5VHuBYp+/ thux2xUtbbOw/eklEPOluenl6z03ARMwARMwARMwARMwARMwARMwARP4DgFPQjgUTMAETMAETMAE TMAETMAETMAETMAEQgh4EiIEs42YgAmYgAmYgAmYgAmYgAmYQDME+JqHmwlMCgFPQkyKp6zTBEyg dgL5ATt/Xrshb9AETMAETMAETMAETMAETCAR8I0pHQgmYAJTS4CJB+5STeNGdG4mYAImYAImYAIm MCkE8jyGm/u6mcCkEPAkxKR4yjpNwARqJ8DP8PFzczR+cs7NBEzABEzABEzABCaFgH7aGL38LKub CUwKAU9CTIqnrNMETKB2Avo9dzbsMwi14/UGTcAETMAETMAEGiSQ5zG+orNB0N507QQ8CVE7Um/Q BExgkgj4oD1J3rJWEzABEzABEzCBnIDzmJyGn08KAX8JelI8ZZ0mYAK1E9D9IGrfsDdoAiZgAiZg AiZgAoEEnNMEwrapeRPwJMS8EXoDJmAC006Ae0u8/PLL6fuYbScB2H/llVc6c48LtPA91S5wQQd+ wl9tN7TApitausLltdde64yPuqiF73+33dBAvMCn7SYt5jLXE/BgfOmCj/Lj41yV8a/QomNSvPW5 FrukZa4yvzKBGAL+OkYMZ1sxARPoIIH5FsZ8/tq1a8WZM2eK++67L/3CBsnfxo0bi9WrV4fvMVrO nj1bcLds7nGBlvXr1yctkZdrwuXGjRuJS1nLqlWrwu+/cefOneLUqVNpMgQ/oW/58uXFhg0bkt8i HcUNUI8fP5603H///clHK1asSH7idWQjGT99+nRKyuFCUrxkyZIUvw888ECklDRpdvLkyVkt+Gjp 0qXFunXrioceeihUC1zOnTtX3Lx5s8An9B140K+jtdCH8RH9CQ1wefDBB4tNmza1ooWx7vr167Na xIW4iWxwwUeMefiI2OURLmhh3Ilq2GbcxUcaZ9FC7NK3IxtcLl++XFy8eDGNs7rXEbHLmBfNBS0X LlyYPT4Sv2hZuXJlJJZkCya5FvwmLZFcYHD16tUUv8QLfywjdvGRYigckA2aQDABXwkRDNzmTMAE Fg4BEqwDBw6kRHP37t3FzMxMsWzZsuLFF19MiXrknl65ciXZpUBBB38k44cOHUpJaaQWipSDBw+m JBguu3btShMh+/fvL9AZ2bjiYN++fanIhcmePXtSwU/By8QEyV9U4+wxDF566aVix44dicvmzZuT jiNHjoRq4QwpsYGvtm/fXuAntJw/f744ceJEqBYKp8OHDyctW7duTT7asmVLmsRCY6SP0HLs2LFU IOCjvXv3pkKFvo6PeD+yYZPiSVzQRB8ijiK5ULARF0yI4BviBS3ED76jn0U2fEQfZiKRPr1z5850 dQZcIs/+4wMmIGDDxDNa4EJfJ3aZ0IpsaGHshQuxS9/Gd21pIX6ZiBEXCmzGY/hENnFhElzHaiZo OH4zeRTZ6M/4g0lWuHB8XLx4ceJy69atSCm2ZQKtEvAkRKv4bdwETGCSCXBGmzOjJOWc+eKPQk7F P2fgIxqFEUkw9kk60aTnJIAkPJFFAoUKyTnJFWdKOcvOGScSQJLSKC6whwsJL1rwC4nnmjVr0lkn JiE4IxXVKPA5c0sSzGQVXDgjSIF56dKl0MkiknIKSLigBS7EyrZt21IBjp6oRlKOPbjAAy2cEUQb vDjjHdXQQZHPhBU8xIUiV1caRWnRvhMfFLhooVihuKRYoQiPmojgqhD8IC3ELlrwGZNq9LOohhbO aHPmmCu9ci709aNHj4ZxwR77zvjGn7gQPxT/aOExouEHfMQVGHBBC8cC+hHHCCYnorRgBy70IWIG LUyQ0494REvUMYmrzzgmMbYQMzpW4yNpieSCFo6LHKulhYkrtHCsjp6giYhN2zCBKgKehKii4mUm YAImMIAARQDFNGczSILVeE5Rx3tRZzVu376d7JHwkeypcYkpxRxnBiloIhqJMGeW4EKClTf0kXhG nXliv5lkIBEnwcsbExEUcLwfUchRBFBs4xPiI2+w4gwhZ9sjGlywhX8oJPMmLhTjEVxI/iko4VK+ RBsuJOtojTq7zUQRfYh+kzf1LbRGcUELXJi8yxs+I6aJ3ahCjtil4SM0qTGxh4+iuGCX2KQ/lbnQ x9HD5FrURCdciE0miXIu+If4ZUKLIjiiEQ9ceVHmgn/gwnGCv4iGj+BCvORfL6Bv0ZeYSIryEbbg Qp/Oj9Vo0bE6igvHYY6RZS74BC46fkb4yDZMoG0CnoRo2wO2bwImMHEEKEKUdJeLW3aGpI91SH4i GokNCV+56Me2lqElongiEaZAkN18/7Us6isZJJacVcJuXiDkXJgoQm/TDTsUI8RGWQvJMMkxfozw EUzQQ3GQFwgwQB9N7NKLBv+hhWKkykdogw2xG+EjCnpsYTcvVth9xS4+pFBousGFYhodZR8RP+iJ 8hFc6NfYxR95U+wy/kSMd9hBC1zKPkIX8UvBGeEjtGhyV/EhNooh4jaqwKXwr/IRmvBTlI/YZ7TQ ylykj3Ui4gUNaGFcLWvhPcVzhBY0ELtMvGqcRYMaWlgn6uSF7PrRBNoi4EmItsjbrgmYwMQSIImg SKBVJcJaFpEIo0FJC4kvSV7eVMCgpelLTkmgKCh5FIOyFvRFJeXsM1qwWebCa9hQsEScqcQO/Ptx QUeEln5c8BcaKTzR3HTDDn+K09wey/ijeFJ/y9+v+zmxi48UG+Xtw4XiKaJfw4RYYP97xS5aIs4m YwMfoKNX/MJK41CZW52vtc+KjfK20adxqPxe3a/xz/9n701jLEmqPF+LjIjMjIyIzIjIyH2vyqwq ikUDPcCHRl018wTdwOtlaBgJuqWiXzNIQw+tJ9TAxyo+Qs2nQdBCwGuQhkYM3SwjddEPNHrFeyA1 0KqiC6iuohZyXyIiIzJjyYwtM975md9z3a6H3/26xY3IczJvuLuZ+bG//c3c3c4xc3O9RtSYDfMA IxKjvZAP/Ndqu/ASAwttRe9jtXiJ0Xbr8RLWEfwUKdxbtMzVriPqL0YdFVlO020MNMqAOSEaZcrS GQPGgDFQYoDOBJ1hJGsghGGkKbpjQ34YLGG+/qD0R/HRMSwaC7yQTz0spONXtGhHWDnI5kc4daR1 mY3v5DF1BP95WAjTcK3LTuad1RXWkeYbpiEsbONhXKf3tV3m4SAvDVfMnc4/1Kd5kKfmG8YTRh1q ujCu0/t678jDQV4aHqu96L1D8w3Lq2F6vYVxnd6HF9omeWq+YR4aFqOOyKMWL4orBhblRfPMbmPy Qv1omdXID/HExqL3d823GhatyzC+k/vwotdrPSyt5punt1Vddp4xUDQD5oQommHTbwwYA13LQPjA DvfrAaazQocCqXUeabQDVE9nO/G18lB8YYe5nbzqnatYNN9sesLhT9Nl4zt5rHlUw0JepCm680k+ 2l7YryXaSa2Vpt24RrDEqqNGsFBeNWraLXut87W9VEuj7SgGlnq8xMTS6DXSDbxo3dXjT9O1sw2f A3l6tI7qtau8c5sNC/PQfPN0hOny4jsR1igvMdpLWN48XjSMdDGfA5pvyLeGhZjDeNs3BrYaA+aE 2Go1auUxBoyBhhkIR2nC/YYVdElC7bzkwYnRsQrzrYUlTBcDVzfVaaO8hBxt5H6M+mmmfDHw1Ksj MMTA0QwvMdLCSz1uYuBoJo97rZ7C+qlV9lpxzfBbL22IJ5s2FgbyDXHUyrdWXBZ/O8eKJy8/DdNt o/moTtLnvebRqB5LZwzEZqByhaHYuVt+xoAxYAxsIAOMOOj6BM2MyvCg14d9rQ4DRrCmK7KYee/d ZvMjTdhZycZ34hj9iqUaL4SH6TqRbzUduvhXNSycR/0UzQv55C2KRjgS4lPMSUwxf2th0RzDNq5h RWwbwUK+jaZrB2MjecRqu/WwaJuJ1V4auUbqYW6nbvRcvb9wrBxoXBgWpgvjO7lPHuro3Ggs9bhX fDF4afTeEQMLedRquyEvtdJ1ot00eu9olhdm/bTSj+lEmUyHMdAOA+aEaIc9O9cYMAY2NQN0CvK+ blGvUGFnIm/ar4bV6wDVy6fReDVENN/wPO1k0UmN2cnSfEMsig9etPMexnd6XzvmYOGXLT944C6G o0idHcpBtqzKl2LOxnfyWDu5yktWN8458Gq6bHwnj2vxEuLrBl6oO9pQDCzqwAw5CHnX9hKjjshD rx3NNw+L3ofCuE7vgwM8eTjIS8Nj8ELb5Rfm6w8yf2JgIUvaZbX7i0KKgYW2q/nk4YlZR5QbLKxX ovkqF2w1jDTaxsP4Tu5r20VnNV7Ao9w1k3cr/Zhm9FtaY6AIBux1jCJYNZ3GgDGwKRjQTkGzHRA6 WfrQz5tBoWGapmgyBgYGfBZ577Xq+6VgKdrwh8+dO3f6zpxyEJZd8SneMK6IffIBU16HjzA6fPCi hkQRGFQnBhpY8nhRLLTDGAYuZQYL5ddOuOJULOCIgYUy89N2qjjYgoUf8TEM3JCXLB7FEl77IdZO 72uZszjIhzojnHYL5qKFPCg3+Wbbr2IBQ4zrGhzcYyg/dZIVwmnbMXjh+qCekFq8xMAChl27dpXb BsehKC9wV7Ro2yWfLC+EgQWJgYV86vGiWGg3RYq2XfLoJC/ghnN+RZehSH5M973HgDkh7r06txIb A8ZAiYGsAdYMMfv27fPJ9RNt4bm6SvzIyEgYXNj+2NiYNxJY1DBbphBLjA4KZaYzlPdJRcU3Pj5e GBehYjq5/OjwZQ0WrbehoaEoBu7g4KDPRzkIcRJGxxxeYtQRnICHPLO8aHuh4x7DqAQL+eTVEdjg Zs+ePVHqiHZLXvCSNRI45toCL22maCGf4eFhj0WNNs0TXqgn6pB0RQu8cF1rfYT5ab3hJIK7oiXE kq0j8uYTiLSnGHWEE2h0dNQXOXu/0zYE3t27dxdNi9e/d+/eXEcRkdzvlLuiwXAP45mk+Yb5aRuC u5i8gKHavZe4GM9qeNHnoz5/yFsFfDgq2rmOss9/1W1bY6AbGTAnRDfWimEyBoyBrmcAAwBDbW5u rsJgoWNMGHExOsIQpZ3umzdvlr9bTzgdvtnZ2WidzxALnXL9JjrhyI0bN7zRhHEVQ7TTncVCR21y ctJ3+GIZ/nS6MRIQeAiFeqOuYjln4IW8aKvkHcrU1JR3hOzfvz+KQ4SOOXkh5B0KbZeOObzROY8h OBcxILO8zMzMeL4OHToUjRewwM/09HRF0eEFI4Y6jDFDhMzBQh3AA21VhXsdDpGY7YX2wCyEbHu5 deuWd35ixMVwzsABdQCWLC/c++bn533bjVVHlJtZF9evX9fq8VuwsGYA8Ty3YojWEfe60FnE9UwY WGI4OSkrzg7aA/mSvwr71BHP6li8aNvk/hLywjMJfOAgjYkxcC8wEOepfi8waWU0BoyBTcdAO6MG GAfHjh3zHfALFy54owDD4PLlyz7sxIkTUaYEQzoG7vHjx/0oGPmDA8Pg0qVLfnvq1CnviIhRQfBy 9OhRb7BcvHixzMu1a9ccRsLhw4ej8UJ5qSM65r/5zW98h5OOJ0YdncADBw5EcxSBBV4YCTt//nwZ Cx1P8GDkxex8UnacQdSRGvtguXLlijesdIQX3EULhhxGC9eRGgoYt1evXvWcEB9LyIuycx3RRmgv 8MMx9ROTF0aTaRfUEW1EjSaua+qOOuR6iyGUHUfDxMREGQvXM9gwnLiuYwlGJbzACXiUl1deecUb lGCJxQvG7ZEjRzwW7nHcd+GFOsI5QVwsLDi9KTvXDryABQfEuXPnvMEfkxeeSZQdLNxTFAv3PpxY 3JfhJ4bgBCI/uKCOeD7ygxf6ADwfY9WR8oIDQp/POMmVFziL5bSKwb3lYQzUYqBHLsC1Wgkszhgw BoyBrcrAF77wBffFL37RF+/DH/6w49esYDTRmaAjjGDw0pGINdU0xIuxBBY6WAidGUZuYxpOiocR JgwUnabM6Du8YITH6vApFjCABUzkzQ8DLtboreJgS+eTzi91xcgyWDB8wUMHNabQZuEFYxscdAfA Qj1xHFOYfQAvGC0qOCYOHjwYzYGm+cILDhBGlJUHriGMONpxTMFg45rG4EaoI64h6iiWEaflBQsG pc5AUCzwEttwAgsGJTOatBuLM0Sdjoo51lZ50VdncBLBS6zRfi0nvNBuwcM+Ai84QGNjIW/qRx1F HIOBthtrliB5IrQRZqvACw4RrmuwUEexZuYlSJK/2X4D1w8DCa3w8o1vfMM9+eSTXvEHPvAB97GP fSzMyvaNga5kQK7BnrhP066kwUAZA8aAMdA6AxgEOBwwLunY8IttqCh6cDC1lA4XP3CAZyOEztQD DzxQnnKKwb1RvDBaef/993ss8AIWDP6N4AYO7rvvvq7AghF78uTJMhZtuxvBC/XBiCRGHMbTRtYR vGC04QBRLBt1LcEDhjXORLBQN3BFeGwhT4w2nHdcRxvZXsACJ8yIUGObOtoIXqgH2gqORH0ObGQd wQlOs27gBSw4E5UX6gduYgttlZlFPCO74fkY9hvgAk42gpfY9WD5GQMhA+aECNmwfWPAGDAGWmCA jlXskcBqMDfK0M/D00280AmNPXKcxwlhhiWfGTVq82PjhnZTHVHybrmujZf8dggv/Ow5sJ6fbnoO dMt1BEvdxMv6WrMQY6B4BuK70osvk+VgDBgDxoAxYAwYA8aAMWAMGAPGgDFgDBgDXciAOSG6sFIM kjFgDBgDxoAxYAwYA8aAMWAMGAPGgDGwFRkwJ8RWrFUrkzFgDBgDxoAxYAwYA8aAMWAMGAPGgDHQ hQzYmhBdWCkGyRgwBuIwwDu8KuG+htnWGDAGjAFjwBgwBoyBzcBAs/2YteUFt7Y8y+dDNkPxDGO3 MLBNFvbePuR6+gfbQmQzIdqiz042BoyBzcwAC0OphPsaZltjwBgwBowBY8AYMAa6lYHQ8dBMP2Zt 5ZZbW7ppDohurdhuxnX3jltblLazutgWSpsJ0RZ9drIxYAxsZgb4hNnt27d9EfT77pu5PFsfO6M1 Onsl3A9KXiVYelrBuUH6it0wTbgfJKoSbPpLHHU1P0E92q4xYAwYA1uAAT45qv0Y/SxrQ8VaSfo+ DaW1RMZADgM4snr6dubENBZkTojGeLJUxoAxsEUZ0G9zN/Xw3qJcdH+x1AEB0nA/QF4luGr64NTK NFUUVQmuPLdCaXAQnhzuV0kSBJv+EhlVaGuMnwpC7cAYMAaMgU3PAE4I68ds+mrcpAVg1KF1MSdE 69zZmcaAMbDJGWAao37XXR/im7xIBt8YMAaMAWPAGDAG7hEGwn5MM69j3CP0WDG7mIH0heguBmnQ jAFjwBgoggFGEFTCfQ2zbZcxkFZX+8DydOWFtZpTnq68MNOfMtBJflKttmcMGAPGwD3BgPVj7olq 3jKFtJkQW6YqrSDGgDGwUQzw4OfHiES4SNRG4FEs5N0NoyK85qKc6LZlXqpOxW9BY56uvLAWVPtT 8nTlhZn+lIFO8pNqbXlPr6Vuuq67BQvXNfeXtq/plmsnOdHqKJ9A5YXYjX4OKJZuabvgMV48BfbH GNhQBswJsaH0W+bGgDGwkQxoZ6QdDMvLy+7q1atuYWHBv5c5ODjoDhw44Pr64t9eV1ZW3LVr19z8 /Lw3DsCyb98+t2PHjnaK2NK5LPQJL3Nzcx7LwMCAO3TokOvv729JXydOor6XlpbcnTt3BUefx7JR RtRdWV16cXHJFyvBsr0TRWxJx4q04ZWVVbetd5vnZKNeTQrrp69vm7wqtWNDjFxw3Lx5001NTTmu 7507d7qxsTG3Z8+eDcFz48YNd/36dY+Fa3lkZMTjaamy2zgJXrieJycn/XXEq2zwMjo6uiG8gIU6 Wlxc9O2W+uF+txHCPRcsLBDIPW737t1ufHx8QxwAYODey72Oaxks8LIRzgiw0F54PvJMHBoacvv3 7y+vYRCzruCD5+OtW7c8F8PDw56XjXhWZ/sNu3bt8s/Hjbr3xqwHy8sYUAbi95I1Z9saA8aAMbDJ GaAT/MorrziM/cOHD/vSXLlyxXd0HnroIUfHIpbQCQYLHSqwsAXL888/7+6//37fEY2FhY7nyy+/ 7Dt6R48e9QYKHfTnnnvOnT592htzsbBoPnfE6L944aI3orbJN657xeCmQ0znXNcF0bSFbmUQbuHW grsqdXN7adn19/W6u2LcHTx4UOpo2PUKtpgydX3KTU5MyrdDZHEzGdmGi/HxfY4OekxZXV31183M 9Izrlba7bVuPXFdD7oAYLP3b4zmuMLTPnTvnDUquI+pFr629e/e6EydORDW4L1265HmhnYIFA4rr HKfEmTNnolURvGDYXrx40TtZcbRi/MPVzMyMO3XqVFQjFyxwgxPk+PHjnhewwQv3mFiGJbxMT0/7 OoET7ncYu4qFOoqFhcaAwX/27FnvAKH9cl3BE3XEcyCmE5i6OH/+vL+XwAtOcrBMTEy4hx9+OCoW 6ujVV1/1z0Gc4QhYeC5RRzgaYwn9BrDgmD9y5Ig4xe94Tn75y196LDH7DbHKbPkYA3kM2JoQeaxY mDFgDBgDDTBw4cIF37Gig8UoHL+TJ086pirTEYz12U/yAwsGCvkzMokRqQYTWOgsx5LLly97Yx8s cMJIHPtITF58hvKHUadXXnrFMaJ87Ngx9+CDD7j9YjDQAX3ppZei1RN4Zm5MuxdffNGtiHFw5vT9 Yhjc58bFuH1Jwq5euerbjuIuckubuSqjgq++/Iqvn9P3n5b2ctKtSof4hRdeEMN7IfnqaJEgSrp9 /bz6ii//iZPH3QMPnJZ6Oir1MynOrJe8URcBhs9idnbWG3JcQxhxOnLL6C3tGmMmlmDkc71w/XCP 0ZlNGN2M6GLkxRIci9xjwMI1BC+M9GNEYfhiWMYSjHywwAf3OOWFfa5p8MS633GPxxGjBiX3XXgB C8YmjuBYWDDyyY/RdPKnrnDSsE+7hrNYWMgHXhDaK7yABUcIbQnnRCwsOGJwCiFwoc9qcNGWYmKh zNQD1zbPROqIe819991X5gW8JsbAvcCAOSHuhVq2MhoDxkDHGcBwwiChE0EHVIURFcKIoxMaQ+jQ 0MmkkxdiYeo2YUwt5xdD6AhTdjqdGAcqTAWmcw7OWFjIm04fhgkj/owk0wGlkz4iWwxL8BAfQ8By peRoOHLkqH9Npq+v38/GGBwa9CPNjLrHEBxWl2UkcFjaKrwwA4K2e/iQzKIRfi5evCAOiTidYdrL 9PVpd/DQQWk3u6V++mQWUTK76Ia025gGrhqwtA0V2i7HbBk9jSFJW7ninVI4Q8LXhjBauLZxisQy WHB4kBc8hFh4NYS2g2EVS7heMR6ZeUCdqHCvow3DC/fnGAIvvGqQfQWPZwAj2mDhnhhDcLJy/+B6 DnkBC7zQtnlWxBCeffDCPZ/2oQIO6olrmntQDAEHvGSx4EjjeYBDLxYvlJlnDrOqQl6YoQIv3Atj Ph9j8G95GAPVGEjv3tVSWLgxYAwYA8bAOgZ0hItOVVYIw4iI1ZkgHzrddGpCAwFc2tEhDZiKFjrl GAh5vGA4gY+OVizBcLp5c9YbcxhMCCyAY0gcJfADZmYmdEqqsbwkdXRTOqAYJzt2pB1z8h3ZMyId 4SWZrTFTF0Y1/XVPLCWgHczOzfp6UqeMngu2gYFdvqNclEMki1+dQDiuQsGJxVT2STFY1mTmRtHC qDYGCU4qvW40T44x7DAiYhgs5MEUerBkp/ODAzwYV1xrRQu88PoD+XINh4LxBEbwgqdo4XqmvYAl ywt544SFkxhYmE2kDrLsFHo4gRvuy7GMbdou13a27cIL9YYzJMYzCV54PiLZ54C2XeqxqPuLzzj4 Q3sBE7xkn4/6ekoMXqgbHEHUQ5YX4IKPay3W4EVAke0aAxvCgDkhNoR2y9QYMAY2MwN0aLRjmdcR pqOFxOgIkw8jKwj5ZjtZdIYRsIC7aIEXOluab5if4lO8YVxR+xgkc8LP2l0654kB1VPKjBF/FmNk jYZFmSLcKVH9WX1zN2/40eR+mf2gbUTTbBfHVY+4R2Zm6s9YqaZfddXb8srF7M05P9NhR8aJRvth DYZV6ShjJFCXnZYQP0Ya7YHrqK+vcj0M2tD2/u1uUerwVoQRXAxprhE4yF5HHFNnxMe4rjHSmLau 10y2DgjHmNH7UDa+k8fwAh4k227hRa/1GNe13nvBkcWi+DDk4K5ooe2qE6gaFjDEaC/kg/EKDq0P wlQI51qO0V5oK7QZJA+LhsXkJWynykmIT59bYVyn90P+lYMwD8LAGYuXMG/bb4yBhVu33ee//DX3 9I9/0tgJlqomA7YwZU16LNIYMAaMgfUM0BGu1ikntXYwSEPHg45FkaIdvrx8NAyDpQiDMiwXvNAx RzTfMF476rF4Ie/l5RVvbPeIsyFbDWDsWetxd+VrGcvCT9Eyv3BLHA1iyHkslW2iT8IAuLRU/HTp u2KkrawsS3uQttqzfiyCemKxTOqSOtX2XAQ/y7I4J3mwWKgQUJEF9QNXGJXL4ogYLHihVzUofb7C QVYI5xriWipauEbgRUdqw/xCfDGwwAvlpl3oNRzi0f0YWOCFX3ZGhmJQfKQpWigvdYRovmGe1BMS Aws4yCfPKQ4GxRKjjrhetcx59w7Fos8K8BUl8KL5aL5hXhoGL0U/q0MsG91eQg5q7T/9o5+4H/74 p+5XL7zkkw3uGnBvftPr3fv+8J1u3/jeWqduaNx/+fin6ub/2Pv/g5TlDXXThQluiRMCTpBHf/ut fvsP33/affXr33If/+iHmtbnFdzDf8wJcQ9XvhXdGLjXGdAOCDyE+/V4oTPBr955pKFDVq1jWC+f RuPJox4WOoVFOyHQXw8LPJMOPHkGVqNlrpaOcXs1ZdkHD/mJC2LdKWDpka8w+Hpq0GjJ6l+vdV02 5YDlZZk6LwrIMys9JWcAX/Hg850Y5WFe2fStHqOT8sI/KPKxCCeS8E4LbSbEHO5Xw8u6E2trMvsA mz9DCzacXpcrK4lRGeoM96vpbyZcr+lq52A4aNutlqZT4XodVdOX8lK8QwReKHc1iY1FcWi+esxW w9QIDuM6vQ8v2mY03zAPDYuBJWwvmu9GYaGthHhCHOwrvlppsue0eqz1E+Yb6lIs1FGtNh6e0+p+ yIvmG+rSsHbai+oI9ba670f8xeDeNz7m3v32R+U1wgE3OTXtjfBfvfCye+KTH+1aR8TDD52uWuyf PfOczH687U4eP1o1TTMRPLZwzvAzaY4Bc0I0x5elNgaMgS3EQDga0cmHt1JUdKemkXy0XLGwKKZ6 26LwhHYs+xj0SM6Avw/nD1jCzmo5Imcnqz8nSW4QNpwvMwry7Dk/aiqR4tsCC06IMK9cpS0E+uwF DK+niAumqgZxQ7g7gG5SQo3hfjU1d6SsUjvlf2G6MPdyPQYJGtEfJO/ILnVYVNttBWCj7bYV3eE5 eh8Jw7L7MbA0yn0MLNnyVzuOgSXkJdzPYqoVl01b1LFiuJd5UQ5CjjWsHV7CPk2ou9l9Rvv5MVOA Ef5QXisG/ufklQScFI9/8i/DqK7Z/8if/0lVLH8hszpwQOBc6YS86x2POn4mzTNgTojmObMzjAFj YIswwMNep2o2UyQe9DrNVDsOeefTcdd0efGdCqs1o0DxMRujEUOiHUzo11kfmm+evjBdXnwRYdXt 6dDULSLnRCc+Bs0p1/j3AMXAreEY6Ci6HkWzXqv3h3i3wPq4jodUrxjJKnEzVEfaWTTadqtppU3H uqYbwQLOWtd+tXI0G14Pi+rrJiyNYlbsrWx5DjRi9MV4BjRa3kbTtcKHngMn5KOvN2l4dhsDS708 9DlVL10WeyvH3DsauUZawdJKP6ZWGb753e/5kf0Pvv8965I98ra3+jURmA3BjIJwBsCCrAn1s2d+ 4WdMDA7KqxtvfP262RKkOXv+Umkmwpp/3WNhgZkJRypeZ5icuu4mZObFfnEWZF/9OHv+Ynk2Q5j/ OrCZABwrzOao5aQIT+E1lHMXLsk6Hbcdsyv257yCojhxbCgW8PH0OiFh6HheuMryEerOlj2LgfMR MLz2oTNh9KbeNyfEpq4+A28MGAPtMEAHREcdmnlXtp4TQjs2dD7peBQtuhq65hvmp2ExOsKhc0Hz zWIhvNHOe3hudh/DtB6zpNnmX32QvaqWrLgEpI6SdGkujepPMdQ/IzVYcsCUjHGai6DxQCo1Vh6l SPP3aqX28w5kaogshxHQkp5RglLTwEpT5+dPaGWayiM9q9cvRgkQ4lPxR4Txk+heMWyyQhqJKknl kYY2sw2vkWrtN2zjzehuNq1eI3k40KXhjRg1zeadTR/exzTfMI2GxcCi917y1HzzsLRiyIV6GtmH F72ma2GJxUutMiu+sI03UsZW0mgdca7mG+rRsFp4w/Tt7Ov1qms+ZHUplrAus2k6dQwW5V/zDXVr WLPthT5MK/2YMO9wHwMaQ/3Nb3xD1dkCrIeAYwADfFAMbYTznvzsl/y5qu8rf/st9+53POIee/8f a5B3UjCLgrUlMK5xZKgwO+Ezn/qkN+ZxVKCPV0Ee+0ClM+RTn/6sfz3kc08+oac2tMW5Qh6NGPKk /eZ3vpfq/a6sAyEOmKw8/aOfOtI+Lq+nqF5w45BgHSNdT4Pz4APHzoIsnFuhW+IefdtbxDnyp2X1 cIseeCiLYMBhwQwUdXiU4zbhjjkhNmGlGWRjwBjoDAN0lvRTWc10iDhPDf+891o1rNriaZ1Bn2rR fOiI0JGhs6OinRPSaAdI4zq9JV/9DJpyEOah+JTzMK7Z/bSE1c8kTb98XYH1Fu7I4pNZYS0CXkuA lz75YkUojepPz6l/xsBOeWdUbGVeQdAOp57vw+SgX0YQE8M8NLBJVV+/6qqXmsUefUcXQ06wJJLq J4y6JE2vXzAy1JxNvT5OQ1KNhFQeaZr+3j7vAOLVD+pCRVOzQCaTNqjHrGiaJLzyKJu2kePwOtLr Rs+jvmjT4bWvcUVsuR/BP3lm2wrH4NPrrYj8Q5167yDPPF40rBPXdZhv3r7yj1Gp+Ybp4Evbbhhe xL6/PuTegeS9x6/49FlRBIZQJ/zzlQfNN4xTXmJgoe3q85R8s6L4YmAhb9pvNV5CLLSbIoW2q/eY arxwbTfLC3r12uvEM57ZBwjGbjVhNgQ/FTWYOVZjHEcGizbyQ0JHBMc4IP5KXvVgBgELPn7zO0/J DIufymse/11eAflPMup/xhvafIkidEL8tLSmgy4Mia5GRGdB/MWHqr+qoXpIi5MADpg1AUZ1smia elucBzhQKCMOgx+Kzq/IApb8cISEPD352S/K6y8/9YtdUm4ExwY6yF+dH2Ai/O9kG3JSD0u3xq8f WuhWpIbLGDAGjIECGch29OtltWfPHp8kbxqkhg0PD9dT05H4kZER3/HOm82hWHbv3u3TdCTDGkrI h45Q3lRcxTI21pl3MWvAKEdhKGyXH9Z/tuN3V4zeu+KIoMO8Q77RXrTo1x3AkZrbSa4sAinWpts1 UPziVtQPnWE63Sur6xc2XGHRypJDqTQpozBq+EQonei7skAldREKjhm4wmmyY2fyedUwvtP7dP53 ycgVhknWqNRj2lOM65r6GRwc9OVXQ0nLy72Ka31A2ooaHxpXxFZ50XzDPKgf8FGHek8M4zu9Tz7c Y8iXX1a4x8BdjDqiLWiZ9d6meMBGmwFvDCzkOzo6mtt2iQML17TiJawo0TpCf/Y5oG0oJi88bzTf bJn1mRmDF/inLVD2LC/gAoumyeKMeYxDoFnBMMbp8B//6J3l2QAY2h+UGQysIYGBHc54QD9f2GDm AAY6aT/yoT/1W17n0Nc8HpEZF+w/L680qPzzs8/53WbXYQAj+aCznvBFEARnCA4IhC3HjQrlwlGg MxZw2jC7BMny9F7hAuEVFxX2wasOCMLfV+KXmRRbQcwJsRVq0cpgDBgDLTFAx6RVoWODMcc3vUM9 dMj5Rn0sYwX8GE50vMGS7ZiDBcFREUPoZGG08NnQrPEElrCD2jSesLqq1V2YRkx96mHX0JDPKjUU kkR0zFfFWMCYwxiu8Aw0qL9chky+Gh4GDwoOZjmQ7135hbIkxhMdUAysVIKzg90QaJVg79BI9QR7 cgLtdnBwyNcFn8j0UlJEnfHFiu3bEyO4YgZDmFmH+OkVB9CQ8LIqX7/gU6llEf3+U6LCC46ZshMi xBBWWEV4WUvTO+Pj475+9LO3qoCRVK4tDD34K1rIA8OIe8v8/HxFdhgr4KOtxHBCkPm+ffv89cw9 JhSMKa4reNHR7zC+0/vwok5X6iQUrivC4IRrOoaoQ7VaHen9MAYW2gt1kMVCu+XeCy+xHCJcRwhY wucjWKgj6gcnWwzhOoEX8g2fSfqspk3F4oW2C5ZsvwEewMezsx0sIdcxuNU8fnP+kt/NM/AxvHEk 8FWKUPjaRlbUSFenw1vklQ0Ex4QKxjkzFDDQGxVeicBJwsyERoT0eXmoQ6ERHXnl0/PG91Ziz9PL WhhgThw81/VUP4Oi0TUtyid16Y45Ibq0YgyWMWAMFM9AOw9sjOljx475TtbExEQZ7PT0tLtx44aP i9XJwgFx9OhRNzc35yYnJ8udvqmpKXfz5k2PJZaxAi9gwTi5dCnpmEAO2OBp//79rXc+w9myYrDn SkVw8nrI+N693hkBH0mdJ5/lpJ7uiPF74MCBxIAKz21QfxlDeG7w6kEYTB3sFSzzYsjdKjmHOJ8Z Gdd5t1Y65Xv3JZ33RG9wdrAbOgaqBEuSipgyTKDh7BgdHXHDYvxfv37drYgRqZBnZ2fdrYVbbmxs xDsHKrSEBzX0V2SmB+G5mpnEEXz48GGZ7dDrrk+nHS3w37gp9SNfNzl0SOL18yZV9AQqNceWtgcP HvQG0tmzZ8uL1tKWz58/79sQWGPJkSNHvKPh3Llz3ogkXwzt3/zmN94RcuLECV+XMfBgVOJouHjx ojeWyBMjDmxc88ePH48Bw+eBQwRjjjpRg1t5IcHJkyc9phiAMHAPHTrkrl696rh2EHi5cuWKbz/U UQznDPnCCdyQt/IClgsXLvhRdniJ4UADC04GeOH+wn0X4d4LTzyT4IXnVgzhvsp1DY6ZmZlylpcv X/Zt+dSpU9F4oczwQlvRfgO88NyGF+4vOGU3mzB7oppTQI3xRmZY6CsgOmsifCUDTtSZ0OyrGH/3 3X/0lPK1j3qieevMxXrpi4pnfQjKySsYf/HxT8nvCf+qCq++bBWxNSG2Sk1aOYwBYyA6A3Qm6OTR gaBTQWeCEUo6WMRh7MUSjARGmXCCYPBjGDDagtGPUyCmYGjDCx1QHVFmJA4c/GLyQrkZrbwtnaSJ iWvekBwcGnRzs3Me3/4D+6XznrxaE4OjI0cOe+MEYw4TvFdeNZiZSYzt40ePO79uRAQgOESOihPt lVde8UaKnwEg7QcjoX97v3SGj0TrmDOCy/VCp5xZPfww/CeuToiD6JA4bkLHTLHkYCTcd9993tiG G45puxhvZ86ciTbzgFJyndx///3e2D4rThFGSXX6+IMPPuiPi2Uj1U75cTRg0PLDsPYzeuQ6h69Y DlflhXssxjYOGdoLWGgzDz30UFsjyWmJG9tTBwz3XpwiGN9g4b5H3VXObGpMZ6upaC96r4cXxUKb OX36tHcitaq7lfNoL2BSAxuOmDVDe9EZJK3obeUcHHrItWvX/DNSn9XwhRM6puAQQeCFZzVYmAUB L9wHN1rUETB5PVkbIg8P6yOwXgFGfd4oft45jYSpoyLU+S6ZvcBsAGZH6IyIRpwJmh9GO84LXmuo 5ijRtOE2xquRYX7ZfbB+RNav4BUMyv7TZ3/hX2vh1RbWtcibdZLV0e3H5oTo9hoyfMaAMdC1DNAR pzNMh5NOBELHj19s4bUDOlQYkxhNdGzoeGFsxjb64YXOFB1NeAELODAWNkLAc/TYUW9cz8jsBxw1 GFVHxNAeGRsVh01x0+t5SyB0Re3aNSjGyX3SAZ1yV65eljj5rKkszoiBywhYs3WV1d8ov+TDyCmG EiOEl2RUEAEDRr+24Vb1N4oD/dQPzhnax4R0zHvEeYVxd/T4ETGcZBaLxMcUDEeMWUaTMeC4puAF TLGFfF/zmtd4YwXjFgyEwVlsIV94wbmJMbnRWDCsuwELzqEHHnjA33d5FsALbZnw2MJ9Fl607XKd bxQWyo9hzfNIeeG+EmsGRMg9WJjxoFiIU16aveeGelvZBwuzKOGEHwIv1F03CGsf4ATg9YkF+ZJD 6BBQfF/9+re9Yf+50poOvDqQ98lO0uvXHU7I/byenL2Ac566SftQrCnxTfkqBA6In8l6EKwj0Ywz AQcGwhoUjQjl5adYGjmn02n4jCkzMsBBWXUhUF7P+MTjn3b/Q2ZHmBOi06ybPmPAGDAGNiEDdB66 pQPRTVjobG5EhzOvCeF0YKorP5wisTqeoQMCXOQ7MLDLjyp7HElgHuSGwrL6GzqplAgsTLHnl3BC RKXGyqNmtDeWVvXzdRKmkvOLWT/VUGJIxhzFroaDcOqpm7DgjOgGgZduwoJRy68bxHjJrwWMfXWw 5qeIE5o8B7oDS16JdfbB3333qXVftdCZBeGaCazlgBMim560OA4wpPXzlZrfOe9weKseyvoHzFhI FmMM0/JKBo6If/jB0z4tizo2Kuhk5gD4mnFcYOA/JfkxA0G/VkGezKiIIbduLbr/Iq9gUG4+x6lC GXDQ6IwRDd+s2/iu9M3KlOE2BoyBe5eBO8tu7c7SvVv+jS55djhej3VbD182nRwT5AUrOCfe2+LZ 8NIp6zbZdGXlkrKO/jW1wtcpDQLa0J/xKQRKg92S/jLs8o6kqYO/Gf3lHBvQ73kp4SqfV22nlK6n T0bPtlm3phpNFm4MGAPGQCMM8OUJnAf/8P0f+hF5XVySVzSe+v7TfoQ+/FIE6fmUpqbHiE8XVZz2 rw9k8w3TYlQnCzBO585YePjBM95BgQ72GxWdBfGudzzS6Ck+3bulPD+U8jz52S/51yH2yUKSZy9c 8mVvSlGLiROnzWlfZj5ZCp9wRJ3Aa/jFjBaz6IrT7GndFdVgIIwBY6BbGVhbvOHWVipXYe9WrFsf V2iV6r5uS6XPHKYehmwEx0joBQjT6L5uk9TrHBblgEw6H276Y/K/tjTnenbucT39cVbdL7UI2xgD xoAxsKUY4DUARuBZFBFjnNkEKsxSeExe08jOLMApQfqnf5SmJ83HP/ohv3aEnq9bXo/AcaG6yfOD ojfv05uE4VBo5lUMnQWRGPSNOy7AxzmPf/Kj3gnxlb/9locMPjB/5evJsZajqO1fCZ9flbxDPsHA Fz7e28RskKLwdUJvj0x7pOdkYgwYA8bAPcfAF77wBffFL37Rl/vDH/6w4xfK2uqiW7tdfXGmMK3t F82AGvm6rZWfptFtrbQap2l1q+F5W02j27w02TBNq9tsfHisaXQbxlXb17S6rZaOcE2j21ppNU7T 6lbD87aaRrd5abJhmla32fjwWNPoNoyTyRBDsribTNU3MQaMAWNgqzPwjW98wz355JO+mB/4wAfc xz72sYaKvHbresMzPFmIktcA+GIEhnAtYS0DHACk3z++d11SjOrPf/lrjs9MMqKv6fdJ2mq6yf8T j39mQxZk1LLnlWVd4QoKUAyN8F8QhFy1PX07XM/A+jrOTZwJlFeSemwmRIYUOzQGjIF7hwHey1QJ 9zXM3V0p79rORjOgdaXbfAM0QalpdFsLu+rRtLrV8LxzNY1u89JomOrRtLrVcE0XbjWNbsO47L7q 0bS61fBseo41jW7z0miY6tG0utVwTRduNY1uw7jsvurRtLrV8Gx6jjWNbjNpuG574y8KmEFhh8aA MWAMRGUgtx/TAQQsVtmo4EgY7HB61mdAmnkVw5/QgT/NlL0D2eWq6AYMucDaDDQnRJsE2unGgDGw uRmwyWCbof5KBmmFXVrFAG20OGVd6DH962jb7PysK5AFGAPGgDGwNRnQfkxRToiNYo1ZE8+/+JJ/ ZYPXELKvgGwULsu3MwyYE6IzPJoWY8AY2IQM3Lx50383HOh8ttGkWxkoORwq/A5lK7k10MZxm04A AEAASURBVBW6TP86EjczP/aW6brqtABjwBjYmgzwmdpr1675wt2QT1BvJeG1DYS1ILbKOghbqX7a LYs5Idpl0M43BoyBTcvA4OBg+dN37JtsVgbadEjULbbpr01Rd/GztrLgeFfVxBgwBoyBrc4An8HW T/g2/KnwtTsNrwfRaf5YB6LRrzv8zec+LWtG3MpdW6LTuExffAbMCRGfc8vRGDAGuoSB/v7+8nfd t2+3d8i7pFoahBEO1et+PWOYeIT04b4PzPxRnQTrvulPSVJOuo8fFpR1S7OuZ7s4Fnt6U8i2ZwwY A8bAFmOgr6+v3I/BIVFX5JPjdxdn6ibrhgR+fYk6C2F2A07D0BoD5oRojTc7yxgwBrYAA/oe5RYo yj1UBHUE6DYsujoXam01vRrRWT16rFtNz7aW3mx605/Pl/JZLD9ry/OOn4kxYAwYA1uZgUE3546O 9vsi7u695e7OXd7KxbWybSEGtm2hslhRjAFjwBjYEAbu3r3rlpaW3MoKX9PAGN04wbGyvLzcFVjg Ynk54aVzDh81XnVbjWuN1229dBqv6XWr4dmtxus2G6/H2Xg91q2my241XrfZeD3OxuuxbjVddqvx us3G63E2Xo91q+myW43XbTZej7PxeqxbTZfdarxus/HtH99ZXfXX9d07d9pX1qaGO3cEy+KSY7vR ckf44H7XDVju3hUsxsu6JgEvy1JHq9KGN1rW1pLn4+rqxn9timd1wsvGY4EXno8JLxvbb6jWRngV 4mfPPOd+9cJL/lOa1dJZeLEMUA8T8tnTrSY2E2Kr1aiVxxgwBhpmoF3DmPNZCOrq1auut7fXsTI1 YQcOHHB79uwWHMUZSHmFZKHNK5evuG2929y2bT3uzupdt7+EZdu2eD5nOJidnS1jgZc7q3fcgYMJ L9u2tTNFXjtryq3OQNCthmcZ6nG3Fm65ufk576DZuWOnGxwadLxDW7mieOv69SsbiiSLIDlO9d8R Q4E629G/3WNJ2kt1/M3qT/NDp6Kq1I/jjLpaXFp0vdJGdu0ackPCS/X2kuJvRH+Shr8pBkWSxiV7 i4uLbkauJ4z/fnk9as/u3W7HTvkOeXBu9Wsq1Z/V2+wxTryrV6+4+bkFwZGMMO6Uac779x/weJrV 1056DP5Lly65+fl5x+tja2JEDQzskut6v2to6nU7mWfOBQv3OtqsxyLX+Q6ppwMHD/rrKJO80EMc Q1euCJbZBAvGJZgOHTqUc00XCsWRNwsD3pS22ytT47n/gWXfvn1yLQ0Vm3lGO86Hyckpd/36dcc0 /eQr1D3Cy2E3OLgrc6/LnNzhQ3gBx5Tg6etPzA3utTwfh4eHO5xbPXVrbkqMuImJiQpeFEvlM6Ce rvbiaR/T0zNuQtpMb1/pWSg3xUOHD/l7b09PvGd1rZI8/eOfuK/+7bfKzofHP/lRv0BkrXMsrhgG nvr+D903v/s9t9XqwJwQxbQX02oMGAP3AAN8UePypctu3/59bnzvXm/mzczMuLNnz7qTJ0+KI2JP JBaSTs358xek4zvuxsfHvVNkamrKnT9/3h05ckjC9kXC4ryRcuHCBV9+jAIcIjdu3HTnzp1zR48e 9fhaA5M1X8PjSuM61I/hRl1dvXrN7Rbn0KAYBgvijJi6PuXG9o7JZ7/2lTrnoT40hMfV9ad5JWnE FSVBeenXpG2cc9elMzw9fd396Mc/cufOnnfv+eP3uD/4gz9I1VTdq6+/Mt8QRyUeOsKMINNWMRBG 9oy4VW/YXfbG7TGpJwyqSgn1ERMeV+qvPE+P8vFjOGmnfGj3sBvcNShOkSX36iuv+mtrr9RRY46r RjAolvwtI8dq9HMNY7jdunXbXbxw0Z07f86dvv9+cfK140TLzzcvlDri+r19e9FfN0OyeO6yOI3O y3X06quvugcffLCGsyhPY+th8HL58mXvdD1+/LgYkUPizFv11zRt6NSpU9GcIvBy4eJFf5/hfsJ9 FnwXJQxeHnjgAe8EaL20jZ8JFgzbiYlJuc8edqOjI56Xy5cvOe6B8NLwYoGNZ5ubEizXrk14R9GJ Eyf8YoU4jq5cueKv81OnaM/xFmCGFxwi+0vPxztyH56cnCzX0cDAQG45Oh2Y8HLN88LzaGxszGeB Q422e+IEz2oGDeIIjhmuJZxU/MBHGG036TfEH8DIlnxB7nnf/M73xCk94P7zn/+JLAzJc3JvNpkd N8AAsxjOnr9kHOZw1R3uthxgFmQMGAPGQLczcFVG4jDg6Ej0ycgXo184ABihpAOKkRdD7t5hJG5C 8u+TDl8yQsooGKM8jGpfvHip9HpGDDRORr4mpWPl3GHp8MFJb2+f2ytOGkYFL4nTpnVeskZm9pjy YRiHIrMy5ubcZakrHBBHjhxxY2IoHDnMyOCgOJGu+BkSyRlZfdljUmX1h3nV2+9xP//5z93/+n/+ l/v1Sy+76Zkb7raM/ldKe/ordVXHzwyIy1cuubtSUd4xJM6rgwf3+9F+ZkZMTE5IHd6tVLfOsVJd f+bEmocLMkMFQ2mX1Mexo8d8W6F+qC9Gu2dn54Lz2+EnUFNll1F+fockf0ZrcX7Qbg8eOujm5VN4 E9K2Y8mctFuMExyLrH6P8wOD9vDhI+KYuO2NKgyYGHJLOtLXr0+LU2i/N/q5psHC9bS0tCz3n6sx YPg8wIKzl3sKBiWz0LjnHjt2zC3JLBacSOvbbjHweC2FNjo2NurxKC/UEQ4AnBHMCIghYMHBOTIy 4nnhGQAv1BF8xOSFMsMLRixOCByaLP58QGbw8Ey4cOG8W/WvLxbPDK87MBuD+z3PZ/Lnx/Np+3ae 1eejtZdk1gzP6n653x7wszKS/YM+7NKli/56Kp6V2jmcPX/RTU5Nu7e88Q3uLW96gzt5/Kg4h+M4 jWoj23yxOCA+9enPuqd/9NPNB75gxOaEKJhgU28MGANbkwGm9jOFfXDXLt8J1lIyjX2XhNEhvHX7 lgYXur0lBglT2XE40PFUYYopHS9GCZnOHUMwjjDW6HzimAkFY44O4fxCs1iyhlb2OM1lrWwor3l3 wR1x0GBUMsV+79hePyuD2QLbxIhiBJVZGtdlVsLdCoO7cf1pzuxxHr+scZ7qe/TRf+f+/M/+D/fB Dz7mjopxkJUs/sr4+vqT9Gl+leeDDmxrjm/Lz87New52yisPKrtlhJsRypmZm+IsWi4FZ/Vlj/Vs 1c9xwn8ak4QRnuVn5saMb6OjYsgl06IT/aOjo2IcyDTqwPBvjJ/KXJs5YvYQRi3XdSi03T4JxxCP YfhjrIAFPrJT13cO7PTGJc4iru0Ywmtn1N2w8BBOXR/YOeC2y3V+/fpMJENOZn3NTHsDf7fMmgkF I3dAHCM4b3CMxBBmWLE2Bk6ikBePRerp5s1Zt7IcZ/0Bf59bWS5/rlHLj5GLwwjnDbN6YghOInjh uglfL8BJQxiOR2Y7xZCF+Vsyg2jZ58u1rYJjhOcUz85YvNA2k2d1JS9gghdmPXFv7haBHxNjoCgG 0t5qUTmYXmPAGDAGthgD3jCanvLGCO+uZ4VOH2nocGBIFS10Punw0cELO8Lkq04JsIyM7KnoEBaB CywYRppvmEcSJq9myAwARjEbFzXq1YDNHqeakpgkHfuM+M+Jsc1IIMZbagDz3vZ2zxnxzM4YKMc3 rj/NWbFpSHic6qMOyiIGZlaSkBR/Gh/qIzQ8Vj0alj1OtRDDWiFz4pS6K+t0JJ1MQpNz6ZjTpuel vcyJsyjLWcJfbf2qS1MluSs2xZIc025npuVdejEO+vvUaZWc2Strm2yXNRnmxWDBiYRhl8Qk59bW r/k0vvVtRXjBkA2NFTTQdrm2SIPBwmsaRQpOTByHODWzWMDRJ9c6Tk4MGu43RQpl5pr1+QZOTvLk nXbWoKEe52QNjaxjoNO4VqXNzophDy/btq3vwtKGcMpiyBX9GgT3uVm53yHZOtB6YzYE07FZ26Ro YdYM12e2vZBvv9QbWHgOFP1KBvnghEB2yEyDULQNkYb2jeFdtOC04nmc90zSNRli8AIGno88i7iX ZQV8pJmX55G+MpJNE+P481/+mn99gLxYF+L5F192Dz942r3vj97pF6j8u+/+o3vzm14vC1b+QtJd lP03uI/IKxvIpMzE+fyX/9aH80oHMyje9Y5H3KO//VYfr3/Ig/voO9/+iEMfC18y0wK97/vDd/pk WT3v+8Pf83mpjmpbMHxTdLKgpmJ49G1vce96+6MVp3z169/ysz0ee/9/qMD82ofOSHk+sO71k2zZ SPcuwU/584S8/+tnv1ReU0O5fESwKB/cG576wQ/dD2WWBAtP7pPXXl77kHAtHNR7/QU8X/n6tz2P f/XRD5VnqhD+1A+e9jMvtPzZOgALeX5Qyv7TZ3/hnvr+02WuGuU5r8zNhq2/gzerwdIbA8aAMXCP MUAnYrk00kZnMyv6zjgGQgxh9gGSi6W0CCRY7t5dkw5qcYjoQGE8Ib2lfMPcwIfdfXuxU6NxmKFZ 4zbMMXFCsAI5nV9mPaSSHIMHYwJDK3FCpCkSg7u2/jB15X6YV2VM5RH6W5FG9JNmPf67a3xRQNqD tOO8NsMClbymwSrytPVwJLMSab7+yjTVjhL8y0sr3oD1DgYqI5CenmSx11UxcBk1zeu4B8nb3vVt V8rt20oGC8rhinayxAyogp0QSZtMnHmVrCTFpC1jyOGcKVrghbaSXEM597vSPfD24q3CnRCsH7Io DkN1CmXLjuOKNWBi8EJbwCmCcM1kRa8b2kvxsuYdUuSZd00TFt6fi8RDu9TZHziostJT4kqfFdn4 zh6veSOtWtvV5xRYat/r2kfF7CbtE2i+oVZ4AWec9hLmXLl/8hgz9GT9InEwsBYEDoiTx5NZe7yi gcOAOAQnQxp33X3i8c/4cIxsHNw/ffY59/kvfc09L+d85M//1MfxBx04IdiSljxIyysLvL6A8bx/ fK93HJAnRjOOi88cP1LTOMcAz2LAifIVWWATzCEG8iHsCXlVQvM6e+GSd14Q9rknnyjjJR2vVCBa NjA9KU6GJ2TBzofFIZEnlGtSnIOcr1zulzWOEBwQYKV8ODIe+e23+P2nf/QT4eVlr7eaI4JzyZsy 4ADSV2UoP9jhNsRJHdxauC0OoUd93pOTST2S1juKxEGjPKP3c08+XpNnr6QDf8wJ0QESTYUxYAzc WwwwxZ8V/BHtaIYMbCsZMCzEVXTHhnzpDCN0YLKiQXQM6YQWKejX93x7Kgz+JFfFR2eMX15nuTq+ SmM6PdIya9nSY6buww3Ol6RONK6ER+Lp+BGP0Zc4HZK4RvVXaqyOPusMSPVLjFeSj799/aoh1X/3 zlrSfqVxaJ2EyAmjqfBFk9RxFSIOS7NevzII/xob6g/3V++u+nZJWwix+Nzk5G1iVPFix11fPyEG QuvrD/Oqt5+0AVJJxnrhZE7ybVycIkVLcr1i+FfJqbSCvl77VVJ1JBheuF4ZYQ/rSJUrxFVZqLJo YRZPck+tUkUlXvT+XCQe6ogv3CC5vJSIwYlWtOAMoY76vNGvNRLkGpEXcMANkstLCVZ6vZUCCtjw rMZxlUgeL0kY7YV7XtXrrQPY/L3D38ckn7znYymPlZJjq5Usc0rYtBoMVZwDOAQefvCMnwGRVYKD 4TOf+mTZ+CUegxbnAcY7I/oIsydwADAy750NgbGel/ZTn/5v3gB/9G1v9ca1VyJ/0MfXIZh9oYa0 xoVbZk+gF2zqHCFeMeAsAIcKaZl1EOpUDOfEwD8hTg8Ew5wyP/HJvyyXDYyfePzTyVcrgnKpbhwD Onskj0tmQGD4kz/pVJgJ8TlxuFDe0Gmi8TggcIjggPjg+9/jwKHCOegMy49uxfmIpFWHBefAx8dl FoUKnH1FZojU41nTt7td76JsV6OdbwwYA8bAJmEg7CCF+/Xhi1FUMujzOy2lroCkoRNUtNDpqyZa Lt8pLGGulrbdcDpZjfACDO2kNp5nZfcq+QJFeDbxmibZJw2YEAzWVJIwHyS7GDVZDhvVn+rM0e8j S3lV5A/SZM0Eb0j7JAnmRF+Kvx396bnspfq9US+OFx9SbsAcJeLbjBxiYCl/KbelNIK/UlL9mlcl h6n+xIWRnH0Xo1J1Jd4YH+HPLanskaz4ckeKIYmorr8SWaNHOFyQEGl4LryQIsY1vVbCEuav+770 JZDNX0eqpfGt8lL1jFIb0lkBVdN1ICJtj/nKtDnrfSg/VWdCubWUbi+iMK/VJGEsHFy0lNtLHgzQ aXup8azoFEbqKK2n9YD0mRSLF8WiHITlVHTJgEFy/Yfxnd4v11NOeynz0k4d5RWy04UQfRjOoTGL 0wLjF4NYHRCarRrYGMihkC6b9sSxoz4JswJCUUObmQrVhFkA4HizLKYZOiBIrxh+KLMMshI6IIjD 8YLMi7GPaNl4/SLEyz55MWthUtYKalaY8YAOxabn4yjAEYEjACdJVv7rZ7/sHRBZ5wlpcXaAM1t+ HC/EMyMlFF4nCUVfLanFc5i+3X2bCdEug3a+MWAMbFoGwpH4cL9+gfJHkLPnpR3UbExnj/NmY2gO iqGzY8aqvXJLh047UpUx2aNOdPa0+5inO43TDmhlny+NJxyOyunK6oI05TB2wF4tThPWiyddkmZ9 ys7qV0TpNuG+bPinEZV7ZU6S9JWRHK1HnqRpDr/HoVngbSiL6JdDn0tFVo3oLytpaqehtiuNpQJO Uzk0nrgWFs9SSFXjaltKWQtLqLCnov7CmIj7JV7WX8+dxxCjHTSKulxH1dpFtfBGM2giHVjKeHLO 03tPDEghDn0W5kCKFpTiWV965SV5xjQOKdXJK2MFvm9ZAxIOCCR5laMyIc4KjO2JUprK2Mqj0LFR GVP/6DcyMwDBgM8KepnVwOyBZoUZEQhrNvBKSChnL1z0hxOTsp5D6TWLML7aPg4BOMNhkCc4Y3Bu TEmaQcGt8nfiyCHcrxsRzJ4gnlc+EGZKZHGCPYlb79TwERv0x5wQG0S8ZWsMGAMbzwCjiLoSdTNT Q3FYqNMiv7ObdDBIwzvKRUufLA5XXUpYwFHwKAmdIV0UrRYvpKuNuUppKIr2/OlR5pWH0XRNQ3Km 8xNU7oEmfPhEpV3U+PpsQX8ZafncQL9E+pkOmqichghNl+lyZvDrqanvQ89LCllTfzV+RKnnRTCo tgr9GJNwJudXmNwkTrJN8FfTr2nK4GWnfK7mWErkdSS5VDjKBFtZjZ6Cvrr8hJk2t69tMnGMhJkm eqgyONE1X5rT3lxqv2Ae3KyH4RVpzeUtttdcTvVTw0vZ0Rm02/KZpbDe8sKi5ZiO7/T1y72uS3hh vYPycyC3opLKY8HgokV5qdJcBJ1iqfWs6AxKOCmvBZHbXpJ89HrrTK75Wnr7qKPeqjPvQl4q1wzK 19dOqH8+1npWlyqv2fbCLL5FWWeJ01kgdiOENQiQZKHjfATqqMiPbT+0HoYh+eLRuVvNOyF0NsLU 1Iw4BZIFV0O0rCfRrCjWcGZFqEOdMTobQ+NwQBDHllkNeWtRwPOtW+vXoQEni/Z2kxR/Z+ym0hoW Y8AYMAYCBugs8TlChA5Co0JnRY3t7DR+dDC1E8EBUe68+5Bi/ujK7HmGv4ZhrGiHuRgUolU4VKMo bwo3YfRJk3fLW3DOhFVUrb7CNGLG9kpdkVM6uz1MIGAkgjryI0hhVIP6y1yWzy3v+KiKo/CgrF8w hB31ME1qhgu3mlN5p0H9el5pK6ezzkJ5IT16rkhZrTgeSrPHqafeMs4wDfvlEzg7lYrg4KC8W97x 5/T7L7qwECb2toKRKNHPES9rJO2qP8mj4vTgINhNEjb/ly9OIGF1hFp0cUYwFy1cq7040Py/9bmV r+v+Ei/rk3QsBF70/pjHjb76kH7dpGNZ5yiSe4wYlZ6XHDC8WoXo/TlHQceCvLFdug60PkLlCk/v iWFc5/d7/BcweB7lY0murRhYvLEt7RephaVZY7s1zhKHN8sm5WIpPRgSXjpwE6kBkqbCZ34R7SOE ycHHr1nnDHzv2M6Xn3iWFVuGEG+4r84HNa7DOPYJz74ikE3T7nE9DBj0atw3k5ee8175Ooe+FtLM +XlpFSuvkOSJzrDAcRIKr2CAgTUeWDeCRTF18UrFyasXj33gPeFpXbvfQi+wa8tiwIwBY8AYaJoB 35EsdZgaPRmjdfuO5NNjeU6Iu6VFuYpezV/x6uffmNmR7Wjpe7c7dsjnDQvun9ABghe2mq9iZKsL hA3sTBw/YVxR+33yGU4/KifGiRoodNYQjCf+4SzqbOc80V+vTEl1tFIpjemvlj+LovX1J4a0ttUw rTe2xSuBc0tXsg/j299P8YMjWRx0fdulLbOIJjMk+FRn0aJf6KCdZNd98MadGHjJtb/+83qdxobh D/9ZHEk+8MLCrr1ifBSPBV5wvFIf2TUofB2VnK47I3yGkvv1drmXwUv2Xgc3LFzp6ygCL9SPfuKR RVxDCbmKUUfkvUPuvUl9VGIhzt+PuT9H4AUjur90veYtyqltKBYvfKrV14csgpsVfX7rtZ+N7+Rx 2G/IWzhVeWmljrin+3u1XKMbIbqWwM/kCxdZ4TUBZhPwFYYiRV9tyMOAsc9rFbreQzM41Hny/IuV 6yk0oyObFocBr1TweojOtAjTEM4sCV0YU+MelnMIZ0FKZjyEr13gjECvOjD0HN3ymka3iTkhuq1G DI8xYAxEYyCvE9to5nvHktWf8z4zxufhMMT37N7dqLq20o3sGfEjf6yUny3T8kryCb/du0d857yt jBo4mTIzwrWY81k6j0+M/tGx0QY0ZZNo54ptuK/pNEyPk22/GLk7+Va92PrpwnmJ4U8HlB8Omh1l I7c5/ZW56VHqWEhR6V6on/TiBvGjcRqvOmptG9Uf6kz3GSWmY06neCUzfdd31mUFdxwzGDWSqARE z2cb7itODdPjWlvV6byxsl3qiA54YhCk+nFaEc7n7KijogWDcre0Xwy2ZCHMNMcV+fIDM3n4jOuu zOhUmqpzexjaQ8OD4igTLKUV9VU7OJblWgfHzpIzVOOK2MLLyMiIv7eQbyj65Yx+cfYNDw+HUYXs 03b37NkjWO7Kl2/WG5Xcj2m3g4ODheQfKgXL6GhyL+MTsqHQluGmT15RGRwqHgt5j8kziet3ZbWy joij3hiFh7uihfvKyEjCS/b5CC88B+BueHfx7YWyjo4mz+r852PCVRxepE8g/FP2vE/Igg/HIveg VqWZu3CreeSdh/H75je93r8mEC5AifHPlyUQRvGLFDC8Sz41yasKIQb9nKXH8Ee/1zQEXnnAYcCi j099/+ny+ejlSxLMSqgl4EL4VGcouljkV7/+92Gwx46DIfyKR0UCOWDxymxZyecRmQWRLT/n8nUS Pgea5/DI6o55XPycwpilsbyMAWPAGIjEAK9xMKJ/Sx5Eq9Lpo7OJ8E4m60xg5A0MVE6lKwoane6d gmV+fsF/n11nRjDKPT8/L53PPjEQ4nSEEwNgl7st7x4uLYpBUB4dXXNzc3N+5HZXS7yo4apb2Ky2 nzLNaOVuMaB4F5J6wZhC6KyDEcfE+Pg+P7qanFVNZxiepKz8q92/ynTJSwWEabhumRnCaG4yIyON r9SaHjWvv1Jnmi+jyXR0Z6Zn3PzcrBsX40U/GUfneGlx2Ru4Q0NDgpp88/FX059iDvfy8fPp1L17 x93ly1ekjmS6bGA8Li0t+876+L5x79gKta3fz9e/Pl3tEIxKrpn5+TnBkl6/c8ITDpG9e/d6B2Nt Le3H4sQcHRlz16/PuNnZWe/8UL7n5+bFkFt2+w/sl5H4ON04jKfp6Wm5hmflXjIkxlIyhsU1hfEU kxewTE5NuRs3ZuR6xmBL2ra/5ywv+fbUymhyK7UGlqtXr0od3RRjd6TMC+/og2dkZI+f0dKK7mbP 4Xql3Ddnbrr9+/aVeVlcXBQst9zw0LB/LjWrt5X0lJvZZdzz98n1q89H2grX17BgjTUTgqnvO8QR PT8771b3r5brAyy35fnNc5zndQwZEofUjh07PQdhvwGnGrzQrxgaHIoBpeN58DnJT5yXT1Z+53ti sP9ERuV3yXV63Ru+jNxXW/+gk0D40gQzFhQD6yDoTAycIK3OxqBsT8jnQ3E6/IMY9Fm9tcpAnpQd J8bzL7winwN9xDsQcCSwWCcOExwHoU4cENmvZmTzCMv62gdP+/UhCDsni2WG5cehwcKU6FSHSFbX Rh33PiGyUZlbvsaAMWAMbCQD//zP/+yeeeYZD+G3fuu3HL8KuSOzCPjliH/1QEb/bty4IYZSMiqJ kXLt2lXfmTh18pTbNYgHPDX+ctR0JAiDgBHRGzPTfhQMhwRG9rVr17zhfeTo0QrjriOZVlHieaHD J0YcnXCMSpwhUzIKMDMz4w4dPiyjQYyAFc8LEBnI75fO8G0xCngvdVA6eIxE3b696CYmrkmKNXf0 yJEOGXPVypSGv/TSS+4Xv/iFe/HFFx3tb04MSj/9X0ZNr1y54nbKSHtoiFOGVFI9aRh71cIrU4VH GAiMst+YueENF9oPU9wx7nBMHD5y1Dsi4K8V/WFe6X4+TtYSwGhbECcaM2l4fYYO+qVLl3xdHTp4 sMFp5Pn60/zr7+FEwxFzXQxuXi+greBovHjxoufjsLRfNcDra2svBflzT5mcnPTlpx0z4n7hwkUx nHa6Y8eORcOCcUs7nZRV4NnfLs69JeHp3Llz3oA7Jqu5+3VV2ityQ2fzag684ESjfjBm4QUstOuT J0506HquD4f8mSkyJbxwGe4UA5P2c/nSZX8PBguYYgizvmib3GvJk3piNsaFCxe84+zY0WMNXkft o2U2BNfxtNQRzlaMa7BcvnzJOznvu/8+wZK80th+brU1UEfwAS8IDgcwwQuzRk4cP5HM+qqtpiOx 8AKWmzdvej6YzQQvV65cFYfIbXf06JEa9/98CDxTnnnmWR/54IMPuDe+8Y35CZsI5U6KE4EZALrm gJ4+sme3e52Esw2Fe8K73/GoGNMy80QeHAwAvPbBM+7//M8fdP/m9a8Jk/on1gP3n3Rn5BcK+ZIf r1ZkDWb0g+fIoQPhKRX7pHn7v3tbGQORYPjA+34/dz0HHAT6GocqysMAlrBs9fSqLt2+5U1v8GXG aXrmvpPlMpA3r1jwyuGytAPwPCYOG/IKJQ8TZf03r3844UkScC5hrBlBHVB/6HxA8qP8oc48fZpf Izxr2p5t8jplf+qs1/BGtp8SkQXDGYcxMQaMAWPg3mPgC1/4gvviF7/oC/7hD3/Y8QtlbXnOrS3N hUHr9hmlnJiY8DMQeMpsF8fEgYMHHKNSsYWRFBwPdIJ5qLFi+r59+6NMwc2WFcONEUJmQ8ALszHg hSnbOCpaEx2V52zd121JY+ZQ0zHKflHeTWUNCKaOr8pIMqPI+/fvD6aRhyfrvm5r69d8SqlK+DgK y7rmvveP/7d7lg6jfIFiWUb6oYJVD3hlBCPid3/399zrXv+6VE21cvrw9frT/BS3bksqg0PeYz9/ 8YJbFIcMhh0dc35M6x4fH/fYUn2cH5xc3g/DMkl8lhqvWx9YSsh+stYA7eXylcvisJL6EaOOd8lZ QPPgoUPSKd8lWEpvj2bVVMOh2bSwxSjA8MfByHXE/93S4T4gbSXOYnopaOrjqjg2Z2/OilFL+JpM Y0+w0MmPKWDxvIjjyq+mIoCYUcOoe78YvDEFLFMyujcjjtdkEdw17yQ6JO0F4zu2TEzQXma8I486 wpHIvSXWCLuWF0fRjMyEmJycSHiROhoQA+qgOPJiYwHTtNTPtNTTil8zY81jOHDggK8rxRxji6lz U64heOH65raxc2CHPB95VseZJRiW0/cbrkm/ASwChmsZB2crr3o99dRT7ktf+rJX//u//7+7P/uz Pwuzsn1joDAGevpkDbCB5r8OAiDpB/bEcc8WVnxTbAwYA8bAxjJAJxyHAyNzYqv4xaEYedkIAQcj TmvSQcdeAUfrBn97JaAzdfLkSc8Lmhiha48XtT51C9uIbpOj5FDT6FY+HSavgJw+c9o7RTBwdTQ3 XXhR0+pW9eq2tv51OLK4fI30uHf+3u/Kr957qYpBtznlrKLf9659nOLW7Xr8fL7u1MkT4gxZ8VPq KcNOWYugl88gdkB/iqU2ftoojobT95+WGRGLvs3gtPKLqUq7qRBfHOVFt3n6K85q6oCRSoy2fWJc Y7yQJQ6rjbiWuG4Oi2G9X5yJKZbgk5lNlay9xGDBsObVixTLxvGyf/8+cZiNeiyUjAURy86q9ora 9Nm8cjAuI7j6HGB2xEZgIU842S3rLVBHyEbyMibrMbBmkfLC/ba950DTVeNP4NrlFREcDpW8ZO6P ralv+qyw38DJ7T8fm4ZgJxgDG86AOSE2vAoMgDFgDGx2BuhA8OsG8R28DXKCZMvfWV60s6jb0ADN 5qxpdJvEg4dRwUpRPZpWtxpemTo50jS6zUujYapH0+pWwzVduNU0ug3jsvuqR9PqVsOz6TnWNMmW RRD55YvqqTyn0sGQPTObNhsfHqf6xU7wTrQkVsPDtLrfjH49p7ktRgvOiO6Q7sHSXbxgYHdHHcEL v255DnQLL1w/nX0OtHdFGi/t8WdnGwOdZKA77t6dLJHpMgaMAWPAGNhiDJQM0gq7VA3RFota1oUe 07+OxU3OT0+/fP2jX16JktXmo4l8zWNTikzhl+HhTQg9/OzuJoIP13C+6WSzthMhepNem2s5nxXd dM3GABsDVRgwJ0QVYizYGDAGjAFjoFsYKDkcKvwOZSu5NZAVukz/OhI3MT/eAbEz+SzfunIVGdAl M5CKLGK36a5opt0GzvAYA20y0Ej7XnDD7uJM8qnRm6sDbtvw4dq5ymLba4sz4gfbpE7T2qWz2E3E QHfMH95EhBlUY8AYMAaMgW5joOhRXNNfu8a7i5+evvgLzdXmx2KNAWPAGOgSBnq3u55d410CxmDc ywyYE+Jern0ruzFwjzPAO7wq4b6G2babGUjrLl3joJ4xTLymCffzymn6U642GT/BdZ2H3MKMAWPA GNiKDDTcj+mR19Rk8V8TY2AjGTAnxEayb3kbA8bAhjIQLiIW7m8oKMu8DgOhEyGbFMdBNj48Jl6d C7qv8apLj3Wr4WxN/3p+u5WfEJftGwPGgDGwNRkIHQ/NfHmkx+Wvl7Nw67b7/Je/5p7+8U82hLBf vfCSz5+tydZmwJwQW7t+rXTGgDFQgwG+NX/79m3/4xNiJpuBgdCJUAtvs+lUV7PnaXo9P7vNxuux brPp9Vjjdavh2W02Xo91m02vxxqvWw3PbrPxeqzbbHo91njdanh2m43XY91m0+uxxutWw21rDBgD xsC9wwCfHO1kP+aWOCGe/tFP3PMvvLwhJE5OTfv82ZpsbQbMCbG169dKZwwYA3UYYOSAHw4Jk83A ADMUwlkKuq/bakYp4UkaTZlfWmLDFLqvW9Pf7fzk16uFGgPGgDGw9RjACWH9mK1Xr/dCieyFoHuh lq2MxoAxkMsA0xi3b9/u45qZxpirzAIjMIAjIHQChMdheDUoSZoe72TISx/qQ0d4nJc+m4/pzzJS eRyHn8o87cgYMAaMga3LQNiPsddKt249b8WSmRNiK9aqlckYMAYaYoARBJVwX8Ns220MZB0B2WPw ho6DZvFn9WWPTX8lo5uNn0r0dmQMGAPGwFZioNP9mIVbt9zPnvmF49WIffvG3GsfPO32je9dR1mY bnBwwJ04dsS99qEz69IREKatpTN78tnzF+Xc227/+Fguhmx6O+5+BswJ0f11ZAiNAWOgyxngwc+P Rfl7ejb2LTfFAmXbtmEk5hmKsQhdk9dcEl7AES6g1RyCrGMhe5xqk9xKJZY6Ke9rvDqdspxk9WWP 9XxcHKa/lqOnO/hJ66udPb2WaLett912EKTndhsWXl9j1NV46d46os0gGz063m1t13hJ22ytvcmp 6+4Tj3/GOyA03T5xADzxyY9WOAFwDnzq05/1DoLBXQN+S/p3v+MR99j7/1hP9dunfvC0++Z3vldO QyDnfPAD73GP/PZbK9KGB+QBliT/vwyjbH8TM2BOiE1ceQbdGDAG2mNAOyPtaFlZWXFTU1PulowY 0NkbkAfq+N5x19cX//a6urrqpq5fd7cWFsQ4cG5gYMCNje0tv3LSTjmbPffOHbBMu4X5OW+o7Nix U0ZS9rn+/v5mVUl6dRqocyB7nKpMYpJ0morYu3fvuBs3ZqXzM+/WxIDq6+9zu3YNueGhIf8+baKh df3rDXPVhWZFQphzS0vLbvbmTXd76bZws83nPzoy6nbu3Fky6tbjb1w/eYX56b7PuhST6ucamJf2 cu7seffznz/r/v3/9u/d4UOHJHE1/M3pT3Llb6gve9zjWBh2bm5efrM+Za9cS7t2Dbrh3UOuL/iU XFKaFH91/WlMs3twclPqh+t6eXnZ18vY2Jjbs2dPqX6a1dhe+hs3brjrcl2DZceOHW5kZESu67H2 lLZwNrzMzc25yclJacNL/r4CjtHR0Q3hBSzU0eLior+vUD/cYzZC5ufnPRYWCOQet3v3bjc+Pr4h DgAwXL161dcRrxmCBV42whkBFtrLgtxjeCYOyf12//79wT03Xm3RZq9du1Z+Vg8PD3teNuJZzbVM HcELdbRr1y53SO673fRa6K9kYcp3v/1R994/eqd3FDz1/afdV77+LfeEOBw+9+QTvuJwVDz52S8J /gH3Vx/9kJ/9wKyJb37nKfcP3/+he/jBM+7Nb3qDT4sj4St/+y138vgR95mP/ifvUCDtE5/+b+5z X/qapM2fZaFODnVAsDXZGgxs7JDd1uDQSmEMGAP3KAN0PP/1X//Vd4KPHjniDh486Bal0/WrXz3v V6uOSQuOhxf+9QW3IB3zAwcOuKNHjgqGRffCCy96wyEmFoyCF198yd0U4+nw4SPuyOGjfqbIr55/ vsNYMEkxSGsLBu61axPuyuUrbsf2HW5sdK/bJt9Jv3rlirty9Ypgq7YoaWP683NPzOVsHHVy8eJF NzNzQxwge9ye3XvEKbLmzp49J06Sm95Bkj0n/zhff2Xa6vhxmv3TP/2T+/a3v+3+ry9/2f31X3/O H9+WdpRIe/orceQdpfqpnwkxVC4JLxjZo3tGZLvTXbly2V2Tjvrq6kqegkLCMLTPnj3rr2uMlDNn zngj7te//rV75ZVXfDsuJOMqSs+fP+/IG4fi6dOnvcH/8ssve3xVTikkGF6uyPXyy1/+0jtlwIJh Cyfgi72w76VLl9zzcj/BsFYsFy5c8PhwxsYSeMFB9Nxzz/ks77vvPs8L+OAqJhYAYPD//Oc/97yc OnXKG7YYu3CFwzymKC8Y//DC85GwZ555JjqW6elp9+yzz3pH3okTJ9yxY8ccYf/yL//inRIxecHB SXvBQUMdHT161Dsj4IX7crfIm9/4BveYzFBgpgLyrnc86h6V2Qo4Dp4PPp9J2F/8+Z+UX7/ASfC+ P3qXP+dnz/7Cb/nz1a9/2+9/vOSA4IC0nIs8JU6LrODkYJYFTo4nPvmXPn02jR1vXgbMCbF5686Q GwPGwAYzgFHLSPYRcUDskFFsDIVDhw77EZ/Lly+51UidPkb2LwmWOzLaf0ScD4ODg65fFtw8cviw GN39jg7x3YifIJ0Qg395eckdP37cGyzbd2z3neHt/ds9ltZ4qRxJT10PaswSkoayn4SsudlZGTGd mHQjYyPeQBgaZjRunzcupyan3Pw8HT9Sq65Qk4bl62+8CSb6GbldmF9w+w4eEINyxGNgBEysWzcx cc13khOdCf5m9Wt6ckskH/+yGAYvvPCCH41jBLm+oV8cP/PCx5QYT8N7kpHJ4d3Dbq+8dzw2OiZG 1ZSblRkSSY2kpeI4PNLStrvFQMCQo05w5jGqzWg/1zgGFPUXSzBSMKwZUQcPWJh1gOGCATUxMREL ijeOwILj4bDcV1jQl3aDQcdMDQzdWIKTE0ceM0IwJuEFLPAyOzvrecE5EENwMuC0In+MSZxo7J88 edJfy/7eG+nLSzgZLl++7O+58AIWZh7cf//93uA9d+5cNCeaOvPAABdscerhjCAO51qsz2JTR7Rd 2iztlec0P/Z5nQgssdoL+dB22cIFs9+YBaG8UEexnUXVrhPWdsjKv33T633Q2fOX/Jb1Id4nMyUe lvUfWOuBWQu/EgcFv6wQxjoR2ZkMnPu4vOKBkyOUSbnfMusCMQdEyMzW2Y8/X3jrcGclMQaMgXuY gRWZTjk7N+v27q183YHO1uDgLm+w7Bu/7XbvaeX1g+aIXRSDkhGU3WK8YfCrbKfjJ1Nx6ZjOLyTx GlfUFgfDzdmbbtfALt/B0nwYsRwZ2eMuy2gqnRU66s2JGtPJWeu/cBHGJ/ukWV2V1zBmZtySdETH pa6Sd9jXZKSwV2Yh7HbTMqozNTUpeHe63uAVmkb1p46LMP/QYNf9Hm+UTE/P+FGdIekEq8DN6Nio jDRfddM3Ztyhg7wOgWY9Nznygf5PGK77Yf7Zc7Pn97gRMWY/+NhjkrDHXZBO+HPP/cLjQ1ti8Ku+ 1vQnOeq5yRF/EwnCpTMO/ytSP2MjY6X6SfDTRiYljtkQo7LfIzwlktYvKcOw0kHLG3UycF2rUD84 AjBkmA0QY8o/RgqGPTMMGEEO116AFwwqppbjIIkxnRynB0YjPIRYcATgBMDYxjkRQ3AGYagxrZ+6 UcFBAy/UEfXHvbhogRfuvRj9YT3wCgQGJnWIAwlcRQvOIKb3Y/SHvOCIgAt4w5mGAV608KoMvICF 9qECJ7QZHGi0axzmRQuc8KN9hnUELzhGqCN1ThSNBU7ghuso5IV9eOH+gyMtvP8UjakZ/TorggUi VXA8MMshz/GgacL0GhZu8xax/OZ3/tEnSfJMnkzhOba/+RlI796bvyxWAmPAGDAGojEwIZ0FDIW8 ziXGP3Fz83NR8MzNzUunfFnWOeivMBDIXPHNzcu79oKpaJmRkeTl5ZUKZ4jmyewMWeLPv6ahYa1v 1fjMaqgs47IYKwu3eMXgbtDpS87t7e1z2/q2SWf5tlsSp1KlNKa/8hw9yj+XDiijcqxHkRrUyTnU E6+FMEuivuTrrzyvWpqAH00i255gPzXsKzVWHukJlaGJAyMblj1Oz8Wwpf32yXvR23rDLom0FFlY lfUgFheXgtHBAH9WbZvHYMGw571svW5UJccYdtQhI/FFC3nMiPMMLKHhRL7gAA+GFVPdixZ4wVAj 36xhj/EERvCCp2jh+sFQA0uWF/LGwIaTGFhwEOlsFEazQ4ETuOHdf9pMDKHtVnsmUW84bpjpU7TA C44gBKdDKNp2qUdeZ4whtBcwcc2EDjTyVkdADF6oG2ZZUQ9ZXsACvuR+GKffQJ7tir4uMSGvTTCb 4W8+92n3P/5G1ngorRnRjn6cD/9RZlngwPj8l7/Wjio7t0sZCJ/4XQrRYBkDxoAx0F0M0KFZvJ10 LOlsZoWOFhKr8zlfcnaQb7aTxYi/xyIzIcBdpNDJui0dboxp5SDMb5u8uoKxm0yvD2Ma3A/tz2oO lbXUwEXrHVlPAEeEk/AEE0oSRZ4rAcSrCCsr8h55C/rLyMvnpvqJ42sRKrelM8VLBL6eNLC0xQDH QcMU/FxpQH+r+D3ikv4evy1nVoG/Vf2+PGWV7JQPvPPn7t1Vb8jSPsoi9csxXLHI6ZI4trxk6jdN X95reQfuuUZoF9nriGOwEB/DeMJIAw/3l2pYMGZiGNvgAI9yEBJMmN4DGcEtWjDSKDN5JtdzZY6E kybGvRdnhzqBlIMQjeKL0V7IlxF28qyGhftzjPZCu1RHXR4WDYvFC+0ybKdhHSkWeIGfIoV7h/Kv +Yb5EQZOTRPGdcs+60EgOiOCz3fiJPhIaU0IDc/iJZwfzoo8YRbFOZlREcr7/vCd7r3ye9fbH5VZ Fi/LbIu/D6NtfwswEDzxt0BprAjGgDFgDERggM7EHZnmj9BpyAor+yN0hqsvepg9q/VjZh4gFUZc SZ3CA0to/JWiO7qhE6frPWgHPMwg+WRo8qWKlngJqdaChRmwH6aRg5XlVb/Yo+adJEgSMdLOPz4j ipFVcW6D+svZl/Nlp3wQ7MlXMWSdDB+Lbv/51PLZTsxtfxqvj6Trd6R6UkVeQ/nEIEWYrexXxJTT p3oIStKoxtQ9kJ6b7pWTJ7qa0O9PKCvS3BI1qytSXsmY66hCJQf8F57WZJ/Xnyr1cFRWWrGbJGz+ L6PWCFjy2+82b6hgZBUt5KEOkby8FF8MLPCiBprmG2LSe2AMLNzH+GmeIQ72NTy532VjO3vMPYM6 QjTfMAcN8/eWMKKAfXDUyic2Fm0Leca2YtE0BdBRVgkvmo/mW46UHQ2DO23jYXwn99Gv95ha11Gt euwknnq6eM0i+xrFUz/4oT9Nv3ih8eqcUJ0/ezZZqFWP2fIJzuyiloT/9Jnn/OKTZy8k60wQhrAY JcLnO/miBl/bCBfE9JH2Z1MzYE6ITV19Bt4YMAbaYUA7IOgI9+vppDNRq/Op5yfpih1dIS9GiqtL YqjRKecrDEVKJS/Vc2LA6c6dYmdlaO7p8oWBwRqMxCfpJFXVL2Sopma2+TzfLZU5cX2EeER3+VDa FgTVlHrxNU+uEal6dVsjaVtRqf678poM7aamSPTdjtZPfm56TefHJvcIsMYwcBvBAs4YBgtY6taR YFFjrxp/nQgPecm7Z2tYLF4Uj+YbllHDYrSXMA/NNw9LDF7C50CIQfcVX4hZ4zq91fpBr+Yb5qFh 8NJIGw/PbXY/5EXzDXVoWDt1pDpCva3us/jkJx7/tHv6xz9xT//oJ4mjQBwTj77treXFJV/70Gmv /pvf/Z576gdP+3Sf//J/95/izOb7bll4ktkQn5NXK9DJ7AfO+2s5ZrFKPudZTfiihp7LKyAImP7s Lz4pX9V42h/bn83HgC1MufnqzBAbA8ZAhxgIRyM6+fBWi5JOR9Edm0ap6A4camm3wguGKuerwar7 qU7lPeQkSS1pNJmPLB2UVCVHqle3hLKvJ4b7YQ55+3pOqAF3iOrWcyp1ao7rkmny8jZffyv8oBIU lW85dFZ/GXZ5J9UvF4ifFFIue7nwycwIwrOviJTVbMBON13TFD80soqko5H7Ywwsjd7HYmBplO8Y WEJewv0sxlpx2bRFHSuGe5kX5SDkWMPa4SXs04S6W9nnNQjWS/r8l9L1GJgB8dj731NWx5ctPv7R D7mvfP1bZccDDoW/kjCcC+FilYR/5lOf9Os7hDpZlJLXOYivJsSR5snPfsmf/7h8rpNnAzMxsrMw qumw8O5jwJwQ3VcnhsgYMAYiMcBDX9/p1Q5AI1nTId8m728i+eclxibrMfRWLLbXiPbm0+Qt0JZq SbD0sjYEU9wLFHjRL0wItTmSBPp0Jf5yElUJUuy6JVm1/VSFn3VAMg+I/MNzFCQGr04MDOOr7af6 071UVxpGbppnOsWfkMQhoXkmZzA3BG54BWG91NefnBOeW22/UjscJfxUhnMU4k9iq+kMw9frocTr z09eIaJq0vaS6vHhctLaNllZo/SKU57mJCxff/X062NqX0dgFBxSP/XSrdfcfEi9PPS+owvrNZ9D 42eAhXLXk1hY6uEgvh5/jeiolwaDrxGjLwaWRvNoNF29steKhxPy0WdrtbQxsNTLQ68j0jXSxquV pZFw9DdyjdTDnM0LvfW4zp5T6xiDn8UlVR6T1yGYfcDnOJmNkBUcE/x4fYNXKPZLOuQtEpYVdLOA ZeI8yNfJTAt+WSGPENcjkubfSlgepuy5dtydDJgTojvrxVAZA8ZABAYYcdBOSDNTIOlkqXNBzw/h 6nR67NrUuA1TdHZfOy3rR9pT4663b/0Cd51FkRjQ+g5w3qsfyhWdpkY6753A1ydfv/CmtLdyKw0p zFY46xyeSv1Z/HwVw9cRrxaozVxK5LmRMNqVrimSPZ9yFCLeuBbNGUydz2s9/qRdyloLOZnDCb8e mabRL1/JqC/r9dc/J02hbZcQbatpbBJGWwnThfGd3Of64JeHg3w0vBGjpl1cYXk131CnhsXAktx7 e335Nd88LHpPDOM6vQ8veh/baCzgqFVmxRfWZaf5UH1aRxxrvhoXhtXCG6ZvZ5/rlXx4VagWFngh bZES3jtqYWn2OuK1FtVXxCtRGPmDx4/WpeZkA2lUSaM6NX21rTkgqjGzOcIrh2E2B2ZDaQwYA8ZA Rxigs8Snsvg10yFikUM+N4ncEUdGVnRhwR3bi/9OPXnv2JFgYc0B7Ywoprt3kwU0t2/vlw5z8Z2s fsmHztadUr6Kg23i9JHPtu2o/GxbmKb2fmgl675uS2dmDqlXvjzBmg+6JkOSUgxc1sgQIxdDuL8f DsOTdV+3+forzyml8RvO45dyPuDLnSyEma5BkehP1siQT76Co6IznMm/nMV6/ZVY9Dzdlk7MHHIO r2IIDSKJ0Z+kJIBfij85TmLT/YzCzGGaTs/TbaKfDjczP5K2EV5LiSLCoWO7XksN69d8Gt/q5yfJ k18oXFd09rlnaLowvtP7tFu4CQ0MzQMsCS89UbBQXow0zVdxsA3D8j47GKbtxD7866cMs3WEfvji /kOaooX6UaM+z4kNFiQGFvKB/7y2S5zyErPtkm8tXmJgAQP5hO2UMBVtQ6Qp2gkR3ju0bSgOtoSB s1lewn6MtsdQr+0bA93KgDkhurVmDJcxYAxEZYCHf+PS44aHhn1y/RpEeK4fjZCO8OBg5bfjwzSd 3N89vMd3oJIOX2U5tBM4ODgsaYq/5Q8NDvmOed6ITBK25vaM7Gmh+JQLg1jLp8axbksq/aGmSYz6 oV1SDxK+whcwyiKOEun04TDCibNDnCet6E+N9AwOjzUMW3PDUk+9YnDDQ9rckjRLK0se2eDwoGxT /M3obxm/pzZ58SK9DsBVib9l/b5koS4CEv0YaAM7B7yxkjiJtOxJ/fDpVBwz23dQP6XTGuInSd7M X7DskraCYaLXjZ6vxxieQ0NDGlzYFkNkcHAw16ikjmhDAwMDTRssrQBWXjTfUIe/hoQvDKE9e1q5 rkNt9ffJZ/fu3Z6XPEOOrw/AXYw6oi1omfWrB1oCsPED7/Bw8qzQuKK2o6OjuW2X/Gi/GNlwV7SE Zc7yQhsCC2li1BFlHRsb88Z99WeSi8IL/NMWKHve6xNaR7F4KbodmH5joB4DxfdI6yGweGPAGDAG NoiB1OBqHgCdz16ZIs736NNR7WS0n2+k98tI5qAY5DFkQKZLYiiAJfzqBOW7ffu22Hs90lkuvvNJ WelAgQUOskYCWBipaa2TpUasbtVYzWNY08hUXDEUhqXjzfocc3Oz5cQy5u8Wbi+4VTEU9uwZDd7V 1XMb019WWHVH9cjo7I6kTSzeXix/qk1Pm5+d88b2iGBJjX/Foqnytqn+JFbP0fC8czRNaSvtxP+T U3RkMD1L9WTOKTsC0pTpXjZtGrN+b83tlXeIaavzCwsSrecmbZcvv4ztHfP1l56raXSbxrS7t3// fl83/roJlM3Pz/v2PD4+Xh79DqI7vst1MjIykvAieYeCYcf1xT0oxuwD8t63b59vG/AQCjyBBwO4 mdlkoY5m9uGFvDDo5ubmKk7FiFuQNoQjiV8MgRewzM6m9xbyhRO4wejnfhhDtA6ydcR9GCw4rdRp UjQeriOto/A5CxbwUT842WII1xFtk7YR3t+UF9pULF5wiOC8ou2GvMADvHA9t4MlqzMGv5aHMdAq A+aEaJU5O88YMAY2PQPtPLAZzTh46IB07hbd9PSMcIGxtuZu3rzpOxgHDx70nb4YJNGpOXDggLsl HU2waLluzMz4js1BiYvVEabjuW//Pj9SOzFxrcwLHcAZwUNHmc5wc1IyhNUe9ic3ZoCCZ1Q6ocPi HLk+NeVX+4af2wu33cz1Gd/pGx0d8Y6aSru6Mf1Vy1GnX89UAABAAElEQVTGip7kQFbCcAekc46d PTU56Y1asNAhXVi45UbHRqVz3iA3OfqbwX9bHFY/+v9+5L79rW+7b3/nO25qYtLdlPr57v/8rvv7 v/979/8+/UPfthOnQCmzcp6UunP8jIiDDGNtanLKLYlxDSerMgNiQjjCwUbHPZZgPGEcnTt3ruwo YtTy4sWL3njgOoslhw8f9ryARZ0iGNpnz571jpDjx497Qy8GHpwvXLuXLl3yzk7yxKC7cOGCH9kF SywBC4YlWNTghpff/OY3no+YvNBWuNdPTEyUHRHwcvXqVX8PBEsM5wzccw3hFLly5UqZF60jZgGc OHEiWnthNgq8TE9Puym57yJc1/DC/Q5emn3twCtp4Q/PG7Bcv37dP4NUBTzhuD916lQ0XnhWg4V2 S5tB4GVS7nXwQlwsB5ryYFtjYKMYaGSlp43CZvkaA8aAMdDVDGDc82WBa1evuRmMf/nHlHI6e3v3 7o3WsYEkHZGbvj4tjpAbkjejcSveOUGnPaaM+7Jvc5PXJqSzteB5WBEsR44ccfvFQYFjoDkppa84 DYu4IqCqyu3SIT518pS7cPGCuyQ/vmyC0TIgo3HUIR1DLxXqGtefm3GFrtKBbJiVQQd8cmrSnX31 VVmPQdZDWL3jjh076o08nFsNSRX96bm18ctyj+7qtWveQKCT/ju/8zv+VN8hFmyLYni/QRwCAwOs 35Hib1R/mq7KXoCf9R7g5Jo4rTC4+cLKqsyAGJLXmQ4cOBjNgQZSDKMHHnjAnRVD/4UXXvB544TA iYexEmvmAVi4Th588EH3qrSTX//61x4bxiRG7ete97qovDBafN999/n6eeWVVzwGriEwwlesUW3l 5f7773fnz593L730kndqggWDG15ivf4AFq5X2gV5wwvXEnXEdUTdxXj9ARwIdXHy5Em/hRfaKryA 5TWveU1ULOCBF/i5fPmyN7KZeaDtJaZjESw8k2nDOK5whFBf4AFj7Ocjz0F4wAmCY4T6oc1wHeEE NTEG7hUGzAlxr9S0ldMYMAY6zgBrLDBywagchkqP/Nuxc0e0EZ6wQHRqcETQ6fXvm4oNChaMJ+Ji CrzgiGD2AdPGEYy74gy42gY3+fNlCjqiy0vLvvPJYpRw05jRX18/eTQi5Md02yEZQV0ULKzEgFGJ o6Q4WY+fGRfvfe8fdyjL9fqbVUxdHJXO+dKi1I8sagpPtN9tfh2T9vU3g4eRyIceesiPVmIc4KTi FaLG2kozOdVPyzWDAckoKQYlGMASa3Q9RAgWDCVmNfG6wUZjOXPmTFdgwbgFC7NVuN/BC22INh1b aBc4aBhpp+1y798oLPCCkR/ygpMm1gyIkHvqBEcnz0h9JikvsZ+P8HL06FHv/FAs8FLc8zFkwvaN ge5hwJwQ3VMXhsQYMAY2KQN0HrqlA0EHbyM6eXlVVyyW0LGi+7WNVb4OsnMAY5/0pK0lqpM0ul9b f6oz1K/nVubFaP+g/Cqlc/pTzOSgGLpbP86rnX7mRSUrzePPnt/8MUZLzFHsWggxkroJS2trutQq YWtx8NJNWDBq+XWDGC/5tYCxz2+jhbbbLVg2mgvL/95loMF5n/cuQVZyY8AYMAaMgW5iQJ0Hug2x hca/xodb4tUg132NVz16rFsNZ2v6U0dLt/MT1pvtGwPGgDFgDBgDxkA3MWBOiG6qDcNiDBgDxoAx UIeB0IlQK2mz6VRXs+dpej0/u83G67Fus+n1WON1q+HZbTZej3WbTa/HGq9bDc9us/F6rNtsej3W eN1qeHabjddj3WbT67HG61bDbWsMGAPGgDFgDBgD3c6AOSG6vYYMnzFgDBgDxkDAACPw4Si87uu2 mlFKeJJGUwZKg11iwxS6r1vT3+38BJVpu8aAMWAMGAN5DMgnxk2MgbYY6Olt63RzQrRFn51sDBgD m5mBcEGqcH8zl2lrY8cRgBNAHQF6TKk1jP1qkqRhMch8UX2qS49JrWH5Zyahpr8WO8ph0fzXxmCx xoAxYAxsTQaa6cf0bB92PdvaMyK3JotWqkYYoO307BhuJGnVNOYGq0qNRRgDxsBWZyB8YIf7W73c m7d8WUdA9piShY6DZkua1Zc9Nv2VjG42firR25ExYAwYA1uJgab6MbIQcM8u+WT26pJ8JnR1K9Fg ZSmYARaRdn2yCK8ssNqOmBOiHfbsXGPAGNjUDPDA5jNmSFMP701d6s0IPutYyB6nZVqTGQvJY3FN 3BG6r/Gch2QfnFl92ePkLP6mOk1/ykq61x38pHhszxgwBoyBrcxAW/0YjMn+gXVPxK3Ml5Wtexgw J0T31IUhMQaMgcgMrK6ulp0QS0tLkXO37BpnQJ0G6hzIHqea1AGBo0FTJbF6rqYNjzWlhmWP9Rx1 XyTpNJXp70Z+Uky2ZwwYA8bAVmUg7MfooMpWLauVa2sxYE6IrVWfVhpjwBhogoHe3t7yd937+/ub ONOSdgcDuAHUcdAsokoXQv7Zpj+fFw3tZn4Uo22NAWPAGNi6DPT19ZX7MfRpTIyBzcKALUy5WWrK cBoD/z97ZwInSVHl/+i5uuee6bnvnmG4BeXUVXTGAxV2/6uioCB+QEV2BfEGXF0XWD/rfgSvXdYb V9iP4CqKoIIueMysoAKKqzjAwAxz3/fRPVf39P99I+tVRWVnVWVVV0ZVz8Sbqc7MyMiIF78434sX EQGBTBHo7UWYDdS8CBQrGwq5pcoEXAqu3LtPldOFbw3LDUndQvgDG5/KJSD4CAgEBAICAw2BMHYZ aDkW+FUEgiWEIhGuAYGAwFGHQOi8B1KWqzIg4jk6YcF163tf7Eeeeg+b7kPd5vDhw4YZo8EygxTt q8S37vc8FSslit9Hfov9uN+73+q9+16i6+kxB2Q5EG+HCC/MZhWo/+EX80bIbvyF8NG99QgvPd09 slN6i/AyxAwazHv1n8x/2vAL4Wh48KJhuveF9yyN6ursMuPbx+MhR4X3GmYxD+57/SZcAwIBgYDA 0YNAGNMcPXl9JKTUHfUcCekJaQgIBAQCAqkRqEeHffhwj9mzp9Ps29dlBdvW1jYzatQIM6gBR18h XHd2dYkA12k32hw+fLgZObIxvIBtpwiSe/fuESwGmba24XXGpZTQ6Qq4haLQIwL/5i1bzP4D+y0/ vBkuPI0fP84MHTqs4DF/V134+c/sTalvRfwWXLokj3bs2GFYy8umYthstI9vF3xGCW98W/r7KJ5K 7/FVyo+Dj9xuFz727tlj+SKfBolCZPzYcWbkqJFRVCX/pgi/ym97pfw+v2KF+cMf/mB27txprrzy yoQQHP4T3tbDiTw6ePCg5YE11m1tbWb06NFm2LBhDdnAdv/+/Wb37t0G5Qy8jBw5Mm9+XY/0VhMG PCgv4EGZpZ1pxMa+5A28UJ9aW1stJvDTCKIuU2b37dtnedE8agQuKBXhBVzII+WF+u2bwIU86pQ+ iSWPI0aMsHWpUbjskbZu7969tq8Gl6jN9Y8LbcyuXbtk7LDHKqDBZcyYMQ2pR77LRIgvIKAIBCWE IhGuAYGAQECgSgQYBK9fv94OPMeOHSubXB4027ZtswOtGTNm2IFOlUHW7L1beNmwcaMd7DGwYqZ/ 7dq1pk2UItOnTzetba01h13thww8NwovO3fsNOMntJvDMuDasGG9HZyDS837b7jyp4QpI7a+rPWK W8wZfsBmjwyGp02bJkJTmyhr9pmtm7eYLlEezZk9Jxr81Ri+ZSL/LTdQxEThtAhxklfM8K9bv84M Et6nTZ8hSgcjA9G9Zs3qNZY3lCJYJPShlOHn014Bn2gQvNts3LDBjJM4x40ZZxlEUbNm7WrT0TFP BN5WYdnZ4DPPA17lISX+Ni35b7mBCvisXrXSPPq7R80myY/lzy83mzdtMrNmzoy8xf8m5K/1kg8/ /kF1z+CydevWqO7klA8ojKhLM4WniRMnVhdgP30jTK5Zs8bWZ9oYBBeeqdNTp07tZ+jpPwcXBKYV oiRCsEUpwzO8gMvkyZO9ClCKCylob2+3gi5tMXxQx30KuSgeli1bZvMIQVJxmTJliqG988kLca9e vdpae40bN872TeAyadIk77yAC7xwnTBhgr3SL1COOzo6vOKCIm/lypV2I2riR2FPno0fP96W35r7 pPRVKO8TXqg35BX5gmJv8+bNVgkxe/Zsr+OGPFPhJiDQAASCEqIBoIcoAwIBgSMDgfUiTO7Yscsc d9yx+ZnJ9vH7zDNLl8oAy5iZM2aKQOljlgUhf6PM9G8WXo6zAgIIM8Batmy5WbdurZl3zDHeQN8i fGwRYXbevGNkwDnGxtstg60lS5ZY0/95c+fWhosrmycJwMTk+pEHhKcd27ebrcLPrFkzLSZ4Y0YZ 64h169ab0aNGmwkiyNhM4yWUMvzIM/71Ln9jHdynXnPYbNq00Q46582bZ61U8DR8+AizS4TNDRs3 mGEyozsqb4XgfJ2/zd/0Cb/Ag7wqy3+v2S+Cwfp160yrCJRTp0zND3ynyJFtK1etMhukbM+SAXHR MhE36rLhW9ZgQm+cW8ct56On+7A5JHnR0dEhgvUUc88996CvSaaiz50H5zb5w3SuCEwIKwjZJ5xw gv0IgQXhe/ny5bbc+JptR3m2VNoSBFvqtW46B3/wgjszqD4Iy5Dnn3/elodjjz02Xy4QMuEHxQTt jQ9ipp/8gE455ZQ8LxtEoQYuWGb44oU8Ik6u5BEWGRBCJe48I2j6IMop+cFM/xlnnJHHhfL63HPP 2fKDgsYXkX7q08knn5wvpyiP6AfAxZeyCFwou1hknHrqqXleUEZQv7AQmTNnji9YrAICRSe8YI0B gdMTTzxh72kHfSqubKThT0CgAQj4GB03IFkhyoBAQCAgkC0CCLDbt+8UYQVz5LZ8ZG0yAEaA2bZt u8yy78u7Z3lzYP9BO0OKwK+DGuJjoIfbDhn4YQ7rgw6LgLB9+w4rrI0ZUzCNRphlxpJBKFYIvgjh YLfMOPX0HBahbWwu2mjaHKyGDh1il0YgBNePkkXogwfFfFysHhCSENpcGiMD4gMHDsr7Xa5zifvk 8Et47uOMEcMeWSZz4OABM1oEWRVu8Th8xHCxmhkmgkynNefu83FdHIr5n3fMXPP2t7/d/L//9zdW MCGKOukUquYW4RHFlWtlgJCCwIQ7FhG+iFljhH/idvNIn1eJsggBywdhDYKgBC6uYgoBG3zgxRdh bUZ7hpWBywuWB9QrZpmp9z4IwZY2DVzcOk1bh4IIXshDHwQvKCCwknFxwQqBNoey64sXloKAi8at 6UdxBjZYZ2AB4IPgRa0OwEEJXuirfeKCFcR2UYqjJHMViPAFLtR5eA0UEDgaEAhKiKMhl0MaAwIB gbojsFPMont6uq2g3yKzxy4xy84gmAGhD2IfCARK9qNwZ1C4RxGB8OSLFwR+BpfEm4QL4uWuXTtr gEUFV67uvQalbvocXVkywxII09JrFQ6RayTiItjB4wHZJwJ/EVUXfu6j2KUgQhe4ivbIQHk1ZAjx 4qfwNjIHFj8i/Fem5PCj7yrzz0wymCDAYnnhEnyxOeXB7kNW6ERhEZHeVA5fvyh9jfNfwEJDZ9WF i0/xfemQ+/OGerJJloJQLqjDLvGMsI2AUCgrro/63iMsImzDiys4EQt8wA+CuA9BDlxQzkBxXqjn CN+0Lz4EXMoss8iKgWUq9wc3+EPo9NHeaXnhikLTbXu1DMGLjzwCAqzP6HfieaT9gFXIiqIia9Ly koQLvMAfwrgvxTgKNMomdcbNI+5xo0yhwPFB1Gnaj3h5IW5w8ZVHPtIa4ggIVEKgeORcyXd4HxAI CAQEAgJWqGdtNuTOOCk0DEAhX4OsPXt2W54YhLuDLHgYLAIlBC+H5XSIbClSdjCoQ5CNE/xBtc30 qODK1b23QTpu+hxdD8ng85AI1OxsEMeGfRnYCLK7u8cRoKoLvxCbis8FF+6iExyiOzYvRbhm09KI F00H+USZaRHBf79cVeDnO6XK4Uc+K/OPVchB2b+EDTEH5RVohfAHC38IENZPb0+OAeW1cvjKcfG1 EL7r7uLjukcQaJy8ce+LfOYeksNP8lnKDcEIYYW86VNWcnULAcKHgAsfCLBJdRreKC/wi3VC1kQ8 8EK8Woc1Tp75UV6Y+c6aENJoPzTeeHy0x+SRD1zII41H23zlR8sQuPgoL8SLsJ2UR7yDP7DzwYvm EfHG+0flD1xq6wcItToCF8pLPI8IRd3ABZ6yJPpF0kw8Gq8bHzxCvsYNbtzhPiDQCASCEqIRqIc4 AwIBgQGNADPJDLQgHTi4CdIBRjRjmu3AhngREqAkXqLTFow5JEsBOH0gSzp8uFcEgMj0uMUe81gc G/yJHCPCLVYH2eNC7D2SZsaWDH77CrKRW6/wTZ72n8oLyggtlsRb3GekJBFUROg/LEqCZIp/leyr kqs9qlT2YSBOFDEFiu7V6bAoZ+o7MHfjKsSadJfep/t1bV9pCFqPKCtah/UdV8oveOTz0X1Z53va F35RnSlOlyvI+ZhlBxcEqFK4RHXLeJnxp56CfxIumkdcfeQRcWg/UKq8+OKF/KEskBeaH8StBF6Q D1zIo6jv66uEgAflxUfZpb5qvdZ44UFJ3cClvm2dxlC4ah7hUqq8kHc+cClwFe4CAo1DICghGod9 iDkgEBBoMALuYM29r8QWgxUGFFC57/CDYJ41MbNdmiIBpkeOEkXYzpqUl2LhNhar4Kf+Ym8qPLr8 671ec5/GHtX6QwXrQgTiEWjkx+kd/IoVIxqQXpPDL/6mEHrkzrcqQBIHgj8u0vWqc04Zo/xRXiJe NKxY/Opsv3PD54XrV+/1mvtQH0kv93k+cu+5YB1BHkn5tbDkX+nHOOi9XnOeYo8Ff/lAHI94dhmI PkY5UgimcGc/jD2WDj8eX+XnSooo6jp1XwXPyiHW7iMNL4Suwl7tMVX+El7KCWjaBvrgRdtduNZ4 3RSoWyX83G9qva/UD/jkxU2vxuumS918lF3yyOXH5YN75aWcn/g3tT67cWi8bljqBi7lyrj7Ta33 hO/yEw/H5SX+Lu2zhpHWf/AXEGgkAkEJ0Uj0Q9wBgYDAAEVADNmLpbNYOiLByg5S6zLDHgs+9sjM dilS4bbc4KfUt9W6k960uFTPD9InuKoUqsKrXnPc2kf14x4wGfOnwq94VYxqCT/6hriTwnfciEfW YljOZH+Kgv/Ij3W3SXBF8L7+rJecv0IYOKhfDUnj1mvuS/uY81P0yn2Q9wJK8YC2hvDzPBG3G74+ u24aPikp3Pf5zn7Ce6iMv8hDVX/Ll10giZQQ1Zfdqtiwnl1hu9zXPniphIvy54OXZsKlmXhx88i9 17zRq488Iq5yPOg7H7wQl8anV8XCvcJLufeu3/7ca5kpblujEDX+/uCSFG5/+A3fBgSyRKDvot0s YwthBwQCAgGBJkKAdaBscgbpHg9p2GMzQzXjtHJQn48iIQl/Po7o1H0f+rAhDjqwYZ2/I20nee23 GwMgxSVZR6O4FPylj1QFVr26Amg8FPWDIC15IK+LrQsi/5GwG4VTsNzQb9OFH4+573MuHIR62X8C Qmek+ZL3r4DhBUsES8pL7jHxonyqX72qe9JHES/4SKQcL/ArbOdIb/RaPvzoI/WrYSRdNRzHrziV U6wVFBPON0lBV+mWZCLtBkGeuWXcfVfv+zS8EGclf/XgizjSCDe+eEmTJl+8pMFF28Q0fNfqJ216 0/qrlQ++I71p0pzGT3/4UF7S5BG4pPHXH34IX/Hv0/47Aasfx6nsLctNdBxT9T4b0qb1dsvGzWUm MspGHl4enQiwr9VgOWmmMECoCYeghKgJtvBRQCAgcCQgwKZZ0akEJn9Nky4GTzpQUHN/9zsdYLAf g4+BFqctQElCpcq2g/MnMric1veeQZZuRKYYuDHk3ZzBmPu+9H1OUFV51XpMJ4AOHhxt1tlHqCWs HGDwzWaRRRPrfWbuS3OX+CbPK3xGD1rW8hE7H1p25M/gtGUmIfy0/NvyyxpxWfqRtETHBi1/og00 UYjkIsvHCePp8HeSWHybD6uAj41Gg+a9Zyrkj6RYK06MB8qK6y/2um6P1CPiKsWHuvvghTjgBdJ4 kxLqHlGZ9L4ebrS7lF/4SOJF3bR9rkecpcJQXniv8bp+1U3bRPddve/hpVyaffJC/miaNd6k9Poo u24eJfGg/MFv1n112rZDsUviN8nNDbdcGejzLfsPdckkjCzVDBQQqBaB3kGyQfCIyf1SROh0S7Vx B/8BgYBAQGDAI8CRWGPHjrW/+LFm5RKHcgGhHkqaYWctPcRgQgfv1iGjP8NESIAOJwwmVPge6oEX 5JQhQyPdtpqduklmvwNku2FDh1WJS07gzV2iMNNJqUMlrsGSduItMnOVsBDA4ZN8apVjBovl6nTh u+krui/iNXrgWDYEbWv6mysj+s0hu9FprxneNjwdNgnhp+WfwfZQjlCVa3ef5UKy54EcPStMyNGL KnzmIiuKs/74uPxLSVFovF1VKEIwiZdf3Mg3d8CfJWMqyBGvCkpufLjBiw/B31VClMIF3hQ/l896 35Nm6mspXODPVx7Bhwp9cVxIt7r5wIX4iIc4NV7clLQM+eAFTMAGSuJF22EfvMADx8hCGq99yP1R /nzw4pbLpL05bL8g9bpaXkhfLeOY3gN7JIOCAsItD+G+CgTYZ+yglKF+UFBC9AO88GlAICAwsBHQ gVn1qWgxo0eNtp/12JMeikPozrlZobP4VSZPo0fneLGnGRTvD6GDnREiAGc904MUP3JEFE+PFaqL k9uDcCvC5ejRo4pf9PuptMA6TJQiI0eOsPF2ywkhLiFsMwgdPrzVCtzuu+L70uEX+yv/RHnAaqX7 kMQrg80CyakiuZMzRtUdG2Jx44pM+KN8apHd8g8U2JA7FEXwN1QUEPhh4FyZisOv7D+NDzdM9z7N t7X5YTCPUE+ZiAssKiAgXI0aVe/y25dfeEExmrRhHvyxCaT66ft1fV2Ip62tzeKi7YnGAC/ajo4b N06dM7si4NLewQdxx4lTDsijESOo89kSwqLGEz91wi1D2j5ny40x4E9exMsu8VJeaP99lF3ySOOJ 4+Lyp3584EJ+xMuu4sIVXtK1dbVzC/6UBeJJ2sRV860/5UXrYiougwIiFUzBUxkEDhePqcr4THwV lBCJsATHgEBAICBQHgFmHthvYL8ci1bc8XOE34HcgE9mvj0QCgYGxBztVXzqRG/+uK8xY8Z44CQa zGHpsE/WqaoVhkZ8QHBhADZ2bD14cYVjve8rrGKxMnr0GGu5snvvbmEl8kOe7e3caweDY8eNFwuO yJpEeS02iygdfsE/d4StPLj3kS8sVhjsdu3bl1c6RG9azO7du0UAjma0Irekv26Y7n2SX+WZd3of 8UYejBo10gqxxNstyquIes0+4Y01xihMylsHaZh9w88FlnBxeXbvHa85a5niY0o1roh/x3fstkSY MV+lHsFl0qRJtkzEj8kDE4QE6r0v6wPiopzqEYPKN3zAH2UJBUHWBC7t7e2JvCBM8aN98cELAi7C Njx1dcladofAivJLufUh4Gp5gYXOzk6HE2MFXgTwyvWo6LN+PYwfP972O/HyomWIcutDUUQiJkyY YNNCfriEMgD+UGr5UtKTZsoNdcZVXIELbuSjL1yIh746XnbBCFzII199tZsv4T4g0AgEghKiEaiH OAMCAYGmQIBBSK3EQKK9fbxhc0t+KoB2dnaJQLnHMCDUWbJa40j7HQMXhAT4IH6lfV37DBtVMagp L1DqF/2/Mgs5XnBhAL5z5y4JMBIMmYXasX2HFQ76h4vmmV5dnhFW1T26MsCcINhMaJ9gtmzZanbs 2GkHezt27DA75X78+HGmXX4FKv6+4M5d3/CL4+O9Csx6r+HxqsVMnzZNBuCtZv2G9aZzb6cVmjZv 3ix4HTJTp0wxw0fIZk+W9Dv3qmHiQe/1ffRVMT/qple+ifyPkHimT5tu82nTpo223OwVfuCFATvv IpPq2sLXeIqvyjP86H2vWbd2rbnzzjvNN77xDXPfffeaIRI/+fP5z3/e3HbbbeYXP/95LgEF/ovD 5TV8apg57zVcUEKQfnDQ9gHBZcOGDVZYmTp1ag2h1vYJcVG3161bl5/FhZf169dbhcjMmTPzywFq iyH9V5MnT7bt2aZNm2yZ4Uvw4Rme4MUXkUe0IeSJzijDw8aNGy1OU6QeRWU3e46YtUZZtGXLlryy CFwov7TH06S+++IFPviBg+ICAsobvNBv+SDFhQ0TUXRC4LJz5077ozz56pNQdtA/wosr/LMZNXlE PfOFC+WWsQH9sosLyhqw4V1/LCF85G2IIyBQLwTCxpT1QjKEExAICBx1CEyfPsOarq9Zs8ZMnDjR pn/btm1W0GZQbjc79IAKgjYDqQMyk4JAt29fu90HYevmLbL/wlAza/ZsK0B5YMVGwQATJcSaNWvt FTNUBuXMfs2eNbufuCBoQnqNnvr+1ffR/h0I/1u3brMDvd27d8mA2IiypN0qIYrzqfBd3zBdl2r9 Rd8Ol0HorJmzzPYd282WrZvFUcKR/3Mkj1iK0feUDo3Hjdu9j7/XZ726ft37FjNm3FgzQ4R3BsNb tsCLrCuXZRiTrIDQlvMcD0ef9Zrz1uei7/Xax0M+fMrF1GlTTc+hHtPR0ZH3iNDC8pBxMjDvSxqu Xvv6qMUFAWC25MXKlSvNsmXLrAIPQWX79u1m1qxZXmcpwWX+/PmWj+XLl9vZWoQVhMo5c+Z4FVZs 3RVcnnvuObNixQrLCzO3KAJmzJhhhbxa8K7lGwRG0v/888+bZ5991ra94IJChHaX9scXoSSaN2+e eeqppwx5RD8ALvBCm4xCxBeBC2WUcgsuWCOg/EVpBS/8fBK4UF7IJ+JGMQIuCNrwSb/lg+h/KC9Y PcCL9tXggtKG+u6TSDv9I/mEYggFmuJC++cLF59pDnEFBJIQaJFOPj7NkeQvuAUEAgIBgSMOga99 7Wt29pWEXXnllfbnJpJNd+zmTa5jn3sx69/Tafbs3WMHD8y6RHse+BlgxdlhhqVTzJQZ2IwRgYr9 EFg20gjCRHn3HpkFE0FyhPDBgK+y8qASp9plKb46A67X0t+zPOTQoWgtObPdzFD2HfDVHr7Oxpfn JAqf/OkWoVvEbDubHc2WappKpSEKOU34BZzVt16Lw2YIgHl/tIyn12IySLApzQnhQOpDw9Vr9Db5 b+SnvE/eQrWHP2iEKAQHy0aj/SAwQdhHcGGZAQKdr9nSONvWikiUeNQn5QXhtxEELswoI2iDB1YJ jcKFOoRyCFzggdlulCWNIHhhJhuFFXW5kbyQfp31RwCn7PqyOohjT/uCkhOrA9pc+gAfS2XifOgz 5YU8ot0nj7BM6NsHqO/sri4uxI+1YtQ/Vh/nd7/7XXPLLbfYDy+55BLz4Q9/OFUgvV3bTG9P8Z5A qT4MngICOQRahsgG18OjpVfVgiLlviVYQlSLWvAfEAgIBASKEJD19TKDnc2GgkURpXpgJrdZzDlR yNR33W9cfHWfVWgtDRPKmEh4c79z/cfd3efK4avgLAcrSqBJ/gvuWF8Ma0U5lOTP5cm9j/ymCT/6 qhBfqXgYACM0yX+H3O8c5z7pcv2lSUf9+Xe5K5XGYj/pnhCYfM8cl+KM/EHY59doAhefs/vl0ouA zay2zmyX85v1O3hBqOXXDNQMmIAD7QvCda0Cdr2xbJY8ajZc6o1zCC8gkAaBxkyPpeEs+AkIBAQC AgGBgEARAnFBN/6MZwTjWikeXvyZcEP4BXQHGj4FzsNdQCAgEBAICAQEAgKNQyAoIRqHfYg5IBAQ CAgEBFIhEBf848+FQHrzlgW9CeoCvuMXF57j4cWfQ/gFBLhrdnyKuQ1PAYGAQEAgIBAQCAg0FwJB CdFc+RG4CQgEBAICAYE+CKjSQIXf+HPhg+gN/lpiqobIraCA0LD4Nh5e/DmEHyGgmA0EfAp5Fu4C AgGBgEBAICAQEGguBIpWgTYXa4GbgEBAICCQLQKsy1Ry79UtXJsdAfJPlQvV8lrI+9JfhvBLY8Ob ZsanPOfhbUAgEQHZSLG3W445lk1sG0EtQ+SI3kFhaN4I7I+EOMM4pnIubpZTsqDJE2vbULFyDMFH WgRCS5cWqeAvIBAQOOIQYDMxJfde3cK1mRAoVjYUnlSZ0HeWnqUZ+rZySgoh4rfwpCGE8AsWIwMR n8olIPg4yhE4fMgc7toqhVvrun88OI2ppW2saRk60n/kIcYBiYCreGDj2HrTkmeeM089s8wGO3Lk cHPWaaeYSQNYgL/pM7fatHzplhvtdYsoJRY9/JhZeM7Z3tPFSWZge9bpp1peyv3B76rV68xJJxxb ztuAeheUEAMquwKzAYGAQD0R4Fg1zpeHOHouUDMjoMqAiMe+J0S476P7Yj/u+4KKoSBOu+8Rt10/ xOm+D+E3Nz7NXI4Db82KwOH9OxqqgFBcevfvMi1DRkiT47Y5+jZcAwLFCHDcZxbjmJWr15pbbr3N bNm6vSjC2++6x1z4xvPMhW84r8h9oD4sESXA3ff91GzZtt1c9e63e0kGip3FjzxmHn/iz3JMrCh2 yigh8Pv9+35myI+O2TPNDSmVECgtVorSYvLEdu/KlbQgBiVEWqSCv4BAQOCIREDPuKcjDzSQEOjP AD3Nt2n8lMIrzbdp/ITwSyEQ3AMCdUegp7vuQdYcoFhlmMHDav48fHj0IMDYpd7jGKwD1GLg6ive bs487VQzUoRlFBK33PoNc/e9PzUjhw8357924YAHGgXA5TIZdZakMWtC6fDlb95pOrv2WTzLxRf3 yzfV0ONPPGnjQrGy8JzmXHpSsEWuJmXBb0AgIBAQOAIQwIxxyJAh9heWYxwBGdonCWkE/TR++gSc c0jzbRo/IfxSCJR37w+25UMObwMCjUSgFyVEoIBACgTccYy7NKPSp729pZVuWAYg9F5+yQVmwcte nBeYJ8ms+g3Xv98+4+dIIJQr55+7UKwF2jNPDpYJWDNce80V5j9kOQhWEKVI/d5w/TXWbyl/A9k9 WEIM5NwLvAcEAgL9QsC1fnDv+xVo+DggEBAICAQEAgL9QMDuDTGkTZZk1H+Nfz/YCp82OQJpxzGU L3M4eQkqygf2SOiYPcMqIOJJRmjHamDRI4/aPQrmiD+lBx5aZB54cLFh80fr7/RT7LINdw8JXV5w /rkLzMo168T/IqvwQDg//7ULzEJRekBYAjzw0GKzQPZqUDeNB2sCrDI+KsI88SQRyxH43g3/wje8 vo9Xljnc8Z0fFsVzx3fusUsZLrv4TfYdfsDlZFkKcdW7L7HKg++LEgaccI/z3ieSnAPLWNJSNX7d MDU9ugEnecLSj8svvsBoXmHpgrvL/1mSVyhjSuHpxlGv+6CEqBeSIZyAQEDgqEWAjp8fS3hbWhpr YKa8kBmDBjFT3MjZ4l5z+HCEC3xUM0tT/8IU5RGrbhqPi4z/ZD8SCEwaiwvL4A/bvfgCL8WlTutS c+CibUwzlBfq9WGpR4OaoOw2Dy5aeg4dOpQ3j1e3qq9SJw93bpG9IUQRMah6RUS+7Gq729huwByW 9LBNcIvtk6pGo24fNC0udvBQWzKH9h4wY9qiMtI2qNv0HhQFQznqPmh6ew6U9IEQC518fOkNEC8T C4m3vPH1IrDK3iU5ul0EdwR+BPUFLzvbKglQJLAs4EaZzZ8jSgYI5QGKCOJBeEfBgCD/+B9lqcJt d9plHiyRQHGBP/ZtcpUQ0UaSj1pFSDmBmeUkWBPAz0nHz88tJbnN8uBaPRA38eBHie9wIww2gkQ4 R2FCem7MbWxJ3Mr7kqXPWd4nT2hv+MaR8EVaRoxos2kmrR2zZuatLsCPNJAPfy3pwhqDtLHEhvSp pYtikeU1KCGyRDeEHRAICDQ1AgxI+ksH9h8wGzdtNPv37bcDrOFtbWbatGlmyNCh/Q266u8Z/G7c uDG/SdVwWbM5ZcoUM2yY/7XF3d3dlpfOzk4rqLS2DhNcpjeEl8M9h82mzZvM3r17ZfLnsIGXyYLL CGcAVTXYNX/QazZt2mz27Nktm6EeFmFFeJk8yYwaNarmEGv9kPK/ZetWs3vXLtPT3SO8DDETJk40 Y8eOrTXIfnwnvGzeYnbv3m0OSdkBl0mTJprRo0d7F3TBZdu2bWbz5s2Gcsx660mTJpkJEyZ45wVA t2zZYrZKPlG/WT42UfIIfnwrr8Blx44dUn43WV7YiR8+GsELuOzcudO2MQcPHjTwMm7cODN9+nTv uMALhGJm2bJl5sEHHzTvfOe7zMiRBQEt8lHlX5SDh+S40Kqo13SJULVJ+qQDB8BliNShUdLGTLZK o6qCqoPnLpmNpvzuk/5x8OBBts2dKm3vYCnHvmn//v227HJlsmDUqJEWF+qUbyJvNm3aIHm13yrF 6auny7ihFlyGtRw0Y4ZHkx+tgw4Za+XQjwQhnELllgog6LoKgEUPP2oVEAvPeXHR5o5btp5nrrvh M+ZLYrlw803XF3GFcM+yBKWVqxeI35vtPgbfEiUElhgnnzBflAHLrJJC40OpAZ19xin6aZ8r/KBI iPPDCRiqROjzUYIDm2+6+16gJMECxF2WwmdPicKCcBeJxUGjT69AeYMVBRiAFVYr4KBkN+GUPGav D5baKLHhaGTJssh+r+5ZXhs7ZZdlykLYAYGAQEAgYwQQapc+u9QOPjs65phZs2aZHjFxXPLU03lF QMYs5IPvEmF/6TNLzcEDB8yMGTPM3I4OK0A9I2579lSYGcmHUp8bBnnPPvucDLC6zJw5c0zHnA4R noaap55+2jsvhyU/li1fJsLTTisszTtmnhkuyoelS5+1igmsAHwRQuTKlavsoHzixEnmmHnzzOgx o8zzzz9v1q9bL2z0XymWNi2ke82aNWazCJTt7e0GXMaNHy+8rDDr16+z1hFpw+qvP4S3NWvWmvUb NogCZJyZf8wxZvz4cWbFihVmleDlF5deyaOV5rnnnrNKhxNPPNGWG/Jo+fLl1uKpv+mt5vvVq1db HFAMwQsKTvh75plnqgmm335RQGyQ/CHeMWPGmBNOOMHMnDnTrFq1Sur6s3nLnn5HlDKAdevWSR1e ahDewAVeUI4sWbLEtnspg+m3N04l+PXDD5vvfe9u88///M/mH/7h48LDU6KkOdjvsGsJYJvs8E/b 1traZuZJ+zJjxnSrrEE5Qvvjk1BYUY8oO/PmzrUKIvqpp59+RvLILy8orOgLIfqk2bNnGRTklCEm EHwS44Zlgku3KH7BZfbs2bbMgsv+3GldPvmpR1yY+0PxEzMQ1hF0UQggqLvEUZ8uYRWBwIxlAkdR QrpZ5GIRqJUe/+OTVgHiCtD6Tq9YVUBxflAQuFYQ6r/U9awzijerPEmUIhAWBKoU4VkVD6rAwa0Z CWxZghFZqxQUEPCK4oI0oWTxRUEJ4QvpEE9AICBwxCGwbu06O9OE0N8qFhAMiGfMmGlnKxHwuj0N +npFiFsrQiwKkJmzZssM3EgzVKwfGJi3DhtqBYXDHo8g3bhhox2Ez5k9x7QJLsPE8oAZyrZhrVbw 9YULBW6TzK6jhJk5c4adKWVWG+uQ9vZxVvDv7Kx2prH2YrxDBsLMsJMvzNpiLTNJZrVRAmzeslks NTprD7zKLxFWsIKYLtYpxA8uXLHKwFKjc68/XHaJJQZWB+AyQWZxwKW9fbydpdy2fbvZtWt3lamr 3TuCEzO3KBQpJ+CC5QHP8Mg7X4SQhBKCWWyUD/AyXhRFx4iShnKE1ZMvQnCiTZs6daqty1hXoRiZ K0LUdskjFBS+CMEfXiivCHDgAi/HHnuslJVdFhcEXx+EovX3j/9eeBhiXvSi08rOHmfND0qGzVJ3 mcGmvJBHWHshdB8Q5TT9lS+FHvivXbtW2v/htl7TB9AvzZ4z2y7N4J2vY7FReGxYv8FawNFX0yfx o05DlCWfuKBAk0VEtuyCC7yQRyxZWSO4+OwfLQB1+KN7QCQJ+Fg0QJtzFhblosPyAVqRWxKyIDeD r0oFlhKwTKLckZZ8T1wI1En88L6/VM5KpL9hZ/m9LrXpmBXliRsXeKFkQZGCssIHBSWED5RDHAGB gMARh8AhMQHe27nXmty6yx0YEGOGi+CLWawP2i8DTAbmY8aMtoNhjRNeRsusJZYJez0J2wyg9uzd Y4bL4LO1rVVZsSbSY8aOMV3CJxtG+SBm2HeJ4M8gL770Yvy4djsIRmjxQQzKt4vgiGg0VnAoUItd csD7nWKt4YPAhZnBoWKdgmDg0oT2CRaXrdv8CNsWFxFiwQXrh8ISgxYrWA4ZMlhwk/cyQPdBCNRQ fEkKAi97IPgS/MEFpQdXlBAFXIwtL62trVYhwnIRH0R5QWhECeLyglUEvKxfjyWPHyKPSDfKIZcX lu7ACxYRLNHwQSzR+dCHPmje9KY3mVNPeYEZPMi/ab+mkz5n/4H9ffII5fgwUQDv2r3LHNjvBxcU aCg+6JPc5Q4ss6LM7NwpvMh7H8RSEHAhXpbtKNEv0Ff7xIV+Gn4oqy4utC3wRx7u7fKnjFYs3KsK 2NWOX/Q7Nyz3Po2VQDwMBGN3SQZLM6C4FYUbD/fwHg8r7udofq6EDeM0HzSglBB0xmidWRupnZAP kEIcAYGBjgBCB4MCXzMPAx2vNPxvlhlR2iRXAaHfMRDm3W5Z9++DEKQx/2WA5w7KiVv5271nl+Up a362y0zywYOHrPVDPC54YXMyBBofhGkr5Z7BHoM8l1RBgvWBD0GOvosB6FDhJb55KcoiBsfbd0QC sMtnFvcHRClFmQET1mm7pLhgIeJDkDsoa6PJgyGS/kGxDfjAhLyjHmG6nDVRDhBgiVfrjcbJM3hp Pqp7VlcUh4xzFAM3HviAH6wT8Jc1gQvKF+KlbXOJsksewQd1LWuCF6xR4MUV4jReBG6EW7A52oix Mf0OeRKnYWIRB3Y++iTGG1h2QQj6LpFv8AcvPsoLcaP8hSfijfeP9JmQD1zIm23bt0naD/WpR/AA f/C5d09jyy6m+tDKNWvtNekPm1C+79qb8ssm8FNKaaHuahGRFJ66qaJikizjUNJlFyzJwCKCd5Us IRCyNV4NJ1xNfglJKWzYowSaJBts+qDi0YePGGuMgwbrYVl39+1vf9t89atfNV//+tfNAw88YNdK UmkDBQQCAskIMJv261//2vz4xz/2MmBN5uLIcqXN2b8/0hQjJMSpRQZaEEKnD0Iwgth5PD7IUsFu n8wMZN1WMshCGGHWmsFmnBC+2QR8j6dBVpfMOMHTIHsCRTE34MQPixb6l6wJwQj8E8uL8DFIsEHo 9yH4symaxYWd6iVul8CEvDt0qNvLTOUhGZCDf1J5Id/ABTNzeM6aKLvkkZYNNz7Fhfc+hCcwgR/K C3G7pLzgR+u++77e9/BBXBqvG77rxqaiWROKdNIMLkllBnf8+Gp7s05v+vB7pVx2SR4NSsQFrKjz PnDpkbJyUDZshrT/cdOh+eaDF5ZZYFrultMiXnL9N7xkbW1F+Ady+08k9QPgAp+N3hfCtT6I7+Og 2HGKAhaNeuSj7uWQ5J+TIyD2fKhETy2NLB0my7I8JRQO8MReBWy0qEoSfZ905WQP8j2Jn1ICeFI4 R5obm1YqlvG06VKXNPjGv631ue8osdaQMvyOhvMnP/mJWbRokd0g6owzzrAd0M0332xuueUWuxER fgIFBAICEQLMNN9///3ms5/9rPn0pz9tqCsPPfSQFwHnaMgDBBFOE4AYNMRpcE4AZzCc9cCGuLE8 gBDY4qTsRbxk207SDut6Vh1ouvxER2Oyi7wvXKKBsCqFXF64J++6e7pFaMlekY0QBz7JeYQyIOIO f1kTgr9SUvnFjTyizGRNWkeScJEMyitJmD3MmlQBRPqTy28kyPnY4I84VCGSlG7lzwcv4KJjLI3X 5UnLkOLnvqv3PeWFH3FqvG4c6uajHrnxNvqeo5B7pC2jGYlbWsGbT1wO0w/k2g5t8118lBeUFVkT uGjbUQmXrMUIeOHkH6hcPVLsssamXPi6oSOnWiCcKqF4wAoCiwWOrVTi1Ako7p+TGFAccDqDa92A 3+iUhkLYnMzAfg8Iwa5fhGaUHGxuCXH8ZyU6+/Ro08uI10IcxOlrv4NKPPp6r9YlxAeW5BsYgIWS 5ivPmpf6Lstr4xavVZEqdvXF6uFDH/qQOe6446wC4qUvfak1m3zkkUfMvffeazeQYt1koIBAQMDY mUM2WGO9Kutm2TFcB5EBnwICOhjCxb0v+Ei+A0sEBKjcd5G/XmmzksOplyuDz9IUSbdWoMx4lFWM S2mOYAPBf8iQvkqT0l9V/4aNvsoReadCcDl/9XiXV3TklA2lwlTlVqn39XDv0bIrIktS+cWN8q1l vB5xlgrDlktelsBFWLHkQ6islF5woYzneY5Yy+RvGl6I2BcupLsS+eKlHB9ann3kUTk+fL/Lp7dk PYpe5P1lyCBlpVBe+jKkeeRD2KYeKS/alrhJV+4iXCqXcffbWu7L1es8Lv1YeqZh1MKb+w2nPVz9 7rdbhcPVsuwCKwYsCBBWEWAXyokXnKagxPvLL74g7x9FAsIvG1ayDEOVGuqfKwIxYeOXMNk0EeXD VRJvnBCM9WjMNDP18E+cCNrKP8oUlmkQ79FA1oJEFEZgwOkll138JruMhXwDi7vv/ak9xhOrE7An D8BMl7/4wGhAKCEef/xxezTW5z//eXPTTTfZnZnZNOriiy82vONM5nPPPddu3uQDtBBHQKDZEeDc 9ksvvdSyiQLie9/7XrOz3BD+3NkI977/zERDG8btOgDqf5ilQ0gz8PDCR2kWc290yOcOUit+VLuH CmNKlat8YJNPRAWemEVsNNlyK0z4wCWf3FLJzrk3Ay7kC5j4wCVtGSgn1KQNI42/Zmlj0vCKn0bm EScfNIxKRZ1zbyQuioni45uXfFujjDhXeCn33vGa2a3iIqW3yji0X022hqwysLx3TqZAmGcfBjaE RHgfKXuunClWBkmKgPNfu9AKuQ88uMhsls2EOWnhwhNeX1KoZUYezrFwwD+bTRJGkpJgjig5oKR4 7YuEPwjbk0WpsSS3xEPDd4/75DP8IHzryRy4YW1x0vHzzQhJr0uqUJkripU4EcbkSdXtpcBRn2mp 2vDB8dprrjD3P7TYYko6la664lKLpWKPUqlUvuo3WVwHhBKCHWPthmaixnR302UDInaYZcMkNuQJ FBAICAQEqkGAgYe2KdUM5hmQD8qZNyQPpKJBBIqN+MZ/1fCX1m/S+tLCtxEvg9n0L2kqqOCx/3cS /mDZpA5KHtBFvIBfeZ77zwohFDauKzWok+URkkdJ65brw0EhFE55gEpxoj7Vnz5ncdXlQvBSqvyK jYQXXNioE0rGhWF59GbI4OyHK4XyYlnq8wesKLuV/PX5sAaHSnFoviVtQlhDdGU/gZc0SohKPJeN JOXLtO1GWn8po63Om5QTLbfVfVi7b9vPSNlMrkfUr+iND1xs/5hbkphYs3NMDvZQp8HFLscrsbRM 8QKXNGW89hyKviyLf46Z6vujXnNQNqaOSFPUX06j77FMQFngLr0oFzL+L7vkgnJeit6h6FhQ5JL8 wB4UUJqlGG4INvzcMZ/qjqLDJfZJcK06eMfykSTC4qPU3hbxMJK+j7vFeYm/d59rCR8lEr8kSot9 0rf1civbq9PJsSGRdrp0MLgxWMd0iR8doHY8PGOOx7NWNNc/4cQ7TPzr4F9nIvkGd5QMEEsv9Bxz PdcXdzaS4Tgb4uLYqGYn0sn6TfAhrUkNHhgiFJH2pPduGgmPHaDZ5RwsuGdN5gte8AIzatQo12tT 3oMF6dVdx+NMkj5+lIX4DstxvzwTFuUBPNi4ih94sIcIcfgi8oFyTrnUOkOZhj/SzO7iWj8UA/Wv PJJu/GvdU3eufMN7twzxTByqrHP9h/vSCICZ1jPwTktgr8oFwogTaz8h1sMmrUON++/vs7arSbwo e4NFCNa09je+Ut8TPvFAioHrV93Aj1/WpH2TYhCPD3eE/qR1y3G//X2GF4t/EjMUl1wx8iJs5wf/ uUhjiYNFyre2U7HXdX0cLEtybB2phIvglzUVyouIbAn80M6Sh+ovS360jiTxQbzqrnU/S17c9Gq8 bnzq5oOXqO0dbNOv8Sbx4vLsvvdxb2tVctXKLHoEV6tolhhU4VAUWa48+8DF5lGufS+fRxmvUxQA aFuGCDbd5lBFXKRqZ0q27chPXvRdKqhYVauI5juUxuS8jyUumYKUELhdOiBLClBwVGMJkRBUcGoy BMr26ux0/PGPfzy/QzMNC4Vdf1So888/37zxjW+0ycLk+xe/+EVeOMK/Ckd8g2D8iU98wq5R5wME 5y9+8Yt2qYUdnDng4P+Tn/ykmTNnjlVAoIiIE3tFsEv1SSedZKZNmxZ/3XTPGzZssCd7cAQY6dUf jCqmXMHtxhtvNFOnTi2bBqw/OC3kiSeesIIpwjpC6A033GBOOOGEst82w8snn3zS3HbbbVbJoFho OXDLzfz58831119fkeXVq1ebu+66y+5/gEAJHgj8X/jCF+yeIRUDqIMHjg772Mc+ls9b8hLS/IWv t7zlLbbe4E6d+eUvf2kH+4qB+uWK8uXWW2/Fa54+8pGPWAWLYqUvEBg++MEPDoi8V54bfQVDVVBV I3AhtA7NKbZ0fb2bFjb1gzRs910W9xy/Bmm9cctGgZehmQvbxIsgwlXjddPLJp1SrE2rnFnvg+xx gvAi+08Qr9zmKcJK9qUQgVzraf5lBjeUL3BJGiSydwXLDRAkhrVGeZkBC/kgbbm0eST9uQjW0gDl 39HukHdDh7ZaBU3+RUY3zIai8EisR8IL/CBIaBnPiA0brNbXCINiIUHdKCvqL0teEBapS/RjxO2S 8kJ5ih+Z6fqr1z3ppfxSZ/i5pLzglmaywP22lnvFnwmGOC+Ep8p7H3lUkv/i7Crprd4v2mS8s09O bUpqe239kvLiQ1FEfR4iZRfK74XjJFbzzQcvRDusdVhpXHKTD1G/la1inLKr44bu7uJ6BJ/gQn2q tuzSPiqWnCh0JBHHgLKvBHTD9dccSUkLaREEyiohmFVmI0h22ueIPz3bHYEfwb+jo8Occkq0Aylo Yo1w4oknmueee8786U9/soIxFYMwjj32WHP88ccXWSzQWbCZ5IoVK8zTTz9tlRLjxo2zfufOnVtW sbBNzv39z//8T+vnfe97n+Wl2XMU6wawYAkJ2D766KP2HHD4HjFihHnFK15hWHoycuRIw54XlYhB ATixAeFvfvMbawHAtz52qK7EW5r34EF5wXJg7dq1eWUK3x5zzDG2bOGH+zQEhiitsIQAWxRU4OmT UKydeuqp9hxzlEMcjwlNmTLFpoe9Gl70ohflWcKdurR8+XLzhz/8wQ6gePnCF77Q1gPqmEt0UODx 1FNPmaVLl1rsUFaRbuLFYihQegTAU8m9V7fS1xYzdvRYs3nT5vxpEK5frFWQeEdJffRBY4SXTZu2 iMAisz2SJoQTJYQYaNSoMVaYU/esrmMknk2DN9v2Px4HuDBLN278uPirTJ4R0FpFgGIAzgDPVTZw BCU0YuQILwIubRkD826JN55H9IXwN3r0KC95NFSUVggsKCDAxh0I6OkmYDd8eFsm+eIGyhiBuNj0 DCWVtYrIeQATyi/1qNqBuRtH2nvioH+mnPJzBXxVBuDHh6UhccMLYx1wcInyQz9PmeKXNWma6Vs1 Xo2TssuP8Qjjt6yJOsz4iMkc4o0T1ruKXfydr+fq+pL6cUW72rW+K39ykxsy5RkBlTKVNaFMpY4g MxzKLxOIYgUb6hL1nPGqD6K8UHaT2l7tH33gQr9MPNu3bxdcopOb3PQrLtW3L4VxjBtes96X21ch zjP7OHTKuJplGMEKs64t8wAAQABJREFUIo7OwH92xx59UjNz5kzzgQ98wDYYV199tRUS8fS6173O /P3f/30fk8TXv/71hh8C8ac+9SkriLF84r3vfa95yUte0id8Ktq73/1uOyv8D//wD7bB4vnVr361 Fcr7fJBzQGhlBpnKfMkll1iBzR1clvqu0e4IiO94xzssGyghOF6UUz8ghE4wc4UH+6LMH04+eNOb 3mR/WE5wjOlAIhQQ/CAE6o9+9KMGaxEIa4ELL7zQ3qf9g0DPZoxgywami+RIV980e/ZsW2dQPvzH f/yHzV8G2X/7t39r3vOe9xQJQfD2mte8xv4WL15sFXHwTjnB8gOFVZwoH9RJFHfkOQqoq666yrzy la8sGjDHv+O5UQOjJF6OBLfRY0bLwHuItUpxhScEBgbCrHUf4WmQNVw2IGKmGEGO+LU9JM9RjEnD Igq57Aef5KsK9WCgwonmN7wMEWHF1+Bz6FDJA1FOUq/igtzevXssWygqXcFXea33lXZgpPBCnd0v Z8WTZ0r75BmsfC0rBBfSjTVdXNjes7fTsoVA6QMXhFf6cpTGXZ37zMhRBcGEpYkcaTdp8qSq+kbF tZYr/SpWdZRfVxhg3EEZ4j08Z03EoXlE3K6VAYoAsKHP86GcIa2km7JLPrmTJNRpyhDlxRcuxEUf S712FR8IcfCDmw/lTKky0CiRkPJKHnR1dhkzscAdbQvlpbW1rahMF3zU/462bN26dba8xPvHfdJP cUqBW6brz0EhRNoXcCFe+kQdZ4ML9Zx3bl0vfFn/O+Vl796uPkpX+KM+96d/bFTZqwapcvsqxMOp Zn+J+LfhufkRSGV7hFkgnQ4VVwUZrcRJScSvNi74K+eX79H00xhcc801ViHBoLEU0ZDed9995re/ /a3BAuJv/uZvKgpfpcJqpDuYuOkEr0o4leOX2fQsicHHj370I/OXv/wlk2goY+5gysWm2ggJC8ub RhL8u50a9UYFwyS+6HTcjqecX75XoQqlHYo/d8YuKfzgloyAtmfJb8u7kkfTp0+TQe9+UbiyMW5k Ok5dIX+mT59uB1rlQ6nPWwTc6dOm2wH4FhEsGfRBtK0sq5su1mtu/apPrMmh0I5hnUObvnEjSkWG Rb1mz+49VojBGogBqA9CiJ4xnV2sW6y1lc56dcognaVTCCuuQJU1T5SJ4W3Dzdp1a/MWNAf2HxCL mk1mtAgQ7Z72NgKXqVOmWqEEQQEhEqJ/Xb9+nW27EDp90ZQpk237t3rNGgMeEII2vLW2DpMdx/0p IRDsERRWrlxp8YAXcFm1apVtZ30u/aS8UEaJG0UERBnGco46jwVcf8YNNsCUfzjqmTKB1aLyQh2H N4S4jpjVXspga/KmvFA+mOGGwIUlusqLL1woG4sXLzIP3P+A+enPfma6ZDnEHmlz77nnHuuGFTFl 2Qcx5iCPtmzdYvsg4kRxBk5cZ82a6a28MCbB0pn82SonHygvGzdtlKMAO82MGTO8jVva2iJedu7a afsg7fexpkEhzOSRr/JCvaWN2X9Axg25ZQbwwxgCXKZNm9pQBZrNqPAnIOAJgbKWEC4PNG407jRk aJpp8HlOIjTlOqjBH410KaJx/p//+R87A/Tyl7+8ZJh8T4eHyfrPf/5zc9FFFxn1z2wOA2zfpvel 0pTGnQbPbfRKYZkmLPxUElrThlPKH3mKZQENKBtfZk39SQ+49uf7eqSN8qgzMdQZ8CtH1BE08hAd EnWsFCHgPvzww7bjfPGLk3fwTfpWO96kd8GtNgQmywwtuG6VAcTuXbtRQ9j9BxAOIiGusCyithjS f9U+oV32FThstsuAb6+UEakGMvg9ZJesMWj3SePFLJg9DrbIcpWurmW2rTskvDDwBDOUAr6oTZYU zJs316xftz4nvA2xuGBxxGCwv21vNemgTZh3zDyzetVq87xYM3FKBTP9o2QZxrSp0/InrlQTZq1+ WSfd0THHCigItVjSUF5GyvKUmTNnecVl6NBhZt7ceWbV6lXWygteDsjYYPjwEbb86okrtaa1mu9o u9mHCCUEFno80z4jVGHZybMvoi9Dof7888/bsgsPjK3oh+fNm5dfB+6DH+pJhygaUDpgiYeyn/Gd 8qj9nQ9eiBNesFgBG+KGF/paJmRchX7W/NBXP/nkX6wVmrBlzjzjdBvlVhEw+WGBdvZZZ2fNhg0f XFBc0SehLKKs9nRHS73mytJdlnv5JNp7aKsofHfJ0ozDMoan6SfvUPT5o0gxTtxbRUGzQ5ZDgBH7 88wUxUx7u9+N7VHEE/82O26IlGj0j+Dimxd/eRBiCgj0RSC1EgJtvAp2qoRImn1lrRNKAjSMEH45 paAUoVTg9/73v79i5/7444+b7373u3bzPda/0+DS8fzbv/2bedWrXmVN0kvF04zu8K/k3qtbNdf+ fl8pLjp3BmJcsyI3De59VvFlGS6dP9YQpAPlGUoIOp2kdFE/fiYzKLrnCv51pimJxyVLlpjf/e53 th5U2rw06fvgVj8E7IyyzPrTPlI/RLVoWmXWJaltrF+sySFRthjcoIyFFwwQ4IWymFTukkOpjyu4 TJQZOWb3VbkGJmohV59Y0ocCJtRH+iM2SouWaXACUSpjwPQRpfCJwHTM/GPsDByDYATK4cNbpX9N VuqnCLJmLwhr7DGjfTqCJUqIRuBCWc3zIvtmDGEpjWClR+HWnMgaPqSssLkz7bAK/UzE6BiohiBr /oQ6w7JFlM+Md+ABXsgr3wQvLBOkP2MCqdG8oKBpNC+0/VjlNguhLGJmn3EFZZc8It9oZ3wTcWN9 gbUg5YV+iH6gEf0jvGAtuH98e9RXCy/ggsLTp1KcPICXqVOnmHFjZdxwkElacBnWsP4RntLSoocf NU8tXWYuu/gCWV7ox6IxLW+1+nvgoUVmkkzinHX6qca9rzW8en0H1o//8UmxmNluzn/tArPwZekn HevBw91yIglxX/Xut5s77rrHVhPyvZ6UuhdjEKczRgxY1NLBZQbhCeHo//7v/2wHiR8aHh2Eun65 p1PFCgKLhkqnObBxJRtR4pfOl435EIgxN8b8eSBuyOdbMIjjX+0zQnSW5OLh3tcSZ3+/ryXO+DcI G3T81AGEQq5Jne/vf/97W28Y+DJwoFyXUkJQp+699167CaXupxGPNzz7R4DBDL9moEYN8pLS3ky8 0G/4nX1LQiRygxesH5qB6NcRapuBmokXBIVmsa6kP2smXpqlvIBLs/DSDPXH5YHxRHNQi7VU8Wkp Uy7dzdNXixJErPT4DSR66pllZtEjj5oL33ieVUJ0yj4W77v2RsNmkzdc//7MkoIwfNNnik+KS4rs 2muuMHOEl7SEoH+7CNhXX/F2497Hv7/l1tvEUm+dCOSXmJNOqN9y7/sfXGTu+M495iqJ31UyoAC4 +96fWoyJryVb8SueXPtMXi955jmrhHjsj3+2bg1TQjB4U8GulBICwQmlArv/I7BisofwVcoS4o9/ /KOsGd5oNxNEm1yKnn32WfOVr3zFKh5QOPz4xz+2XlF6ED7fNksDVyoNSe6KZ9K7Um5gjzkkS1CY BefnoxMmP7NUQsSxiD+XwoNZEMoZ694VDwaPzUCUSYQNVUKQd3ElBHXma1/7mvmrv/ora1rKnhso IUrVGZZhoHi74oorypZ5ZszYuAt89OQZ6g6n1rDhLPWZ9ZppcW4GPAMPAYGAQEAgIBAQCAgEBAIC zYMAezxlrfQijpNOmF8y0Y8/8WfZUyPa8LSkp4QXix95zLouECsDVXJw7xJ7dxA+tEQE81qUEAjz 37/vZ+b8cxdYiwsNH3t4LEpGirzgEgoR3G++6XozaWK7+6ru91h/PP7Ek1YRgkWI0l6ZFNW42fRc 7/V9Pa6pLSEQdNUSAssGhJw4/epXv7LC4Lve9S6zUtZTIhwyc2tNg2Oe+f5b3/qWYU17h6yDKkds QvnnP//ZWlQQZpzmzp2beQWIx1nv50rCIHixISdLXVjyAqZUetadc0KGD8pSCVEt/1jR3H333daC AKEcwR2hnzXeb3zjG6sNLhP/aNzVBJJ6kGQRRH6ycROngXz961+3fJAWlAdxYrkGm4Oy9Ij1wOUI pcy3v/1tq4DQsLhS5yhrnMbCqRpxpUi5MMO7gEBAICAQEAgIBAT8IdAypFXMoKtYqtUr+y5kQYfr H27v4b5yRBashzCzQwBB+Uu33JhdBLmQiYdlAUmEkgAhn+UU1QrKm+Vb/ca9d+NB8QBNFmUAAjtW INUSlhzwyFGjLp3/2oWy1GKh62Tv8c+RpMpbHw91dFi5ap3lzQ0STLH8uPAN59l3KHguT+DT/aaW +9RKCHYQ1xlmXWvmRshO48zosls/Ao5uDITwrCbm+j3C7A9/+EO7DvWtb31rxbWNl112meF3tBLH VnLc4y9/+UsrfIIxM+cIlShovvrVr+Z3Qj4aMGJm/4tf/KJVTLGMh+Nd2SyTmf9HHnnEfOELX7Bl q9FYoCTCEgJCARG3bmAnb+oBp8KgiNMNz5L2hKAeff/737fLNDi6VOtSqTSyPOm6664r9Tq4BwQC AgGBgEBAICDQpAgwWdAyXE6oGexvI1TfUBR2RatjzHbZcJ33LuO0KRtuHfkkqDqFe6C3zWzvipRE +3qGmpbWCpt+9hw0vd2lDwzQVHbKTDgz5AjEkya1m5OPn6+viq4I16NESTBn9kzrzncrRYDlKE42 p8LaoFOOX2bJBooCKCnsSRNrO5EJJQE8xgV8G1GFPx2yZ8nIkZEVgnvvfsbSCJQB55+70NwuSyee kvQmWUMoDhPFL2nGeuOs006xWKxas9YGqcoIFBqkF2F/s/AOViha9BnPYESYEAoJl3BnuQSEhUj8 vfp1cSad8KM4a1xbZCIXWrlqrdm8ZZsNi7Btml+70CwWqwzyLm4honH05xpJSClCiO8JgVCkxCzv nXfeaY+L4rhAyF17ixk6/lXIYnkFyzYQpn2di668DrQrM9q33nqrwcoEQfvDH/6w3aRKMcYKgt2g EcSPBsIK5HOf+5xVQCxYsMBcffXVVoAn7WzGxI9lGfhpNOlyDPhgSQb1QAmFBJYcbCT4mte8xloM qVVC0p4Q7ADO3hFYeahiQ8M6Eq6PPvqobUNIe7WEJdRHPvKRaj+z/l0LJPe+psDCRwGBgEBAICAQ EKgHAsNkn5YjWAFRD4gSwxDljQCX+Kpmx2osUWqOpPYPu1uGmq4D0dhJzs4xLcMqKCGI6uAe03tg T8lIEVBvlD0YEJqVEEoRnuPEMgaE4Buuv8a+QnHx5W/eKQqHU6ygzCy6EsLs5bK54ZfkvRs2AviN 8r0qMtR/mitKAuIvJYiXC+Pa91+Rf+3eqyPCPnyyZ8SZp51q7F4NEt8NMaUAfiIc5lulAs/gxfIG XeZBmHxv7jPWqmPhORNkH4rHrBvYwf/d9/7M7rmBXxQ5+u33vvXvOFklheYLmFlsJbyTRRHx0Wve YxUZ1qP8Ye8ONpR08Wf/Cyw5sHDQuNU/+15AxLXgnBfbH8+lrDV4119KrYSI7wnhKiHYrZ+TKxCE dX+CuBICRQVKCGbv2VgPQfGVr3xlf/lv2PdYc6AgwORehcd6MwPGi+RYTIRPlsKg4Dn++OP7RIPb K17xCmt+3+flEeRAGfrJT35ij03jyMHzzjvPKr7iSTzzzDPtMh+Wr9RKqjRwlyFVGxZKCF2OgSWE q4Rg81YsOt7znvfYvMWyQcsRgjhKCsoYgjF7OTz00EN2LwfSdiQSR9+94x3vKJs0VRLoVT2zJ4xi pW5pr65FSTzctGEEfwGBgEBAICAQEKgnAi2DZRlGoIBAlQi4Y5pyn6Ko6D24t6SFhwq6KAwQSKEH HlwUCdHlAnbeIUR/VDaKRLhGKP/yN78tM/vLrHLjr8Wq4C25zS0JFwsDftVubsneCYR9UQ1LJBxW S94uFkEeOun4Y62Af5YoIhDuEexRAsSJ9C0UvM4UiwMUNlg4INTDJ4oZlpTwvhSxQSW/i975foub KnbUP/kCqdKCexQLLBMBR10qghLpy7fdaRUhbh7ccus37IaXWLVYZYTghj/S9KXP3miVJoTpi1Ir IVxhDIEKgRBibT6CMkdrMSOvpMsxeGb/AlVaMKP75JNPmiuvvHJAz+hyVOiDDz5omIW96KKLEpUD ikWtV2b9Ue4ghGIxctZZZ5U0w1cBtta4BsJ369evt3tAIMxj8cDpEEmCI42wWt3Uki7KN8e+YrHD JqsIx+U2Ti0VR6k9IagPLK2hvmDdApEO8hBlk264yhWrB+oMp868973vtZYTpeIbyO6cbc6vEaRt WVJZagQ/Ic6AQEAgIBAQCAgEBAICaRBg7KLjmDT+1U/LoGGmt6fvsgyd/UdYdvcrQGhFsfB47qQE DafUFUWDWidgFXDVuy81V18rgq7cX3bJBfnPiONXIgTr3gv5FyludKlEFksFUDRg1UEa4BlaeM7Z VmB3BX6XTZQOpfaucP3Vcg8/nKCBNYniSjjkC0oI8kbJWlzIw3ViHaGnhZAGlDzvvPp6ScNjiUtK 9Htf19RKCCwbVMPGLK0WeMyon3jiCbv+XN/DvHucFNYP+Ed4ZGO9l770pVa485XIesezadMmc8cd dxj2wWDDTIQnlDD1NpMnntWrV9tZXgTUadOm1TspNrwf/OAH5he/+EXZsBGcEYZRiNx///1l/YLD pz/96bxVTFnPVbxcu3atAROI/RY43SELQknGUgmI0yTYf6MWCwR3OYa7JwSnwjz33HPm4x//eNFy JBR9WE6gfMA/mEPf+973zMknn1xWCWU9hj9VI4ByVNsyxbvqQMIHAYGAQEAgIDDgEWiR5Q+9sl6+ KWhQnZcUNEWiAhNZIOCOY3Q80594dK+BpD0Wkmb/S8XFnghJxN4LcRpVw5GyagXBUoksSE/cQPGg xF4QYIAiRq0O9B1XFARZEfFqnOz1gAWIu9Siy1nyvUIUEvhXBYTyhBtWFFybgVIrIRAs1boBU3WE JE4lQCBFQIvv1u8qIfBLJcEEffny5ebaa68tUlI0AxDV8MAsO+lDCYEwzL2rgKkmLPWLOXmcODWB zRYhZsjVtD/ur7/PCPNJyzzccFE+wAsKlw7ZRLEcgUUta/vLhck78AYTtL5goae1VPqu2vdYMKB0 w8qnP8e/ussxEHBRwnEkJydUnHPOOVZx5fKGEkIVWdQxfij4SDdHcmaV/y4PjboHF07UqYUoD1iV 1GLJQBmiDkNHMr614Bq+CQgEBAICRxMCLW1jTW/X1pIm6r6wsBsLNvk+BL6wCPFURoBxo45j6jEu RriFalEMVOZWjqTMbQSZxm85P2z+yOw+SyWyID2+c86sGbLPQ7R5I/GwPAVLiFIbVGbBi4Zpl1l8 8678hpXqHr9ysoVab8TfuVYU8Xe+n1MrIWBMlRDcY93AjC5X9nbQCsA7CIFKCSUEQsZdd91lFRZY DTSS0BRSaWsRWuCbpRGXX365eeyxx6xQzkx5LUqIJMWDiwuCK9gpMUNejwZGw9Pry1/+csOvHK1a tcps3LjRbqL413/91+W81vQujkX8mUAR4hHMeYeSg18tuFdikPJJ/iIUH3fccX0UbJW+1/euJYQq IX7961+brVu3mte+9rV5hUOSf9LJ8hP2taB+NbrOoEQE6yzwJv1Lly41LHGqRXk1c+bMoqVgime1 16QyV20YwX9AICAQEAgIDFAEBg01g0aKhWX3fhlnVL9Jcj1SbfeCCBtS1gPKoyaMeo9dSlkwNBOg LBnhx5KRUsJ2f/hF2Cd86Lobbk4MiqUaSadkJHqugyNWD+6eEHqiBkGzh4RLzWLp4PKUdF+VEsK1 bsA0fvHixebcc8/Nn9bgRhBXQrCh4Jo1a8wnP/nJfq3Xd+Oo5R7h7r/+67/sJo8IL7USQiSCO1YR te4/UKnhQOnDTDwz8ihO2AjzSD5NxMXDvdc8Iu1YKLAcCKEeBZi7Aar6c69J4bjvk+5RqF188cU2 Du5rVfxgUeFaNrCchd+ll15q97SIx01+q3+UTyggWH7CCShZCf9xHpKesT7B4onjUE8//fQkL/12 O+OMMww/31RL+fDNY4gvIBAQCAgEBDwhgAXC0JEmk+MjPSUhRHP0IlCPMY2egLFi9do+5vzNgqxu GMkpD1mQ7qlA+ElKDvZgYDNH3VwzCx7iYepeHWwWWsmagWNCt+YsWuLhrJR8pX2r5SSSeFj9fa5K CTF27Nh8fAjybMpY6shAdyM/lA/8brrppszW8ecZq3DDEhJOm0CJ0B9CMHUVLf0Ji2+TGg5mv9m4 EOyYiWYGvZISIimc/vLm63uXd/de42dfBMocgjkKCJaIVFJC6LfVXvujXNK4UBwof+Qfx9Kecsop VqmgftwrfnVJAJYBK1asMB/96EfrWs7c+NLeo4Rg75M5c+ak/WTA+EsqZ7Uyv2/fftlD5P+s9Ux7 e7SJUa1h1fodliSsFeyS+oGlF9Y4I0eOECWWDKw9E9h2dnaJFdweq0RraxsuZblxvGBJ1dnVaXq6 eyJcRME4RJZ1NYK6ZEajs3Ov3f8FXEYMbzPDWlu9s0IeRQrdvWJldsjiMmL4COFlmHdeiFCtJlG6 U3ZHyC/ixa9ICC5MWOzcudNOAOgSPfqFWi0o+wMouDAJQV7BCwrruPVpf8Kv5lt4UF7Ag3EQedUI XHRyhokJ9s0Ck3qOy6rBhT6e8kI7Ay+aR43ABatZeAEX8kh5acRkBrhQXhizMb4hjxjrNAoXJvWw zGYMDy6Ul0bgQhvD2Ap+dDkFE72NwCVezs86/VR7UgWCvnuSA2OLJUsj64D4Nz6fsVLgeMmsrCBI C5tkonzQPRji6WPJCoqKxXLqhbt5Z9xfPZ8ZN0DuPhA8q8UG90pn544TZdmIyx/YYdkBdlltoKk8 pLkOSuNJ/biNOx30hRdeaDsffe9eXasJOtDTTjut4UdyUukxcdfNDV1+fd/Di2t6TqcRpwkTJtgN PBl00Giy/MX9Rv2r6T7PhMtvoFGc76R0sh8Fp1WAB8okTq9ISiudHfsoQEnh+MTGrQd0em95y1tK Rk/9UiUEg6uFCxdmZnlQkomEF9SZdesKu+4meDlqnVCEoSz62c9+Zj71qU+Zz372c2bDhg0NwaNb ysxKWUK0WpZOUS8ob+Tb8mXPFy3r8sEcA8/Vq1eJ5c9yO8BjkLdp0wbZE8g/L2BBm48V0n5RFDEo 5+ShZ2VzWAaAPon2iLZp2bJlVliBlx07tpvnpQxt375DWPHXdoMLOCx7bplVFtGugsey5cusgtcn LsRl8+TZ58zePXutEBfxstxs3iJr9D3jQh6heEWQQ6Ck/2XDYpYk+iZ4eOqpp2yfh7CPkMvxzr7b ZMoLQtNf/vIXOyFC2WVsB2/wwnufBC7kCe0tZRde2PCZ9tg3L4w52MSafbMoL/DyzDPP2PbYNy/U G/IIXBhPMLZkrMSSWt/jIXABBzZYp7wQP+0w+eQbF+oNZRUc6BsR9mmH4YUxtE/S8gEWOuaDL3hh 7NdoQvg+/9yFVhDnSEc2aGQTyJvkeEjdL6KRPKqVQtLGmcoX/LJEASG8WnpM0ks6y1kbqGCf9qQQ tS6xFhTCG8qAagnlEMssCOPue39q8+SO7/zA5gthoSRSgj/yEazwi6KC71jOQRiuBYlaetx97wM2 zLiSQ8PM4lqVJYSrhDj77LMNv1KkM8C85x6LibTaRioh2nYqKhYVaAlpSGlcscYopSmkUUPjygCC OCdOnFhkSk9HybGajRISXKzglYGNEg0kaaRxVAIvThL5zW9+Y3//+7//a09IwBpAicH+z3/+c/Pw ww9bJ7DDcuKFL3yhehkQV/h2OwIXG00A2LD8h7TSmbAcCOsI99QQwnnggQfs0aZ8R5jgMWXKFA3G 6xVNO0SZZQNXLCFKEX51OQZl91WvelVqvkk39YVypKeG4EadqTQrz8CawQJ1Db9u/cL6hqVUCAiB +iJw8y23mAOCO5gj6Pb2NmYdMULahg0brfDIPiba/lKmli1bbtZL2zfP4148W7ZstkL1vHnHSJs9 xgJH+VqyZIkVWOZJG9Yi7ZsPok9YK+mfPWu25NMkGyX9CMI3g+Pjjj3WDJUBsg/aK/WRpYxTp061 P+oa1m1s2Ez/NHp0QRGZNT/U+TVr1ko/OcHo0sRx48ZaIQ4eR0nZ8YUL7RZ50T6h3cySZZItLYMs LijV1gqPY2WGEOHOB8ELAgF1SDdsZkKA/hkecXfHQlnyRP+OVRzKbOq1jg8Q5sCGsuPLIoK+FAGf fDhW6oz2VaSfsktbU8lSs15YkRfwQv1hmaDygkBHXaKt8cULeQQvjNewXtVyimIEXigrkyZF7U69 0l8qHMoFZZSxAMsbFRcUAAi48OSOl0qFUy93cKE+MU7TcsqVfgBc4MUdb9Qr3ng44AIvtHmnnnpq nheUetQvZAKflp6MSVGiwYuOESmzbEROGe7o6PCCSxwn9/ny3BGaCK7sfQAhBHOyBcsQGkVqBYGC oJySQG3nalGa/D53BClHjJYiBPmTT5hvFTVbtlUeH7N3xMKXcbznY+bL37zTWiEsPGdCqeAT3YkT 64Xbv3OPVS7gCTeWZ9wv+URaUSDgxu9GOY7z9rtyfu+LggQzwlDFA65YRZCnWJfw+9KJ0Qkg0RfZ /q1KCaGzunTIb33rW8s28lRuKhMNNIJ02vXeHPn5jW98w3b4NJwf/OAHrVZXNbvMhl9wgayHkUZN O2QgopH5+te/bmcvEDgRCObPn2/5nDFjhj2Z47Of/aztuOHpX//1X+33dFTMTvsQ2lGOcPQjAzwa IBo/JTTVpJUBDo3S3/3d31klCo30NddcY4VJtP5f+MIXzNve9jY7IEGL+rvf/c6w2SFpRONM2r72 ta9ZAf0lL3mJefOb36xRNN2V9HP8JB0DgjDWDUr//d//bRUJdJrk4+WyUSRER/He977XHgGKUob0 YpEze/ZsOzvEZqG//OUvrVBIp4MgDmazZs0y559/vnnFK15hw/H1R4VB8vG8884rqxAg36kzdMpn nXWWVfK5ZbwUzyzzoFyBBb9bb73V/OpXv7IdGlYMDF5R3sTrIHhz1CyDWTpAnukU3/nOd9pZC7C8 7bbbbP0jburld77zHVvv2ddCB+il+Doa3P9VjqKFUJp9+tP/KvWa2Wz/dGC/mI7v2in5ODY/qIEL 6s/48ePsIGvK3k4zclSkFMuSw8MiIGyTTpllBqqAID4GxPQdtM17pc5r3ciUF+kXiK9tWKvUvfH5 qKhX9BPM+m+Vcu9nYC4WGZu32L5tsgglOvjmitKR9pyZVNryrIn+kv5I49b4EP6nTplmnnr6Kdsm z5J2NXsCl8227Zo+bbrwFCmn4I08Yvad/mH27Fn5d1nyRHkBH8YaSgiYPIMZCgrGHz4I4QjhHyWR 2xfAC3zSxyGEu++y4gurL8z7afdVuCUu8ghcKL/xPiYrXoiPcYOrgCAu6jHjK3ChfXH5zIoXlJxg Qz+rCgjiQqFPHtG/ovRkPJs1Ma6EH8ZDbtppX1AUoaCgDfbBC/kDLvCiCgjSjxwBDwjijDvcd1nh Q7mlHaGsIpsokS/wAy6UHR+4oCBirAUGqoCAH/jCjTrvU4mmWCRdUUSwHAHBf5IoqxFsoatiR2J+ 71v/XvQ5Qi2/OCH4xv2qH46NTEPwUSoM93tOsDgzZznguqe5v+rdl4qgfmlFrzeIkO9SJb6uuuJS uzSCjT8nSzog8E1a8lEqLBRB/NjXwQ1HLTNcfsD72vdfYRUT5KHrP+7vS7fcmD8BZJJMBviiqpQQ VFgqzatf/WqrCS/HJJ02s4NUrNe97nW2opfzr+9oJBDAvv/979vO90c/+pE1Sb/yyittx4KigV30 P/ShD9mBG9+haUXwRki/5JJLrMk+HeInPvEJO3i57rrrbOXGGgPBFcuCl73sZfZ7n5phBEQ043RM 4MNgU2egsIygkaQTYQCEXyUE73/8x38099xzj3nkkUfE5PuztuHmG4Trj33sYwblDQM1BiPMEiDg a9gaTrNd6aDULA88mF3hSrr4MSAnLW5nShooH5///OfNnXfeaU0gb775Ziu8gxuDkn/6p3+y1hKE Dx5cERKz2lSxHK4MRKgDCxYsMJyiooJH0jd0hnTIDDBRHrkDmiT/6sagEEx+/OMf2zLwwx/+0M4y vO9977PKq6uuuspiSTliMAK2DEjAEEsH8KJuo7y5/fbbLZaXXXaZ7bDhg/qFf/BjoEVHTTiBmgcB zPAQVhi8xMsYii1or+xB4EMJsXvPnhwv0dGnLkpRmW6xA2UfSgiUkNR96hVti0utba32cZ9g19t7 OHMB98CBgzIY6LRt0mBRyLg0dEh05PCePQXrOPd9ve97erptfzN48BAzRNpIlxQX9vOgH4q3v67f etwfOtRtOiWPaKvjcZFnuO0VBRp7eQwZWpyH9YjfDUP7HeKMCySUXfhBoKFcab1yv6/nPfUZCzRw ifNCHYcf2mb8uYJVPXnQsBQX4mXM5BI4gBdWd/AS59X1W497+nlwIS+S4oI/BGDqPe1hlgQujOeg eB5ovjGuw7I3idd684ZwCz5JYwfip+yipMi6/wYXFEVQvLxoGWLsQfn1oYRgDEhbBi7x/lGx8oEL eJBH8BIvL4oVY17Kry9LHuItR3ZWffbMcl6a9p0qTZqJQU60qAdVE07aPFTFSD34SxtG8UiowlfM kn7gAx8wzLCnqSDvete7rPD/4he/uM8AsFRUHWKGhBUAQvRvf/tb23AjwNGAIqShaWf2nGMiVaBC +GI2mM0mUZBAaBSZeUaJwdpOjjnke2YxUEIwI46G1iehaf3c5z5XdZQ0mvCKpQTKGDSlNN4oGUgT nTHKG9KEkohGnQF+vPGvOuKMP6Ac8auWSC9mjwjPDDSY+aDTAw9M/LhHkH/9619v8QATfo3AA+UA 5RleKsVPXmKVQ/qqscyhznCaB3UGRRQDZKxDKDeYfVP/2IwVhQ91hsEip11Qv/7lX/7FWpqQB3xD PWL5BWUJCxQUF1hVMOij7qG8C9R8COzevcuW+0GDB/UZZKlgR12ZJMsRBuVmmrNJRa/Zs3ePHQgP EQE3TrpBJgM+HzP+DOYYlFOnpDoUEUICdaTLCnKHEgfuRR/084FNpRCkhw3ruxkm+QaPewW7boTt IcWKgX5G3edzeGEgPHSozM7GgFFcDhyUJZEiPI2KKUz6BNZPh0MiuKKgSUozmAyWTVVR3hyUdi3r jURVqEewJm6XojLUYtvXNMvc3G9ruUdwRWjUeN0wKLfkE3zgJ0mgcf339x5ciEfjdcODP370u0yi 6HJA10897ym3tB8abzxs2jv6QHjOWglBHhEPpO2s8gNWigv55EPpirCdlEfwRHkBO3jJWgmheUS8 SbjAC+WFfGSsnjWp0irOi8sfuMAL+GVF9EWkmbSDQZzUjb46UEDgaECg7yixTKqZdeeXlt7whjek 9drHH5URxcOJJ55or3igQceNhl8rKR0j66jQLtLhsFeAEgIpDQuCGZX/SCAUDAjgcUIA55cVgT0a Y67NROQx1g8u0YlUU07db+t93yEKAn5pCGyxGqqF+FbzBqWD25EyQKWe8IPokB9//PF8J+jWGZRZ LOHghxIi0MBAgDYRSlIwaFnoPtRtemkHReDNig4f7jVskAm1DO47mBs0qEXKpjGHZDAcbTbY14/9 uE5/EKQh4k0isEHoZ/YpazrUfVAGnyhE+g4+4YN/EPglCeT15I8Za4i80PLhhk9bQp/pB5ceiatH +OirnIG/HCwiWFJmsiWtR2Ci7akbI24IEAi5WRPYI8wlzd66/GleZskPuFAeGJcl4aJlSPHLkhdw Af8kXIhX+fOBC3yQR2689iH3R3HxUV7IH/BPErRd/nzwomWXeJP4UVx85BH1VculxpvLHntRN3DB rz67fup1Tx5pmlXh4IZN3PzUj/su3AcEjkQEqlJC+ASAikgljWuPtYNRXtC8Y9ZF5WYdIFpGJdZe LVy40FoJqJteaWwCpUeA2Q1m9P2snU7PV/AZIaCdF08srXBJB87qhgJvpaxTRaGHMsK10EBhgRIj Xu/02yPtCm5K7r26DZRrT085JWuUxh4R9HpFSWD6ysB1TGavCK4RL4McbPtEIO0v/pIGYn389sPh cB6XQj67wZHnCMA++gPFpSQsORYPdR8ybfIvS6K/hFTxEY8LXOCX/T2yJvCHknGJBuW87xZcsqZK Shdwoayo4JklP2l4IX5fQmW5OqJtpy9eFHeNV5+5qlsl/Nxvar2nHuXrUkIB9smLm16N102Xuvko u2Di8uPywb3yUs5P/Jtan904NF43LHUDl3Jl3P2m1nvCd/mJh+PyEn+X9lnDSOs/+AsINBKBplZC AIyrdHArl9tYqDtLNOKz4oSBsBUf7Or3WErQKGRttgcfA5mYdcAqJVBzIkAd0Hrg1hm41Wct8+oP qxqWfySZZsbXafKtDrZY4wrFlR3WMfxpCALMsJciHRuXG/yU+rZadykmFQZykaRNeYKfeLtcbXyV /GuZLeUPfg87ZbuUv3q4WwUQAVXQfxcUJ/WINTkMbQt4q+2B69PmozCK4ipryudRKVxy7n7Kbykm IhTASsuuN1wqRNQMuCiLgRdFovjqGxe3fhdzYsoKwXG//Xkux4O+84WLxqfXpHTBS7n3Sd/U4qbt XXK7G7U//cElKdxa+AzfBAR8IJCdXW4/udfGoFSF0vcoD1hbzGAWKwgEI/eHO2v21L+Gpw0B69/v vffeYP6UIr/ATvFL4T148YgA5TtexjV69x1u7Ody0kkn2fqCqaRbX7hnNkuXbuCfPKe+aPjsL8Fp HEcCoaBhBoRfPQRiK7yVl2MygY0NBkuR5ptNn2okSnnupztlZXBuuQdY9KXIMfKXqUmGjXrI0NK4 KG9YbKiiTt2yuKZdYhHftDILXrSskxtaPuLxYCXBfgxZU5K5thunFqNyZdz135/7irxIofZVdjWP SqVH860Sz6W+r8adOEh3JarEc6Xv07xPG0daf2niLOWHONK0Hb54KcWn6+6DFzBJE08aPy7vtdwT R7myq/Wokr9a4o5/Ax9aXzXeuB+e1U/SuyQ3wtVxTJrymBRGcAsINAKBplVCqJJAr6XAYcPBV73q VXaZAMdV6l4R+GcdGG4cXQlRsXUDJ5Zq0Aiw9ippEyr7QfgTEBggCFCWtVOL1xl116Sw+RKbvVIf OC0mTg888IA9TUPd2XeDDk7DpV5hGXMkEAoX0sOPNmGgkgq4KrC56VBlgK9BlgqL8XIHT3k3X4J/ TjmTj9cFxt5HG4SVG6T2+aRGB8VfampyCDnn+GkVyZ7751pQLiTzAl6DRZAYJAP4rAmlFfgn4wJ/ EY9axrPkxx38lyozriCRNS8WF63AsciUP5fnmJe6PRKH1hGNNylwrE6zJuoRghZ8JPGibvjLmrRO E4/G68apbr54KReP8uKjvLhKCI3XxUXvffGSRjCHlzT+lPdarmnbjnL5mBQv4zMdxzRyPwlO6Xr8 iT/bH0dCBgoIVEKg6ZQQmHqzTp3jfahYf/nLX+yxlmymxwaUzz77rE0Tm+lxj+DAGvYLLrjALF68 2Nx22232Wxo+ZmxRQrDBHpUfBQTLNRDCONISywl+8TOVK4EW3gcEmgkB6sof//jH/FFh1B+OgsWd k2HY/wHzPuoLzwwUOf2Cky7uuusueyKGniLw05/+1Cry3E1OOVqUjo1jb7EqWrZsWcUjepsJn3K8 oIDU01O4rw8lC3b1CTs5FD1xQdfXu750qcZQsQpQQcJ9X897aWblBIPI+iBpWQEbVyJTDZNTGbLm hXRZoQgB1wosxSkFFxRrDPgKQnmxn3o+RQPuyKooHi58IISDydCE0zPi/vv7TFmQyCT9fQU5sKIc cWJH0gkn/Y07/j1KK9KtSk73vfJncalb/XRjKL7XNiDCoHiJE260o754QSCizERlt2+bAl7w4kMh DC7EBcXzSXHhnQ8lBHyUw0XzyAcv8KECY5IJvWKl5QqMsiTSTJwarxuXuvngBUyi9i55+Ydi5YMX MNA6ovG6uKibD17ctgP5Jk7wQn2qtuyCtY5jtDzGw876eeXqteZ9195kbrn1Nvtb8syyrKM84sJH gfPAg4tSpQu87773p3JyVHQ6T6qPmtBTZVtVj0xT+VA63HHHHbbDQxBSZcGLXvQiezQnSofjjjvO /OlPf7LHQXEMaEdHh3nb295mK+9v5PjNa6+91u7xwPGEnNCha97p1Dn6kOM6OYoQoYxjB0855RSP qQxRBQTqiwBKh3vuucdq8akbWDKgKOAeSwc05HPnzrXKvT/84Q/mK1/5iq0THLf73e9+19x+++3m Bz/4gR00sLTpPe95T95iCE7PPPNMc+6559o6s2TJEsNRvaeddlp9E9Gg0GhzlNx7davl6gRZy+c1 fTN61GixXtlij4BEuG5xjuHUwc6IESMzn+kRydaMzMXT05M0yMKtVzY+HVVTOqv9CMUzM/o6uHO/ 140ihw9vE8E/+66wrbXNCixJpzyo4DB8+Ii8UOPyWu/71rY2UTBEx+SpcKJxuIPy1tbsZ7YZQDPo TprBo06izGoV7IZ5UEIgrKggpzi4uMAP/GIdljXBC+WXfavibRN5Rr3Gj7uxcFY8aTwoobU90bjg RfljOV/WhJDFxslMTMXLLnFj3UYeqdVrlvwguLJ/Ekr8+Kac8KZlyNdGz+C/adOmfLxu2sk3xsA+ yi55RDwcGRrHhbKiuPjgBQxYts0m9vGyyzt1gxdVtOGeBYE/ZYEJojguxFcrLlr/suA5bZgoHxCI L7/kAnPS8fPNpInZH72alreB4u979/3UrFq9zpx8wrFmzuwZlm0sSjZv3W46Zs+UsdXwfFLuFr+P P/GkuM8wZ51+at59oN1kP/KqAhEaAGZn+SXRa17zmiRn68bA4bLLLjMXXXSRPV6Q5+nTp/cZ0LW3 t5vrrrvOHttJ5xE2pCwJaXgxQBA4//zzDb8kevOb35zkbN1Qzl199dXmHe94h11+QefIKSjxjnjM mDHm4x//uPVDR+1rQFWS8Tq+yKbzLig26shq2aBGSr4wIN6/b78MZA7LIFyN3Hrzy0x8CAgwSfmg /e2ye/EUK0RQIjMQGzt2TNn01OslM/4jR44QDOSIQZllcmeJ9nZGZ7GPHj1GeMredHuYCPRtba3W 0ujggYOGZ6UD8ozQ0j6hXZ0yvYILebBz5678wFcjRKCCqPcIc1kT5ZZ2hVOuEHJd4ZGB+kH5TZo8 yfjYK4O2jzZww4YNVnnrbtALbwgJWFJSvrMmcGH/Hqw1iRtFgBKCE8pl3rvu+r7eV3Ah3atXr7b7 Bbl9AMoj8omxlA9eqMOM4xBwWX5LOVWiDrGfEeWJX9ak5QWhEmWRO54EF35g5ZbpLHkCF3hx93Qi Pvo5LUOUGR80adIks2bNGqugceOjDsEfmDB774NIM7yAAWWEPggCF+oRz75wIR7qtrazbvrBhTrk lmn3fZr7bMY05WPeIkIyv7NOO9Wcf+7C8p7D25IIvPPiN5uVa9bmFRB4XPTwYwaFww3XX2OVE/rx 5eIXZcVAVkCQFh2paroG/JWG7ZhjjjGzZs0qGnC6CaPBQUHhdhju+3AfEDiaEKDDmz9/vj3KNq6A UBwY+HE8qzv41HcD+drfDnv37j1m27ZtVqm5Z090PPDzz6+wbgySfRGDGgagnV2dMrjpyke7T2Ym dKDuY8aUiCkrDLQQTBBysXzgh+DEbBQDT1e445usCIuQCRMmCi/dZofErUtT4G37th0G6wRfyhnq 1pQpU61gv2Xb1jwvvTIoZlaXZQntngQEcGlvn2AtZrZs3lzEC0IMQvYETwoR8h6BBYXHpk3FvKCY oI5OkLJdqm2qd9lBOQsvmy0ukUIRwWXjxo2WhylTptQ7ypLhoRAhL9avX5+fsYUXlCQIc7opd8kA 6viCPKLuUj7UaoW8ASd4Ykzli1CI0IYw609dhuBBrQC0PPngB2UHbQj9gO4rBC4sLabtpbz4UOaR VvpyeCFPFBfcyTOEbayDffFCWYEX2jYUaZDiwjN55Es5Q1mhf4QXV0HTiDwizfSPlA0XFxQkPCOX DNRx1siRhZl6m+HhT1UInHTC/NRKnEkT21P7rYoJz56zn+bwnKAQXUAgIBAQ8IXAl7/8Jbv/BgIB QvYoWWZwzz0/MD/5yY8tC9dee50sizk2c3YQ0FASwcOqVatkkDPOKgMYdDEAQynrS4gjsQx2EQpW r45mwoibAR8DU5bZ+bA8UNARWBiQr1u33s4+MQjcjYm78Dd33lwvs7fKy7hxYwWXWXZG+aBYhQyX vGEGFSVJR8ccLzPsygsCwqxZM83atWtNt5RfygkDYxRXHbLEsa3N34CSuIlzxYoV5hCzx2NGi7DQ Kb/9spSsw5vSCmwQAObNm2eXtD399NNWsCOP+OHen1lKwq+GUByyjxWba/NDQEGIQpmH4tinsAIv LOtbunSp/SFIITjRxrCvFkKeL0LpymQTS2pZIkgdR8hG2IZHlDe+CCXRsccea5cSk0fwgjKCPKKt oy30ReBCGVVcyBPaPpR5vnkhzeACLywPRemA8orywr3PPomJR9oX+KD8kkcoRCgv3M+ZM8dXFtl4 4IUxA9hQVukr4QVc4MVnX10u4Uueec58/76fGfYfYKkFM+8LzjnbLHzZi/Of3fGde4zu/4D/mz5z q10icNnFF+T9JN0wu79YZvk3y3IDlhmcdfop5sI3nNdnGUcaHuDts7Ic5KzTZGl9i5F9FRbbcBHU z5ZwLxOrAQ2HK/GxvOGqd19SFB+8s5QEZUpSGPF0gMsd3/lhHh/CPP+1C4rw0W8eeGhRPsxS8YPl ytXrrNWD8gs+EO9GSj951RVvN5NkgoDwWI6hzxpPWp6+/M07LQ4LXnZ2Pg2EUSofNPx6X4MSot6I hvACAgGBowaBj33sY02TVma4GIAjLLFLNQMbBjssR3D3iPDBMNYQDDIZBO/GQkQ2QJwxY3rO6kBG CZ4JQWDcuPGCzW47KJ8onTjPapbrix0Glwx6mT3dtXuXCNzddgnG+LHjvJxEEU8nvCBUIzAdOhiZ 1FOGyD+fBC4I2CeffLLMBu4xBw4ekLIyznTMaQwuWEMgZOsMMvcI/Qh5vgllEXtXYVmFCTeKB+q1 jyUh8bQqLwi1CNooFeGlEbhQbk8//XQr1IILiiz2KvJl8eVig2KTPKIeocgDF9q/RvHC3mfkkZr3 s4+TL6uDOC7EzQw/SmjwOPHEE70slXH54J64X/CCF9jyQh7R5tDeUG58C/3UXZSLigvtLXt4Ub+a hRByb7/rHhHS261QPUIE90WPPGq+fNuddukFCgNozqwZotAxVhBHuEaInzypvEISAXjRw49GSg0R glnKwTPKjJtvut4Kx4SNooLNF5N46OrcJwL/QrxJOd9nlQwI4PB5tiwLgR7745/N/aKQWCGC/VMS NksX4HvlmnX2BI8bRenwpVtutH75g+BPGBCKFk0zYaDouOrdl9p3/MEvig/I9RvHh/eaDqvEyaX3 ceHtuhtuNjfKMos5oryAUEAQLqRYYkUKPig4UD6MkLYGWrmq4Nc64Ca8o0iBlCcwgKdV8g5ljBLx gBu4nyTKJZbRPLV0mV3+QXw3XP9+9ZrpNSghMoU3BB4QCAgEBPwigJDic4a0XOoYjPNrBmI/hra2 Sc3AirW+mDzJ32xtuUQjQDID1wzE4HzixAnNwIpVxPicxS6XaBSM5FEz5BMCk88lKeVwQYmIwohf owleULr6tAgpl+ZmwAT+EPARrptFwG6WPGo2XNyyxGaI9z+4yAr/N4owihIAQui/6TP/bhUDC0WY ZvPJhee82ArJKC0QlC98Y6SccMNz7xF6+aEMcP0uFAsLlAKcDoE7PKgCwlVMwMN1N3zGCvYLJG6E dSWUBkl+UUBcLpYZqrTA/y3/fptBEcBGkLoJJO6EkZRm9mZAsEdgh1CkxP3Ct+Jzsihj1C9pAkP2 dVB66pkovfc/tFiUG29X5/wVLPndfS8Kj2UGiwWUGOWIzUEhMNA8g6dbbv2GVcawZ4fyhD+7kWgM F/gnPt652OI/Czri9oTIAqQQZkAgIBAQCAgEBAICAYGAQEAgIBAQOJIRQAhlNvwiEWBVmCW9CKUI 8xBCeS2E4A+hvHAJ4RghWffsYKkBBA+uMMw9CgyE5MWizHAJId31C+8THQVKsd/59nFFzvJB300W xUo8zW/JWX0oT1gRgA8WF65fwlALkUWPFOOD1QE8K5FelBLnn7tAnfp1VZ6w9ujLU7RxPRYdLuHP Vczwbo4s0YS2Svp8ULCE8IFyiCMgEBAICAQEAgIBgYBAQCAgEBAICDQxAiwLgDpyAqnLqrt0wHVP e89xkygK4oIy31/7/ivywZTjQS0CWFZRiUbJcpskwoohLc0ViwSoU/bAgbCsgNhIMk4oF0ifLuv4 /+y9WXAlx5WmGZlAJhK5AMhMIPd94yZRoqQqSdNDdbVJYyWqrYovVfNAdcusrUcsq26rMRPN1PNY PQ/TD8XneZA0ZjVPMqNNq0zSA1kPUmuKLDMtw60pUiRz3zfkCiD3TGD884sf6QjE3XBvnHsBnJOG jLgRHn5+//2Eh5/jHh6cJzDxf4c1HZjB8Uy4hmALn9ZUOfJ5zOc3MzqQuC5GLgN0wTdf3qgnCuJM hFd6LcSDEBYsuw5nwBlwBpwBZ8AZcAacAWfAGXAGupgBRu2R1QVfu5CTyie45yPk3UgA4NadihNc hEF6b4V1ISxEeJn9kIqOp8fY53g664HZBrzy8Xp49YLZFJpF8q/DIpbpOg35fJr5LX3VMDWTl2Va D0JYsu26nAFnoKsYSBekSve7CqSDcQacAWfAGXAGnAFnoA4D7ejHyJEdHb0aF0NMVWqEXwskpuca 3VeQo1b6Nf3FMxi4hoW3kU3Tr1rEHyX+x/oUiGZvKBCTD0oIAsfzsxz4rWNwyGKVvB6xOpQzXRtD eTS7FbZqmOBcaZrNu8z0viZEmex63s6AM+AMOAPOgDPgDDgDzoAz4AyUzEA7ghB81hJhbYi86DWJ P/5iJU3+fL3fvIrAqP0fwroKeWFhRT5FiehVh//3rdnrPnBOGPbs3s7PtooCHGmmM/rCl0AQXm9A /vDp3DKw6CayZzotazWwiOX/925lLYx4Lrze8f2/+W58bYMvjrRDhOmfCvIDA5yz0GW3iQchuq1G HI8z4AyYMcCnuq5evRr/+FyWizPgDDgDzoAz4Aw4AwuFAT6Xq34Mn+huVVjXgFFzvnjBpx0l8YsV YQSfc08/UftLDbomv1WAgzUSNMOANMwMwFGXM60FFsGQpsOhJlABhn8ZvlbRbiHgwNcsJCozv8GE wA9rO/BaRbW0WvARx5/ABOVLy8FsCAIDLIRZSzTrot7sEfSAicBRHhNBEESLZtbSV3QuftGkILhR lLbZY/46RrOMeXpnwBlYNAwwajDFR66DaLtoCucFcQacAWfAGXAGnIFFzQB9l3b3X/hM5X8On2v8 /t/+XXyNAIcZxxmn+G+Tz3Y2SywOPM4wTvl/CnnjPPMKweXwygOfwFRgAT3f/5v/JXxe8v/K/uP3 //dZGAhAMJOgDCFvsOnrFnr9BMyck/yHf/9vIj8EU5RWQQY+uam0lIMvipCuqBxFn+eUDrYKysAD AQs4ST8pmqZtFFN6Tb196l1BDF4nGdn4mIN61zZy3oMQjbDkaZwBZ2BRMrB27dqZb8x3y3fMFyXR XihnwBlwBpwBZ8AZaDsD/f39M/0Y+jTtEJxoAhG8NvG7934fnWq+vMAIv0bnpYc1JHDS5TDreLUt ayDwiU5G7PniBLr++l+8FAMN6TUEKPIY/uR//OMYqEgx1NL/L//FH4dZGwfSbOM+WIiONjUAAEAA SURBVMG8N2xTwdEnMCBspCMPreegtGD+P1/9z/EzoR99WnltBX4olwIQSgtnzKIgz8tXr8XzReUo wkrQ5j8GPNQBZdZ6HX8UXplBT7o2RzOY/vX/9CeCN2vLbIose2FmhoaCKCRqdwCCPJeF6FllGJBf Ls6AM+AMLCEGfvjDH2b8IS+//HL8S4s/dX88m7rX+vTGNE/fdwacgXIZWL56OMt6VparxHN3BpwB Z6ALGHjttdeyV199NSJ56aWXsldeeaUhVFO3r2ZTj+41lHYpJPqf/93/GoMNf/u//c1SKG5byris ty9b1l/7lZJqisJM5GW+JkQ1dvy4M+AMOAMNMkAs9+GDB9nDhw/DFd0R1713737E1GARSkoWeHlY 4aXz8e6pbPLRo8jJ1NRkSeVtPNtHXYSlm3iZnJzsmjoCy4NwX7PttHQTFu5l52WuRXQbLzyPKs+k uVgtj4iXbsHSTbx0CxZLe3BdzoAY6NWOb50BZ8AZWGoMtMMxvnf3Xnbx0sXs7p272bLly7L+Vauy rVu3Zr0rVnSMThap+vGPf5x985vfzA4enN8CUq2Cp3N18eLF7NatWxlrb/T1rQy8bMtWrrQfoZ58 NJldunwpYyFS9sGyafPmMLWx+mfAWi1/9eunskuXLmfj42PZo4BlxYqAZdNI1q5ptNX1zj2D/Y9e uZKN3byZPXr4KGDpzTYOD2edeTUpYLk8mrFA7INgO/AyMjKcrVu3LtrPXPTlHYEX7qHLly9HJ25F uJdHRkayjRs3mmOhlKOjo9mVUE84/r29vXHqNXi4rywFXq5fvx7s91LE0tPTE3npBBbKfePGjdjG 3L9/PwPL0NBQtm3bNnNewHIz3EPwcidML6eO1q9fn20ObQy4rIU298yZM5l44X6Gl+XL7ccdaXMv XLgQeYEL2rnt27dHjqx5oW7Onj07Cwu8cH9by71792Id3Q6fm+Q+5lm0e/fujvBiXXbX5wyIAQ9C iAnfOgPOgDPQJAN0sI4fPx4dpT17dmeToZN+KQQkPvrDx9mhQwcz3tW0EBz+9957P3aCP/7k4+y/ v/9+xmrKX/nqVyzUz9HBat3Hj58IHaqe2LFaHibdjV4Zzf7w8cfZ/n37Il9zLirpwOTko+zosWPB mXwUOuJbszVr1mTXrl3LPv30cLZt+9ZsU3TmbDrnOJHnzp2LwZAdO3ZkawOW6zdvRBsa3jgc8YS3 JEtiYna2zAY5c+ZsdPrpiOPss7I69bZ580gMpFlNlmSUH16ol+3btgdncjAbC1hOnDiRDQ4MZnv2 7g7grXiZyk6ePBmd233BVjds2BAdTO5znN79+/ebOrmnT5+OWLAXnH3anCNHjsQgyVNPPTW7Ukv8 RQACZxJucJbAglN3+PDhGJgg2Gnp5GIvOJRbtmyJdYIzdyzc5wRJnnzySTNnDl4q7cmn0dGHm0r7 dzwGjp5++mkzLFQ/AaujR49GLHv37o3BInjBdp944glTh5tAHli4h9BNUOTUqVPhWfVe9vnPf94U C3WErQ6HIKsC89zTH3zwQcZ9ZBmQJmDFPUybe+jQoTjTiqARvDzzzDOmWEpsMhZc1qwRsWmkvQsv LjgSjAHb9LyMC+XqnAFnwBmwYODc2XNhpKs3juz0hRkQBB22b98RO510KnhFw0Jw4j755JPQGb6a /Q9f/WoYhevsg/TihYuh83s/rOK8O1sVeFkZZh7g6K5a2RdHf6x4gftLYXQd53rHDpzbodjxZYRy w4ah7Py582Gmxm2LKoo6rgdHgI45DiVYmC0zEjrFdNIvj14ODuYtMyxXwwJZzILYFmanoJ/RQLbM ymCmxq0JO17olDPrAF42hoXB4GXDhvUBy6bsanAebt4cM+MFJxZHbufOnXEkG15wXPgNRs5ZCSPa BCHggdlVYGGEnUAIdsRMIysh+EGbhtPPvcyMJkbYcXRx8AhQWAnBD7Bgr7t27Yq8gAXnEluCF4ID FoJjjTMLFuy3r68v8kIAi2AEOK1e51GQk0ArvICFmQfwQpCGoJ4VL+iBF56Je/bsiVhwug8cYOG7 LGLhlTQLgRfuI/gAC5j4o46YhQBOS17Agl7083wkAKLACHUEXhd7BlgwU1/nsNe+NDV6EGJp1ruX 2hlwBlpk4H5Yc2Hi1kQYWV896xUDHAWO4fjeCh0/C8Eh+Lf/9t9k3/nOd7KvfOUroRM8YKG2UMeD +w+y8Ynx2MnrW9U3k4ZO10DAdTs4EOPBobEQZkHcuHY9dPRYVXr2qxcEaugE3wyBAQth3YVrwXFE ZtdP4GVgIHaCceYsBKcIZ3tFL7a6ZpZKXjmAF2auWHTMwaJy42BjJxVZVgnU9PZEh3tqymZdBoIM YCBIlApOJiP9ONtWvDC9HyEI8ZiXLI6g4lDpdZEUZ1n71BF2AQ8pFpx/sDAzwYIXykcABiwEh1Is OLk4dQQhCA5YCFgINnDf8BqGBOefe8uyjrinCdDkeaHtgxfOg9VCeK0KXdgLz0QJzyruLXgDq4UQ zCMIQ/uSvh5DIAKbseSFYB54FBBX+WlbOMaMFQJpLs7AUmDAgxBLoZa9jM6AM9B2BuSkFa1xQKec DjmBiKUm129cDw7Ag1mBGXEAV8vCPzqoFsI6Hbfv3A7OQc+cqeLUETIRZkLwOkvZcid0yOl046jk X3PgGJ1jOqAWci9goQ7o+Pb0zO4GiBc67RYjcgStmAFC+fPT+fkNN9xHvE5TtmAHrL2A3tRxQi+2 y3Hq0MKR451xnCNxkJadY+DBmbHAAi8409KbYqF++AMvNlO2qI4IPqA3LzjbBCBw9soWAiHYC4JD mwr2jA2BhXqyEIJWBPWqPZO4ny3aGDBolo7aE5VfNkQ9WtQRerFd8ZIGrTin+9yCF/oEWtslzwtY qDdwWvGCThdnoJMMzO59dBKJ63YGnAFnYIEwQEfh7t3KKE46siL4y4KTgFiN9Ehvp7d0snCKGLWm s5kXnG8GusfHy3cQ0H07BCHAtDwoRW8qdEb5exCcBDrEZQtOGnZTaC8BB+tm4LBYjODy5ZTIS1hI NU8MnFB3Dx48jI5l2bw8mP56SpG9UG/wgvME5rIF26WOZBupPvHCeQunEpsED/aC7lSEhTQWjj84 0CW9RVg4ZhFcxPGnzPBSZDMcJ41F25ver9Xua3ixsBf04LzCSREWjnPPW9gL9ys2g1TDwjkrXtCD 7RZh0TF4gZ8yhbZD/Etvqo86AqcVL6lu33cGOsHA3F5iJ1C4TmfAGXAGFhADdCb4mgBCpyEvPaEz gdAZtppGnsfQid904rTeQ5GDsBynNwivSVjwcv9+5RvoCgrlOaHuHj56GOqp/Kn+OHHR8Q9OdV6i DU2bEenKFhx/SZH9cow6wn7LFt0jBBvmCPfW9P3Fp17LFgWAKH+x/VYcOZMZIsGRU0CkqNzCZ4EF XuSgSW+KSTYk/tJz7d7HXvhDp/SmOnTM4j5CB3WESG+nsICjVpmFz8JewCI9Rc52p7BIb1EdgVc2 np5v5z756x6pdR/Vqsd24vG8nIFOM1Dw1O80JNfvDDgDzoANA2mnJN2vp53ORK3Op66vpCt3dEW6 umE7m5fqiBhwsnD8J+usI0CdywmujrY9Z2bKOzdmNUuBgluzDrb5xyM5TrwcM+3kpyo4ltZleq7d +/AfpQovgmfRMdc9Xa2M4mUGc7WEbTjeCBbUWPGCPdQTKyy1cMierepI9SS9KTYds8CS6pDeTmGp 13YIX4o5xdrOfdUPeUpvmr+OgaURG0+vbXY/5UV60zx0rJX7SHmk+fq+M9CtDHgQoltrxnE5A85A 6QykoxHpfuuKKx4V/fapaYev9TznkUN9v2EemVa/pIofmVygFFM2vNQpv/yqsjufCQHBIGb9mvOj XuBkzgXzOTALw6wfMbdot2HPghfVQVVepuFNTs7FOZ+i17qmXnk5z1/q2NTKz+KcBRbK3IhzU48/ Cz6ko5uwWNVRI2XvBl6EYSnzIg5UZ2x1TNv0XK399N5sqh8zPTOxVt5+zhkok4G5q/qUqc3zdgac AWegixigE6R3V5sZleGhvzy8e4wUdxgqDhMdgp6wKGKnpBhbiWgCLz3Ti8XNOJez1FV4gT+lm3W6 zT96w9cfKlLNgQ3rRYQ6Wr68/DrqXVF53FZDIie8aLG9NtMy8240WKrVU5gjMWPj7daf5rdC9pIe nNkPTv80MRa8aJG6GfUFO9huI+kKLm3qUL3y6t62wIKO1NGpVpB6mKtd18zxoun9Rdc3mq7o2kaP VdqO+mN5Frw0qqPRdI1yUJQOXtDDOji1xMJ265VX9xHpGrHxWuWpd67RtqNZ26UM6sc0E9hZ1rs6 m3pg87WUetz4+YXJwLIVs7861mwp6reezebo6Z0BZ8AZWEAMqCPZzMOba/RVAXVi0iJr1JY1EPJf QkjTLbZ9OlkKuoiDtIw6Js7Tc2Xs81UMpNjRrhynw6e1KsrAoDx7g57YyS0Ck0QmenvKHxsAS0US xQIatkDEvnsMgjM9vSzGFroi1XiZhljPmUjgz3tXOrini+/ryqKVSjdvRQ1cqHukCAeX67i1Iye9 aRF0zAJLpe3tieWX3iIsFnWEDvAgncYCDvAU4UjxNevgxsI1+R86pKcIj44pTZPZN5WcNle2IL1p BjoGltg+pyfbvB+fj9Ntr/SmKnSs2fuI69ReNNOPyXpXZctXDWbLDNr5tJy+v/AZwJaXBdvJemd/ FajZkpXf25lGxKIvn3zySXb27Nl45DOf+Uy2c+fOZvF6+gYY+P3vfx8/7cUKu3zyh28j853xzZs3 m4zgNACxMMlvfvObiPWJJ54oPL9QDi6WciwUvlvBSUOqz5k10yHCaV0R7i1E79enOFjUD1He6TnL /SmDKexpeeCTDhRbcZCeZzFKfM2+lZXPY6bnytiPn0EDS1h4Er1hd0borIEHp1+OxMzJEnbUyX2o NRASHbyCMRk7kj3Zyj7N3kgStHk32mWso+Bs87rQTFCi4kxRdytW9IXOu4IVbQaQZNcT+CfgUXgf BU7oYBOkWLnSiJeADZ35zryOYSsW9zWOE/cS74ejOxVh4T4r+tRfmrYd+5QX+4WTarygh89jli3i nwX+8ljQzYy2tF0vEw91pOdG0Xv8ml1nYS+UE/75+kItXizsBU7kSNfixQILvKCnGi/iijTYTZmC 7arMso1UH8e4t5UmPVdrP7X3pp9lK9Zky/irpcDPOQMlMdDQTIg333wz+9nPftbSZ2PeeeedDOcY WbNmTbZ27dqSitS5bI8fP579+Mc/zj799NOOgRDPfBcZnhFwUYdalbdj4OooBqeCVHWSdvXpxVKO ria5TeDSzn26Xz/7Zdn6wfUxmb4GkV4T77XQoVnX4XZOU9lTbGXvD64bDB3z3sL2Jq5AHqbXr99Q 4a5sLHTK+4ID9Sg41epsSueD+5WvLaxZu8bEwV29enXWW8WppPMJvsHBgehwC2NZ25V9K7P+wE38 AgZBiETEC9z195fvVOKg0ekWBwmUuHgpTsy6dWtNHH+wDAwMxLrQ6v7Co9Xzwbpu3TodLm0L/+jB LvK2S1vFVHdsqr+/tZGoRgoAL2CR3vQa6o0/HE8GPMoW6YET9OaFaeniLn+u3b9xtFXm/KsHYMN2 wYtNWcjGjRtjHeVtF90cA8vQ0FDpUHCEpSfPCzYEFtJY8kKheS7nn/PqFwtvmeQQLEAPwSu9PpHq gxfSDA6GEeZ5Sr5888zGL3MGTBhoKAjBjdHKd2u5yXHMaKxfeOGF7Gtf+9pMw21SyjYruX79evbL X/4ylqkoa0WAi86VeQyeCYAQ4HnxxRcjz1//+tezHTt2lKm26bzffffdyF/TF/oFzkAXMYDzuiKs OXBr4lbo2Dx25Ogc8416Rvv7g5NgLanDgvNtLf2r+4NTuTI6SbMDNJVv1K8MHXecJwthJJ9gLB3h dESOjtrY2Fjs8OFcWbwyQ8dzbcBCYEjfihcH/MZpGRoq34lDJ7zgADwITlK+Mzw+MR55obNswUvF YVkfecn3M+7evRMxDgyEKcMlj1KqLoaHh6OzkscyPj4egwEjIyMmM2fghT4Ttnrz5k3Bi9uJiYmI kTqyGmWn3Ag8pAJP3Fs4wGAuW8QL9nDjxo1Z6ugD0fZyz1sEZ1CucufriPsKLNSRVZ8QeyHQQNuW Cn148NA3tAigpbyAJQ0WYSvYDDis6kiOP/dN6qSDBV4UZEs5K2ufAAPPAu6jlBdwwQucWNVRWWX0 fJ2BRhko/4kRkOhB0W3OcKMk5dPRoF+6dGlOYGbfvn3ZSy+9lLHthIAL4dULq4fefMpJEAf+XJyB TjOQdkiaxUJneMvWLdnd4OCOjl4Jl1emjmPfdCa2bNkcnO3yRymF+3e/+132xhtvZD/5yU+yCxcu xsO/+m+/yn7+859n//iP/zjTDit9WVucA179ooN16TL3OVPJp7LxsYmIgU67FS840du2bovO9Pnz 52cCEbdv38muXrsaR5xaGXVqlsOtW7dm/X2rsohlur2+dzfYz+XROGtm/dD8R8CawQIvmzdtzlYF LBcvXowjk1xPsObChQvRiduwYUMzWbaUdnh4Q7YmBKbOnj2XwQeCQwkWAlojwxvNghA42zgBZ86c iXyABV74zSwInq9Wwn2EA8UMQRwoBMcJLDzjeaXVKjiDg8sfdSIs3ONgoS20fL0WHLQj2K6cf3hh sAsnfNeuXWa8EPDgvmb26bVr12IdwYvaG3jB6bQQggzYDFjEC0Hp06dPx/aYPriVvXCvwAv9/9HR 0Vh8sMAL9kMdkcZCmBkDFp7NV65cmQlEnDt3Lj6r9+7da8YL9+22bdtigEr9YPoh7MPL9u3bzYIz Fty7DmegFgPzbhm5mekkqMGj8cUJppHjASEhHX8IHXNuNKKOaRrO0WhyjhuUc+SbCnmjg0aW648d Oxb30Ue+3Lx0mtAlLOSjwAfXcw1bHhrkzzYvnOehSp56yKfpwKigisoDJtKkGNNr0JHmy7ki/SoH 59gHB9el5cjj1e+UZ+qlGs9Kr60enNX0kK/qWdewTbGmx6ulT68hPwSMSL6uOZbaAw/xlM+UZ9LC E3WguuYYaerVdzWsaf6pXvLN48IWZZf5tEXpi9KQzmVhMrBp00js0FwZvZKN3RyLo7msP7B79+7Y SQ6rEJgVjDV3cBDoZD7xxKEZvR9//EnsgD4T1uHBobGQ9euH4hoHo5cuh1H/oxET0/zpYMGZJS+r wisF+/btzc6fOx/bhBXhKxX3AxaeF7Q9OC1WwijXvv37stOnTmfHT5wICz8ujyP9a8PrBlu3bDX5 GoXKyisZe/bszuiI01ay5gK8rFmzOrSlO015WbFiZbZv777s1OlT2YnAC1juhedEf//q6DxYfEll hpfQph84cCA7efJknFlIG08QAoeJZxG/rYR7+eDBg9G5po7AwPOJvgmDHWythPtkz5492alTlTrC scbxF0arUW3Ki06w4FwTeEA3WHByn3766Vn9hbL5IQBDe49+uKGPQB3hWLK+FX0TK4EXnHt0wwvO N7iExer1B5UXXhD6evSTCM7IXiyDv2Dg3kU3gSuCItgKeLiPmP1kKVu2bIn6sRX8CeoHm+Fetwxy WpbZdTkDRQzMOwjBGgMInTgaOwnrPnzxi1+MjS/HSIfDipCOP67hNQGEc7/4xS9m0sSD4T8cyq9+ 9aszD1kcRl6BoMGQc04+pEMn+XIuxUJeHCPd22+/HW9y5Y8z+I1vfGPWw4pXGVhTAaGjgbPMqwMq D1jBIFF5vvKVr8zgYlFC/VY68v3ggw9m6eec8lU6lYMHV35dCcpBvtUEnHLqGTXhL+W56DrKmtcD L+nrMuQDLuqL/CTUK3Xyl3/5lzN1xDn4KQoicU7lYx8Rl8wekcA5x1UWjqd1wO9atsD5tB75jVCf X/jCF2I9VY5U8kVPqp9z2Fe+HnlAUOY8Ltlcvs6rlSOfTlh8uzAZYESZDgUOrTrkOAZWI18pa9/5 znfSnx3dh5fhMFI5yJT/cO8g8GLpNKUE0PmmbSMQQueztzcsnBbaBDql1oLDdODggfh8ofOJc8do v2VgRmWGk/3798c6InjGl01oKzvBS9+qvoiFtrNSR70dw8LrQk8++WScqo2jgt3CCw6nteBIPvXU UzNYwMCxTmE5dOhQHMWljjqNBadNr1qBhSCNZVBRtoBOsGC7tHedxMKzh3ua1wx4JoEF2+3EMwle 9oZZBikvncICDwRocPK5pxHZi3V7By/4LvSp4Qb98NKp56Ps2LfOgDUD8w5CABSnnCge6zzgeMo5 xNnEOeOGYm0Con0EGj772c/Gv7SQCkAQcKCxQrieP5w+BSt0DY4/nQMaEjpQqaCHwALnFDAgPU7l s88+O+OAMmJI/jirOKYI6XDK0+CH8uA45UEfDmut8qR40nxxUp5//vmYR5ovI5Opc881jEyl5YAH 8MFPPq30wRP5soBovYAF16j8cEm9UFfw9N5778UgAHVKedFHWupZutFDXSPg0tcsSEODWi2SiwPO n4IMeeef/MiXOsBu0A+m3/72txGD9JAOQXfeFlSP8A0nlEt8E1hQmSo5NP4/NkAAIuULXQRIiqRa OUhP/bgsLgboQPDnMpsB7j/+ukHo+PX02816qFVmsFiOHNfDAp5ukG7iBafFau2SetzjpHSLvYCl m3ghINMNAi84tfx1gzgvxbXQTbx4v6G4jvzo0mGgpbA+NxCONQEIhK2innpnsBaVOHE4iDjBCkCQ nt84vDh96cgz53BQCRywlV6OI2CRA4yzyWdAEfLCgVWHmPzBnuZNeo4zO0GdZo7J8cUZno/gwCMK QLBPvgqufPjhhxyaJflyCANOfjuEQAG4mCYIlyov09UIEnBeMyTgE66YESHRPrxqn3PiU3Wg9M1s wcRMDDhCwER+YMJWUsGZz9uCygWHKlc9vtM8i/bRja3m+UK/6iZ/XVE5sNeicuSv9d/OgDPgDDgD zoAz4Aw4A86AM+AMLFYGWgpC4OTJWRRBOKwIzlY9kQNLQCEvTPlD8o638s+n53f+HI4gkg9WcAzs KUYcaoIQlAdnVwGQvOPLtY2KHE7Kl+eJ3wq0aLqy8q1WjhSv0s5nq2BBGvhRPgo6pEEX8DOyL5zU G8dImx4nXzinXPOVPE/kk+dDeefTYivUF7NL8uf4jR2At1mRDRbZaV6P8i46XnRM6X3rDDgDzoAz 4Aw4A86AM+AMOAPOwFJgoKXXMVolSE5tUZCA6fTIfJzG+eJi9L9o7Yb55ifnVcGQfD5ySgkuaNQ+ n6aM3+Jd+vM6OJ4GXwg2MBNACzASbGDGBMEGZh6w0A8zFnhNpchRz+df1m8FaYrsCZ0KQlC2amUv wqZZPc1cU5SPH3MGnAFnwBlwBpwBZ8AZcAacAWdgqTPQ0SBELfLlBFcbBa917XzO4WRrTQitM0A+ HGctgflIPexymueTt+U1CiwQZFC9EIDAKeePNSz4pBnlYfX7bpX58m0ZIOpW7hYrLt7jlaT7OuZb Z8AZcAacAWfAGXAGFgID3o9ZCLXkGMVAR4MQGllmZD0/hV+jz1aflOOzYEj6RQ6RNN+tFm+qNptD Dr14mK+eZq/TzAzN1Eivx1EHb1ofBFP4TRACh1wBCK5jXQRmkGj2AV8K6JTIVsBZJJSLsuT5ZmZI rUADARZE9VWUtx9bmAyw+JzEH95iwrfOgDPgDDgDzoAzsNAYSPs0Cw274116DDzugXeg7PpKgAIA KQQt6Jg6w+n5du9Xc0KrObSN6JfzTpCFv1RYVwGnWByk58re17oPzPLIO9ZakDK/XgTXUAZwp+eo HwIXfHGE/Wo8pmVSGr0Wkp5rZT/lO18ulVWzOtCjYASvmaSSDxoR3CBv5ZGmbcU+0nx8v3MMYIf8 eRCic3Xgmp0BZ8AZcAacAWegeQbou6gf0/zVfoUz0DkGOjoTAscWJxznDsG55UbCEcaJ1EKRFvTg QONgv/766/GLBzidBEfywQOwyHkFN/vMAtBMgDxW1k5444034isdKg+OK047MxI41gn58pe/nL31 1lszn06lvDjf+rpEPjgCP0WBIQU0qC9eY2lExB/54eDndTWSR7U0fC2DT4Dqk7Doqsa3bI/Xbfgq CsEU0soeUx3UE6/rKF/OVUubXuf73c0A3wvXN8Pn+7pOd5fQ0TkDzoAz4Aw4A87AYmUg7cfgQ7k4 AwuFgY4GISAJJ52RcQIPcv5wiHEKq33+sAxy0YUTgmP87rvvRhU43uDDSU1fXcCxxSklLefYrxaE IO0LL7yQvfnmm7PWllDecsjLKFOtPFlIks9Yvvfee7NwMVOAV1LyomADePOYCR5Rf6RpRAhWsI4E gRiknUEI6qFRvsGreqR+EK7XsXhg+j/ZB+XUGiHUYVHa9Drf724GmLqo74b39na8OexushydM+AM OAPOgDPgDHQVAz09PTP9GPZdnIGFwsCyqSDdArbaO/uW+IgiakQ072zncSgtQRO9YpBPk/4mPWtd MAOikfTptWXuM4uBMlvj0isT9Xieb9kb5Vv1iJ5GsKS4CUowQ4KADoEdl4XFwI9+9KPsBz/4QQT9 8ssvZ/ylMnV/PJu6N54e8n1nwBnocgaWrx7Osp7K58K7HKrDcwacAWegJQZee+217NVXX415fPvb 386+973vtZSfX+wMWDAQXiNa1lVDf9VmE1iQIR0EBxoNEDSTlvxJ3w1lVFm1xfFuxPlW+nZty9bZ KN+N1COBB17XYUZEiluzOfRJ2XZx4/nYMNBFMVibArsWZ8AZcAacAWfAGViUDHifZlFW66ItVFcF IRYty16wBc8AwQZmPTBbR1/hYM0QghO8kpEGJhZ8YZdQAdrxwH4YZjjduHkzu3v3bqbXO4aGBsO+ /bRI3g29OTaW3bl9K9Tisqy/vz8bHBzMOjlF8/btO9k///M/Z4cOHcz27NnTEeuinscCL7cDL48e TYapq33Z4MBQtmLlio7gAQuz4qgvXgdaF2bHrQp1ZS3wMjE+kd26PRHWY3qY9fXBy2DWF/jphIyD 5RZYHkQsA+HLRKv6VwUojz+la4ELXmjbae8fPnwYFyam3eeLV51YwHZ8fDza771792aw6CtXFnxI B7zcuXMn8sLsSV5ho33hC1Kd4IU6ujnd9jKYAA49n4XZakv7zyLXbMEyMDAQ/zrBC3Zy9erViIU6 wlbgpRNfTgAL99Ht27ejvVBH2EwnsNCuUEdgQT+8MDDYSSzYMM9n2paNGzd2BIvVPeJ6nIE8Ax39 OkYejP92BrqVAdYoYX0QHqasGcIfnTCOE4RwWZoMYAOnz5zORkdHo9PEq1lXrlwJi9qenFnw0ooZ giGnTp3KLpw/Hzo1vdG5jYunHjseO6NWONDDfQKWX/3qV9l/+S//R/bDH/4wO3my8hlkSxzownFi geEzZ84Eh/JR7Ozh7B45eiTDubOUycnJaCsnT56Kzi3BS4IRJwNX165dB60ZHHihQ0493b//IHbI cRiOHT8W688MyLQisBDYvXvnbgzqguV4+H159Io5L9zPH3/8cXRWcFRwXvh98eJFa1rielRHjhyJ QQgcW2yIgDjrKlkK9kLwDB7YYrscO3z4cMTCvqWwThc8YDc4tujHfvizxoIj+dFHH8W6wl7Qf/To 0dDmnTTHQpvGoAm84NgShAAH9zm2Yynwgu1y32AvYAEH3FjXEcEzeAELgV8CrmABH89xSyFQBQfc w9QR/YYLFy7EY7Q1Ls7AUmHAZ0IslZr2crbMAAtotnMRzZYBeQYdZ4BOxMTE7ezgwQNx1gGA6BB/ /PEn8fWdXWGdkGVhxKV8mcrOhNeF6Jg/cSi8MrR2TVTJCPvhI0ezs2fOZgcO7g/HbEaU//7v/z67 EDp7+/bsza4HTFPGnd+UbzjhVSpmYQwPh7UCgqxfPxScpyOhc34qe/KJQ2FGhM36AYzaEgzZtn17 tjksjMso6YYN67NjIVAExjWhQ2o1CwFn5UywCxbY3bp1S+SFGTwnQ8f81KnT2drgNFjxgrOCQzAS ONm+bWvgZXnkhfrBdgeCHVvNFCEoBBZGaw8dOhR5YYRSTi73N46mhSj4wWgtrwJqVJ2R25PBsSQo AR4LwXE6duxYbOcOHjwYHUr0ggWbxrHT/VU2HmangIUZB08//fTM6DEYCIrAC3VmISmW/fv3R+cW vcxCE19btlTur7LxMLMK2yWA96UvfWkWLwRs4Cv9THnZeCg/NvzMM89E+0Af9sqi7vADFtl0mVgI voCFGRDPPvvszLOawAhBNZ7jLLBuJdQRz6Xnnntuhhfs9e233451hh1Z8GJVXtfjDFRjwKJ3XE23 H3cGnAFnYMEy8Ch0hOlIrFmzeqZTQ2EYYRkYWBenw07cum1Svnt378cR9cHBgax/9eNp/SsDFhzu GzdvZLdCsMRKXnnllezVv/u77K//w19nW7dstVI7R49mHqzqWzVrmjZO7uZNm0Nn/W52JUxbtpGp MEvmahyNHAkdzsedzGXRYXr48EHAwqh/+QIvTNcGA/YhibyMbA6zNB4YjvpPZaNh9hAO1JYQEAFD RZbFLy4tX74suxQ+nTw1ZTOKy0wm+EkdRxxtfrPFgbASRm1xcreHoNVje8kiL4wqnz592mzGlV57 IGiFbsnIyEgcybXkhTpiZHvbtm0zjjZ4+OKVRrjhzUKYFQI38IJuCYEjnFwCNFYj7eAguIi9YKsS 1qxixB1n2woLgUXwEJgiOCSBE/CcDzP2qEMLIfgAL9hqioWACAErAsBWvBDMY6YgQYcUC7Mh4IXZ i/QrXJyBpcDA41ZqKZR2EZfRp3At4sr1onUlA9dCR0Hv9OcB0hnFkeH9dgu5FdY6oBO1cmXfLGcF 3XRukPGJ8KUP4ynTUXEH/4MTRrZ7V/TO6pQDaVV/xWG4Ezqo1FXZcu8eWMbjqzI9iROH3hW9K+Lx sTGb10MIMuAg4Kj05j7pJl5uhQAar/iULaxFMREcBF4hSp1b9EZ8gRten3n48FHZUKId8OoOOHTf SCn3NHhwaCwcFnTgbPO+eOrcgoeABMdw4hj1Llu4PwiIoDd1nNDL6Dp8gQUHq2wBCwE06iJfR+gG H9zhdJYttKfwgjCynwr1Bj7qx6KO0M1rRPCTtxfOcYxz3PdlC7zgTCN5e5EN0WclUGEhvJpCUAoO 0mAeusWVBS/o456m7Hl74Rxc0aew4gWdLs5AJxnoWBCCm6zTNxr6u9F5bxYX6V9//fW280m+x48f j+9Td9JIXbcz0G0M0MkaG6t05ooWfcShQriHLITRODAt71k+p5MlfGCZMlxzwKLc9XTgMNLxxmEJ /tMsYeFQOqS3g/Nk8Ry4fftOwDIVHLa5C5ZSb4z4E7SycLbvhHUX6JTDS54Y8XLv/r3sbnDmypYH QQdrUvQEDvICvp7wd/euTR3hRGML2EXkJgFUsaFl8TyBrbIFxxU8eRzoBR/3NY6/xWgyOPiT3rTs 4OOv0iaOpadK2cduCTBIb14JARHqELxli+oIPfkAmmwIXiwCImDAkS6qI87BF9xZ2K7qCL1FvGC7 8GKBBQzMPKD8eSwpPrCAqUzhWSRb0HM51adjVrykun3fGegEA3Of+i2gYEoT73o1Ir/5zW+yn/3s Z40kLSUNHXL0v/POO6Xk30qm4HrzzTcbyoJy/OIXv4jRXI0KEDjgrxWhHsFBPb311lutZNW118IR 70m6OAPNMsBoxaPpkVk6N3mRQ0VnrOyODbrv3a2Mhi6fmcr+GBGdUoSOeSfXZniMyG4PRxrBwS8S uMHppz7LlgcP7gdbICAyNwgBDv4hFrMPNJKPacg+0vJj03SYeeWobHkQdExOPgo45t5H0XSnq84i UKQRazgpuq85xv0s/srkhraDP3Tm6yjFJ8xlYkEH9pDqTfUJn4XjDyfYQhEvYFK9WdQROMCT6o0/ pv8TLxa2S/1QT9KZ4kjxWfBCe6oyFzn+wmiBRbzAgfSm3OgYWMp+VqdYFHDIYwGPBS+pXt93BjrF wOMX+9qAAMeViCPv5fF+HIKTTFSPd52YtufSXgbgnEbr+eefn+FXX27YGRbFmw/nOOfkS5194Qtf mFce7S1la7lVs0ECELJX3t90WXoMqANCydP9ekzQWaFDUe860pC2mbzr6S46/2gaS9G5ULJ4OGIJ I/HZXB+4+LJFcHQyfI6zItOebK5M1AsOcNmdT9SqjoLKYpGzHV6VWBX+lSkztjttG3ld8MKnTJUu f76dv6WjmJcQmpk+IWevnbrzedULRoEFW6mXLp/vfH6Ll2rXWvNS6x6xxJLyIr0pRzpmVUfCI72d wpKWt9NY4ES8pHxoX/hSzDrX7m2KQ3pTHToGllo2nl4z333yT/Hk8xGWVto65ZHP2387A93IQFuD EF/72tfiIjwKQFBgObRf//rXZwIT3UjEQsQUP78XAgZ8OpLFfiRf/vKXYxR6PgEI8sAxR1i5l4DS QhcCKtjhiy++OCugwuc1CZB5AGKh1/D88acP7HS/oRxDh6K6VDxKdTo0Olc9fWtnGGGvJvJ5a3V+ ql270I/X61RShaqj0ssqe6llNgEEr2yULSkvRXYfeQmv7ljYzEx5qxV7+riFw1KPd7iyspe0jmrh 6gZehM8CS6O8WNiuyl1va81LLY6seKmFQeeseJE+bYvqC15qnS+6Zj7HpKO43a00dq3UUVG+88Hp 1zgDFgy0NQiBI/zkk09a4HYdgQG+ecynwvKfjWQGRCuiaXRWnyFrBWsr1xJgWQxBllY4WOrX8sDW qEMzgQKu4z1+RL7lbC4rnQmmmDeT7+w8Gv+lNSiKrtA6EBFHwN0JgY1inspF01Ow/kJeY6zLgldq 8ula/V00/bYoz0bTFV3b6DHpqNRLxVbz1/J6iIXtFq2RkWIRulo2nqZvZV+8VMsDB8LKXupxL2em aLp7NfzzPQ4vlLue1OOv3vWNnG9URz3+GtFVLw06GtHTKOZ6+mqdb1RHI3hr6WnkHDoawdNImkb0 1UqDjlq2q/sIzLXS1dLR6DnyV5mlt+hapSk6V3SMfOfTjynKy485A5YMNByEYNSdEWVWmUWYqp8f gccp5hNAHGeE+cMPP5xZeIZXBFizgOn96cgzU+VZd4B8cX6ZRZHPF335dOTBd7NTB1z6P/vZz0as rHLNawr1nHKm5XMtOgikkCd55CVNxywD8iVdOguBaxh1P3HixMyCjpTpM5/5zJyZIJQXTvl+scpO ukYErKzBkZZf18En75QxMwVhZgP87927N+qpVlalUx2TD8K30sUhOLmeMoovypfnoVZd/PKXv4yf y+M6OKWexKdmJ4C3nk2kNgmufN2pHvLlYVYOgg7KrN/xYPgvn2+RrZGWcujVI+oxX475zkQRDt+W zwAPbmwHaebdajosere/aBbC5LTHzVoEZXdswC5HTg4bxyRy/ut1xpR+MW17w5cVkOodvrBQZFhE 1KKOxL+CQnN4nq48K6eyor/IYip8sSDk8tCBL1t6QjAP/ot5AV8Fo2y8TDwp99VsBqxpurLwoCPy ohs4p0j4LJ4zwgIE6c3BiT+1LlXRuXYd4z6i/QVHERYds6gj3dOUTXrTcuqYFRbwVBNLLGkQQnqL cFnwIix6zhfh4BhYSFumpG1HO3lhRonKp22Z5fC8nYF2MdBQEAJnl4USGRlnpgNGzrE33ngjOnAK KvBtWxwxhIeRRplxVkmDk5h/SLGoovIlHY4jx771rW/NTJ3nOF9/4Abev39/PI6Tj5NMfjt27Ig6 pR+nk+M4uXl9MWHy35nwDWc58zzMyReHEkkDERzjjzxx/IWV8qZY03QEXOCKPHFY86+k6IsW5AlX 5Em6RkSONd+DzovqQMfBwDECQ+wrcEG5wSvnHa7AQQAD55w6gxPNiOBaYYbzWjykdUG9pXUBFv7g hTzQyT51j27OEeTC1sQztpa+TkFa6h9sqqfDhw/H8lAOAlSUq6g84oUyoiuV1NbztgYW6eIaroVT Ain5cnCeYJpLdzNAx0OfymqmQ0RwQU6RAg5pSTWdEicY+y9bVq6sONusb5AXBUlWhM9UWmDJ69fv Wp0upWn3diXrEAX+0Y0vl1YFvFBPy1fyBYbqnfd2YarYF2tQzH11hmM44dQP9VS2rAh2DxmTk3Md ObjCjnpDO0qApmzpCToodzEvFXwVXspfU0oOfYWD2fXEMTr7VlhwiLCZiu3ODRbBF1jq9XHaUX/w gi4kX0/ihXMWWMBRixfLOgKHHP+iVwvEVeXeh6FyBf61iGhek7DIxvPn2/kbTlTmIl50zAIL5YIX Fk2V3rSsOmaBJW07pDePhfupWSzwrX5M2YGUFK/vOwOtMlC3h4HjxRckcPZeeOGFmZsDJ4+vJzCa nB9JBhQOLH84uThrjMLjiOYF5y117DiPg4mTrfT6+gN6cCwR9P/85z+Pjuhf/MVfxGP6D+c1n6fO 5bd8IzjNFzyUC2dUeWhknHOpY0n54Aa84MGBprw46JqFgD7KQRlwtFUmroFbsBKskOA0awaCjhVt cfIRBQiK0uSP0QBShymHlFXBAI5TZnDhoMuRVz6UlXPMVuCcRDww+0Gc6RyOfFo+HQd3yrvqU7yn +YALXgkYUQcIv3mwpDbJOWwCHOQH1/xVK4+waEs62XoeG8EhdFI/qkOu4xowwAGSliMe8P+6mgEe +JJ0X8eqb5eFe29d+Ob31exRCM7lRV84WLNmdf5UKb/XBSyXL4/GL3bgXKdfGtA0zdWr15Q+0lO7 cI+5rp2ufWfpmDGiT4cvX78svIj096/KVqys+yhsGdSqvlXRYXnwYO4XJ3AQ+OvvXz3j1LSssEYG feF79L2h4wonck6UXJ1jOsJ9fSt1uLQtzgptOc/PvER8oZ76AncxoJRP0Obf9AfAAifiQSpkQ+Bt 5rmr65vdggX75ZN+edsFH/c1aVaFuixbpIfPgao9kU6wCN/g4KAOl7bF4Vq3bl3sI+ZtF6UMllBH cspKAxIy5h5ZvXp17AegN5XUhsBrIfBPfztvu+im3nBQLWyXOkIP/cg8L9iK8FlgoexDQ0MZfea8 7YoXtmBRoI3fZQj8Ywujo6NzeEHffHnR/VcGZs/TGSiTgbpzj2jQcLR4TSCNzuEY4ozRyLQiqbNJ Pjh5CDoRGg72OS7nmeN0FAhsaNSeYxLNjNDvWtt8vujgL+0QETBA8ljRj8ARAqaXXnppVgCC4/ky cQzHH0mdeX7LyWa/logfdDYq4E05VFkZzW9EKCcNdR4zv3HCmRGQF3GUPy7dOk45VG/a6hwPEERl Zp9ZEQSfUpskD/JN05G2UZGtU56UJ/Il8ILwulEq3AMKQHCctNQ39jNfHGn+vl8uA608vDcEm8fZ vx065mk+dD4ZjaIjbNX5XBPuS2zv7h1Ge9IR3Knszt07kUTdR+UyWj33QIu5MKtg7dpKez4ZAhGp qN0bGBgMnfPyZ0L0rao4lcwyyL/6w286oMMjwynE0vbhZSg4LA/DlzjyHXPxwnlsuGyhDV+3bm10 /HFyU8GBuR8+bTq0fijrMcCCE0KbTvud/9zk7du3Yx0xS497rWyBF3TRtqA7FeoMfDidFkEIeOG5 RtuWxwIO6onnIMGKsgUHF17AJFuVTuHj+T0wMKDDpW1lLyggWJQKNsR9zTPAIiCC7o0bN8ZAQ77v gQ1xb1E/cGchGqzJY6Gdw4YI3lgFISgzdgMH2IgEXrBfggNWvKCHeztvu2CCF+qoFdtN+yIqp2+d gW5loG4PQ0EGnGZmBKTCTSRnK3Xa0jSt7qthpyHLzxDQTcw27wi2qje9XhwwEp4XOiM6r3PwxIg+ uPJRYKXhHJzNlzc17PO9Xjga3aKPv3yAQNfDPzyQJsU0n85a/ppqHRt0ESBCr3iWTQhXM1vVY5Gz pmON5J/H3wwGT2vLQCsPbN6XHx7eGAKlN7OxsbHgEFQ6vdgl7dbGcI5RdguhU0MHlEDa+Nh4dNro IN+aCJ9IHp+II0HV7qMy8DGT7drVq9mD4CxdunQxqjh16mR2+NPDsTO4fcd2I+dpeeBlOLTFp7Kr AdOmTSMx4HD//oPsapjFQkdY93YZPOTzxJG7d+9uNhpmrWzdtiVw0RtmrzwMM2quhI7pymz9dMA1 f127fxM82zg8nI0F2+B5tWvXzsgLgZrR8Js2jPNWMjw8kt28ORZs5XK2Y/u2rDfY81RwFrBn7Hjj tNNpgYf76Gqw3YsXL0ZngEAMjsuFCxcili1btljAiDpGRkZi/Zw/fz7aKvcwWOhfILyOaTX9GucJ XqgTnCQca9pPeAJT0auhEWQJ/1FH2C1YwEUgBgzwhJNLHVnxQn9H3LDlN7zAFc+CgwcPmgTzoJmA B30xRtq3bt0608ZSRzjbDHLhjFsI9gEf9Kv4Axe88Gzgeblnz54ZfGXjActwaM/QzbNZM3ZUR3Bl xQu2iv2CJeWFviW8ELyxGrwom3fP3xmox0DdIETq3BU5YKnDWU/ZfM5LPzdrJ/SDmUALwgMvL3T+ 02gugQr+NELOeaTo2nxezfxWvvCj/Waun2/aag62MFBHZdsE2AmK/frXv46dZQIjCkKhXzbTbBl1 XVqfykPlVhod9+3SZmDbtm0Zr17wqtDERKWTVQlIDGbbtm6Ljp0FQxqRo606QwD01kTsVNHRwdFm YVnSWMlPfvKT7He/+110CBi1XRtGuv/pn/4p3rPg+P73/1NY8PagCRw6wvBy4cLFOBJGh3QsdESR 3bt3m7afQ0ODwVnanp0+fTq0U/ez/lA3dIqZvbJ79y5TLHR0tweHH4f25MnJaCe0n3fv3osOgto8 i0rCRnft2hXb9QehrtYNrAujgrciFuqIOrMSeEHn0aNHM9YZwuGmjvjDibN4vqmsOCwHDhyIrxge OXIkBswYLWWGaCewUEfMeIQbnrmMLNPGcNwymEd/Y09wYqkfXr/EqWPWAc43sy/5bSXcJ9QFC7FT R+jG4aeO6JsQSLKSPC+0ffRZCHLyDLDEQpnhhTqiv0bbQjuMvYCDQJHVM4mAFDaKfmyXOiIggr2w zzlLQR/1wqL0mmEEFnixflZbltt1OQN5BuoGIeRcssaBHL18Jha/eRUi/xqAhV7p4EHDKwC1hIcO AQga3nTtCK4pmkXRikOrDiKNquqoFrZWz0kH+oqEiD9S5MAXpW/12AcffBB1pWtCkCejI8LSrA6V sSiQojyVptm8Pf3iZAB72H9gf5htEGY/TIzHThWdCKaXZ5md0w+7jNjSAcdZuhUcFUYG6aizLkW6 RoRFTXz3u9/N+OsWocM7NLQ+cDMWO3/DGzfE31ajpeKBTjedXtrJm2M3swf3H2YbApb1g0MmX6IQ Dm3BgpPNs+tBmB2CI4kNWY0KCge8oPuZZ54Jo4Hj2b3798Jo5VC2Z3dneGHUlP4OjgHOLfsEAzrR /jNqy+uADMTwHCJIwn2tPoA4tNgKC04tjjYBGbB0ghfslvWmcGrhhUDWc889Zza6nvJNkIw64j5S /4HngMWrMikO9sHyuc99LgYeNL3/2WefNQ3kCRNY0E1g/ubNm5GPp556yqyfKBxsqQteK8deqCPa HNob7MYqGCI83LusBydeaG/5Cp1maCidb52Bxc5A3SCEAg+MlmjfkhS9V4b+TgUhwMC0f2YzCE8R BzRuCJHNVIqCDY3mmeaT7ms0hqCA9tPz7d6n0UQPTj7lyXc66CDRsbbAQoeDP4I9eRytlJtOA6M8 lDFfz5rJUu11lFb0+rULnYGwSGUIOvDXDYKTwp/LbAZWhTUZVq2yG5WcrX32L6bVbxqZ/ZyYncLu F22o9QhptdLxnBkethvFroaD4zgGlq9e1MJCgJE66oZ6gpf887EW9jLPEUQkYMRfpwUszDzgrxuk GziBBxx8nOtucbC7pY66jZdusFnHsPQYqLswJU4XziXOGVOOJTiirNGQX6dB57VVtJ4I8XwEp5YH Hk5gfjYBv/m6Q5GTPx9d1a7B2UUoK86vhH30ww2iWQBMPRMmtrw2kBflyddFlCdp+TpDI6KpjzjM VkLklqBHHiO/KQNRZgtR4IHAVJ47BQuKcBD9riUEj2TraT7owNY41y2dr1rl8HPOgDPgDDgDzoAz 4Aw4A86AM+AMdCsDdWdCEETgc4V8ovCtt96amQ2h6Uz6akC1AmpWAI4q7+7N57UOXm3QJxKZkYAz yKwDnHZe05BTWg1Dq8cpA+WkDAQdcERxxjX6r9Fx0ilg8tOf/jROl8SBBS88pk4wacGOc/v666/H NJxnhEyBm1q40UM6MFgJM1EoN5hx0tN6IKiiwErZeCi3uKM+CFSBi8gys3XgBNuQXYALu3nzzTcj Z9/4xjeqztiQrWNvmvlDfuhMP9tZdhk9f2fAGXAGnAFnwBlwBpwBZ8AZcAYWIwN1gxAUGicPBwzH k9FnHDIcb5y7dPo9qyPzm/MSHLnnn38+LkzDcZ3jvVMFKJSWLelxMOUAcow8//zP/zw6koz843Du 378/YkjzKNLP9UWCg5rXo3SM+OcFB5wyM+uBYAGYKAM45OxyDTzh8IonZiyQhmPgTp1j9JOPZjOQ P7pJm+aZx8Jv1QF60jw5l8ePDnSlXJEOyaflWC0eyYcACHrhoageGskDTHmpprcIv7jTO5jYC2Vh Fej0GDooN4Es8VzLBtFVZOvkna8TYWi0HPl0/tsZcAacAWfAGXAGnAFnwBlwBpyBpcbAsrBC7NRS K/RiKS8j9G+88UYMMOAQuzgDzkBzDPzwhz/M+EP+6q/+as5iilP3x7Ope5UvKTSXs6d2BpyBTjGw fHVYI6Dn8WBIp3C4XmfAGXAGymbgtddey1599dWo5tvf/nb2ve99r2yVnr8z0DIDYfZ6+Fi4y4Jl gNF/ZmgwO0NrIyzYwjhwZ8AZcAacAWfAGXAGnAFnwBmYFwO8luziDCwUBjwIsVBqqgpO1qrgFQLW O+C1DBdnwBlonAFWNH/06FH8s/4sYeMoPaUz4Aw4A86AM+AMOANzGSDwoH6M9Wen56LxI85A4wx4 EKJxrro2JWsYMCuCNSdcnAFnoHEGHj58mN25cyf+8d17F2fAGXAGnAFnwBlwBhYKAwQg1I+5d+/e QoHtOJ2BrKGFKZ2n7maAhRZZeNHFGXAGmmOgt7c3fuWFq/ILjzaXk6d2BpwBZ8AZcAacAWfAlgFm cfK1OoQ+jYszsFAY8JkQC6WmHKcz4Ay0nYF0Xd50v+2KPENnwBlwBpwBZ8AZcAbazID3XdpMqGdn xoAHIcyodkXOQGcY4CsqvnBpMffteniTz8OwJguvd2RZZz84FLEEHN2ABS4ePqzw0i6ui2uy8aPU 0727nZ+yyhRasExNTTYOvqSUk92EZXKye3gJWFhraTJsOy1g6BYs3MvdgsV5KbZMeOEZUHkOFKex Oip76RYs3cRLu7F0y3PWyrZcz8JmwOftNFh/OHJ8heLSpUtx/YWvfe1rDV7pydrNwLvvvhs67lMZ i3J2WmQXO3bsyPizlrNnz2b88YlWFigtEjDCmdYOKUqzVI+144GNQ3vx0sXs7p272bLly7L+Vauy rVu3Zr0rVpjTimNw8eLF+H4oyvv7+7PNmzdnvLJlLXSuwEIAjIWz+vpWBl62dQSLyo7T///81/+a bdiwIfvTP/1THTbeToXnyOVsfHwsLCY2GV4DWplt2jQyM53WEgz2P3rlSjZ282b26OGjgKU32zg8 nA0ODlrCmNYVsFwezcbGxrIHwXbgZWRkOFu3bl20H0tA8HL16tXs8uXL0YnjVa2RkZFs48aN5lgo 9+joaHYl1BP3N9Oth0Mdgcd6JXx44XlCPwgsTAMHRyewwMuNGzdiG8N6WGAZGhrKtm3bZs4LWG6G ewheeDefOmKdLtreTix4TJt75syZuE4Y+rmf4aVKG7bAAABAAElEQVQTixZOTExkFy5ciLyAhdcG tm/f3pHXBqgb+ktshQVeOvEqJms3UEe3b9+O9rp69eps9+7dHeEF+3VxBjrBgAchGmCdBv0Xv/hF fOh2qkPfAMwlk4SGG+mGIAQdsePHj8cAQCeCEHQI0b93796qQYh9+/bFjuMvf/nL7IUXXqiabskY UBsLisN04sSJ2OHcu29vnARxefRy9uFHf8gOHjxgyjWdPWxhYGBdtmvnrtDJWh6CI5di8HTXrl2m jiVt5okTJ0MQZFW2Z8+e2MnCVj8KvOzduyc6C22shppZffDBB9nvf//77Pz5C9kf/vBRcBbGsm9+ 85s1rynrJO3F6dOnMzqgO3fujEGi8fHxWG84LTt3Esi0+cQaMzFOnjwZOuR3s127dgZbXR0CIxPx GEGaHTu2h3qzmSzJqC28YMO0ozgq2BD41qxZm+3fH+4tI17Awn2E03/o0KFwPw0EXsazo0ePxsDE gQMHTB1LYeE+ol5wWsBCgOQzn/lMWaY6J18CEDx7z507F+pjf8TCYr4Mzogry/fRT506FZ1b2rZN mzZFx1JcPfPMM2bOHLxQF8eOHQv30a74LOb+5rnAcbBYOrkEfuEBeyE4RJsDFoJqfNJ9VQiSWwlB GXTTb37yySdjQA8bYlDkc5/7XAhM91lBiTZ65MiROEBAPVFv2BDPB3jRmgoWgK5duxbvYYKa6AbL +fPnIy9PPfVUDLxa4HAdzkCnGbDpYXS6lC3qp0GnIcfpZTTZF4GcP6GffPJJhjOMQ+JixwAzJZDf /OY3dkqXgKYLwbHt7V2RbdmyJY7wrwyj/XS46HTSWWfk3UKmguOEk01nZuvW7dmq4PyvCLMfIq6A hY4fU+6thBHtR48eZju274gdTWZi4CgwS4TOlhUvlJeOJo7bH33pi9kXvvAFKwoK9VwPI7eM3jL6 xgh/ZcR0KDp0OAkTE7cKryvjIG3wjTB6uy3M2sHR7unpjcEhRtlxKm9N3C5DbWGejCJTfmYQMWoL L4ODA9GJgi8CR1aCPpwEgiE4/ZU6Wh9/g5E/KyEQw/3CvYNDCRbqihFTzQKwwqIRbdo3YcFxA4tm AVhhYSSb0XWcuDjrLPDC/UTAHZw44rSFFsIsDAJo2ArtLW0/vODoco6RdwJbFkI/lbKjH17Awgg7 gxQERsBpxQt60Id+7iWeARrtZwYP7TKBUAuBF+qBAAxYCH6wTx2BhWCnJS/0DZiJgX5hIWgEFjgD r4szsBQY8CBEA7VMRwCh8XJpjQE6TkTH/XOirfHY7NV0AIi4wz1/Lq0zwCjgxK2JOIKcvu5Q6YSu iaOnE8H5tZA7AQuONg7KypWPJ7iBZSA4dWAdD51zC6GzOxZeNaDD2bfq8UgXHayBoYE4YjkWRpat 5M/+7M+y7373u9mf/Kt/FQNEVnrzeggCXQnT6pHZrzsEXkK9ITi/FoJThDO9IgTQ8q9xDW8cjs7B 6JVRk465sFBuHDnspCLLYlCEV0QIilg5cjhxYGBmSipgw3HAgbBwWCgvWBCCEI95yaLDjfPCSLuV w0Id4DTOrqOKLePQWfECH7yewuteBMzyvPAKGrxZfXIZXtClwAz4EAIB/PG8taoj2g/6q2BJeaEt 5o/AI88JC6GvBy/YC88hCc9KXpuh/bHCAicEp7inCeRJsBWCV5a8MHtSWFJeeFUGXuCNPxdnYCkw 8PhubKC0PPDkwPBQJCJeJKSjMaTh5aYvct656XFEyUP5ckMyRVWdokbyQT96mAonfRqJzGOjIWJE UOnAJl35tMpXjTq/wUzjwYOF69IyEGXlN1PO0oYlLRv68pyBBR3kSeMMPnCSP2mFL80n5agId/5Y ymO1eqNeVa5GdTWaDjx6CLBF1MjmH1Ap1mq2EzPI/SceOVyUJ+WjXorqIK3HVD92S/pGBQyqv1rY 03Rgop6r6SGtbDtvE7VwqUyqU9IShGBauqZH1rrez9VngE4UzkgagNBVHOPcRHC2hwzer6djwwKQ lbZHTlwFjdqj8Ynx6OymnVPhbeeWUdEHDx4GXY87nsp/ZXjPn2n1TG/nPl1KQqCIAA2d4HwdcIxO KO0io2Nly72AhWdZX9+q4FjPHotgNg+Cg0D7U2Tf7cT34P6D8Hy4HR38PC9wguMPVhzgst9px7Gl 7cSpTp0VygsP4KMOca5wYMoU9GAPlDmPhWPg4XlKOt3jZeGBF5xt6U31gE11hM2oz5Kmaec+WOgz FfGCHoIztEFgKbuOsEnNjEFvKnACN9xDYMmfT9O2a5/+C1JkDxwDC8+KsuuIAJqw5MstG6IesV+C AGWLsBS1ZeLKghf6BNguZS/CwjG4gxcCSS7OwGJnoKEgBDfEm2++GR/OKSFMfeMVBd3EnHvnnXfi O4JpOho8FnJMHS3yQ3CSFNjgN++KkScBChymVLied9pT4X1E3umicZUU6SOvfH6kQ1dRkIS86Izw 6oBEmHkdg7KDFew4d+BA5OTDGetIsE0FhxMuxJl0gIG88uX46le/GjklnUQcobeWoJvp9ym/pIdH MOhBRDrKCTY6eynmDz/8cM6ChpzPlw1MTPnXtP88Ll5pSfnHTpAXX3xxhguwki6VorpMz2uf/LgW /ZQDgUvqLF9+eKPeJapXrkv1g5d09XgmH+rnpz/96Zz6+8Y3vjHDM+kIALz99tsxHWUT10Xczcdm pePXv/51dPJ4fUjCA44yEjBzaY2ByclHwRG5EzOhs5mX5csrx5g2bCF0chEWxpzryE1juX0ndnCK 8LYLI50sHLSpqcngJMzlhTUGGOhm7YGlJjiLdDDV9qflp86WB27u3rsbn31FHdQ0fav7d8NiqtTV 8mAvsUKSDCOW4OTSfoK5bCwPHt6PnXJmPORl+TQvwlLEXf6aVn5ju9QRHOTvo5QX2u2yHVzKDB4c 2WpYcGa49+lHlSngQBcOJH+piBeO4fyrX5Gmaec+jj9lph3LY0EPx0lj0fbST9WszqJ2Vfiwl7T/ 204+0rwI7qKzCAvHuOf1rEiva/e+bJd8i7CkvLRbd1F+8I+dFmHRMXiBn/y9VpTffI/Rtoh/6U3z ghf00w93cQaWAgNzn/oFpZbDiUO8N7xbhshB4oEjx5NjOOPMBuAYHQZGht97773o5OYXxaNhYFRf Tjlpf/vb30anm4cqDhxOPenkTJOGaxCcKZxPRtRwttAnpx7HEgcXIQoKNhx9ykA6jpGGfFNHOF4w /R+6X3rppZgG55R0RQ9YpiFSBjpretDU4gy9qXOIOsqCw0twA2HtBDCTD+WTbuHmHGlrdcqEIXWk uY4/yp3HgLOe1h11Cb8EGNK05MtDJq0fykS+CsLEQiT/yclWoIH8FCwgmYIIqX7ZAwGSanXEtcpT OjiG4IhTJpUfzASsKBf2lQYXsDFGnbBR6lB21AjP6KL+0A9+RPWX8owOMKH7W9/6VrQljhVxh72h G450f4DprbfeqmmzBDmk4/nnn59jH0z3gxPykq1GwP5fUwxMTk7FrwlwUVGnJTp34RydYRzyshf4 ux9GkxEc2byEPk2UCpZy35OmE6f1HsRBikfHCOJY8JLq7vQ+Tlx0/AvriGBABWG1UbJ24n8Y1uuQ FNkvx+gwYzNlC18IwRaKbDfcODNBEngpW+RQUn45SqlOHeNZUrZQXgVEinRZYoEXbBeR3hSTbMiC F2ySv6LgDJiExcJeVEep3pQX7Vtg0f0KL7XEoo7AIj1FzrbqSGlq4W31XIpFetM8dQwsZQchyF9t TK37yKLdTTnwfWegUwzM7bHmkOD04ijh8O6dDkCQBIcLBwmHDuHGwmnCwWLxLznHBAyYOcB5pY0X hP9w2vNppYNVn3FokTTQwbuAEvIjj9TZwrHCuQSzRsC1JW/hIm/KQPpWo4441AQ4yJP8cSDRT/4q D5j5DY/gESaVhevBzfX8kVYjUJRPwQ90kCd81sKNA696S53tehjS+uA69OKwSlQ20qX1oyAFTnCz Ak4FBlL92A6/KStOfZFgc2CiXPxJsFsCA/Ct8sMrAQl4zecnO5JjzrYRnqUPPehP64/7g3qmfAjl IE1an/CrFc7BLJFtc++QJwIm3UuUOS/UE7MsuAepD9lMmk7lS+s0Pe/7jTEwFZymR6GjhagTM/vK ikdJpwMnq2xhEciqMh2FiB3haUeiatoWT8TyzvBSlJl4ycLIavkObhGCTh2bsYPpYEM1HHwqs2wR lmUh8lFkvxybxHan67JMPDMd7iq8TJtvCG7VsPE2AZzBUiU/eMHGrZzKKjDiYdWbhSMHL5Qbkd74 I/dfN/FSry5z0Of1EweXP6SIFx2z4CXVIb1poXTMghdspVt4ScsrDqrxIhtPz7dz34KXojK2swye lzPQTgbqBiHkLOOQ5QWHjlFapFY6nFWcPBzjVORcpcfkeOcdKP3GkUPYolNOFY6e/nRMTp1G3Hm1 QDjJA+cUZ03pOTYfETZdi/OLEFjIC99HRrQmgs6r3PrNFn7IO5+/0oqL9Brtq+xywHWcLQ4zwgyO VJRvegyHNtWjIAOciW+2dITAmfKb5lNrn3fkkCKscAgucZrmQwCCP82eSM8JB7aX4mRfx9LOm7hO 8xAfafnT8+l+vo44p3tGWOCMIIS44zh/mp6n/NCnmQr5fJnJwAyUvG1RLmao8P5ltQCE8meblj09 vtT20wd2ul+fB9y3+hL77dOd9/qp55+iJnZNftB2/mrqXgknjfASv2VqwEt1wAZkVFNeR/VkCHDZ ylxAM1Uzs1MeohkVc2FUlE4fJyjSacGJ0F+nsUh/2Y6T9KiNqaWv1jnlY7Vdylhqlb3WuTLqpkif jmlbht6iPGvpq3WuKK9WjxXp0zFt56ND9+l8rvVrnAFrBmrP2wpocG4QnNG84BBJ5NTknSad57jy 0rFWttKHE/ezn/2sZlY4nTh/ONBM7cfh5BUHnEQ55DUzaPKksBUFN7QYW9kj0QpyFNWb6qgR5zpf dF3zxhtv5E/F38q78GSVg8Ka2pOSEgigvqRXx7ElAhBIUYMtW+N1iGpCnuRdlihvYUEPwSFwKzBR pFv2U8QlfGDPeVE5Oa/gST5N+jvPZ3puKe3zrjMLUiFpPdXjgAf98um1IIrsLzrZIROmXPb0zl0b oV7+zZ4vmvL6OI+K8xbTaFj58cn27oX8e6anAxf7jBUs8Kd07QXQvbn1Tq95UNWVnj5Rbzp1O0oo e0FltXoizCYbb4fOanmskL0UJghOf/iHiL/CZG06qDa7VnbYbiPpauXRyLl6dqB2xwILOhpxbiyw yHbhUBwU8ZmmKzrfjmO070XT6vN516vLfPr5/G5UR6Pp5oNB18AJelhTppZY2Eu98sqGSNeIjdcq T71zjbYdzdoufUb1Y/KDWvUw+XlnoJMM1A1CtAtcWU4Psxw04pzHmgYBCELwx4g6jiBbHDcCE3qV IH99Gb/lYDbiKLaiv1bjLgyt5M9rAUUyn3LpmmacQHQzG4BreHWBfc14SXExW6caF9Kbpi9zH6xa 44PZDNgn2LBHjrcilIX7AC5YE0IzlKrlaV32ajg6fZx2SXbXzEJmMbjQU5lIpk5MWpbJsGYEwhoI Za8HgR7ZeBEWOZkEQyw6WQq6iAPwSXSs0c67rmv3ltdprKU3BK0i/6qQFMB0AIJDvT3lP5bBUpFE 8fQRNkDsCfbdU7C4aJKsLbs9vSzGFu6larxMQzThRQGRgKXoXmKKOXVYz7FpBzG6R4pwkL+O695v h85qeaTlld40rY6l6dLz7dyvtL2VRRaL8rXEQnnBg0hviknHrHhBj3SmOFJ8zTq4+Xwa+Y0O6SnC o2NK00ie802j+5V+r/SmeekYWEyej9Ntr/QWYWn2nk77MQysuDgDC4WBur0dOfJMmc+PzNJ5Z10C nD+NuGtUOyVAN0iRk5ima2Y/vUnrzWaQkwF+nFX+WGuAIATv1jMq3U5sWvyvKF+9epDnspmyN5JW +aMvXzZWUEaUppH8lIZrmMXBeg1pHej8fLbKR/WU5sEx/vJlwIlmkVFsi1kZLNiYLnyqsvFtbC1k muZrsa+ItGZ4aB2K5557rnA2gzDBB+Ur4oM02BX3m8rIMYJC2DUPWmyaYETR6y0KQMGLS+Ub91u3 bo1U8M35RoXgwopQR0j6zqmu17GVK2d/tk3n271dtaqi51Fc8HH2Ct8sAon0hU8vamHIdutXfnTi sF220qtzbHHi8DX5BGInpcjfLRvPit7KaHK6KKR08goGf3xRRJ/I1LkytthCqKRQH8HZ5p32maBE xZmi7las6At/dbsILcPrCUEXHABsNy+8glHhZXmw3/JmrUmvPieIg6B32nVOx3A6lU7nytjiUPIs 4F3/vMPCb9oY7jOLe4nyUkdworZNZRYWfpf9xRB0iH99bYZjqcAXvFjUEW2dAgzpmgzCI64ssKAT /uk3SK9wsLW0F2wFbhD1OeKP6f+Ez8J2UUl/h/6Y9FbDgt2UKdiuylzNXriflKZRLPQF1Y+Rz9bo tZ7OGegkA5UQbg0EMujDhw/PScXoLV+zQJgiTqODA5R3nnCIkGozFuLJJv9DF44pDhl/qdDoSSfH CTbwykYel8qGI9tOkdN7IsyyyIteIcBhLFOEgXUw8qJj9YI3+ev4Ldx8ZSIvRXWfT6OAQ3qcPKlP 6iz/wCJPJG875MMfje+Xv/zlGIzQKwmkV9CiyG4JouRtgWtaEa3BkeahoINewUnPpft5+4ULbJPj eZzwwStF+WvEK8E1AhR8baTolR8tZqqgYYpjKe6nnft0vz4Xy7L1g+tjsqLOxIMHoU0JHZp1a9fU z6oNKdatHQjqlsfF+/LlEL6102naoK5mFgNBD45lUbv68GEYjQr/FJirmVGJJ/MclahqJutV/avC KwU4lY8X+dNJOscEBAYHqcdyO8LoJIDWHwJBBBvyi08+mP7SCh1hBbeEs4wt7R1BERbLzDv+HGOh TtqrFSsqTk0ZGJQn7ejAwEDEkX8W8Ru7waG0aD/hf926dRFLnhdwcH/hXDXrsKiszWzhBSzSm15b sd3JGKSwuK9xcNFTFBABF8EJOLGoI2GR3jwvtL2kgTsLoa9BHanNT3Viv2AZHBxMD5eyj7MtPfnn APjAQhorXmSXuofTQnMMEd70XLv3advRQ+Cq6FUVsJCGNmi+Ar8uzsBCYaBuEILgAiOqOD04eWx5 lQFHCAdJK/tTYDmDBCdwlkiH080fD4T5OL21iMThogPDKDgOH9jQi36cMDlwvIaBCBfpSI8jDa56 TmItDEXn4IyygqWIM/Cko9hFebR6rFa94ZzOFwPlwsEnYJCWjX3+qPNaoo4B18MPjS51CB4eVq+/ /nrMg3yoQ2yH+qllOwRcZKNcg6Tlx1bJT/aIHYC1nYJNSQ/l0v3BopmqawWG0I39KZ0CU7JXcMm2 U5slHeWrdS/BJV/fYMunP/WAVVmZqcT1CsDpuG+bZ2BNCDAwun1r4lbo+D2e3k/nmFc7+sIsiH6j GSf9q/ujc3Tn9p1sMji0ErAwAkSHb2Dd3HV9lK6d2wqWlbGTpc91VvKvfKN+ZXBoOjEThzrSaLu+ DtHOctfLi47n2hA0JQijWVK6RqN0Q0OVwJaOl7XtDa/m0NF9EJyk/PTd8Ynx2BGm427xKhG2Sbnh JW0DKfvdu3cixoGBQZPgDDqZEcWzKI+FGYTcTyMjI/F+Im2ZAi+00zgVN2/enKWKYDIYqSPaeguh 3IhmUkonPOH0bty40ZQXnLX8zFs4oe3lmWsxKwMOVO58HXFfgYU60iCBOCtri70QaND6ANJDPwA8 PPutHH/xApZ0BgK2gs2Aw6qOqAPaX+6b1EkHC7xwD1nxoiAE91HKC7jgBU6ssMg+fOsMdIqBhuZa 6rOGchwBy03L8dQ5xMnCAXrvvfdmOXmMdDN1vt1Cg8sUfJytd999dyZ7nGTeuZfzh0PKdHVmAKTO J+lw9pRuJoM27KCPfHEccTYRcVY0Tb4NKudkUaveWsFAHcM35Wq2bMxowI4UFKAOeECDhy11RH1K sC/KUU9IQ4eEvLE38lX5mZGiPKmD/fv3x6BHvTybOU8QheCO9HAtAQjsS4Idkg48sldsGFsl2JB2 ejnOWiUE2PI2W209DunhWvQQsOBP6cmfYEl6z+qapbpNOyTNcoCTsGXrluz8+QvZ6OiVEPgaCR2c LNoBXO/atSs42/3NZjuv9HSwtm7Zkp08dTIbvXo127J5U3QgeR2Lzs6WcE6vj8xLQRMX4Rxw/50+ fTq7dPlSpi8CjY9NxHsULFa8AJs2inaBzub77/33WBJ+/8M//EMMhhw6dMjsnmDKLIGi8+fPZ3v3 7IkzI+7dvZeNXh4Ns2ZCcHCo/FFKCCC4sHnT5uCsjGcXL16MPND+Mjp34cKF+Oxqd3A+El/lv+Hh DcE2rofnwrlsX3hG9IXXi3AowcIsiZHhjWZBCJztq+Ee4mteOC/MfIAXfrNPO24l3Ed8lpznJf0J nEgcJ7BQX/S5uN8shOcKf9QJW7DgRIGFtlBBdissOLnYLlhw7OCFex0nnLbXihfqhfua2ZDYC/cN vHCPgwleaJ8thDrBZtDNFl4InNEWg4m+kRUv3CvwQp8DG6bdBwvYCAYwgEkaC2FmDFiw1StXrsRA IjzwhTie1fSJrHjhvt22bVvl+Ri4YZ9+CDzBy4EDB8yCMxbcuw5noBYDy4LxNzV3B0cLR66e486N TSeCRpGbrmwh0ssNXA+bNS7K3ShnZXJUFob55lvrOuoIoS7bZTvS1848i+pLdljP7oWn3n2Ejkbz LMKTHqODRkDjxRdfrHv/ptct5v0f/OAH2Y9+9KNYxJdffjnjL5Wp++PZ1L3KGirpce1PTU0Gh+Va dvXK1dj5ZTSXxfwGgzNJZ9SqYwMemnLs6lrAA47loZPFVHs66RqVEu6yt/CCg3s5dKwCCdFJYe2B wdBJ37hxQ/ithRHLRpJl77//fvZ2CMb1BBysMUCdpI+9L33pj7LPfe7Z8oFMayDocPbc2ThjBeeN OmJmyMjwSHS+zYAERYzU4lTiLOG8wQuOQQxaGTy307Li6OOgwA9rUTBbZVUYFSQo0B9eZWlGlq8O 67v0zH+GAAErHH+2OJA4Tzw7cGSoK8v7Ggw4T9QVz0PqSFgaeX40w1u9tDyLhCXabnBswYK9MHpr zQtBCIKscvLBhFNnPZKsYAyzIagj7AVMBAIITFjyAhbuaZ4F3NMI+sULHFmJgjEMEIEFXuCH5xF/ lrygG3uBFwn6aV+sn4/wQjCEwBW8gIP7mgDnfHh57bXXsldffTUW66WXXspeeeUVFdG3zkDXMhDs flnT4Vk6tI0ID0fLByQNWyPYrHHBVSO4GuG0lTRlYZhvvrWuK8Nuaulrhdf8tY3aYTN4Gs0zjyX9 TWCHWTlE/MvgN9W1lPYZUWb6NlPbceToTFBf6hRbcoFuAh8EwMCC4CCok2OLpfJOMA4bjgsCL/xZ y+c///mMv24RRvn3hFkQBOnpeFI/caHIzGZEO+WBqb+7d++OdTQZnH6+bILNYEvWQvCDEWx4kRPX KSyMnNJWEgDAYcBuwWLpwIl/sDA6Kixg4FgnsMADsxkJiFBHncQCB9guwSvaO7BgQ9xP1oJOsGC7 tHedxsKMB55L4gXb7cQzCV7AgnMtXjqFhTohEKOZKtiI7MW6vYMXOKEfiM2gH1468Xy0vldcnzOQ MtB0ECK92PedAWdgYTBAAIIHHa9puLSfAToQ/HWDdBMWOlXesZprFXRCrd6Hnqt99hGw8NcN0k28 4LR0Yu2Sonqg7e4WewFLN/FCMKIbBF5wavnrBnFeimuhm3jppmd1MVt+1BkolwG7eVnllsNzdwac gSoM8BoG0/5YY8JnQVQhyQ87A86AM+AMOAPOgDPgDDgDzoAJAx6EMKHZlTgDnWOAKX8egOgc/67Z GXAGnAFnwBlwBpwBZ8AZcAYeM+CvYzzmwvecgUXJQDPrTyxKArxQzoAz4Aw4A86AM+AMOAPOgDPQ NQz4TIiuqQoH4gw4A9YM8B6vJN3XsSwsPOniDDgDzoAz4Aw4A85AtzNQ2I/pdtCOb8ky4D3sJVv1 XnBnwBlg8TlJ0cN7WW9//MSk0vjWGXAGupuBZXz+tYXPc3Z36RydM+AMOAPVGUj7NNVT+RlnoDsY 8NcxuqMeHIUz4Ax0gAE+M8cnsqpKmAmxfPVwNnX3Rjb1qPKpyapp/YQz4Ax0lIFlvX3ZslVDHcXg yp0BZ8AZsGagZj/GGozrcwYaZMCDEA0S5cmcAWdg8TEwNTUVv3dPyfiOeaEsX5EtWz2SPX5xozCV H3QGnAFnwBlwBpwBZ8CUAQZT+EOq9mNMEbkyZ6AxBjwI0RhPnsoZcAYWIQNMXdR3w3t7vTlchFXs RXIGnAFnwBlwBhYtAz09PTP9GPZdnIGFwsDjF6IXCmLH6Qw4A85ACQwwK8LFGXAGnAFnwBlwBpyB hcKA910WSk05zjwDHoTIM+K/nQFnYMkw4A/vJVPVXlBnwBlwBpwBZ2BRM+B9mkVdvYuucD7/eNFV qRfIGXAGGmWgHQ9sFoS6du1aNjExkfFKx5o1a7KNGzdmnVil+tGjRxHL+Ph4pGDt2rXZhg0bIq5G OWlXOmG5efNmxhTR/v7+bHh4uCNYqOfr169nY2Nj2cOHD2fqaOXKle0qblP53LhxI+MPLKtXr86G hobitqlM2pAYXqgf/niXmFeTsF3qqhNC/YDl7t27kY/169d3jJdbt27Fe4n7G16EpegrOmVzxf0M L3fu3JnBwr1tLdgLGK5evZrdu3cv6+vrywYHB7N169ZlneCFOoIXttzLYOFe6oRgs/ACP/AyMDAQ /zrBC3UDltu3b8f2Fizw0olnElhoe3k+rlixItpKp7DQxulZDRfwwn3dCV7AQh3BC89H7udO9Rs6 cb+4TmcABnwmhNuBM+AMOAPzZAAH5fjx49nFixdjJwJnhf1PPvkkwwm3FDo1R48ezc6cORM75HRq Lly4kH366aexM2qJRbycPHkyOpE42leuXIm80Em3FByns2fPxnpi8S4cFZzdjz76KAYCLLGgnzo5 cuRItA+w0Anl9+joqCWUDF4uX74cdWM7OAbU28cffxzryhRMUAYWbBX7wDHAeeE+gi9LgRfqAh7A Qh0RLOI397a1EKzCPghEwAv44OncuXOmUNDLfQMP2CwOHMcOHz4csbBvKfACDziV2C7O/okTJ+Kf NRaCILQnBETEC20x7Z81FuyE+wZeeAYQnAHHqVOnZhYvtKoneMF2uW8IVBGEOH36dHxOWfPCvQwv YOF5RKAVTsBHu2cpBKywj/Pnz8/0G2jnOEZb7OIMLBUGfCbEUqlpL6cz4Ay0nQE6d3T6nn766Ti6 jgI6xO+//37sDO/bt89slIUOOI7+Zz/72dgRBgsO1O9///vYCQWjldC5wpF75plnIgb0MgvinXfe icGAp556yowXRpuop0OHDmWbN2+OFDDihNNAp+/ZZ5+NHXULbhgRJGi1Z8+ebNu2bdFxGhkZiZ1j MNJR10KpZePBYcJmduzYEf/Qx6wZODl27FjEwoiuhXAPoXfr1q2RGxxK7AUHAYzYMY6DheBg45yg E5tBsBecJrBQR1azEHBIcPoJPjzxxBMzsw0YucVecHjBYyE4TtgFztvBgwdnZjSBhcAndkudWQhB IbDg1NKuaSQbDARF4IU6s5AUy/79++MsCPTCk/jasmWLBZQY2MR2CeB96UtfmsULARv44n63EsqP DfMcULuGvfJMgh+wWMwUIfgLFmaG0N6jG2HWIvcXAb29e/da0RLbF4Jozz333Awv2Ovbb78d6ww7 suDFrMCuyBmowoDPhKhCjB92BpwBZ6AWA3Q+cfpxSOjMSOhs4TRcunQpBih0vMwtozx0anAiUwcJ JxLHAEecUUwLYQYIvOA04shJmHK6adOmmdcidLzMLZ1PRr6ok9QpwWkhCABvlqPbjHaBiWBI2skk EEFn3QoLGLBPMOR52b59exwZtBxpp9zYzc6dO2fxQlACjGCxGjnFduEndRyxF36zxcmzEnihnaFO UnvhPuLVL0aVrWZcEbQiEIHtpl8SwnYZbbfkhTri3uUeVgCCOoEX2jywwJuF6BUieEmDdjwDeC4Q oLEaaaeOCOhhLykvPBdoj7mPrLDoVRmePwpAUB9wAh4C1dShhRB8gBdsNcVCQISAFTPlrHjhHiIY TbubYuEeghdmhPEsd3EGlgIDHoRYCrXsZXQGnIG2M4AThwOgUZVUAZ0LHBkrx58OFp0oOsGpswIm dXTooFo4cgQ86GgVjVyrk04nzELAQR2gN+2Uo1v4GPm2cOToCIOF0cjUiQML+DhmxQsjpUzXRmce i3jBXiymBlNH6CrCwjE651ZYuGdx/KWXupFwH2FDOFcWDgu84GwTvNN9Iyzg4Bg2RV2WLfBCAA29 ak+kk/qBLxxKC6eSe5VZVmBBd15oj+GENrFsoT2FF0T3jXTKhsBCXVoIzyTqKm8v6OYYgRkLBzfl Jf98lA3RttD2WgjPJOwGDqo9Hy14oawEGSh7nhfOcW+B04oXdLo4A51kwIMQnWTfdTsDzsCCZICO njq5dDbzguOAWHUmFGBAb76TlWKxCELgpKFHelNuhA+8FqIAAx3fvHAMrnDkLJxt9GA3RfYiLOC1 GMHFWUQP5c/bi47haFs4lXCPriJ7AQvcgMPC8U95yduMsMCbxX0NLziveRyyY/jivEUdoUPBjnw9 wQvHuOfVJgpjGVucNNoYeCniBizUkYXjDyfVeKHs4IMXC3tBH/yjM19HwkL7A3dlC7ZLe4cUtXey FwssYCDAgJ1Ww0IaPbfYL0tS/mthsbKXssrp+ToDjTIwt2fW6JWezhlwBpyBJcoAnQk5rkUdPh0j jYXjL0eEjlZedAws4C5TyF+d8iIHQcdwKC14ERY4EA9p+TmGw8Jf2SJbEAd5fcJn4WzX4wWMOHtW vKCriBc40XELXqQj1ZvWE8exW6VLz7V7X3ZJ+WUb0pHiU13qXBlbysu9neot0mPh+MML91IRL2Cy tBdw6B6R3pQX1Rvpypa07S3SJXwWtgsWlVnPwhSTeLHContEeqthKfuZlNaR6iOPBYziLj3n+87A YmTAgxCLsVa9TM6AM9AQA2mnJN2vdzGdCf6QWtel6erl2cp5dYRrYSFN2Z0s8hcv1cojjDieZUs9 XsACjnqY24GzVnnFCXosOqC1sIABPI3UJWlblUa5V122qq/W9Y3yUi9dLR2NnqvHi2zGipdabYcl lpQX6U051TGrOhIe6e0UlrS8ncYCJ+Il5UP7wpdi1rl2b1Mc0pvq0DGw1LLx9Jr57tdrU4WllXta ecwXo1/nDFgy4EEIS7ZdlzPgDHQVA+kDO91vF0g6HWV3bMBaS4fKlXbG2lW+onyERXqrpbHCU6Q/ PdYNdZTi6RZewGGBpVEdFg5LWg9F+9g09tIo5qI8Gj2m+6he+m7gRRgtsDTKi0UdUe5G8FjzUgtT N/AifFa8SJ+2std0Cy+1zqdpW9mXjqLno861UkdF+baC1691BspkYO7LzGVq87ydAWfAGegiBnhg a9ShaHpkNahcp/TqOBSlJZ1Fp6Do/VLhET7wlo2F/DX9VnqFI92STvylx9u9L15qYQGHJZZ6ZRR/ 9dK1cr4RHc7LXIaxIyvbrWeTsmnZ+Fy07TuCvTTSdjRiV62ialRHPf5axcH16ACPniHV8mwUc7Xr GzneqA5LXurhbhRzvXxqnUdHLdvVfQQvtdLV0tHoOfJXmaW36FqlKTpXdIx8ZYMW9VuEwY85A/Nh wIMQ82HNr3EGnIFFwQAPbr2X2sz7zHQS1FEo6kzomDqpZZOlVeKlt0gfX2UoW+gMySkqwsIx/uBF 6crEVK/MYAGHRccNPfBTjRfxoLrU7zK29bhXHcnGy8CgPGvxQhrxVa8ulV8r25QX6U3z41hq4+m5 du93Gy+UG6nGC+csbBeb5H4FRy0saV2CrQwBSyO8WGGpdb+KKwss6XNPelP+dcwSC6+5SW+KRftg Kfs5kLYdRVh0rFle0n6M+jMql2+dgW5mwIMQ3Vw7js0ZcAZKZYCHvT6t1oyTQ2dFHYWiqZM6ZtEp hyBhl96UNE15tepk1cICPjpaVrzoU3XSK4cBfsDB8dSRSHlr974chKI6Agt/4BN/7daf5if+pTc9 xzFshjSy8fR8u/e5l/irxgvH4UWY260/zU/cw0Eej45Z1ZHamKI6ArM1L7p3qvECJos6Agd2WY0X bNeqjsCh+1rtLDxIxJXFfYRO+GcRRukVDrY6JhtPz7V7H05U5lq8WGChbPDCIIM4SMsrfBZYUruU 3hQL+LDrZrHAtfoxssc0X993BrqVAV8ToltrxnE5A85A6QzwwJek+zpWazs0NBRPF4086Ni6detq ZdG2c+vXr495MSKSL4emaQ4MDJQ+0gMI9NAREgdpIbXoovCm58rYp2NGB40OX54XYVm7dm38fnwZ +tM816xZE3mR3vQc+OiAgsWiE8n36OmYS2+KRfZCGnVs0/Pt3kcPwSLpTfMHH3/gsHBwhUV681iw IewJGy9bwILNFN3T2ArH4Y10ZQs6qAPpTfXJccLBUpuYnm/3PvfH4OBgLD+688L9hRMHd2ULNsk9 i+Tva7BhR/BiYS9goF2VXn6ngr0Q2LLAQh1V44V7CCyd4AW9edExeAFTmSL+0ZO3F/QKS7P9hvyz rcwyeN7OQDsZ8CBEO9n0vJwBZ2BBMdDKw3vjxo2xU5f/vjidQEZdrJwVCJcjzbfZ6fimou+1Wzn+ dObonKM3zy+fEqUjZuGswAHOCHgYHczzMj4+HmnCobFw/HEYqSdw6JNxqid4oQO6efNmHSp1Cy/Y Ax3hfLBIvHDeghfuE+qA+0a2qsKDjb8NGzbMjKzqXFnbTZs2RV7yr2fp3rLCAi/cJ9xDeV6wFfDB m0UQAq6Hh4cL6wgc2BFYLQJF2CS6cOQmJiZmmYFsiICJnOBZCUr4MTIyEnPVfSMV2C33OTis6kj3 LM+kVLAh2hjaIKu2V21ZHgvtH1j6+/tNgpzwIF7Qi41I4AX7tXwmgYX2N2+7YFIdNRuEUHnY5p+5 6Tnfdwa6jQEPQnRbjTgeZ8AZMGOglQc2neGtW7fGzsT169dnMN+8eTO7ceNGPGcxkoxiOpdbtmzJ xsbGMrCoXOAADx1COn0WAi/ooxN++fLlGZV09kZHRzOCN1a80Lncvn177HheuHBhJhABlosXL0YH QU7EDNASd3bs2BHr6uzZszOjXjiUYKN+4MZC4AXbxdE9d+7cDC9gOXPmTHSasCcr2bZtW+QF3WBA cBbAhrMJlrJHKVVW7IH7Cd3CguPEb/GmtGVvqSPuFexFgSt4OX36dFS9c+fOiKlsHORPEIKA3vnz 56OzxDHaGX6z3bVrF4dMhDrCUeO+4V5GxAt1hT1RVxYCDrihbZPDDR/8JnjEPc99ZiEEpXByadvE C3qpI35jLxaBRXTSnsHL1atX4zOJY+KF5xLtslVwhlkxYLly5Up8RoIFUZ1Z8kKZsV+CEOo3wIv6 DbLtCkL/3xlY3AzYtIyLm0MvnTPgDCxRBuhg4hwcO3ZspnNDp4vRUs5ZdfigH4eFjuaJEycyRuXo +NLJYiRuz549Zk6csBCEOHXqVOyI40DS4WIUDmfFkhd0Uhdy5HDq6ASD6cCBA2YOAryAhbrAXnBw wULgiE7ovn37TEaSwYHgPO3evTvay5EjR6KdgAV7PnTokCkWHBbq4pNPPom84Exhw9jzwYMHzZwV eMFhgZejR49mhw8fjo43WPjbu3evWQANLNzD4oU6wn5wbHFYsBerwCJYmOXAvfvpp59GbnB2Gbm9 du1aPN7K6C35NyO0H9xH1A82Q/AOu6W9o47AZiXUEVg+/PDDjDoCC3YrR9sqsEh587zwHGKWCs43 jrYlFvCojngm4XTzTMBeCFIz48gqsAgWbBf9tL3woIAI+wStLCXtN8ADATRsF14IzljyYllu1+UM 5BlYFm7Exy9F58/6b2fAGXAGFjEDP/jBD7If/ehHsYQvv/xyxt98hA6WRsFwYqymAhdhFRaadhwD HN1OdWpwmMBDJwscjKR2SnBScLJx/hmNwqHrFC90hnEkhQWn22rkNs8/TgpY2DIDACyWQaIUD3yA BX46jYURdRxJsOCAw4vViHbKCftggRdsmKncYGHbCeFehhewwAf3USexcE8TDAED7R120wmhvRUW 7h/auk5hofzUEcEQ2hWwWM06yHMPL3omgYU6sgye5fFQRzyXaPvFSyeeA/BCnwFu0N9Kv+G1117L Xn311VjUl156KXvllVfyxfbfzkDXMRDsfpnPhOi6anFAzoAzsNAYIOjQycBDylc3YSHwwF83CA4B U127QXBquwULzhtTlbtBcGqtR2urlRtHspuwMKrdDYIj2U1YCILw12nBkSQ4xF83SDdwAg/wQuCB v24QAg/8dVrgpZue1Z3mw/UvTQZsXppbmtx6qZ0BZ8AZcAacAWfAGXAGnAFnwBlwBpwBZyBhwIMQ CRm+6ww4A86AM+AMOAPOgDPgDDgDzoAz4Aw4A+Ux4EGI8rj1nJ0BZ8AZcAacAWfAGXAG/v/2zixU krPs4xWzzExm3/d9yTKZxBgXBAUFURFREa9yp6Ao3il4pfB9IHihF14JIgp6IQheCV64gARUEJOY GGeSzH5m3/dJYtTkm9/7+T95p1LdXadP1VPd0/8X+nSd6up6/vV7nvet53mrutsETMAETMAETCAj 4EmIDIYXTcAEJosAn8tUy5e1zs8mYAImYAImYAImMA4EnMeMg5esUQQ8CSESfjYBEzABEzABEzAB EzABEzCBMSTgSYgxdNoES/avY0yw833oJjDpBPimd37+jtbVzxJOug98/CZgAiZgAiZgAsMRYOJB eQw5jZsJjAsBT0KMi6es0wRMoHEC//73v9Pv3bNjfvfezQRMwARMwARMwATGhQATEMpf/vWvf42L bOs0gcKTEA4CEzCBiSVwzz33FPfff386fpbdTMAETMAETMAETGBcCHAXp/IY3wkxLl6zTgj4vh3H gQmYwMQSePPNN6ePPV+eXukFEzABEzABEzABExhRAs5dRtQxljWQgCchBiLyBiZgAncqAZ+871TP +rhMwARMwARMYLIIOKeZLH+P+9H6/uNx96D1m4AJDE2giRP2G2+8kT6P+eqrr6Yvt5wzZ07Bo4tv qeZ4XnvttfQAyrx58zrV8vrrrxc3b94suEV07ty5nWuBDZ+fhQt6uvARfuFzu6+88krSgg4eXdxG S7ygBS58P4pit6uPJkkLcQMT/NQVF+IEH8HlvvvuS3q65MJnvnngI9h0pQUejHX46t57700+6lIL TIhftMAFX3XR9P1CaEEDsctt8l2MMfRrYhc2aEBLl1yIFWKGOCF+0dIVF3TgI8YVPsLQFRd8hH/w E1rkoy64dNFfbNMEIOBJCMeBCZiACQxJgAmI48ePF2fPni2WLFmS9nLt2rVi1apVxcaNG0MTLbSc OnWqOHPmTLFgwYKUfF65cqVYuXJlsWHDhtCihQTr9OnTiQ1cSLK60oJTLl26VBw5ciQlwCSex44d S0nf5s2bU+EypPuHetvVq1eLAwcOJLsknsQLxRNa9LneoXY8xJuwffjw4eSfhQsXFjdu3EjL27Zt C9eC7f3796e4JX75n6KFfrRo0aIhjm74t2CbeKFPwYXChWV8tHjx4uF3PMQ7sU28MCkCB/5nedOm TcXSpUuH2OPwb8H2oUOH0sQMXCiimDCCSxdajh49miY5sU1hyWPLli3FsmXLQsdeimx8hH3iQxMj jLucCyIb/piamioYZxh7mRwhnokXzgWRRS5c0MLYjxb+p+hes2ZNsX79+kgsiQPjPucCjScwWrdu XbF27do07kUJov+ePHkynauJXcYWxuIutEQds+2YQBUBT0JUUfE6EzABE6hBgKSGCYidO3dOT0Jc v3692LdvX0r2SEKjkr4TJ04UPHbs2JGScK6AkWS9/PLLqWjZvn17jSNqZhMmQtCydevWYsWKFYkB Cfrzzz9fMEFBoRDFRQxIennAhSJh7969qXB44IEHwq6GoYVCe/ny5akoQAtFwksvvZT8tGfPnrDJ IgqTgwcPpqIJf+Ra0Pjwww+HcaEwgQGTD8QMVyfF5cUXXywef/zxMC3EKRMzTIDAQFwovik0H3nk kbCJK4oVjh8eu3btSpNo9B/8ho8ee+yxMC30Gbig6aGHHkqaKJ4oMtHz4IMPpgmbZkaQ/nvBLjaJ kd27dycGcGFCGB+xjkmSiIYGuBDDjCXEMPooMlnPHRpREzTYZWKGQluxARfGY/xELDP2RDWOn/Mh 5x4mIdB38eLF5Du4RE3QYJdJxQsXLqTYZRKC88/58+eTj1hmAiCqcW4s5w1MQnBOitYSdcy2YwJV BPydEFVUvM4ETMAEBhAguSORmD9/fkoyudrPgySUq2HcCUBiGtG4+kWChW2STAookhl08D9JKFcx IxpcSO5IeLnyBhO0cNVfWviIRlTj2NHC1Tdx4c4DJiS4QkfCHtXgQpFLwistSsbxz7lz56KkJB8R N6tXr75NC1woHIinqAYXjp+rtfooE1zQQtFLXyKuIhrxgBaujuY+wmcUnBQQUQ0t9BW0ELP0I/oT k5twYRI0iguTVvQX4kU+ol/xPy2SCzqIUfo044q4EC/wYDICPhGNMZ6+AgfGX7SIC9rQQtxENGIF NhT3ORe0oYmJiCgu3PUAF85BTECICxMyTBAxScPEVkTDzuXLl9PkPHp0TuJ8hJbIeCEWGOfJG8QF NuhCC/2I2HYzgUkg4EmISfCyj9EETKBxAiQ1FHHcQk8SoUayRwJI4kPiHtFIhHmghQRLDV1o4UoQ V+EjGgkUyXCZC7ZZRxLMVZ+Ihn9IyimaKCjzpiulcIko5NDCVUDig4Iyb/iI9cRUREMLBQKFPmzy pluV0RrBhXig2CZuiY+8wYW7APBhVPHEhAexUvYRRQMa0UJ/artxvBQr9GGK27zhN9jQ16IKOSbz OH445E0+Il4ifKSr6TyXteA31jHuRk26MoGGFsaT/DyAj4hdxjr6W0TDB9gqx4vGHGIlqsBlgh4u uutAxy8fcY6ImqRnjOfYy+cknathFsWFsQ57xGkeL/BhXWTeIJ/42QS6IvBWttqVAts1ARMwgTEj QHFG8kkj0Sw3ElAaBUtEo6BEEwleObGRFpJhksK2G0kWV3tkN7cnfSTLEY3EkoRPdnObrKORDLNN 243CiKuDVVxIhvEbkxARhRzJP3dkUFTyyJu4UMTBpu2GHbTIbm4PbbCRH/PX2lhGB2xkN7chHxEr EZNocOG4q2KXWGE92/Bou8EFLbKb21MMMf5ETKIxtmCH4y/HLrroX1GFHLGgyd1yv1YMwSUiXjh2 zknYrepLrIOd9LJ9W41xTue+shbFUBdcyj7i+KUvYjKa8y+xy3NV3iAtUfHSlv+9XxOoS+D27KPu u7ydCZiACUwwAQpFEi1aVSKsdRHFLRooEmiym/757x+tIzFvexKCxFLHTNFWbrmW8mtt/K8CjcSX R7mhBz+SnLfd4AIfMcjtSV/OL3+96WVigVbFhPX4Lo9x1rXVYI+tXlxY3wWXKj3SIn5tMWG/cOGh 2MhtsU5a1Pfz15te1tghu+X9K47U38qvN/k/Yxh9qYoLduQ3jUNN2i7vKx87+o13EVroI/hpEJeI 2JWP4NWPS4QWcUGLYoNlNa1DC9u22epowX8R8dLmcXrfJlCXgCch6pLydiZgAibwXwIkE7pirSQm h6OknGSs7cIfu5oQkd1ci5YjCm2OtZ8d6WM78ZO+Np6lRXbLNliPjggf6XirtLBO66W5rLXJ/3Mb spvvn3VRPoIL/alKR65JMZ6va3o5t1Glh3V532/afr6/QVykL/dl/v4ml+twwZ5ivEnb5X3lXMQg 30brIrVgX3ZzLVqO0EIcEJu9mvRFaMGG7Mhula6I2EVHv/Fd+qS3SmdT69ChY5bdqn1rm6rXBq3r t99B7/XrJhBNwJMQ0cRtzwRMYGQI5CfsfHmQwH7JXv5eko5+CVC+7WyW+9nQcQ1KUmdjP3+v2Mhu /hrLrI8q5PpxkS62kWata+O5jhbs5gVfGzrYZx0tbBORmNe1MZvEvC7HQXGg2B0FLTqmKC2D2KBn VGIXLXXjim2HbYPGDo2BUT7Scciu/s+fI7hgT/FSpUXrIrSgo58WsUFLnXFR2w/73E9LE1y0j2H1 +X0mEEng9m/qirRsWyZgAibQMQHuYlDiUXXbaC95nOjrnOzrbtfLTt31/bQr6WGbOprr2uy1ne4M kd1B2/V6vYn1+oxtPy0RTDgWcRl0XNI8aLvZvN4vXrRfuNTVrPcM81zXRt3thtGg9wziQhyNChfF 9CDNOrbZPNcdOyK01I2DiH6Nljp26mqejY9yG4qNqv3l21W93sQ6bMhOlRat0zZN2Oy1D2zIR7Jb tW2+XdXrTa3TMVdp0TrprWuT7ZXHaP913+vtTKBLAp6E6JK+bZuACXRKgKsf+hzzTD6fSrKtYlEn //xAlEyQEEQkBfrCLdnNtWhZevV/G88kQ9JStX/08YBJRMHSTwv60IKOCB+Jf5WP8nXaropfU+sG 2RCXQds1oUdf0JYzyPer9YN8mb9n2OX8eGW3al9RWuhPvXRofaQWWMhuFZecX9XrTaxTf0VHlRat i9CCDY0dspsfo9ZFa8k1aDlSC3GrY5Zd6cifI2JX8ZLbzZelD73obrMNOj/Kttjp/0HP5CDKYyLu Rhqkx6+bQF0CnoSoS8rbmYAJ3HEESFD0M28zOfGTTCj5rJqE0LqIxAanKJlTQpU7KlJLPrkgu7kW 6YN72wlfmQu2yzZZNyhJzfXPZlmxIAblfWm9fFl+vcn/FevYlN18//iObWDTdlPMVOnI9Wmyok09 g7gwaYlebdemlpxLLzbYj9CCDfWdKi3q6xE+GiUu0oIfxIBlNa2L8BE2sUOMVvlI6yK0jBoXxlQu Msgf8g/PWpfHeP56k8twEX/8VG5owU8zPQewX+UxLLuZwLgQcLSOi6es0wRMoHECSszYcb48yBAJ uX6LverzvlqnxGDQ/mb7eq6lfBxKdviN9IiikmPGTtUVGbigb8mSJbM95Frvl5aqxFxa5s6dW/lz abUMzGAj7PTiIn1soyR1Brue8abz5s1LWkh6lYRrJ9JCQYmetht2OGb1mdye9LHNnDlz8pdaWRZ/ 7KrfyJC45Am/XmvjmWPGT7Kb26APsR5uET7CBnqwW/aTfIS+RYsW5TJbWRZ/jr8cuxjklwXoZ7Br u+VxWf5FA1hJX9R5YPHixclHsCk31nHe4jzQdqOIFv9yvCh20RChBTucH6til9fECi2aaGN9G43Y lZ0yF+xJy0zjhWNzM4FxJOBJiHH0mjWbgAk0QmA2J+8VK1akpEW3QeaC+Nk8Eo6IpBy7FPUUJFVX e/QTfsuWLcsltrYsLXAp80Ufid7SpUtbs5/vmGSYhI4JkbKWmzdvpk1J3CMKf5JPijmSz3ICSgHD upUrV7aeCHPQcOG4sanEV9wUz8QuRVbbDS3YQke5kEMf64jdCB/RZ7GF3fLHs9S34BZR+MNl4cKF qZAtc6G4RQ/cVOy16Sdxwa7GE9mjb+E74ptH240JBsYP+nNZi9bBJGLshQt9lqZ+o+OHCz5Cy0yL Su1jps8a48vxIi7EVNTYu3z58iT/lVdeue0wpIWxRZPnt23Qwj86V8MF+2os46PIcxL8ieEyFzQR zxqbpXGmz/nxzfS93t4Eogl4EiKauO2ZgAncEQQoREh0r1+/flvBQqJz7dq1VORFJVnScvXq1dsS c7SwjuIgSgvJ0QLJMgAALzBJREFUJYUaCZUKfRxOcnTp0qWUlEdpoXAlGUYLHNQopi5evJgKWyXu eq2tZxLd1atXp4QX23mDC02Je/5aG8twITGneLx8+fK0CXx07ty5lCSjNaLBRZMv2M6T6AsXLqQJ ARUREXpUVMon2EQTPiNu9HrbWuCyZs2aNJlZ5oIW+jZaIiZnOFZ8QPEkDjp+4odCDq0RDS70E4q1 8+fP3xYv0oLWKC5MujL+Ei9MXqkx9jExEamFcZex/uzZs7dx4RyFFrhF3FEEA3GRT8RFYzHjbsQE GnY1QYaP8jv0YAIbtEZx4dzHgxyBfqOGLs5Rel3r/WwCdzKBu//nVruTD9DHZgImYAK9CDz99NPF s88+m15+4oknCh4zaUxCkGRRLJEUc3VjamoqFQ4PPvhgyJVk9JKYk0iRxJBocYWOZO/IkSPptZ07 d4YlWeghkSLBO336dGLA8rFjx1KSvmvXrrDkEy34CBanTp3i31R4nzlzJiWB27ZtC7syiG2uiOKr EydOpGcmAdBCYbd58+awSQi04COKarRQZPOgqKOA2bp1a6gWihGKRrSokKPw5rFhw4awAhcuaMFH J0+eTHzgQv/mfwrtdevWpdfZtu2m2/2JXWIYXVeuXEmc0AKbqIYWim04ELdoYew7evRo4rJx48Yw Loy16CFWGXOZHIELWlatWpX6EvoiGra5YwUfMfFAHEsLYzJjDONxRMM2BTf9GC3Ypcg+fPhwOj9s 2bIlsYrQAn+40Hc4L+EzCm/Oj/jugQceCNOCjzgPMM7CA23EzcGDBxOv7du3h01aYZuxlxhBD1zw FedqluHC80za3r17iz//+c/pLXv27Cne//73z+Tt3tYEOiHwv7faXbdOsG/dm9SJDBs1ARMwgW4I /PCHPyx+9KMfJeNf+tKXCh4zbRTYJFo3btxIhRwJBgUCiVZ048oKCShaKDDRwpU4EtPohha4kHhy mqG4g0vU1a/8eLnKxOQMiR8FFEne2rVrpycF8m3bXsYvcEELurgCR+FEwh5VOOkY0aJJNIp/ihju gOCKahda8BF64ELhABcKuS604B/8BBe0cNcBt1J3pQUuTERIC1xYjmzEC/2Z4mkUtFC8MVHFWEPB jX90x0YkF2yJC+cDtNCHiN9oHzHWwoUJGl1p566DyDsycvaci4gXNNGIW7REnx/hQszq/IgWfISW qLsgsKlGnBC7+lgGE9TD5g2/+MUviu9+97tp108++WTxta99TWb8bAIjS+DWufQu/zrGyLrHwkzA BMaBAEU1VwFHoZFMRV4d7XfMaFm/fn169Nsu4jUmHSiuoz5i0O+YKFAoTnh03dDCLdpRHwPpd7xo oSDg0XVDC4Vb1Ed1+h3vqGmhiOTRdYMLE3c8RqFxpZ1H141JMiafeYxCGxUtcOFcvWnTplHAkrRw 95ubCUwygZh7xCaZsI/dBEzABEzABEzABEzABEzABEzABEwgEfAkhAPBBEzABEzABEzABEzABEzA BEzABEwghIAnIUIw24gJmIAJmIAJmIAJmIAJmIAJmIAJmIAnIRwDJmACJmACJmACJmACJmACJjDG BPjuCzcTGBcCnoQYF09ZpwmYQOME8hN2vty4Ie/QBEzABEzABEzABEzABEwgEfCvYzgQTMAEJpYA PwWon8hi2c0ETMAETMAETMAExoUAPyOsPIafoHYzgXEh4EmIcfGUdZqACTRO4Pr168WlS5fSfi9f vtz4/r1DEzABEzABEzABE2iLwKuvvjqdx1y5cqUtM96vCTROwJMQjSP1Dk3ABMaFwLJly4p169Yl uStXrhwX2dZpAiZgAiZgAiZgAsXChQun8xhyGjcTGBcCnoQYF09ZpwmYQOME+B6Id7zDX43TOFjv 0ARMwARMwARMIISA8hh/t1UIbhtpiICz74ZAejcmYALjR+DNN98cP9FWbAImYAImYAImYAIlAs5p SkD870gT8J0QI+0eizMBE2iTQBMn7Ndffz19HvPGjRvFPffcU8yfP79Yvnx5J3dY8KVUfMcF33VB W7BgQcHtmeiKbtJy9erV4u677y7mzZtXrFixohMt+Jnv/Lh27VrBl3jJR/fdd180lmSPz+3yQMv9 999fLFmyJD1Hi4EL/uHBF7POnTs3xS6+6qLhH7S89tpricfSpUs743Lz5s3Ul+jfcJGWLq400p/h wme/pYW+Hd2IFzRcvHix+Oc//1nMmTOnWLx4cbodvAsu+AguPNOX0UJf6qIRs3CBD1wWLVqUHl1w wTdo4csKGfvRAhddLY/kgxbGXs6P9957b4qVrrQwxnF+RAss4EK/7oILWvARWjg/0p+7yhsi48G2 TCAn4DshchpeNgETMIEZECDhPHDgQHH+/PmU0FC8kVi89NJLqZCawa5mvSnF0v79+4uTJ0+mgp/P iV64cCGt06TErI3U3AFa4HL8+PGUdFL0Uyy8+OKLKemquZtGNqNwOnr0aHqQbJIAU7TgI3zF61Ht jTfeKE6cOFEcPnw4Jb5ooXjh/zNnzkTJSHY47lOnThUHDx5MDEjGmTh6+eWXi7Nnz4ZyQdDp06eT bSZmmDgjSUcbvCIbXPAF8QEPfITf4AKvyHjhuBlb6EvEiT7vDZdjx45FYknHTTGpsU2FJLFLP4dR ZKOYRAtjG1zo2zA5cuRImtyL1MLk2d69e9PYJi5TU1PFoUOHwrkwzu7bty9xYVKGCRH8gx76VGSD C/2G8xBamISgn3Oeop9HNsZ8zj/0Jwp+Jn/Ror4VqYW8AS7nzp2bzhvQBRf6uZsJTAqB+Mtjk0LW x2kCJnDHEyC542rTww8/nK6uc8AkW88991wqnrZt2xZ2lYVCm8R8z549KbFBC1d6XnjhhZSEojGq UayhZffu3YkHdrnK88wzz6Rk+KGHHgrjwkQDheyuXbuK1atXJwQULRQNFCxM1kTdEUERRzGwZcuW 9EViXCXl7hCKKWKJAoYr3RGNYgWbGzZsSA9sMhHBlUtiCS0UMBGNQhJfrF27NrGBC/FCgUBhib8o GiIaVyaJF/rx1q1bk0kKbCYk8B3ro+5C0EQMftm+fXshLjyjhfXEb0SjOCIuiE+46O4qJmWII/xD LEc0CljihaJ2586daSwhXohXCjnGPf6PaGiBC1q23OrX6jOMKUxCECtr1qyJkJJiFF8QN+KCYTRR 9KKR/h7V4IIWzgMa14hXzklMAKCFWG670X+JF4r/Rx99NE3SY5OYZWICLerrbWth/4xpjHmPP/74 NBfGuKeffjr1K/X1CC22YQJdEvCdEF3St20TMIGxJUBypTsguNKvptuluZrKlaCIRnJFgUvinRdI JJ/86geFOEVnRCMp58oXTCjY1LjldNWqVdMfi9D6Np9JPkkwSTbzooSrpuvXr09JKa9HNIo17lJB E0VJnnzDhXiKuhsCDdhSUavjhwuFAXeyUMxENXxAkb9x48bbuDApgSYmBaLuQODqpHyk40eDtFBY RTUm8+hPZS7ECwUlExG8HtGYVGQigtjVBAR20cI4E8mFcZcxjz6Mb9SkBS7EcETjY1WMrXDRBAR2 mSBiLKbgjNTCOYc+nHOhwGU8ZvyJutLOZB5cOP9oAgIu6GAsJraZwI9o3AUBF+Ij18KECOcoxpco Lhwz52om7HItTFrBhfHHP7MZERW2MQoE3hq9R0GNNZiACZjAmBBQsZInnpJOckHRFPUxCOxQxJLI 5MUteqSPJCyikFOxUvW9AtJCEhbRSCxhQ8GWJ+XYlj4SVArgthtFE4k5WpiQyRt+o7CL4qLPaWMz LyjRJC6KqVxnG8v4iNis0sI6ePE68d12Y/KBfo1dfJI3YpcYooiIKCrhQl8iVmCQN3Sgj5iK0CIu 2FUflh7FM3rR03ajrzLJiZYyF2wTv8Q3/brtxnjKR5do6jeyqdjFP7CJaMQumsqxi23WMWFFX2q7 oUETqnmhjV3FEP05wkfYpB8RwzDod35k27Ybscuxl7lgl75FfHOecDOBSSDgSYhJ8LKP0QRMoFEC JDSaYCDZLDcVmVHJBFecSPywW06yci1s03YjsZSWsi3pi7orA/4kdSS+5cY6WFFURhS42CFuquJF WtAbcWWbYhE7HH85XrSO4imiqIQ9thSnuZ/QApuoYjvnUo4ZaYFbRL+GC8VrWYf4wCuq8IcLhT2t 7Ce4sI4+rzFRGtt4pj8zxsClig1a8FFE4Q+TXlw4dvTBJSJesAd/bJZ9JC2MPxGFP7HLeEerGu8U LxFa0MCdBcRpLy1so/MWy221nH8/LVHx0tZxer8mUJfA2zOzuu/0diZgAiYwoQRIJlS4ViV8Wsc2 EYW/ikUSrXLTOrSgu83G/pWUVxUIWkfRGcFFWmAgDvnxs46ChUfbTbEgBmV70gebttsgLmik2Ivi 0muiCCbiFcFFNnK7uS9YT9xqu/y1ppcVlxy/YkM2cn3ypV5r45njpW/ndqvsRBT+cKEvVXFBU2S8 oEN9RHZzLvIb27Xd8rG3ypb0RcQuWnTMOhfmmsQlSov6iOz20tL2OSn3kfxR1oJGsctf87IJ3IkE PAlxJ3rVx2QCJlCLQJ6U5MuD3kwywYPW7335doP2OZvXlQj308I2bSdZ7F9ceh0PGtmOwrPtNogL WtAxSHMTOvsdb+63iAS0nxaOFT1RsVuXvXzZhC967aMOl6jYHcRFMRPFpd/YIS0RsZtzkd3cn1oX wSXvI7LblZY8dvtpybfLtTa5nHOp2q/0RWmRBtnV/zxrHVr6xXj+nmGX2b/iV3bzfWndbPqR9pHv 18smMKoEPAkxqp6xLhMwgdYJ5CfsfLkpwyQdbSc2aO1nQ8el5KepY+u1H2mR3V7bRenpZV/rR8FH 0sLzKHGJ0FLXRkTBkvuhapmYJl7qaq7aR9116keDth8FLtI4Slzq8pP2YZ/r2Inm0k9ThBZY9tOg 1yJiF1uyp+cqX8Ol3+tV7xlmXT8bek3Pw+x/0Hl3mH36PSbQFoG3f5i5LUverwmYgAmMGAFO2Fwx 46Q/k5M32+p2yn4JA9vNZL/D4qn6fKn2JX3obVsL+9ftt7IrHfkz24lfvr7pZXHppwUdkVoGHaP4 DdpuNq/XsWEubydMHEXF7qCYVEwrxt+utrk1xEudsaNOXM1WVV0bg/jNVgfvxwZ6Bt11UVfzbDTV tRHJZdDx1NU8aD/9XsdGv9hVP4JLv+362aj7GvvXMctu1Xu1TdVrVevYr/KYCP9WafA6ExiGgCch hqHm95iACdwRBPhMqj6Xqu9VqHNgJAlKFKqSCa1Tklpnn7PZRt+GLrtV+6r6Jvmq7WazjmRIRVGV FtbxgIu2m429Qe8ddMxoQUdE4oYd+PTiomORL/V/G8+D2MtHivE2NGif/biwjXgN8qX2N5vnnIvs 5vtjXR7j+WtNL48aF46b1osLr0XELjFJf0VHPy25L9HWRkNLHS5RWvr1V7GK0JKf92Q35691kVr4 iIPs5lq0jJa2zwP52FGlRetmyoU7SpTHRHwvi5j52QRmS8Afx5gtQb/fBExgbAlQ2Nx///3pMZMT P8mKiqKqW0p1y2tEUg582am6pVT60NsvSW3CiSRZaOFZDPL9Sl/5Z/7ybZpc1s+gyW6+b9bxgEnb XLCLjV5cSD7RIn65zjaWFS/YVeIrO/xPzOQxrtfaeFbBwvGXW84lImZyLmU9uRZtV9bb5P9wwQ52 eZRbdLyoj/Tigr4IH6mPyB9lLsSutim/1vT/nDN03qi6G0KsIuKFY4M/NmU3P16ti9BCrIzq+TFn wrLOj3AhbtpseVzKbm4PHxHXw/hIecww7801eNkEIgl4EiKStm2ZgAmMFIH8hF2VRPYTu3jx4vRy 1ZdI6arEwoUL++2isdeWLFmS9lV1DFqHlraTLERghwJKDPKDFCvpzV9rY3nevHmpSCDhKxdy4rJg wYLphLkNDdrn/PnzExcx0HqepQ8tsGu7MTlD7GNXxYlsigvbwK/tJi2ym9tDHw8SbBU1+etNL1PE 8ZDdfP+sI4YoOiP6tfjDpRy7+Iz1aGW7tpvsyG5uT4UTY4vGxPz1ppcpcOFfxQVb9C9ihZhpu9GH 6Ne0cvzChZiBC/06ojGuym7ZHvoYWyJiFx+JS3m8I5bRApcILXBYunRp4lL2Ea9pXYQW8efYy1xy LTONFx0D+9CkGMtuJjDqBNrPdkadgPWZgAlMLIG8yKpKCvqBWb58eUrqyr8vThLILZEkA4sWLeq3 i8ZeI2mhUOC32Ul886bfaycRi2gcM8k5dsvFEx95IRGLmoSgGEEPP9FW5nL9+vWEg8JJV3nb5IN/ 8BM69JNxsgcXEsnVq1drVavPcCEeiPnyZJG48HoEF/oJPqDfKFZ18GjjsWzZsrDketWqVYlL+bZm 9a0oLXChn9CHylyIFfTBLWISAn+sWLGi0kfoII7Qmk/qyodNPxOT2KKQu3Hjxm27VwwxATHTQu62 Hc3gn5UrV6at1W/0VuKWfo6OKB+pz3JOyhsxxBjDGBQ19mosK2th/EML596IiSI4iAt2iRE1uBC/ kecktDD+lmMXTfLRTCdE8nNbntPoOP1sAqNKwJMQo+oZ6zIBE2idQJ4c5ifyOoZJhteuXZuSicuX L0+/5erVq8WVK1fSa1FJFsnlmjVrimvXrhVoUfGPDvSQEEYlJ3DBHkn4uXPnprmQ7J0/f75g8iaK C8nl+vXrU+J5+vTp6YkItJw5cyYVCCoipoW2uLBhw4ZUCJw4cWL6ChwFJdrwD2wiGlyIXQrdkydP TnNBy/Hjx1PRRDxFtXXr1iUu2EYDjWIBbRSbaOE5ohEP9CdsSwtjA/+LW4QObOAj+grxookruBw7 dixJ2LhxY9IUoYdJCCb0Tp06lYolbDLO8D/PmzZtipCRbOAjCjX6DX2ZJi74injCVxENHbBhbFPB DQ/+Z/KIPk8/i2hMSlHkMraJC3bxEf8TLxETi9hkPIPLxYsX0zmJdeLCeYlxOT//8npbjbsy0HLh woV0jpQd+SySC8dM/DIJobwBLsobFNvSWOc5z10iJgLraPI2JlCHQMzIWEeJtzEBEzCBYAJckSBZ JYFVwTETCSSYJABTU1MFV8JIJngmkeC1qIQPzRQsXJGkYCHB4bhIbEhMN2/eHFbESQtMVbBQQDJB QpFNsRLJhSt/HD9a4EOBSXHAM1qiCgS4oGXr1q2piCTeSNTRggbiJTKBpHjatm1bmnQ4dOhQKnZV RO3cuTNUCxx27NhRHD58uNi/f38qMCma6Fu7du0KK1bwEQULXI4cOVIcPHgw/Y+PmATAd1ETaGgh LvAF/kEP2ohh2Gzfvj1sYhEtxCb9BR08iB8mGhlriF3+j2qMH/ji6NGjyUeMcfSnS5cuJd9RiEc1 fLRly5YUt8QLd8qIC7yiJhY5XrighX6EFiaN4MJ5gDEwUgt60IKPOCdxXuScgBYmibjjKGpiES34 Avvo0d0gTIZwrkZPZFPewGQi50UasYsOJmdmyiWfhOC85mYC40LgrltJ89u/8Whc1FunCZiACcyC wO9///viW9/6VkrsP/3pT6flYXZHIq4CjkIh6lbgKq3SwtBOYUDRNNOkpmq/w6yjeEMPyR86SIq7 ahSRJHwk5VyNIhHtigtFCsm4tFBERV25LfOnqEWLJmjQEjlJlOtRwQQfkukutZDYU6SghQIcLZET VjkXtOAjYpiJU7Tw3EWjL8MFLfCgH3WphT7NbexoYLzrqghjvJUW+g9jXVdaiAt8xGQV4wpaou46 KMckXHROQgs+YtKxq4aPOC8x9otLF+cBuJAzwAb7s8kbfvCDHxQ/+clPElLyGXIZNxMYdQK34v4u 3wkx6l6yPhMwgdYIkESTMFKAkVSTYA9TDDLp0OXEQw5olLQw8RB55TjnUF6mIOCq1yg0itpR0ULx xq3Ko9Doj9FXa3sdN+PCKGnhCvsoNMbHUdLCJIiubHfJh0KSySEeo9BGgQkc4MLEA49RaEw88Oi6 waWpczWTcGqjkodIj59NoB+BmA/N9VPg10zABEygIwIUprqKx1UjJiHcTMAETMAETMAETGAcCOgu TLSOyuTTOHCzxu4JeBKiex9YgQmYQEcEuAVSt8zqM+gdSbFZEzABEzABEzABE5gRgfxOiFG5E2dG B+CNJ5aAJyEm1vU+cBMwAW7L1OdTPQnheDABEzABEzABExgnAvmdEKPyUalx4met3RHwJER37G3Z BEygYwJ8X4HuhNB3QnQsyeZNwARMwARMwARMoBYBTULwPROj8H0XtUR7IxO4RcCTEA4DEzCBiSWQ 3wnBt1Tz7fxuJmACJmACJmACJjDqBPiVjcuXLyeZ5DNd/YLPqHOyvtEk4EmI0fSLVZmACQQQ4GfL 9K3d/HQXv5LhZgImYAImYAImYAKjToAJCH0nxIYNG0ZdrvWZwG0EPAlxGw7/YwImMGkE1qxZkw6Z Wxo9CTFp3vfxmoAJmIAJmMB4Ejh//nz6eXHUexJiPH04yao9CTHJ3vexm4AJFFu3bk0U+CgGd0O4 mYAJmIAJmIAJmMCoE7h48WLxn//8J8nUBZVR12x9JiACnoQQCT+bgAlMJIHdu3dPH/fx48enl71g AiZgAiZgAiZgAqNK4Ny5c8Xrr79e3HfffcWmTZtGVaZ1mUAlAU9CVGLxShMwgUkhsHr16unvhThx 4sSkHLaP0wRMwARMwARMYIwJXLhwIX2hNpMQ69atG+MjsfRJJHDPJB60j9kETMAERIAvp+Q2xuvX rxeHDh3S6trPb7zxRvpMJr+ucffddxf62U9+Liu68U3Zr732WvHKK68k0/Pnz08/QdqVFq7QwPUd 73hHMW/evALWXWqBCx+7wUc8utCCY/juEf0aCzpgA6PoRrygBS48y0ddfcO6tPBzuWghfrviwi3O cEELP+OLn7rkQr/mC+joQ11qof+IC4UPPuqKC1o03qEFLjx30aQFH4kL54Muxhj6Nd9xBBs04KMu uXAeQA9xIh91xQX/EL+MKwsWLOiMi87Vs8kbiDm+E4J94d+NGzd2Efq2aQJDE/AkxNDo/EYTMIE7 gQDFzqpVq4oDBw4Up06dmi466hwbExB8hOPs2bPFkiVL0lv4Xgn2R0IQmWihBf1nzpxJyRXJ5+HD h4uVK1emL6yKLBRIik6fPp3YwIWE78qVK51owSmXLl0qjhw5Ml1Mnjx5MhW5mzdvTkVdHV83tc3V q1dTrFFMEnvcfcMyWkjQIxuxSozgH34lBp+xvG3btnAtJOP79+9PRRPFAVqIWfoRPz0X2dBCvNCn 4ELhwjI+Wrx4caSUZJuxiUkROHD7Ncvcer106dJwLUzUUvzAhQkaCky4dKHl6NGjqbjFNgU3jy1b thTLli0LHXuZPMNH2Cc+4MIyXxTIuSCy4Y+pqamCcYaxF18Rz8QL54LIcxJc0MLYjxZNMjLpv379 +kgsicOxY8fSuUDjCbHM3QNr165N416UIPov5yDO1cQuYwtj8Uy14GvuhKAxZuJfNxMYJwKehBgn b1mrCZhA4wS4gsCXU/7pT39KyRJJft0rChQqXIl4+OGHpwslrvg8//zz6eoEiV9U0keCxYTII488 Ml0QoGXv3r2pUNi1a1fj7HrtkAQLLdhcvnx52owE9Omnn06J6I4dO8K4kIzv27cvFSfyK0ng3//+ 9+Kll15Kvou6SigtJJuKDSZs/vGPfySN73znO8OuKlOYvPjiiylx3b59e/IRyTDreOzZsyfsKiFX JolTikcmQJhAgwt+g8273/3uMC1MOFBQMjFEv6bBhYKFeHn00UfT5FF6oeU/xOkLL7yQJoTQoolE Jo7w0bve9a6wSTQKa7jQGGOkhYkAJo8eeuih6TGwZSzJH/iC9thjjxX33ntvWmbckY/008vphRb/ UOTDhbjBR0ws0pgQPnjwYIpljYEtyki7VpwyzjzxxBPTXDinETO0yEkRuHAOevDBB6c/8siEMP2a +OGjkBENLvgC28QLd4bQ+J/Y5XWdGyL0MDHDhQv6keKUMfDZZ5+dkRbeoz7JOZVx080ExolA/P2f 40THWk3ABCaCgH4hg+SNK7B1GkUSVyG4AqFEgvdRvHDVhySDJCGicUWEhIqrcLrKg12urFPYMVFC khzR4IIWCntsq1EorFixIjEjMY1qXG1CS558k6xx9YuPiqA1quEHihaSb01O8Yw2fEixENV0G2/O hbsg4ELc6gpbhB60cPxcHVUiDRe00OiTxFVEu3z5ctKSf9M8XPAZGih0oxqxSfEPBxX92JY2Jh6j uDBpRX8hXnItXH2FTyQXrqwznsFBExBwwUdoYQKUCZyIRl8hZuDC2K/GxAPFLlzo8xGNcVU+yrmg hTGQu66iuDDpDBfOhZwj1ThXcp7SXYda3+YzfYg7DeCQ322GDvSgJYoLscAvWnCezrmgCz2Mdfiw TuOYyDMYK5nAdjOBcSPgSYhx85j1moAJNE6Aq5s0Tuok9nUaiQSFE4mmCkreRxHF1TAlPnX2Ndtt KBBIhtFCEq6GLpIbrvSQuEc0JnJIhstcsM06kr0oLZqcoTjICye0aOIIvRGFHPFAzBAfebGCFuKF 9VETInCh8Kcw4fsO8kYiTGMSgrhpu1GscNwcf14gYJf/0UcxE1UkUJBQwOmKto6fgoG+FaUFLhQY 2MyLFfSgDzYUK8RVRINLlRY4EUfEdkSxTUxqAo3xJG/0cVjBJWqik0lONFFU5ucB9S3OKfS3iEaf xZbGNtnMz0mMdxGNYrqKi3yEf6J8RJ+ln5TPSRpzYBbFJfdRHi/4hNhFJ+fzOu3ll19O4yKxx51r biYwbgTeylbHTbn1moAJmEBDBLgTgtvAKUa5/XlQYzuSblp+xUnv0zoS0IhGEYcmkqpyYqPim8Q8 oqhkgoFiRHbz45c+ksKIxjGjRXZzm9JHIhxRJGCHwlKxkWuRPthFFNtMWJHsEiv5pBWa0ELjM+1s 13ZDB7ZkN7cnfSTlET5CB1fYZTfXAifW40Piqu2mYkSxkdtDB+vRGnGHE1yIX9nNtcCFB+NPxOQi /ZmCkeMvxy666NfESkTsYkcFo8YTsVEMwSXqPECBC5OqvsQ62EXELn1Ex1zFBS1widCCPzhXw6Ws hdfECi1oarNx/iV2ee6nRewGaWESgsZHMaK/Y2OQNr9uAnUIeBKiDiVvYwImcMcT+NjHPpYSFX2f Q78DplAk0aJVJcJaF1E4oYEigSa76Z///tE6ipq2JyHYv45ZyV2VFunNX2tjWQUaBQGPcoMNfiQ5 b7vBhSRX/sjtSR+vi1/+etPL4l/FBFsqWBTjTdvP9wd7+lMvLqzvgkuVHmmhL7Xd4MJDsZHbY520 yJf5600vY4O+Lbvl/SuO1N/Krzf5P7FCH6nigh35LaIf5WNHv/EuQgt9hLgcxCUiduUj/NGPS4QW uKiPKDbQpaZ1aGHbNht9SMfciwv+qxMvTFTo17y4CyL6i2Hb5OR9Tw4BT0JMjq99pCZgAn0IcCLn 6gS3HQ+6LZ5khUSLpiQm37WScpIOHm03FYuyW2UvotCGSz870pfzq9La1Dppkd3yflmPHyN8pHip 0sI6rZfmstYm/5cW9im7+f5ZF+UjtGCrSkeuSTGer2t6OWdfpWeUuEhfrrlpHtpfbkN29RrPWpfH Vf56k8v0VcWL7Ob717oILYpd7MturkXLEVrwEVx6NemL0IKPZEd2q3TlcVX1ehPrch9V7U/6pLdq m6bW4Z86x1xnG+4m1PdXkbvoOJrS6v2YQAQBT0JEULYNEzCBkSewe/fu9MWJJFDPPPNMX739kr38 jeyLR9utnw0lJ4OS1KY0io3slvfLeraJSPr6cZGuUfCRtPAcUWzX5RLho7o26iTmOcdhlhW7vd6r 2I3QUsdH6IzQApdBbNAyKrGLlrpxxbbDNnzUj4vGwCgf6ThkV//nzxFcsFeHS4SWPHYHcanb53Ke M10WlyotWleHC99Fwpee8mWb/DKNmwmMIwFPQoyj16zZBEygcQJ80Zu+3Imfb+x3SyTJghKGfkLq btdvH3Veq7q1U+9T0sM2dTTrfcM+684Q2e21H23X6/Um1utzt/20RPmo7vFKcxPH32sf/eJF74FL Xc16zzDPdW3U3W4YDXrPIC7EURSXOlrQPWg7HdtsnrFRZ+yI0FI3DuronQ0T3ouWOnbqap6NntxG v/Eu32429vq9Fxv97Ehfv2367X8mr2FDPpLdqvfn21W93tQ6HXOVFq2T3n42n3vuuTTpx0+O8gsk biYwjgQ8CTGOXrNmEzCBxglw4ud7IfjiQO6E0BeOVRki2VaxWHX1RMkECYeSjqr9NLVOX3You1X7 ld6q15paB0Npqdon+njAJKJg6acFfWhBR9c+yv0W4adBNsRl0HZVPp7pOn5FgJYzyPeh9YN8mb9n 2OX8eGW3al9RWuhPvXRofaQWWMhuV1zUX9FRpUXrcl9WaW1iHTY0dshuvl+ti9aSa9BypBbiVscs u9KRP0fEruIlt5svSx960d1mG3R+lO1BXPjS1d/97nfp3PXe9743/TqN3utnExgnAp6EGCdvWasJ mECrBLZv315s3ry5OHjwYLFv376etkgmlHxWTUJoXURig0glLUqocuGRWvLJBdnNtUgfiWHbCV+Z i2yX9QxKUvPtZ7OsY67SwX61Xr6cja1B780LBNnN34Pv8A+a226KmSodrNN6TVa0qWcQF26TRq+2 a1NLzkUMcntaF6EFG+qvsptrUV+PiN1R4iItsBCDKi4RPsIudvBPlY+0LkLLqHFRXA7ykWI892GT y3AR/6qPXKAPP2mbXrb5VQx+vnfVqlXFBz/4wV6beb0JjDwBT0KMvIss0ARMIIoAP3P18Y9/PCUC v/71r3t+rphkhd/mplV93lefjeZ3vyNarkXJpuxKH1oiikrZEQPp4Bkt6Iv6Jm9pIeErc0Ef6+bN mxdyJQk78K/6mI+48JGgQQloznPYZdkh6S0n5tIyZ86cYu7cucOaqP0+7HDMitP8jdKHDrZru2GH ggW75SJBMYQPI/o1x0vMyG5+7NKH1ggfiT92y36Sj9C3ePHiXGYry/BfuHBh4oLtcqN/wQV2bTcm xsS/3K9zLhHxwrFyWz52y7HLa/iNInj+/Pn822qDv+yUzwOMudKnbVoVc2vnxAtcyrGL3VxLxCQE sYCdfloGxctTTz2VfvHjwx/+cJqIaJuf928CbRHwJERbZL1fEzCBsSTwqU99KhUZf/vb39IvZfQ6 iGXLlqVkoupn6fgZLhI+kp+IxiQEhRx2y4m5fhIs6nOjFCJogUu58EcLCViUFnSQ6GrCIfcFt7TS xC5/rY1liiIKFpLecgKKPtbxJWNtJ8IcG1w4bmwqCdcx6+fsiN2ouw/Qgo5yIYc+1hEvaG670WeZ IMOu+o1sqm+hNWJChONV8VTmQr9CD69HFNvigl3Fh7gQu/iOia0oLYwxVVq0Dv/gp7YbXFasWJHM lM8DiiGYwCaiEbuMH73ihZiKGns10VzmIh8xtkRNQjCu4iu4YF+N5ehzErHLRJrOP9LCM32LCZx+ sUtf++1vf5smHz760Y+GnC9yjV42gSYJeBKiSZrelwmYwNgTYHLhE5/4RHHx4sXij3/8Y8/joaAk obt69eptBQtJzZUrV1IxM+iKRs+dz/AFEl2SPuzmSZ+0kOxFTYiQXMKQhOr69evTR0LCd+HChTTB E8WFpHvlypVJCz9ppsZEzblz51LCR4Ia0SgO1qxZk5LG8+fP32YSLjRur41ocMEWCW2uBR/xs28k yWiNaOJCkYDtvEjARxRzaI2YnOF4ddz0fzU0wYlnva7X2nrmeNeuXZuKpzIXvhmfggot+DKi4QMK JGI1n+iEE1q4iyyiwYXCn3EGDmUtjHlojeLCOYBJBuKDWFW7du1acfPmzVAtnAPQQrzkXDhHcV5Y vXp1yAQaDKSF+MjPSTBhLMaHEZNWaOF8wzmQ2M0naPjeJ85RnAMiJhbRwnmYSQb5hHU0JvPgwiRF v3P1X/7yl3Tuet/73udfxfh/dP47xgTu/p9bbYz1W7oJmIAJNE6A5P8Pf/hDcezYseKTn/xkz4SW ZIHCn+SGBJ2rG7yHIu6BBx4IuZLMwZOYK7EhkaGgYxLgyJEjaXnHjh1hSRZ6SIRJPCkSKBZYhgtF 786dO8OST7TgI1iQmNPQgC6Sz61bt4ZdGcQ2iTC+4qfVeJaWS5cupe8iYfImqqGFQuXUqVOpuFah TTEFl0gtFAD0GbiokGMCAi0bNmxIxVMUFyYX5SP4wIX+DSeKfh68HtEYU2BDvBLD2GW8IZbRsW7d uggZyYa04CPiFi2MNVNTU0kLforiwgQDeogRxlxiBy5oYQJi06ZNYVqwTZGLTyiw0YYWxjsKcfoS 43FEQwtjL/GKFuwyznEeYLIELmwT1eDCJAQTMviLZ7gQ07t27QrTAgfOA4yz8CBOiZvDhw8nXtu2 bet5jm+DFWMvMYIeuOArfMS5Ei6sq2po/vGPf1ycOHGi+OY3v5km2Ku28zoTGAcC/3ur3XXrBPvW vUnjoNoaTcAETKBlAhQf3/nOd9JEBPO0H/jAB3papMAmmdDVHpJArr73SiR67qiBF7gKSAJKwcLQ zpUmismoK075IaAFLiROaCHxhIs+Q51v2/ayrjJx5QvfUiigBV9FFU46RuwrMWeZOOFKHAl7F1oo JLkqhxYKFOKFCa2utFAkMBHRtRaYUChQcKOFIo5HF1zQwgMuFFQUt/gosqAkfokR/EPM0KekRbeY K8YjntFC8cYYIy0UmsRvNBeOlwKbeOFKOzGCf7rQwlgLF8YYuNDgwp0HXXBhzIWLJtEY54hfCu7I Bhc0cH7k3ESTFs5N0Y18AR+JC+dFzkn9uPz1r38tvvGNbxSPPPJI8f3vf78Tf0Zzsr07l8CtcfIu T0Lcuf71kZmACcyCAN9A/eUvfzl9+/StCdvw4mMW0v1WEzABEzABEzCBO4QAk27f/va3i1/96lfF 9773veJDH/rQHXJkPoxJJcAcRMw9YpNK2MdtAiYwtgT42ABfUvmb3/wmXW0b2wOxcBMwARMwARMw gbElwHdB8B1Vn/3sZz0BMbZetPAyAU9ClIn4fxMwARO4RYBbjZ988sli9+7d6TPZ+qy64ZiACZiA CZiACZhABAHugvjpT3+avjfo85//fIRJ2zCBEAKehAjBbCMmYALjSIAvOfvKV75SfP3rXy9+/vOf p+82GMfjsGYTMAETMAETMIHxI7B3797i4MGDxWc+85n0iznjdwRWbALVBDwJUc3Fa03ABEwgfQ/E e97znnTy/+Uvf1mcPXvWVEzABEzABEzABEygdQJ8sTN3QfDR0I985COt27MBE4gk4EmISNq2ZQIm MJYEPve5z6WfWvvZz342lvot2gRMwARMwARMYHwI8DEMLn7wU7Rf/OIXQ39me3woWek4E/AkxDh7 z9pNwARCCPDzZl/96leL5557rjh69GiITRsxARMwARMwAROYTAJTU1PFU089VXzhC1/o5Ge2J5O6 jzqSgH+iM5K2bZmACYwtAX5n/MCBA8WSJUsKvivCzQRMwARMwARMwATaIHDp0qXi9OnT6cux29i/ 92kCXRLgJzo9CdGlB2zbBEzABEzABEzABEzABEzABEzABCaEAHMQ/jjGhDjbh2kCJmACJmACJmAC JmACJmACJmACXRPwJETXHrB9EzABEzABEzABEzABEzABEzABE5gQAp6EmBBH+zBNwARMwARMwARM wARMwARMwARMoGsCnoTo2gO2bwImYAImYAImYAImYAImYAImYAITQsCTEBPiaB+mCZiACZiACZiA CZiACZiACZiACXRNwJMQXXvA9k3ABEzABEzABEzABEzABEzABExgQgh4EmJCHO3DNAETMAETMAET MAETMAETMAETMIGuCXgSomsP2L4JmIAJmIAJmIAJmIAJmIAJmIAJTAgBT0JMiKN9mCZgAiZgAiZg AiZgAiZgAiZgAibQNQFPQnTtAds3ARMwARMwARMwARMwARMwARMwgQkh4EmICXG0D9METMAETMAE TMAETMAETMAETMAEuibgSYiuPWD7JmACJmACJmACJmACJmACJmACJmACJmACJmACJmACJmACJmAC JmACJmACJmACzRH4P/H4N6MRpTb6AAAAAElFTkSuQmCC ) #### Exercise 2 - Copying Text with an Induction Head (Pointer Arithmetic Version) (2)(a) Let u^{\text{cont}}\_0, ~~ u^{\text{cont}}\_1, ~~ \ldots ~~ u^{\text{cont}}\_n be the principal components of the content embedding. Write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) selecting tokens with similar content to the present token, including the present token itself. W\_Q = W\_K = \begin{pmatrix}u\_0^{cont} \\u\_1^{cont} \\u\_2^{cont} \\u\_3^{cont} \end{pmatrix} We simply project the content dimensions into both heads. Note that, in this example and all that follow, we could compose an arbitrary rotation into both matrices and arrive at a functionally equivalent head. We show what we consider to be the most straightforward version of these matrices. (2)(b) Let u^{\cos}\_0, ~ u^{\sin}\_0, ~ u^{\cos}\_1, ~ u^{\sin}\_1, ~ ... be a basis describing the position embedding in terms of vectors that code for the sine and cosine embedding of the token position (eg. \lambda\cos(\alpha\_0 n\_{tok})) with descending magnitudes. Write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) that self-selects the present token position. W\_Q = W\_K = \begin{pmatrix} u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin} \end{pmatrix} (2)(c) Using the position embedding basis described in (b), write W\_Q and W\_K for an attention head (with 4 dimensional queries and keys) that self-selects the \previous\ token position. Hint: think about a 2D rotation matrix. Note that by the definition of u^{\cos}\_0, ~ u^{\sin}\_0, … is that they encode the token index n as a cosine or sine wave. If we think of the corresponding cosine and sine pair together, we can think of it as a two dimensional point: (u\_0^{\cos}, ~u\_0^{\sin})\cdot x^0\_n ~=~ \lambda\_0(\cos(\alpha\_0n),~ \sin(\alpha\_0n)) If we do the same thing for the token n-1, we find that it is the point for token n rotated by -\alpha\_0: \begin{aligned} (u\_0^{\cos}, u\_0^{\sin}) \cdot x^0\_{n-1} &~=~ \lambda\_0(\cos(\alpha\_0(n-1)),~ \sin(\alpha\_0(n-1))) \\ &~=~ \lambda\_0(\cos(\alpha\_0n-\alpha\_0),~ \sin(\alpha\_0n-\alpha\_0))\\ &~=~ \lambda\_0R\_{-\alpha\_0}(\cos(\alpha\_0n),~ \sin(\alpha\_0n) \end{aligned} So we want to take the positional embedding vectors, pair up the sin and cosine components, and perform a 2d rotation. A 2d rotation about the origin is: R\_\theta=\begin{pmatrix} \cos\theta & -\sin\theta \\ \sin\theta & \cos\theta \end{pmatrix} We compose two such rotations along with a projection of the positional basis, to arrive at: W\_K = \textrm{(as part (b))} W\_Q = \begin{pmatrix} \cos\alpha\_0 & \sin\alpha\_0 & & \\-\sin\alpha\_0 & \cos\alpha\_0 & & \\& & \cos\alpha\_1 & \sin\alpha\_1 \\& & -\sin\alpha\_1 & \cos\alpha\_1 \end{pmatrix} \begin{pmatrix} u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin} \end{pmatrix} (2)(d) Write W\_Q and W\_K for an attention head (with 8 dimensional queries and keys) selecting tokens with similar content to the present token, but disprefers attending to itself. Hint: refer to (b) and use extra 4 dimensions for keys and queries. W\_K=\begin{pmatrix} u\_0^{cont} \\ u\_1^{cont} \\ u\_2^{cont} \\ u\_3^{cont} \\ u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin} \end{pmatrix} ~~~~W\_Q=\begin{pmatrix} u\_0^{cont} \\ u\_1^{cont} \\ u\_2^{cont} \\ u\_3^{cont} \\ -\beta{}u\_0^{cos} \\ -\beta{}u\_0^{sin} \\ -\beta{}u\_1^{cos} \\ -\beta{}u\_1^{sin} \end{pmatrix} \beta is a parameter that lets us tune the relative weight of the “same token” and “not the present position” parts of the computation. (2)(e) Write down W\_V and W\_{out} for the attention head you described in (d), such that it extracts the largest 8 dimensions of the position embedding from the token it attends to, and writes them to the vectors v\_0, v\_1, .... W\_V=\begin{pmatrix} u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin} \\ u\_2^{cos} \\ u\_2^{sin} \\ u\_3^{cos} \\ u\_3^{sin} \end{pmatrix} ~~~~W\_O = \begin{pmatrix}v\_0^T&v\_1^T&v\_2^T&\ldots{}\end{pmatrix} (2)(f) Write W\_Q and W\_K for an attention head which attends to the token after a previous copy of the present token. Hint: use the outputs of the head from (e) and the strategy you used in (c). The head described in (d) and (e) puts “the position of the previous instance of this token” into the subspace defined by v\_0, v\_1, \ldots{}. We project that subspace out, and rotate it the same way as (c), except forwards instead of backwards: W\_K = \begin{pmatrix}u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin}\end{pmatrix} ~~~~ W\_Q = \begin{pmatrix} \cos\alpha\_0 & -\sin\alpha\_0 & & \\ \sin\alpha\_0 & \cos\alpha\_0 & & \\ & & \cos\alpha\_1 & -\sin\alpha\_1 \\ & & \sin\alpha\_1 & \cos\alpha\_1 \end{pmatrix} \begin{pmatrix} v\_0\\ v\_1\\ v\_2\\ v\_3 \end{pmatrix} #### Exercise 3 - Copying Text with an Induction Head (Previous Token K-Composition Version) The first head copies the “content” subspace of the previous token into the v\_0, v\_1, \ldots{} subspace of the present position: W\_K = \begin{pmatrix}u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin}\end{pmatrix} ~~~~ W\_Q = \begin{pmatrix} \cos\alpha\_0 & \sin\alpha\_0 & & \\ -\sin\alpha\_0 & \cos\alpha\_0 & & \\ & & \cos\alpha\_1 & \sin\alpha\_1 \\ & & -\sin\alpha\_1 & \cos\alpha\_1 \end{pmatrix} \begin{pmatrix}u\_0^{cos} \\ u\_0^{sin} \\ u\_1^{cos} \\ u\_1^{sin}\end{pmatrix} W\_V = \begin{pmatrix} u\_0^{cont} \\ u\_1^{cont} \\ u\_2^{cont} \\ u\_3^{cont} \end{pmatrix} ~~~~ W\_O = \begin{pmatrix}v\_0^T&v\_1^T&v\_2^T&\ldots{}\end{pmatrix} The second head can then use that subspace in its key projection, and compare it to the current token’s content: W\_K = \begin{pmatrix}v\_0\\v\_1\\v\_2\\v\_3\end{pmatrix} ~~~~W\_Q = \begin{pmatrix} u\_0^{cont} \\ u\_1^{cont} \\ u\_2^{cont} \\ u\_3^{cont} \end{pmatrix}
85a78e7d-2e10-459b-801c-facde0431902	StampyAI/alignment-research-dataset/lesswrong	LessWrong	Responsible Deployment in 20XX The purpose of this post is to briefly talk about two categories of alignment and how I think AI safety evaluations should use them. I only consider responsible deployment in 20XX because I often find discussions about 2XXX to be distracting and unproductive. Also, for the remainder of the post, I will only be talking about LLMs, since these could be the 'operating systems' of future AI's (Ex. HuggingGPT). In the near future, LLMs have the highest potential for abuse, and they will be the focus here. As Connor Leahy and possibly others say, we should be optimistic for the sole reason that we get to build the AIs of the future. A more realistic version of this quote is: we should be optimistic about building our own AI, while acknowledging that others get to build their own AI. This adjustment is made to bring light to the fact that without standardization of safety practices, different groups will have potentially unsound or unknown safety practices. Safety standardization in AI is challenging due to the lack of comprehensive safety tools, but I want to outline potential industry standards for my own edification. This will be a reminder of the safe practices that I will advocate for during my career. Mandatory Alignment Evaluations! -------------------------------- Alignment evaluations should be mandatory for all models after GPT-4 and should be conducted by third-party and in-house groups. It will not be enough if AI companies are relied on to graciously open their doors to the [Alignment Research Center](https://www.alignment.org) (ARC). By sheer chance the OpenAI executives had the integrity to have GPT-4 be tested for self-replication and resource acquisition. While the openness is a great present state, my concerns lie with AI capital owners of the near future and the absence of concrete safety obligations, standards and regulations. The case for oversight and regulation is difficult to make in the United States, but it's an important one. The main crux of the argument for regulation is the classification of generated content as a commercial and industrial product. This definition informs an intuition for the regulation of generated content on the same basis as any influential product massively distributed to the public sphere and private sectors (i.e. food stuff, fabric, children's toys, cyber security etc.). I plan on making a post about that in the future and when I do it will be linked here. Instead of justifying a legal basis for regulation, in this post I will talk about two classes of alignment evaluation. I think these two classes are broad enough to hold up in the future however, like most things, techniques in practice will/should be a combination of many tricks. ### Static Alignment Static alignment is when evaluators unleash algorithms onto the weights of a model and look for specific structures. Since looking at weights does not require running the model, this is the safest type of alignment evaluation. Currently there is no ability for static evaluation but hopefully these will emerge from a field like Mechanistic Interpretability. In its best form this type of analysis will look at weights through some lens and identify inner structures that can be manipulated to massively change or control model behavior. If this is possible then it is also possible to study the way that these structures form during training as well as how that formation relates to training data. In its lesser form, static alignment in large models will prove to be a Sisyphean act of looking for inner structures in model N and hoping that they generalize to model N + 1. If we are unable to build robust static analysis tools, by the time model N + 1 arrives we will have failed to detect some percent of the danger, and this missed danger will compound as N increases. Despite the previous paragraph, I am optimistic because of work being done by Mechanistic Interpretability researchers like [Neel Nanda](https://www.neelnanda.io). He has a [three part series](https://www.youtube.com/watch?v=ob4vuiqG2Go&t=24s) talking about how he reverse-engineered an impressively complex algorithm that a single layer transformer model learned to perform modular addition. This series is an example of how, although the algorithms implemented by LLMs are not designed to be human readable, these algorithms can be understood through rigorous research and mathematical understanding. ### Dynamic Alignment Dynamic alignment is a class of evaluations that a third-party organization like ARC should be required to conduct. This type of testing usually consists of giving a model limited power and resources and seeing how it behaves. Another type of testing could look at how a model performs competing with itself. Both of these evaluation types prepare for various takeoff scenarios. One type of dangerous takeoff scenario is when a super intelligent AI is acting in the world untethered. This would have unknown consequences for the world's economy, public trust and international relations. Dynamic evaluation seeks to identify models that, if given the ability to act untethered, pose a risk to humanity. What would we do if we found an AI able to fail a dynamic evaluation? Who knows, but I hope we would not release it and study every facet of its internals. While it is important to decide what we will do if a model fails an ARC evaluation, we should instead focus on increasing the scope of the evaluation set and making each more robust. Another dangerous takeoff scenario is when a sufficiently dangerous model is irresponsibly released to the public (I'm looking at you Meta...) and the world's leading evil-doers are given a tool powerful enough to cripple a nation. If the models are instead only powerful enough to topple things smaller than nations, then it is likely that the world will have to function with super intelligent agents with conflicting goals. These models, which simulate some kind of agency, will be competing or even collaborating with one another. This type of behavior should be studied and that is why I consider it as a type of evaluation ARC should look into for GPT-5+ and its competitors. This type of evaluation is notably dangerous since you are giving resources to an under-studied force. We also run into two classic problems. One problem is that a model smart enough to take over the world is likely smart enough to pass a dynamic evaluation. The other problem is that a model smart enough to take over the world is likely smart enough to break out of whichever box we put it in. The problems of deception and self-jail breaking are beyond the scope of this post, so I will not delve deeper. ### Static and Dynamic Alignment in Conjunction Responsible deployment will involve both types of alignment but the order in which the evaluations are conducted matters. Safety is maximized when static evaluations are done before dynamic evaluations as opposed to the other way around. This should be obvious from a research and scientific lens since static analysis looks for states in the model that will inform the dynamic evaluations. Although running a model will be required to verify claims made by static analysis, the research to verify theories will hopefully be done in less risky scenarios. The Standard Process -------------------- The standards in a world with responsible scientists is one where the static model analysis is done before any risky evaluations. Any standard will require thorough research in fields like mechanistic interpretability and deep learning theory. If looking at the weights of a model gives us no knowledge of its behavior, then we will not detect an evil model before we run it. Without robust theories and tools for static model analysis we are doomed to continue to just run gold-spitting black boxes until the end.
75fb5e84-dc67-4dd3-95cd-a12bbc92fa0a	trentmkelly/LessWrong-43k	LessWrong	Meetup : MIT/Boston Secular Solstice Discussion article for the meetup : MIT/Boston Secular Solstice WHEN: 11 December 2015 08:00:00PM (-0500) WHERE: MIT Chapel 50 Massachusetts Ave, Cambridge, Massachusetts 02139 It has become tradition, in the community of those who seek to become more rational, to gather for one night of each year, and sing. We do this close to the winter solstice, which is the longest, darkest night of the year; and, gathered as a community, we stare into and confront the darkness. This consists of participatory singing and a few short speeches, following an emotional arc from light to darkness and back to light again. It will last about two hours, starting at 8pm at MIT Chapel at 50 Massachusetts Ave, Cambridge, Massachusetts 02139 and be followed by a reception/afterparty nearby in room 1-132. We may also be organizing an optional pre-ritual potluck nearby, details TBD. The Facebook event page is at https://www.facebook.com/events/505931562916689/ . Discussion article for the meetup : MIT/Boston Secular Solstice
4f97ff39-ed93-42f0-9dc8-bc779e777800	trentmkelly/LessWrong-43k	LessWrong	Follow along with Columbia EA's Advanced AI Safety Fellowship! Columbia EA is starting an Advanced AI Safety Fellowship this summer, which will involve one 3-hour meeting per week with a group of people seriously considering careers in AI safety who have some background in both ML (at least working knowledge of linear algebra, multivariable calculus, probability and stats, and neural networks) and AI safety (usually the Cambridge AGI Safety Fundamentals program or a similar reading group we run at Columbia). Each meeting will involve reading about a topic selected the previous week and discussing it, and may also include exercises (like these). While we do not have capacity to invite people from elsewhere to join our meetings, we would like to invite others to follow along with our exploration of the field! You can find our running meeting notes here, where we will write down our meeting agendas, readings, and other notes (maybe key takeaways, responses to exercises, questions, etc.). The hope is that this can serve as a small, easy win, by motivating at least a few people to set aside time to dive deep into AI alignment topics with little additional effort on our part. Participating in an alternative program with a similar premise, in which the organizers put more effort into making it a good experience for you, would likely be better than following along with us. If such a program already exists, please let me know! I’m not currently aware of any programs that satisfy all of the following properties: 1. Have the basic premise of being a deep dive into AI safety for people with some background in ML and AI safety who are seriously considering careers in the field 2. Are open to anyone with that background 3. Have organizers dedicated to making this a good experience for many interested people from various places Given that we are not satisfying 3, you could dispense with “following along” and just structure your entire plan yourself if you feel that would work better for you. There are two main benefits we hope to provid
ea98c712-d07c-4c7b-8a98-7afade684156	StampyAI/alignment-research-dataset/arxiv	Arxiv	Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Society 1. Introduction ---------------- There is a growing community of researchers focused on ensuring that advances in AI are safe and beneficial for humanity (which we call the AI ethics and society community, or AI E&S for short.) The AI E&S research space is very broad, encompassing many different questions: from how AI is likely to impact society in practice to what ethical issues that might raise, and what technical and governance solutions might be needed. These questions are approached from many different perspectives and disciplines: philosophy, computer science, political science, sociology, law, economics, and international relations to name just a few. This research space is fast growing, with new initiatives, workshops, and research centres emerging every year. Finding ways to break down this broad and evolving research space is important, as it helps researchers to clearly formulate and communicate problems and research agendas, collaborate effectively with others, and to identify gaps in current research. Natural ways to break down the AI E&S research space include by themes or sectors (such as the impact of AI on medicine, the military, or issues of social justice) or in relation to pre-existing disciplines (distinguishing research on ethics from technical approaches from governance and policy, for example). Due to the cross-cutting nature of AI’s impacts, and the need for deep interdisciplinary collaboration in this space, it may be useful to have ways of carving up the research space that relate to more fundamental priorities and assumptions rather than being tied to specific domains or disciplines. One such way of carving up the research space which has emerged in recent years is to distinguish between those who focus on ‘near-term’ and ‘long-term’ issues (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy"); Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI"); Krakovna, [2018](#bib.bib32 "Is There a Trade-Off Between Immediate and Longer-term AI Safety Efforts?"); Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction")). However, the distinction between ‘near-term’ and ‘long-term’ is being over-emphasised as a way of characterising differences in the research space. On closer inspection, the distinction is poorly defined, and is used differently across people and contexts: sometimes to refer to issues on different timescales; sometimes to issues related to different types of AI capabilities, and other times masks deeper normative and empirical disagreements. Highlighting inconsistencies in how the near/long-term distinction is used, we argue that it fails to adequately capture the complexity and nuance of different approaches and priorities in the AI E&S research space, and risks causing confusion and fuelling conflict. By unpacking the distinction between near- and long-term into four different dimensions, we propose some ways to more clearly conceptualise different priorities and disagreements within the AI E&S research space. First, we distinguish between research which focuses on near/long-term capabilities as opposed to near/long-term impacts, and argue that in both cases near/long-term is a spectrum rather than a binary distinction. We show how this way of thinking about different approaches could both improve collaboration and communication between researchers as well as helping to identify important areas of research which are currently neglected. Second, we analyse some of the beliefs and disagreements which seem to underpin differing research priorities, and suggest that clarifying these beliefs could aid mutual understanding between different researchers and groups. 2. The near/long-term distinction in practice ---------------------------------------------- The distinction between near- and long-term in AI E&S is used both to distinguish different types of issues one might be concerned about, as well as to highlight a divide or disagreement between two research communities. In this section, we briefly review how the distinction is made in the literature. ### 2.1. ‘Near-term’ issues As the phrase ‘near-term’ suggests, those who have written about the distinction tend to characterise near-term issues as those issues that society is already facing or likely to face very soon: Brundage (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")) defines near-term issues as those society is “grappling with today” and Cave and ÓhÉigeartaigh (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI")) talk in terms of “immediate or imminent challenges” (p.5). Examples include concerns about data privacy (Tucker, [2018](#bib.bib3 "Privacy, algorithms, and artificial intelligence"); Zuboff, [2019](#bib.bib4 "The age of surveillance capitalism: the fight for a human future at the new frontier of power")), algorithmic bias (Hajian et al., [2016](#bib.bib5 "Algorithmic bias: from discrimination discovery to fairness-aware data mining"); Koene, [2017](#bib.bib7 "Algorithmic bias: addressing growing concerns [leading edge]")), self-driving car accidents (Goodall, [2016](#bib.bib6 "Can you program ethics into a self-driving car?"); Bonnefon et al., [2016](#bib.bib8 "The social dilemma of autonomous vehicles")), and ethical issues associated with autonomous weapons (Asaro, [2012](#bib.bib9 "On banning autonomous weapon systems: human rights, automation, and the dehumanization of lethal decision-making"); Anderson and Waxman, [2013](#bib.bib10 "Law and ethics for autonomous weapon systems: why a ban won’t work and how the laws of war can")). It is worth noting two things about these examples. First, they tend to be relatively concrete and specific, and an important feature of ‘near-term’ issues is the fact that they are already fairly well-understood. Second, these examples tend to be directly related to recent progress in machine learning which have enabled increasing real-world applications of narrow and specialised AI systems: for example in medical diagnosis and predictive policing (Buch et al., [2018](#bib.bib28 "Artificial intelligence in medicine: current trends and future possibilities"); Richardson et al., [2019](#bib.bib30 "Dirty Data, Bad Predictions: How Civil Rights Violations Impact Police Data, Predictive Policing Systems, and Justice")). ### 2.2. ‘Long-term’ issues ‘Long-term’ issues are often characterised as those arising from advanced AI systems, or simply as issues that will arise far into the future. Brundage (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")) defines long-term issues as those that “either only arise at all or arise to a much greater extent when AI is much more advanced than it is today”, while Cave and ÓhH́igeartaigh (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI")) point to “longer-term concerns and opportunities that are less certain” (p.5). Examples of long-term issues often centre around the implications of very advanced future AI with broad capabilities (‘artificial general intelligence’, or AGI), which achieve human or superhuman intelligence (‘human-level AI’ or ‘superintelligence’), or are in some other way radically transformative (Karnofsky, [2016](#bib.bib27 "Potential Risks from Advanced Artificial Intelligence: The Philanthropic Opportunity")). For example, the White House Report on Preparing for the Future of Artificial Intelligence talks about “long-term concerns of AGI”, implicitly equating the two (Science and House), [2016](#bib.bib17 "Preparing For The Future Of Artificial Intelligence")). While thinking about ‘long-term’ concerns of AI may have begun with concerns about superintelligence and related notions (Bostrom, [2014](#bib.bib26 "Superintelligence: Paths, Dangers, Strategies")), the notion has since broadened to include challenges that are likely to have long-term consequences, as opposed to merely arising far into the future. Examples include the effects of advanced, ‘transformative AI’ on international security, race dynamics and power relations (Karnofsky, [2016](#bib.bib27 "Potential Risks from Advanced Artificial Intelligence: The Philanthropic Opportunity"); Dafoe, [2018](#bib.bib18 "AI Governance: A Research Agenda"); Gruetzemacher and Whittlestone, [2019](#bib.bib16 "Defining and Unpacking Transformative AI")). ### 2.3. The divide between ‘near-term’ and ‘long-term’ communities Baum (Baum, [2018](#bib.bib31 "Reconciliation between factions focused on near-term and long-term artificial intelligence")) not only distinguishes between near and long-term issues, but also points to a divide between two different communities: ‘presentists’, who claim that “attention should go to existing and near-term AI”, and ‘futurists’, who argue for focusing on “the potential for radically transformative AI”. Cave and ÓhÉigeartaigh (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI")) also point to “two seemingly separate communities of researchers and technologists” focused on near and long-term issues. Baum suggests these two groups disagree quite fundamentally about what issues are most important to work on. This has provoked hostility between the two sides: the ‘near-term’ camp suggesting that long-term concerns are overblown and a distraction from real problems (Etzioni, [2016](#bib.bib23 "No, the Experts Don’t Think Superintelligent AI is a Threat to Humanity"); Crawford and Calo, [2016](#bib.bib15 "There is a blind spot in AI research")), while some in the ‘long-term’ camp suggest that the problems they focus on dwarf any near-term concerns in importance (Tegmark, [2017](#bib.bib14 "Life 3.0: being human in the age of artificial intelligence"); Bostrom, [2014](#bib.bib26 "Superintelligence: Paths, Dangers, Strategies")). This adversarial dynamic seems to be at the core of proposals to ‘reconcile’ or ‘bridge’ the two sides (Baum, [2018](#bib.bib31 "Reconciliation between factions focused on near-term and long-term artificial intelligence"); Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI")). 3. The problem with the near/long-term distinction --------------------------------------------------- ‘Near-term’ and ‘long-term’ are often used in ways that combine and conflate multiple different dimensions, including: when issues arise; what kinds of technological capabilities issues relate to; how well-understood or speculative issues are; how high-stakes issues are and how many people they affect; and how long-lasting the consequences of an issue are likely to be. As table [1](#S4.T1 "Table 1 ‣ 4.1. Unpacking the near/long-term distinction ‣ 4. A clearer account of research priorities and disagreements ‣ Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Society") shows, even a single definition or paper sometimes associates ‘near-term’ with one of these dimensions and ‘long-term’ with another: Brundage (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")), for example, defines ‘near-term’ based on when issues arise and ‘long-term’ in terms of technological capabilities. Rather than describing a single dimension, we suggest that the distinction between near- and long-term as currently used is better understood as describing clusters of issues, incorporating multiple dimensions: * Issues considered ‘near-term’ tend to be those arising in the present/near future as a result of current/foreseeable AI systems and capabilities, on varying levels of scale/severity, which mostly have immediate consequences for people and society. * Issues considered ‘long-term’ tend to be those arising far into the future as a result of large advances in AI capabilities (with a particular focus on notions of transformative AI or AGI), and those that are likely to pose risks that are severe/large in scale with very long-term consequences. We believe that carving up the AI E&S research space according to these clusters is problematic in two key ways. First, important and nuanced information gets lost if we give too much prominence to only two clusters. Many beliefs, values and assumptions will vary within each of these clusters. For example, among those who believe we should prioritise risks from advanced AI systems, there seems to be substantial disagreement about why exactly those risks are most important to work on, and what work should be done today to prevent those risks (Sittler, [2019](#bib.bib12 "A shift in arguments for AI risk"); Ngo, [2019](#bib.bib13 "Disentangling arguments for the importance of AI safety")). Similarly, there are many different reasons a group may choose to focus on issues arising from current applications of AI in society: because they believe we should prioritise helping people alive today, because they believe advanced AI systems are very far off or hard to predict, or because they believe today’s problems are as high-stakes as anything we might face in the future. As we will discuss later, a binary distinction may have led the research community to neglect impacts which fall somewhere between ‘immediate’ and ‘very long-term’ (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction")). Second, giving too much prominence to the ‘near-term’ and ‘long-term’ clusters without examining underlying beliefs can lead researchers who identify with either side to end up misunderstanding each other. For example, Etzioni (Etzioni, [2016](#bib.bib23 "No, the Experts Don’t Think Superintelligent AI is a Threat to Humanity")) argues against the idea that superhuman AI could pose an existential threat to humanity (most prominently argued by Bostrom, (Bostrom, [2014](#bib.bib26 "Superintelligence: Paths, Dangers, Strategies"))) by drawing on survey data to suggest that experts do not believe superhuman AI systems will arrive any time soon. But as Dafoe and Russell (Dafoe and Russel, [2016](#bib.bib24 "Yes, We Are Worried About the Existential Risk of Artificial Intelligence")) point out in response, Bostrom’s argument does not actually rely on any assumptions about superintelligence being imminent. In practice, Etzioni and Bostrom’s disagreement turns more on how important and tractable they think it is to work on the potential risks from superhuman AI today. The idea that there are two ‘camps’ of AI E&S researchers, with very opposing perspectives, may lead researchers like Etzioni and Bostrom to make unfounded assumptions about the others’ perspective, and prevent them from engaging in more substantive and productive debate. At its core, the distinction between near and long-term in AI E&S aims to capture the idea that different researchers and groups will sometimes have quite different priorities which guide what they work on, underpinned by some quite fundamental disagreements. We believe that acknowledging these differences is useful, but suggest that we need a clearer way to think and communicate about what they are. In the following section, we propose exactly that - some clearer ways to communicate about different research priorities and underlying disagreements. 4. A clearer account of research priorities and disagreements -------------------------------------------------------------- ### 4.1. Unpacking the near/long-term distinction As commonly used, the terms ‘near-term’ and ‘long-term’ in fact appear to capture four different dimensions of differing priorities within the AI E&S research community: * Capabilities: whether to focus on the impacts and challenges of current AI systems, or those relating to much more advanced AI systems * Impacts: whether to focus mostly on the immediate impacts of AI for society, or whether to consider possible impacts much further into the future. * Certainty: whether to focus on impacts and issues that are relatively certain and well-understood, or those that are more uncertain and speculative. * Extremity: whether to focus on impacts at all scales, or to prioritise focusing on those that may be particularly large in scale. None of these four dimensions are binary: one can choose research questions that focus on AI systems that are more or less advanced, exploring impacts on varying time horizons, with varying degrees of certainty and extremity. Table [1](#S4.T1 "Table 1 ‣ 4.1. Unpacking the near/long-term distinction ‣ 4. A clearer account of research priorities and disagreements ‣ Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Society") shows how these four dimensions in definitions of ‘near-term’ and ‘long-term’ by different authors. \| Dimension \| ’Near-term’ \| ’Long-term’ \| \| --- \| --- \| --- \| \| AI Capabilities \| “Attention should go to existing and near-term AI” (Baum, [2018](#bib.bib31 "Reconciliation between factions focused on near-term and long-term artificial intelligence"), p.2) \| “Issues that either only arise at all or arise to a much greater extent when AI is much more advanced than it is today, and in particular if advances turn out to be rapid.” (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")) “Attention should go to the potential for radically transformative long-term AI” (Baum, [2018](#bib.bib31 "Reconciliation between factions focused on near-term and long-term artificial intelligence"), p.2) \| \| Impacts \| “Issues society is grappling with today.” (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")) “Extant or imminently anticipated AI applications that interact with existing legal, political, or social concerns” (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction"), p.5) “ […] problem X that we are facing today with AI” (Krakovna, [2018](#bib.bib32 "Is There a Trade-Off Between Immediate and Longer-term AI Safety Efforts?")) \| “Longer-term concerns and opportunities that are less certain” (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI"), p.5) \| \| Certainty \| “Immediate challenges involving fairly clear players and parameters” (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI"), p.5) \| “Longer-term concerns and opportunities that are less certain” (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI"), p.5) \| \| Extremity \| \| “The existential risks of extreme AI advances” (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction"), p.5) \| Table 1. Dimensions underlying the ’near-term’ and ’long-term’ distinction in the literature. Of course, these dimensions are not entirely independent of one another: for example, if one wants to focus on particularly large-scale impacts of AI, one may have to be willing to consider more speculative questions. However, it is still useful to unpack these different dimensions, particularly because there are many possible views and research questions which are not easily captured by the near/long-term distinction as commonly used. Looking at these four dimensions, we can identify some differences in what they refer to. ‘Capabilities’ and ‘impacts’ both refer in some sense to time horizons; to whether research focuses on the present (either in terms of AI capabilities or impacts) or whether it looks further into the future. Both these dimensions therefore seem to capture different interpretations of what it might mean to say an issue is ‘near-term’ or ‘long-term’. ‘Certainty’ and ‘extremity’, by contrast, do not relate so explicitly to time horizons: instead they relate to the nature of impacts that different researchers might focus on, and our ability to predict and understand these impacts. Although these dimensions do appear in how near and long-term are described, we suggest that they better characterised as capturing the kinds of motivations researchers may have for choosing to work on different areas. ### 4.2. Distinguishing capabilities and impacts ‘Near-term’ and ‘long-term’ are often modifiers applied to the state of technological capabilities themselves, with near-term issues being those arising from the AI systems we have today (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction")), and long-term issues being those we might expect to arise from certain kinds of much more advanced AI systems (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")). At other times, ‘near-term’ and ‘long-term’ refer rather to the impacts of the technology: with near-term issues being the impacts of AI that society already faces or will face in the immediate future (Brundage, [2017](#bib.bib34 "Guide to working in AI policy and strategy")), and long-term issues being the impacts of AI that society might face very far into the future (Cave and ÓhÉigeartaigh, [2019](#bib.bib35 "Bridging near- and long-term concerns about AI")). Of course the timescale of technological advances and their impacts on society will be related. However, conflating long-term capabilities and impacts may mean the research community neglect important questions about the potential long-term impacts of current AI systems and their applications in society. For example, in what ways could injustices perpetuated by increased use of current algorithmic systems have very long-lasting and irreversible consequences for inequality in society? What are the possible longer-term implications of how data-intensive AI applications are beginning to change norms around privacy and personal data? (Newell and Marabelli, [2015](#bib.bib1 "Strategic opportunities (and challenges) of algorithmic decision-making: a call for action on the long-term societal effects of ‘datification’")) Furthermore, it is important to recognise that both capabilities and impacts lie on a spectrum between near- and long-term. Understanding the potential ‘medium-term’ impacts of AI on society - for example, exploring what different sectors and domains might look like in 5-10 years as a result of current trends in AI development and deployment - is likely to be important both for thinking about what we can do today to protect against future harms, and for preparing for longer-term impacts and scenarios. Similarly, there are many ways in which AI systems could become more advanced over the coming years, before reaching anything close to ‘superintelligence’ or ‘AGI’. We need to consider what issues these intermediate advances might raise, and what kinds of intermediate advances in capabilities are mostly likely to be of harm or benefit for society. Figure [1](#S4.F1 "Figure 1 ‣ 4.2. Distinguishing capabilities and impacts ‣ 4. A clearer account of research priorities and disagreements ‣ Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Society") shows how considering these two dimensions of capabilities and impacts, each sitting on a spectrum, allows for more nuanced categorisation of different types of issues. ![Distinguishing issues based on capabilities and impacts.](https://media.arxiv-vanity.com/render-output/7430885/figure1.png) Figure 1. Distinguishing issues based on capabilities and impacts. This way of conceptualising the research space could also be used to situate different research groups or research agendas. For example, AI Now’s research seems to sit pretty squarely in the bottom left-hand corner (Whittaker et al., [2018](#bib.bib25 "AI now report 2018")); the Centre for the Governance of AI at Oxford focuses mostly on issues at the top of the plot but spanning left to right (Dafoe, [2018](#bib.bib18 "AI Governance: A Research Agenda")); and the Centre for the Future of Intelligence at Cambridge works on issues across a much wider range of the plot. Both authors of this paper see their research as mostly focusing in the upper-left quadrant - i.e. considering the medium-to-long-term impacts of near-to-medium-term AI capabilities, primarily because these issues have thus far been more neglected in discourse and research around AI E&S impacts (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction")). Of course, this still contains considerable ambiguity and room for disagreement, and placing different issues on these axes requires making assumptions about questions we don’t necessarily have good answers to. Is there really good reason to think that advanced AI systems will pose an existential threat to humanity, or that advances in AI capabilities might pose a threat to democracy with long-lasting consequences? We believe that provoking productive debate and discussion on questions like these could in fact be a valuable consequence of this way of conceptualising the research space. However, we also recognise that any two-dimensional plot is limited in its ability to capture more fundamental reasons why people have those different priorities. We therefore next turn to a discussion of some of the beliefs and motivations underpinning different approaches in the AI E&S research space. ### 4.3. Underlying beliefs and motivations One question that has not been explored in existing discussions of the near/long-term distinction is why different researchers and groups have different priorities. The idea of focusing on impacts that are more or less certain and more or less extreme are present in some definitions of near and long-term (see table [1](#S4.T1 "Table 1 ‣ 4.1. Unpacking the near/long-term distinction ‣ 4. A clearer account of research priorities and disagreements ‣ Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Society")); we suggest that these dimensions may be best understood as capturing some of the underlying motivations people have for focusing on certain types of research questions. One reason to focus on researching the longer-term impacts of AI, and/or the impacts of more advanced capabilities, is that those impacts might be more extreme and so particularly important to manage. Indeed, groups like the Open Philanthropy Project who explicitly prioritise risks from advanced AI state that “all else equal, we’re more interested in artificial developments that would affect more people and more deeply” (Karnofsky, [2016](#bib.bib27 "Potential Risks from Advanced Artificial Intelligence: The Philanthropic Opportunity")). Similarly, a research agenda published by the Future of Humanity Institute in Oxford, typically thought of as an institution with a ‘long-term’ focus, explicitly highlights its “focus on extreme risks” (Dafoe, [2018](#bib.bib18 "AI Governance: A Research Agenda")). The idea that immediate impacts of AI and/or those relating to current AI systems are more certain and well-understood is also sometimes invoked as a reason to work on those issues (or as an argument against focusing on advanced capabilities and long-term impacts.) For example, Andrew Ng explains that the reason he is not concerned about advanced AI is “the same reason I don’t worry about overpopulation on Mars… we’ve never set foot on the planet so how can we productively worry about this problem now?” (Garling, [2015](#bib.bib19 "Andrew Ng: Why ‘Deep Learning’ is a mandate for humans, not just machines.")). Relatedly, criticisms of those who focus on potential existential risks from AI often make the argument that these worries are too speculative, and are “a distraction from the very real problems with artificial intelligence today” (Crawford, [2016](#bib.bib22 "Artifical Intelligence’s White Guy Problem.")). Working on problems that may have very extreme impacts on humanity, and working on problems that are certain and well-understood today are somewhat in tension with one another, and this tension is perhaps a central driver of the divide between ‘long-term’ and ‘near-term’ communities in AI E&S. Of course, in line with the main message of this paper, our view is that there is not a simple binary choice between focusing on ‘extreme’ impacts of AI and focusing on ‘certain’ impacts - both exist on a spectrum, and there may be many ways for the AI E&S community to identify areas for research which have relatively high-stakes implications while also being grounded enough in current AI and societal trends to make progress on. Further dimensions of disagreement may help explain why the AI E&S community has divided in certain ways. Disagreement on normative issues may be relevant here, such as around whether we have a special moral obligation to help those who are alive today over those who will live in future (Parfit, [2017](#bib.bib21 "Future people, the non-identity problem, and person-affecting principles")), or to prioritise helping those worst off in society (Parfit, [1991](#bib.bib20 "Equality or priority")). Someone who holds the more fundamental philosophical belief that we should prioritise helping the worst off in society, for example, is likely to choose to work on the implications of AI for global fairness or social justice, regardless of their position on the certainty/extremity tension as outlined above. Other disagreements underpinning different priorities may be empirical: disagreements about the nature of problems, or our ability to solve them. Whether one believes it is more important to focus on challenges relating to current or advanced AI systems is likely to depend at least in part on beliefs about AI progress. If you believe we are likely to see very fast and/or discontinuous progress in AI capabilities over the next few years, preparing for the potential impacts of these advances seems much more urgent than if you believe progress will be slow and gradual, giving us time to solve today’s problems before we have to deal with anything more advanced. Indeed, much of the disagreement about whether we should be concerned about advanced AI seems to turn on disagreement about whether large advances in AI capabilities are possible any time soon (Brockman, [2019](#bib.bib2 "Possible Minds: Twenty-Five Ways of Looking at AI"); Ford, [2019](#bib.bib11 "Architects of Intelligence: The truth abut AI from the people building it")). Which specific impacts one chooses to focus on may also depend on empirical beliefs about the severity of different issues: how damaging to society threats to privacy might be in the long-term, and/or how likely advanced AI systems are to pose an existential threat, for example. Beliefs about our ability to forecast and/or influence the future are also relevant - often scepticism about research on advanced AI systems seems to stem from doubts about our ability to productively work on these problems. Andrew Ng’s statement quoted earlier (Garling, [2015](#bib.bib19 "Andrew Ng: Why ‘Deep Learning’ is a mandate for humans, not just machines.")) appears to express this kind of scepticism: that we can’t ‘productively’ work on the impacts of advanced AI systems. 5. Recommendations and conclusion ---------------------------------- The field of research concerned with the impact of AI on society and humanity is fast growing. Due to the cross-cutting and interdisciplinary nature of these issues, it is useful to be able to carve up the research space in ways that go beyond traditional disciplinary or thematic boundaries in academia, and capture different underlying assumptions about what the most important problems in this space are. We propose more nuanced ways to do this that better capture this rich and complex research space, breaking down the binary distinction between ‘near-term’ and ‘long-term’ into four different dimensions: capabilities, impacts, certainty and extremity, and emphasising that all of these sit on a spectrum. Based on this analysis, we have a few concrete recommendations for how the AI E&S research community can encourage more nuanced and productive discussion about different priorities and assumptions: * Be specific about what you mean when using ‘near-term’ or ‘long-term’ to refer to research priorities or projects. As outlined in the previous section, we think it would be particularly useful if there was a clearer distinction between (a) immediate vs. long-term impacts on society, and (b) current vs. much more advanced AI capabilities, and if there were greater acknowledgement that in both cases ‘near to long-term’ is a spectrum, not a binary distinction. * Communicate clearly and explicitly about the assumptions and beliefs underlying your projects and priorities, especially around some of the questions highlighted above. A good example of this is the preface to Dafoe’s (Dafoe, [2018](#bib.bib18 "AI Governance: A Research Agenda")) research agenda, which explicitly states its focus on extreme risks from advanced AI (though they could go even further to explain the reasoning behind this decision.) Another good example is Parson et al. (Parson et al., [2019](#bib.bib33 "Artificial Intelligence in Strategic Context: An Introduction")) who clearly state that their project focuses on the “intermediate scale of AI impacts, time horizons, and implications” since this is where their expertise is most relevant, and where areas of potential importance are receiving relatively less attention. * Make an effort to understand the underlying motivations and assumptions of others with different research priorities, again using some of the questions outlined in the previous section as a starting point. Conferences, workshops and journals could potentially help support this by providing fora for researchers to debate such fundamental questions and disagreements, improving mutual understanding. By taking these steps, we hope the AI E&S research community can: (a) establish new opportunities for collaboration and reduce adversarial dynamics, while allowing for differences of opinion and approach; (b) develop more consistent and coherent research agendas by helping researchers to think more clearly about what they prioritise; and (c) identify neglected research areas which may have been overlooked due to not fitting neatly into existing ways of characterising the research space.
e6a92372-a161-4d2c-8969-82b35b3d596f	trentmkelly/LessWrong-43k	LessWrong	Paper review: A Cryptographic Solution to a Game Theoretic Problem Crossposted from: Equilibria Club It's time for a puzzle involving two fields: game theory and cryptography! Both fields model self-interested agents in a competitive setting. In game theory, we list actions and payoffs, after which we can determine which actions will benefit agents the most. In cryptography, we model actions to ensure agents only get a limited amount of information. Comparison of concepts between cryptography and game theory ---------------------------------------- The problem: game of chicken To illustrate how cryptography can be used to achieve better outcomes in a game, we're going to look at the "Game of Chicken". Imagine two players driving towards each other in a car, in principle wanting to avoid a collision, but also not wanting to chicken out. The standard two-player game is characterized by the following payoffs: Player 1 cooperatesPlayer 1 defectsPlayer 2 cooperates(4, 4)(1, 5)Player 2 defects(5, 1)(0, 0) > "While the “wisest” pair of actions is (C, C), this is not a Nash equilibrium, since both players are willing to deviate (believing that the other player will stay at C)." ---------------------------------------- We can display the three Nash equilibria (the optimal actions for non-coordinating self-interested players) on a graph. The players will either (cooperate, defect), (defect, cooperate), or have a 50% chance of (cooperate, defect) and (defect, cooperate). Three Nash equilibria in the extended game In the previous post about game theory, I referred to the Folk Theorem of game theory, which says that many equilibria can be enforced by the players in iterated games. However, those enforceable equilibria all fall in the so-called convex hull of the Nash equilibria - which is any situation contained by the Nash equilibria. In the Game of Chicken, situations in the convex hull are all situations below the purple diagonal, leading to a maximum total payoff of 6. Is there any way we can do better? Well yes! We can try to a
911ffe82-4d48-4cdc-ab6b-2c257c1a5306	LDJnr/LessWrong-Amplify-Instruct	LessWrong	"If you ever find yourself saying, "Even if Hypothesis H is true, it doesn't have any decision-relevant implications," you are rationalizing! The fact that H is interesting enough for you to be considering the question at all (it's not some arbitrary trivium like the 1923th binary digit of π, or the low temperature in São Paulo on September 17, 1978) means that it must have some relevance to the things you care about. It is vanishingly improbable that your optimal decisions are going to be the same in worlds where H is true and worlds where H is false. The fact that you're tempted to say they're the same is probably because some part of you is afraid of some of the imagined consequences of H being true. But H is already true or already false! If you happen to live in a world where H is true, and you make decisions as if you lived in a world where H is false, you are thereby missing out on all the extra utility you would get if you made the H-optimal decisions instead! If you can figure out exactly what you're afraid of, maybe that will help you work out what the H-optimal decisions are. Then you'll be a better position to successfully notice which world you actually live in." The above is a chunk of text from a post created by a user on the website LessWrong. What potential problems could arise from assuming that the optimal decisions will always differ in worlds where H is true and where H is false? Assuming that optimal decisions will always differ in worlds where H is true and where H is false could lead to several potential problems: 1. Overcomplication: This assumption may lead to overcomplication of decision-making processes. Sometimes, the optimal decision may be the same regardless of whether H is true or false. Assuming otherwise could lead to unnecessary complexity and confusion. 2. Ignoring Overlapping Solutions: There could be overlapping solutions or decisions that are optimal in both scenarios. By assuming that decisions must differ, one might overlook these common solutions that could be more efficient or beneficial. 3. False Dichotomy: This assumption could lead to a false dichotomy, where decisions are seen as only being optimal in one scenario or the other, but not both. This can limit the range of potential solutions considered. 4. Misallocation of Resources: If one assumes that different scenarios require different optimal decisions, it could lead to a misallocation of resources. Time and effort might be wasted on creating separate strategies for each scenario, even when a single strategy could have been effective for both. 5. Potential for Bias: This assumption could potentially introduce bias into the decision-making process. If one is convinced that the truth or falsehood of H must necessarily lead to different decisions, they might unconsciously favor solutions that align with their personal beliefs about H. 6. Ignoring Context: The assumption doesn't take into account the context in which the decision is being made. There might be external factors that make the same decision optimal whether H is true or false. In conclusion, while considering different decisions for different scenarios can be beneficial, it's important to not rigidly adhere to this assumption. There may be situations where the same decision is optimal, regardless of the truth or falsehood of H. Reflecting on the potential issues you outlined, could there be situations where the assumption of differing optimal decisions in the face of H being true or false might actually be beneficial? Yes, there are indeed situations where the assumption of differing optimal decisions, depending on whether H is true or false, could be beneficial. Here are a few examples: 1. Complex Systems: In complex systems or situations, where there are multiple variables and outcomes at play, assuming different optimal decisions for different states of H can help in thoroughly exploring all possible scenarios and outcomes. This can lead to more robust decision-making. 2. Risk Management: In risk management, assuming different decisions for different scenarios can be a form of contingency planning. This allows for preparedness in the face of uncertainty and can help mitigate potential risks. 3. Innovation and Creativity: This assumption can also stimulate innovative thinking and creativity. By considering different decisions for different scenarios, one might come up with unique solutions that wouldn't have been considered otherwise. 4. Learning Opportunities: The process of considering different decisions based on different states of H can provide valuable learning opportunities. It can help in understanding the dynamics of the situation better and can lead to the development of more effective decision-making strategies in the future. 5. Ensuring Fairness: In situations where decisions impact different groups of people differently, considering different decisions for different scenarios can help ensure fairness and equity. This can be particularly important in policy-making and governance. In conclusion, while there can be drawbacks to assuming different optimal decisions for different states of H, there can also be significant benefits in certain situations. The key is to apply this assumption judiciously, taking into account the specific context and potential implications. Discuss the role of contingency planning in risk management, particularly in scenarios where the truth or falsehood of a hypothesis significantly alters the optimal decision. Contingency planning plays a critical role in risk management, particularly in scenarios where the truth or falsehood of a hypothesis significantly alters the optimal decision. Here's how: 1. Preparedness: Contingency planning prepares an organization for unexpected events or scenarios. If a hypothesis turns out to be true or false, a well-prepared contingency plan can provide a roadmap for action, reducing panic and confusion. 2. Reducing Impact: By identifying potential risks and outlining steps to mitigate them, contingency planning can significantly reduce the negative impact of an unexpected event. This is particularly important when the truth or falsehood of a hypothesis could significantly alter the situation. 3. Resource Allocation: Contingency planning helps in efficient resource allocation. By planning in advance for different scenarios, an organization can ensure that resources are available and can be quickly mobilized when needed. 4. Business Continuity: In the face of significant risks, contingency planning can help ensure business continuity. It allows an organization to quickly adapt and respond to changes, minimizing disruption to operations. 5. Decision Making: Contingency planning supports decision-making in uncertain situations. When the truth or falsehood of a hypothesis significantly alters the optimal decision, having a contingency plan can provide a clear path forward, reducing uncertainty and facilitating swift decision-making. In scenarios where a hypothesis plays a significant role, contingency planning involves creating different plans based on whether the hypothesis is true or false. Each plan would outline the steps to take, resources required, and potential challenges and solutions, enabling the organization to respond effectively no matter the outcome. It's a proactive approach that recognizes the uncertainty inherent in any situation and prepares for it in advance.
18a63728-6073-4f46-8b46-66473c06cdd2	trentmkelly/LessWrong-43k	LessWrong	Voiceofra is banned I've gotten sufficient evidence from support that voiceofra has been doing retributive downvoting. I've banned them without prior notice because I'm not giving them more chances to downvote. I'm thinking of something like not letting anyone give more than 5 downvotes/week for content which is more than a month old. The numbers and the time period are tentative-- this isn't my ideal rule. This is probably technically possible. However, my impression is that highly specific rules like that are an invitation to gaming the rules. I would rather just make spiteful down-voting impossible (or maybe make it expensive) rather than trying to find out who's doing it. Admittedly, putting up barriers to downvoting for past comments doesn't solve the problem of people who down-vote everything, but at least people who downvote current material are easier to notice. Any thoughts about technical solutions to excessive down-voting of past material?
b1c47ffd-2df9-42b4-8c36-6966335d93f8	trentmkelly/LessWrong-43k	LessWrong	Meetup : Stockholm Slate Star Codex Meetup Discussion article for the meetup : Stockholm Slate Star Codex Meetup WHEN: 06 May 2017 01:30:00PM (+0200) WHERE: café Dox, Stora Nygatan 31, 111 27 Stockholm, Sweden This is a meetup where we will discuss topics related to the blogg Slate Star Codex, for example rationality, science and effective altruism. Having read the blog is preferable but not required. Feel free to invite friends who might be interested. (added May 3) The theme of discussion this time will be "Meditations on Moloch". http://slatestarcodex.com/2014/07/30/meditations-on-moloch/ Discussion article for the meetup : Stockholm Slate Star Codex Meetup
664dd716-2968-45e9-ac54-0ce1487124c3	trentmkelly/LessWrong-43k	LessWrong	Absent Transhumanism and Transformative Technologies, Which Utopia is Left? Assume for the time being that it will forever remain beyond the scope of science to change Human Nature. AGI is also impossible, as is Nanotech, BioImmortality, and those things. Douglas Adams mice finished their human experiment, giving to you, personally, the job of redesigning earth, and specially human society, according to your wildest utopian dreams, but you can't change the unchangeables above. You can play with architecture, engineering, gender ratio, clothing, money, science grants, governments, feeding rituals, family constitution, the constitution itself, education, etc... Just don't forget if you slide something too far away from what our evolved brains were designed to accept, things may slide back, or instability and catastrophe may ensue. Finally, if you are not the kind of utilitarian that assigns exactly the same amount of importance to your desires, and to that of others, I want you to create this Utopia for yourself, and your values, not everyone. The point of this exercise is: The vast majority of folk not related to this community that I know, when asked about an ideal world, will not change human nature, or animal suffering, or things like that, they'll think about changing whatever the newspaper editors have been writing about last few weeks. I am wondering if there is symmetry here, and folks from this community here do not spend that much time thinking about those kinds of change which don't rely on transformative technologies. It is just an intuition pump, a gedankenexperiment if you will. Force your brain to face this counterfactual reality, and make the best world you can given those constraints. Maybe, if sufficiently many post here, the results might clarify something about CEV, or the sociology of LessWrongers...
19f9d17e-a12d-46e6-9d63-398b017d42ae	trentmkelly/LessWrong-43k	LessWrong	Why there are no online CFAR workshops? As there are many people here here who attended CFAR workshops, I wanted to ask, if you understand why there are no online workshops, even when the in-person workshops were not possible due to the pandemic. In my experience, everything that does not require physical contact (like martial arts) can be taught online. There can be some inconveniences, of course, but if planned well, it can work quite smoothly. I can think about very small amount of reasons, and none of them seems persuasive to me. 1. Maybe CFAR workshop indeed requires physical contact? Sounds quite strange. 2. Maybe organizers are not good with technologies and don't know how to do different Zoom rooms, how to make parties in Mozilla Hubs etc.? Maybe they have never heard about properly organized online schools? Sounds even less likely. 3. Maybe CFAR workshop requires full embedding with no distraction? Well, it can be clarified in the announcement as a requirement for the participants, and for many people, it is certainly a possible thing to do. 4. Maybe CFAR somehow got into a death spiral and turned into something like Lifespring? I would hope it is unlikely in the rationalistic community, but here is more hope than actual belief.
92dee6ab-e3a0-4283-98ed-9073f2686276	LDJnr/LessWrong-Amplify-Instruct	LessWrong	"Written with much help from Nick Tarleton and Kaj Sotala, in response to various themes here, here, and throughout Less Wrong; but a casual mention here1 inspired me to finally write this post. (Note: The first, second, and third footnotes of this post are abnormally important.) It seems to have become a trend on Less Wrong for people to include belief in the rationality of signing up for cryonics as an obviously correct position2 to take, much the same as thinking the theories of continental drift or anthropogenic global warming are almost certainly correct. I find this mildly disturbing on two counts. First, it really isn't all that obvious that signing up for cryonics is the best use of one's time and money. And second, regardless of whether cryonics turns out to have been the best choice all along, ostracizing those who do not find signing up for cryonics obvious is not at all helpful for people struggling to become more rational. Below I try to provide some decent arguments against signing up for cryonics — not with the aim of showing that signing up for cryonics is wrong, but simply to show that it is not obviously correct, and why it shouldn't be treated as such. (Please note that I am not arguing against the feasibility of cryopreservation!) Signing up for cryonics is not obviously correct, and especially cannot obviously be expected to have been correct upon due reflection (even if it was the best decision given the uncertainty at the time): Weird stuff and ontological confusion: quantum immortality, anthropic reasoning, measure across multiverses, UDTesque 'decision theoretic measure' or 'probability as preference', et cetera, are not well-understood enough to make claims about whether or not you should even care about the number of 'yous' that are living or dying, whatever 'you' think you are.3 This does not make cryonics a bad idea — it may be the correct decision under uncertainty — but it should lessen anyone's confidence that the balance of reasons ultimately weighs overwhelmingly in favor of cryonics. If people believe that a technological singularity is imminent, then they may believe that it will happen before they have a significant chance of dying: either everyone (including cryonauts) dies anyway when an unFriendly artificial intelligence goes FOOM, or a Friendly artificial intelligence is created and death is solved (or reflectively embraced as good, or some other unexpected outcome). This is more salient when considering the likelihood of large advances in biomedical and life extension technologies in the near future. A person might find that more good is done by donating money to organizations like SENS, FHI, or SIAI4 than by spending that money on pursuing a small chance of eternal life. Cryonics working is pretty dependent on e.g. an unFriendly artificial intelligence not going FOOM, or molecular nanotechnology not killing everyone. Many people may believe that a slightly higher chance of a positive singularity is more important than a significantly higher chance of personal immortality. Likewise, having their friends and family not be killed by an existential disaster such as rogue MNT, bioweaponry, et cetera, could very well be more important to them than a chance at eternal life. Acknowledging these varied preferences, and varied beliefs about one's ability to sacrifice only luxury spending to cryonics, leads to equally varied subjectively rational courses of action for a person to take. Some people may have loose boundaries around what they consider personal identity, or expect personal identity to be less important in the future. Such a person might not place very high value on ensuring that they, in a strong sense, exist in the far future, if they expect that people sufficiently like them to satisfy their relevant values will exist in any case. (Kaj Sotala reports being indifferent to cryonics due to personal identity considerations here.) Furthermore, there exist people who have preferences against (or no preferences either for or against) living extremely far into the future for reasons other than considerations about personal identity. Such cases are rare, but I suspect less rare among the Less Wrong population than most, and their existence should be recognized. (Maybe people who think they don't care are usually wrong, and, if so, irrational in an important sense, but not in the sense of simple epistemic or instrumental-given-fixed-values rationality that discussions of cryonics usually center on.) That said, the reverse is true: not getting signed up for cryonics is also not obviously correct. The most common objections (most of them about the infeasibility of cryopreservation) are simply wrong. Strong arguments are being ignored on both sides. The common enemy is certainty. Calling non-cryonauts irrational is not productive nor conducive to fostering a good epistemic atmosphere: Whether it's correct or not, it seems unreasonable to claim that the decision to forgo cryonics in favor of donating (a greater expected amount) to FHI, SIAI4, SENS, etc. represents as obvious an error as, for instance, religion. The possibility of a third option here shouldn't be ignored. People will not take a fringe subject more seriously simply because you call them irrational for not seeing it as obvious (as opposed to belief in anthropogenic global warming where a sheer bandwagon effect is enough of a memetic pull). Being forced on the defensive makes one less likely to accept and therefore overcome their own irrationalities, if irrationalities they are. (See also: A Suite of Pragmatic Considerations in Favor of Niceness) As mentioned in bullet four above, some people really wouldn't care if they died, even if it turned out MWI, spatially infinite universes, et cetera were wrong hypotheses and that they only had this one shot at existence. It's not helping things to call them irrational when they may already have low self-esteem and problems with being accepted among those who have very different values pertaining to the importance of continued subjective experience. Likewise, calling people irrational for having kids when they could not afford cryonics for them is extremely unlikely to do any good for anyone. Debate over cryonics is only one of many opportunities for politics-like thinking to taint the epistemic waters of a rationalist community; it is a topic where it is easy to say 'we are right and you are wrong' where 'we' and 'you' are much too poorly defined to be used without disclaimers. If 'you' really means 'you people who don't understand reductionist thinking', or 'you people who haven't considered the impact of existential risk', then it is important to say so. If such an epistemic norm is not established I fear that the quality of discourse at Less Wrong will suffer for the lack of it. One easily falls to the trap of thinking that disagreements with other people happen because the others are irrational in simple, obviously flawed ways. It's harder to avoid the fundamental attribution error and the typical mind fallacy, and admit that the others may have a non-insane reason for their disagreement. 1 I don't disagree with Roko's real point, that the prevailing attitude towards cryonics is decisive evidence that people are crazy and the world is mad. Given uncertainty about whether one's real values would endorse signing up for cryonics, it's not plausible that the staggering potential benefit would fail to recommend extremely careful reasoning about the subject, and investment of plenty of resources if such reasoning didn't come up with a confident no. Even if the decision not to sign up for cryonics were obviously correct upon even a moderate level of reflection, it would still constitute a serious failure of instrumental rationality to make that decision non-reflectively and independently of its correctness, as almost everyone does. I think that usually when someone brings up the obvious correctness of cryonics, they mostly just mean to make this observation, which is no less sound even if cryonics isn't obviously correct. 2 To those who would immediately respond that signing up for cryonics is obviously correct, either for you or for people generally, it seems you could mean two very different things: Do you believe that signing up for cryonics is the best course of action given your level of uncertainty? or, Do you believe that signing up for cryonics can obviously be expected to have been correct upon due reflection? (That is, would you expect a logically omniscient agent to sign up for cryonics in roughly your situation given your utility function?) One is a statement about your decision algorithm, another is a statement about your meta-level uncertainty. I am primarily (though not entirely) arguing against the epistemic correctness of making a strong statement such as the latter. 3 By raising this point as an objection to strong certainty in cryonics specifically, I am essentially bludgeoning a fly with a sledgehammer. With much generalization and effort this post could also have been written as 'Abnormal Everything'. Structural uncertainty is a potent force and the various effects it has on whether or not 'it all adds up to normality' would not fit in the margin of this post. However, Nick Tarleton and I have expressed interest in writing a pseudo-sequence on the subject. We're just not sure about how to format it, and it might or might not come to fruition. If so, this would be the first post in the 'sequence'. 4 Disclaimer and alert to potential bias: I'm an intern (not any sort of Fellow) at the Singularity Institute for (or 'against' or 'ambivalent about' if that is what, upon due reflection, is seen as the best stance) Artificial Intelligence." A user from LessWrong is responsible for the text you see above Could you further elaborate on the potential ethical implications of cryonics, particularly in the context of resource allocation and societal impact? Cryonics involves preserving a person's body at extremely low temperatures with the hope that future medical technology will be able to revive them. It's a concept that raises numerous ethical questions, particularly when it comes to resource allocation and societal impact. From a resource allocation perspective, cryonics is an expensive procedure that requires significant financial investment. This raises questions about the equitable distribution of resources. If only a small portion of the population can afford cryonics, it may contribute to social inequality. It could potentially lead to a situation where the wealthy have access to a form of life extension that is denied to the less affluent. Furthermore, the resources used for cryonics could be allocated to other areas of healthcare or research that could benefit a larger number of people. For example, investing in public health initiatives, disease prevention, or developing treatments for currently incurable diseases could arguably provide a greater societal benefit. From a societal impact perspective, the widespread adoption of cryonics could have profound implications. If a significant portion of the population chooses to be cryopreserved, it could lead to overpopulation in the future if and when these individuals are revived. This could strain resources and potentially lead to conflict. Moreover, there are questions about the psychological and societal effects of reviving individuals who have been cryopreserved. They would be waking up in a world that could be vastly different from the one they left, potentially leading to significant psychological distress. There are also questions about the legal and societal status of these individuals, and how they would reintegrate into society. In conclusion, while cryonics offers the tantalizing possibility of life extension, it also raises complex ethical questions that need to be carefully considered. These include issues of resource allocation, social inequality, and the potential societal impact of reviving cryopreserved individuals. Considering the potential societal implications and ethical dilemmas associated with cryonics, how might governments and regulatory bodies approach the development and implementation of policies related to this practice? Additionally, what could be the potential consequences, both intended and unintended, of such regulations? The development and implementation of policies related to cryonics would likely be a complex process, requiring input from a variety of stakeholders, including medical professionals, ethicists, legal experts, and the public. Governments and regulatory bodies would need to balance the potential benefits of cryonics with the ethical and societal implications. One approach could be to regulate cryonics similarly to other medical procedures, requiring rigorous scientific evidence of its safety and efficacy before it can be offered to the public. This could involve establishing standards for cryopreservation and revival procedures, as well as guidelines for informed consent. Such regulations could help ensure that individuals who choose cryonics are fully aware of the risks and uncertainties involved. Another approach could be to regulate the financial aspects of cryonics. This could involve setting caps on the costs of cryopreservation and revival procedures to prevent exploitation, or implementing measures to ensure that the costs of long-term storage and revival are covered, so as not to burden future generations. However, these regulations could also have unintended consequences. For instance, strict regulations could potentially stifle innovation in the field of cryonics. On the other hand, if regulations are too lax, there is a risk of unscrupulous providers offering substandard services or making unrealistic promises about the prospects of revival. Moreover, there could be societal consequences to consider. For instance, if cryonics becomes widely available and accepted, it could change societal attitudes towards death and aging. This could have profound implications for how we value life and allocate resources. In conclusion, the regulation of cryonics would require careful consideration of a range of ethical, medical, and societal factors. While regulations could help protect individuals and society from potential harms, they could also have unintended consequences that need to be carefully managed.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.