sentence-transformers How to use kaamd/ll with sentence-transformers:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("kaamd/ll")
sentences = [
"Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery: Does borrowing from my 401(k) make sense in my specific circumstance?",
"I'm not sure why you think that it matters that the distribution goes to an S-Corp vs an individual tax payer. You seem to think it has any relevance to your question, but it doesn't. It only confuses your readers. The situation is like this: LLC X is deriving income in State #2. It has two members (I and S) residents of State #1. Members I and S pay all their taxes to State #1, and don't pay taxes to State #2. State #2 audited member I and that member now needs to pay back taxes and penalties to State #2 on income derived from that State. Your question: Does that mean that member S should be worried, since that member was essentially doing the exact same thing as member I? My answer: Yes.",
"Interest rates are market driven. They tend to be based on the prime rate set by the federal reserve bank because of the tremendous lending capacity of that institution and that other loan originators will often fund their own lending (at least in part) with fed loans. However, there is no mandatory link between the federal reserve rate and the market rate. No law stipulates that rates cannot rise or fall. They will rise and fall as lenders see necessary to use their capital. Though a lender asking 10% interest might make no loans when others are willing to lend for 9%. The only protection you have is that we are (mostly) economically free. As a borrower, you are protected by the fact that there are many lenders. Likewise, as a lender, because there are many borrowers. Stability is simply by virtue of the fact that one market participant with inordinate pricing will find fewer counterparties to transact.",
"\"You're getting great wisdom and options. Establishing your actionable path will require the details that only you know, such as how much is actually in each paycheck (and how much tax is withheld), how much do you spend each month (and yearly expenses too), how much spending can you actually cut or replace, how comfortable are you with considering (or not considering) unexpected/emergency spending. You mentioned you were cash-poor, but only you know what your current account balances are, which will affect your actions and priorities. Btw, interestingly, your \"\"increase 401k contributions by 2% each year\"\" will need to end before hitting the $18K contribution limit. I took some time and added the details you posted into a cash-flow program to see your scenario over the next few years. There isn't a \"\"401k loan\"\" activity in this program yet, so I build the scenario from other simple activities. You seem financially minded enough to continue modeling on your own. I'm posting the more difficult one for you (borrow from 401k), but you'll have to input your actual balances, paycheck and spending. My spending assumptions must be low, and I entered $70K as \"\"take-home,\"\" so the model looks like you've got lots of cash. If you choose to play with it, then consider modeling some other scenarios from the advice in the other posts. Here's the \"\"Borrow $6500 from 401k\"\" scenario model at Whatll.Be: https://whatll.be/d1x1ndp26i/2 To me, it's all about trying the scenarios and see which one seems to work with all of the details. The trick is knowing what scenarios to try, and how to model them. Full disclosure: I needed to do similar planning, so I wrote Whatll.Be and I now share it with other people. It's in beta, so I'm testing it with scenarios like yours. (Notice most of the extra activity occurs on 2018-Jan-01)\""
]
embeddings = model.encode(sentences)
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4] 
