Wow, that's an old model. Great that it works for you, but have you tried some more modern ones? They're generally considered a lot more capable at the same size
theterrasque
joined 1 year ago
Increase context length, probably enable flash attention in ollama too. Llama3.1 support up to 128k context length, for example. That's in tokens and a token is on average a bit under 4 letters.
Note that higher context length requires more ram and it's slower, so you ideally want to find a sweet spot for your use and hardware. Flash attention makes this more efficient
Oh, and the model needs to have been trained at larger contexts, otherwise it tends to handle it poorly. So you should check what max length the model you want to use was trained to handle
Like when under Arab spring the Egyptian politicians tried to get the military involved to stop the protests, and got back (paraphrased)
"Our primary job is to protect the Egyptian people from violence. You really don't want us involved in this"
Sounds a bit like worldwar series by Harry Turtledove
If I go to a restaurant and order risotto, I haven't made the dish, I've only consumed it. I want you to focus on that word "consume", it's important here.
If I buy a bread at the bakery, ham and cheese in the grocery store, and make me a sandwich, who's the creator?
Hmm.. what about pendulum painting? Where you put paint in a bucket, put a hole in it, and let it swing back and forth over the canvas?
On one side he chooses paint and size of hole and initial path and so on, but on the other hand he let nature and physics do the actual painting for him.
AI can be art. And you're like the people criticizing the first photographers saying what they did wasn't art. This is what I think.
And it’s going to have to be okay.
And woman a combatant factory?
What do you think is “weight”?
You can call that confidence if you want, but it got very little to do with how "sure" the model is.
It just has to stop the process if the statistics don’t not provide enough to continue with confidence. If the data is all over the place and you have several “The capital of France is Berlin/Madrid/Milan”, it’s measurable compared to all data saying it is Paris. Not need for any kind of “understanding” of the meaning of the individual words, just measuring confidence on what next word should be.
Actually, it would be "The confidence of token Th is 0.95, the confidence of S is 0.32, the confidence of ... " and so on for each possible token, many llm's have around 16k-32k token vocabulary. Most will be at or near 0. So you pick Th, and then token "e" will probably be very high next, then a space token, then.. Anyway, the confidence of the word "Paris" won't be until far into the generation.
Now there is some overseeing logic in a way, if you ask what the capitol of a non existent country is it'll say there's no such country, but is that because it understands it doesn't know, or the training data has enough examples of such that it has the statistical data for writing out such an answer?
IDK what did you do, but slm don’t really hallucinate that much, if at all.
I assume by SLM you mean smaller LLM's like for example mistral 7b and llama3.1 8b? Well those were the kind of models I did try for local RAG.
Well, it was before llama3, but I remember trying mistral, mixtral, llama2 70b, command-r, phi, vicuna, yi, and a few others. They all made mistakes.
I especially remember one case where a product manual had this text : "If the same or a newer version of is already installed on the computer, then the installation will be aborted, and the currently installed version will be maintained" and the question was "What happens if an older version of is already installed?" and every local model answered that then that version will be kept and the installation will be aborted.
When trying with OpenAI's latest model at that time, I think 4, it got it right. In general, about 1 in ~5-7 answers to RAG backed questions were wrong, depending on the model and type of question. I could usually reword the question to get the correct answer, but to do that you kinda already have to know the answer is wrong. Which defeats the whole point of it.
Temperature 0 is never used
It is in some cases, where you want a deterministic / "best" response. Seen it used in benchmarks, or when doing some "Is this comment X?" where X is positive, negative, spam, and so on. You don't want the model to get creative there, but rather answer consistently and always the most likely path.
https://learnprompting.org/docs/intermediate/chain_of_thought
It's suspected to be one of the reasons why Claude and OpenAI's new o1 model is so good at reasoning compared to other llm's.
It can sometimes notice hallucinations and adjust itself, but there's also been examples where the CoT reasoning itself introduce hallucinations and makes it throw away correct answers. So it's not perfect. Overall a big improvement though.
No, all sizes of llama 3.1 should be able to handle the same size context. The difference would be in the "smarts" of the model. Bigger models are better at reading between the lines and higher level understanding and reasoning.