llama.cpp for fun and (maybe) profit Citibank 2024-03 talk @IanOzsvald – ianozsvald.com

Strategist/Trainer/Speaker/Author 20+ years Interim Chief Data Scientist By [ian]@ianozsvald[.com] Ian Ozsvald 2nd Edition!

Engineering & Testing for Pythonic Research Fast Pandas & Higher Performance Python Successful Data Science Projects You – post in Chat if you’re using LLMs? Ian Ozsvald By [ian]@ianozsvald[.com] Ian Ozsvald

No need for a GPU+VRAM Llama.cpp runs on CPU+RAM Nothing sent off your machine llama.cpp By [ian]@ianozsvald[.com] Ian Ozsvald X

Experiment with models as they’re published Use client data/src code – no data sent off your machine Why use local models? By [ian]@ianozsvald[.com] Ian Ozsvald

Prototype ideas! By [ian]@ianozsvald[.com] Ian Ozsvald llama-2-7b-chat.Q5_K_M.gguf 5GB on disk and in RAM, near real time

See the wackyness early on What’s your strategy to catch varied outputs? Why use local models? By [ian]@ianozsvald[.com] Ian Ozsvald

MS Phi2 can “reason” (IS IT RIGHT?) By [ian]@ianozsvald[.com] Ian Ozsvald I had confident answers: 125.2m/s (good Python) 17.2m/s (partial Python with comments that had mistakes), 40m/s and 31.3m/s (as teacher) Which one to believe? My model is quantised (Q5) but random variation exists anyway… The MS post didn’t disclose the prompt they used https://www.microsoft.com/en-us/research/blog/phi-2-the-surprising-power-of-small-language-models/

Similar to JPG compression Shrink the trained model 32→16→8→7/6/5/4/3/2 bits Fewer bits→worse text completion “Q5 generally an acceptable level” Quantisation By [ian]@ianozsvald[.com] Ian Ozsvald

Experiment with multi-modal e.g. OCR and checking photo meets requirements What about image queries? By [ian]@ianozsvald[.com] Ian Ozsvald

Llava multi-modal Extract facts from images? By [ian]@ianozsvald[.com] Ian Ozsvald llava-v1.5-7b-Q4_K.gguf 4GB on disk & RAM 5s for example llama.cpp provides ./server

Trial code-support Code review? “Is this test readable?” Can they help with coding? By [ian]@ianozsvald[.com] Ian Ozsvald

Can you explain this function please? By [ian]@ianozsvald[.com] Ian Ozsvald codellama-34b-python.Q5_K_M.gguf 23GB on disk & RAM, 30s for example Can we use this as a “code reviewer” for internal code? codellama answer: “The function test_uniform_distribution creates a list of 10 zeros, then increments the position in that list indicated by the murmurhash3_32() digest of i. It does this 100000 times and then checks if the means of those incremented values are uniformly distributed (i.e., if they're all roughly the same).” (surprisingly clear!) https://github.com/scikit-learn/scikit-learn/blob/main/sklearn/utils/tests/test_murmurhash.py

Give test functions (e.g. Pandas) to codellama Ask it “is this a good test function?” Try to get it to propose new test functions Check using pytest and coverage tools Shortcut human effort at project maintenance? My experiment for code assist By [ian]@ianozsvald[.com] Ian Ozsvald

Run quantised models on client data locally Experience the wackyness – mitigation? Use Python API to see tokens+perplexity+more Why try llama.cpp? By [ian]@ianozsvald[.com] Ian Ozsvald

Let me know if training/strategy chats useful? Discuss: How do we measure correctness? What’s the worst (!) that could go wrong with your projects? Summary By [ian]@ianozsvald[.com] Ian Ozsvald

Appendix – Ask Mixtral to challenge my Monte Carlo estimation approach By [ian]@ianozsvald[.com] Ian Ozsvald Mixtral gave 5 points and some items I should be careful about, ChatGPT 3.5 gave 7 points, both felt similar

Appendix – attack via asciiart By [ian]@ianozsvald[.com] Ian Ozsvald

WizardCoder is good (tuned llama2) By [ian]@ianozsvald[.com] Ian Ozsvald wizardcoder-python-34b -v1.0.Q5_K_S.gguf 22GB on disk & RAM 15s for example You can replace CoPilot with this for completions

Quantisation By [ian]@ianozsvald[.com] Ian Ozsvald Original fp16 models Better Bigger models with higher quantisation still has lower perplexity than simpler, less quantised models Choose the biggest you can K-quants PR https://github.com/ggerganov/llama.cpp/pull/1684