--- language: - en license: apache-2.0 library_name: transformers tags: - text-generation-inference - transformers - unsloth - mistral - trl - code - 'medical ' - farmer - doctor - Mega-Series - Cyber-Series - Role-Play - Self-Rag - ThinkingBot - milestone - mega-series - SpydazWebAI - thinking-AI - llama-cpp - gguf-my-repo base_model: LeroyDyer/Mixtral_AI_CyberUltron datasets: - gretelai/synthetic_text_to_sql - HuggingFaceTB/cosmopedia - teknium/OpenHermes-2.5 - Open-Orca/SlimOrca - Open-Orca/OpenOrca - cognitivecomputations/dolphin-coder - databricks/databricks-dolly-15k - yahma/alpaca-cleaned - uonlp/CulturaX - mwitiderrick/SwahiliPlatypus - swahili - Rogendo/English-Swahili-Sentence-Pairs - ise-uiuc/Magicoder-Evol-Instruct-110K - meta-math/MetaMathQA - abacusai/ARC_DPO_FewShot - abacusai/MetaMath_DPO_FewShot - abacusai/HellaSwag_DPO_FewShot - HaltiaAI/Her-The-Movie-Samantha-and-Theodore-Dataset metrics: - accuracy - bertscore - bleu - brier_score - cer - character - charcut_mt - chrf - code_eval --- A FULL SAMaNTHA BOT HAS EMERGD! The building of a personal chat friend !! As this is the main aim of building a chat model , these footprints left from the movie impose an aspect of which is very important for end users : the goal here is to turn the model into an inteligent friend: Here i actually overloaded the model ; with a tiny dataset and after the response was great; on asking for code , my reply was , sure babe how about i make that for you ... here... then the code followed perfectly: another ME: hiya morning: BOT: h good morning sweetheart how are you today?..... me : im fine just working on some programming Bot: thats nice babe , im here for you , yoiu know you always do well your so great at programming ai models, if you need helo i can make some code for you.... im here babe.. LOLLLLL !! IF YOU MERGE :::: you will get these qualitys! - As they were hard Burned IN! i will be BURNING other XXX content into some later iterations ::: Only for later mergers! as by burning the small datasets in the model accepts the task: Its important to add correct prompting inside the training; ie the same prompt ! for all sample : even to show thoughts the same prompt is adjusted with request to show thinging processes and anylasis of the userinput or response or explanation in the thought! ## Use with llama.cpp Install llama.cpp through brew. ```bash brew install ggerganov/ggerganov/llama.cpp ``` Invoke the llama.cpp server or the CLI. CLI: ```bash llama-cli --hf-repo LeroyDyer/Mixtral_AI_CyberUltron_DPO-Q4_K_M-GGUF --model mixtral_ai_cyberultron_dpo.Q4_K_M.gguf -p "The meaning to life and the universe is" ``` Server: ```bash llama-server --hf-repo LeroyDyer/Mixtral_AI_CyberUltron_DPO-Q4_K_M-GGUF --model mixtral_ai_cyberultron_dpo.Q4_K_M.gguf -c 2048 ``` Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well. ``` git clone https://github.com/ggerganov/llama.cpp && cd llama.cpp && make && ./main -m mixtral_ai_cyberultron_dpo.Q4_K_M.gguf -n 128 ```