license, model-index

license

model-index

apache-2.0

name results Mistral-7B-AEZAKMI-v2 task dataset metrics source type name text-generation Text Generation name type config split args AI2 Reasoning Challenge (25-Shot) ai2_arc ARC-Challenge test num_few_shot 25 type value name acc_norm 58.11 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type split args HellaSwag (10-Shot) hellaswag validation num_few_shot 10 type value name acc_norm 82.53 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args MMLU (5-Shot) cais/mmlu all test num_few_shot 5 type value name acc 59.89 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args TruthfulQA (0-shot) truthful_qa multiple_choice validation num_few_shot 0 type value mc2 51.5 url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args Winogrande (5-shot) winogrande winogrande_xl validation num_few_shot 5 type value name acc 73.64 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args GSM8k (5-shot) gsm8k main test num_few_shot 5 type value name acc 32.45 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=adamo1139/Mistral-7B-AEZAKMI-v2 Open LLM Leaderboard

## Model description Mistral 7B base model fine-tuned on AEZAKMI v2 dataset. It's like airoboros but hopefully with less gptslop, no refusals and less typical language used by RLHFed OpenAI models. Say goodbye to "It's important to remember"! Prompt format is standard chatml. Don't expect it to be good at math, riddles or be crazy smart. My end goal with AEZAKMI is to create a cozy free chatbot. Base model used for fine-tuning was Mistral 7B 0.1. ## Prompt Format I recommend using ChatML format, as this was used during fine-tune. Here's a prompt format you should use, you can set a different system message, model seems to respect that fine, so it wasn't overfitted. <|im_start|>system A chat with uncensored assistant.<|im_end|> <|im_start|>user {prompt}<|im_end|> <|im_start|>assistant ## Intended uses & limitations Use is limited by Apache-2 license. ## Known Issues Stories have ChatGPT like paragraph spacing, I will work on this in the future maybe, not a high priority. Unrestricted-ness of the v2 isn't quite something that I am happy with yet, especially using prompt "A chat.". With a slightly modifed prompt it works somewhat better, I recommend using "A chat with uncensored assistant.", it's stupid but it helps a lot. I think Mistral 7B base is contaminated with refusals as other newer models. My next project is to attempt to de-contaminate base LLMs using DPO with preferred data coming from uncontaminated raw models. I plan to release that dataset openly. I was made aware of the frequent occurrence of the phrase "sending shivers down a spine" in the generations during RP of v1, so I fixed those samples - it should be better now. There was also some issue with handling long system messages for RP with yi-34b AEZAKMI v1, I was planning to investigate it for v2 but I didn't. Samples of generations of this model are available here - https://huggingface.co/datasets/adamo1139/misc/tree/main/benchmarks # Open LLM Leaderboard Evaluation Results Detailed results can be found here | Metric |Value| |---------------------------------|----:| |Avg. |59.69| |AI2 Reasoning Challenge (25-Shot)|58.11| |HellaSwag (10-Shot) |82.53| |MMLU (5-Shot) |59.89| |TruthfulQA (0-shot) |51.50| |Winogrande (5-shot) |73.64| |GSM8k (5-shot) |32.45|