--- license: cc-by-nc-4.0 tags: - mergekit - merge base_model: - mistralai/Mixtral-8x7B-v0.1 - jondurbin/bagel-dpo-8x7b-v0.2 - Sao10K/Sensualize-Mixtral-bf16 - mistralai/Mixtral-8x7B-v0.1 - Doctor-Shotgun/limarp-zloss-mixtral-8x7b-qlora - mistralai/Mixtral-8x7B-Instruct-v0.1 model-index: - name: BagelMIsteryTour-v2-8x7B results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 72.7 name: normalized accuracy source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 87.36 name: normalized accuracy source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 71.16 name: accuracy source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 74.54 source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 82.64 name: accuracy source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 61.33 name: accuracy source: url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=ycros/BagelMIsteryTour-v2-8x7B name: Open LLM Leaderboard --- # BagelMIsteryTour-v2-8x7B [GGUF versions here](https://huggingface.co./ycros/BagelMIsteryTour-v2-8x7B-GGUF) [AWQ versions here](https://huggingface.co./ycros/BagelMIsteryTour-v2-8x7B-AWQ) Bagel, Mixtral Instruct, with extra spices. Give it a taste. Works with Alpaca prompt formats, though the Mistral format should also work. ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63044fa07373aacccd8a7c53/lxNMzXo_dq_JCP9YyUyaw.jpeg) I started experimenting around seeing if I could improve or fix some of Bagel's problems. Totally inspired by seeing how well Doctor-Shotgun's Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss worked (which is a LimaRP tune on top of base Mixtral, and then merged with Mixtral Instruct) - I decided to try some merges of Bagel with Mixtral Instruct as a result. Somehow I ended up here, Bagel, Mixtral Instruct, a little bit of LimaRP, a little bit of Sao10K's Sensualize. So far in my testing it's working very well, and while it seems fairly unaligned on a lot of stuff, it's maybe a little too aligned on a few specific things (which I think comes from Sensualize) - so that's something to play with in the future, or maybe try to DPO out. I've been running (temp last) minP 0.1, dynatemp 0.5-4, rep pen 1.07, rep range 1024. I've been testing Alpaca style Instruction/Response, and Instruction/Input/Response and those seem to work well, I expect Mistral's prompt format would also work well. You may need to add a stopping string on "{{char}}:" for RPs because it can sometimes duplicate those out in responses and waffle on. Seems to hold up and not fall apart at long contexts like Bagel and some other Mixtral tunes seem to, definitely doesn't seem prone to loopyness either. Can be pushed into extravagant prose if the scene/setting calls for it. __Version 2:__ lowered the mix of Sensualize. This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). ## Merge Details ### Merge Method This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co./mistralai/Mixtral-8x7B-v0.1) as a base. ### Models Merged The following models were included in the merge: * [jondurbin/bagel-dpo-8x7b-v0.2](https://huggingface.co./jondurbin/bagel-dpo-8x7b-v0.2) * [Sao10K/Sensualize-Mixtral-bf16](https://huggingface.co./Sao10K/Sensualize-Mixtral-bf16) * [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co./mistralai/Mixtral-8x7B-v0.1) + [Doctor-Shotgun/limarp-zloss-mixtral-8x7b-qlora](https://huggingface.co./Doctor-Shotgun/limarp-zloss-mixtral-8x7b-qlora) * [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co./mistralai/Mixtral-8x7B-Instruct-v0.1) ### Configuration The following YAML configuration was used to produce this model: ```yaml base_model: mistralai/Mixtral-8x7B-v0.1 models: - model: mistralai/Mixtral-8x7B-v0.1+Doctor-Shotgun/limarp-zloss-mixtral-8x7b-qlora parameters: density: 0.5 weight: 0.2 - model: Sao10K/Sensualize-Mixtral-bf16 parameters: density: 0.5 weight: 0.1 - model: mistralai/Mixtral-8x7B-Instruct-v0.1 parameters: density: 0.6 weight: 1.0 - model: jondurbin/bagel-dpo-8x7b-v0.2 parameters: density: 0.6 weight: 0.5 merge_method: dare_ties dtype: bfloat16 ``` # [Open LLM Leaderboard Evaluation Results](https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard) Detailed results can be found [here](https://huggingface.co./datasets/open-llm-leaderboard/details_ycros__BagelMIsteryTour-v2-8x7B) | Metric |Value| |---------------------------------|----:| |Avg. |74.95| |AI2 Reasoning Challenge (25-Shot)|72.70| |HellaSwag (10-Shot) |87.36| |MMLU (5-Shot) |71.16| |TruthfulQA (0-shot) |74.54| |Winogrande (5-shot) |82.64| |GSM8k (5-shot) |61.33|