Quant for 8.0

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Image+Title.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -14,62 +14,93 @@ pipeline_tag: text-generation
 tags:
 - medical
 - biology
-quantized_by: bartowski
 ---
 
-## Exllama v2 Quantizations of BioMistral-7B
 
-Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.14">turboderp's ExLlamaV2 v0.0.14</a> for quantization.
+<p align="center">
+  <img src="https://huggingface.co/BioMistral/BioMistral-7B/resolve/main/wordart_blue_m_rectangle.png?download=true" alt="drawing" width="250"/>
+</p>
 
-<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
+# BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains
 
-Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
+**Abstract:**
 
-Original model: https://huggingface.co/BioMistral/BioMistral-7B
+Large Language Models (LLMs) have demonstrated remarkable versatility in recent years, offering potential applications across specialized domains such as healthcare and medicine. Despite the availability of various open-source LLMs tailored for health contexts, adapting general-purpose LLMs to the medical domain presents significant challenges.
+In this paper, we introduce BioMistral, an open-source LLM tailored for the biomedical domain, utilizing Mistral as its foundation model and further pre-trained on PubMed Central. We conduct a comprehensive evaluation of BioMistral on a benchmark comprising 10 established medical question-answering (QA) tasks in English. We also explore lightweight models obtained through quantization and model merging approaches. Our results demonstrate BioMistral's superior performance compared to existing open-source medical models and its competitive edge against proprietary counterparts. Finally, to address the limited availability of data beyond English and to assess the multilingual generalization of medical LLMs, we automatically translated and evaluated this benchmark into 7 other languages. This marks the first large-scale multilingual evaluation of LLMs in the medical domain. Datasets, multilingual evaluation benchmarks, scripts, and all the models obtained during our experiments are freely released.
 
-| Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
-| ----- | ---- | ------- | ------ | ------ | ------ | ------------ |
-| [8_0](https://huggingface.co/bartowski/BioMistral-7B-exl2/tree/8_0) | 8.0 | 8.0 | 8.4 GB | 9.8 GB | 11.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
-| [6_5](https://huggingface.co/bartowski/BioMistral-7B-exl2/tree/6_5) | 6.5 | 8.0 | 7.2 GB | 8.6 GB | 10.6 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. |
-| [5_0](https://huggingface.co/bartowski/BioMistral-7B-exl2/tree/5_0) | 5.0 | 6.0 | 6.0 GB | 7.4 GB |  9.4 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
-| [4_25](https://huggingface.co/bartowski/BioMistral-7B-exl2/tree/4_25) | 4.25 | 6.0 | 5.3 GB | 6.7 GB | 8.7 GB | GPTQ equivalent bits per weight, slightly higher quality. |
-| [3_5](https://huggingface.co/bartowski/BioMistral-7B-exl2/tree/3_5) | 3.5 | 6.0 | 4.7 GB | 6.1 GB | 8.1 GB | Lower quality, only use if you have to. |
+**Advisory Notice!** Although BioMistral is intended to encapsulate medical knowledge sourced from high-quality evidence, it hasn't been tailored to effectively, safely, or suitably convey this knowledge within professional parameters for action. We advise refraining from utilizing BioMistral in medical contexts unless it undergoes thorough alignment with specific use cases and undergoes further testing, notably including randomized controlled trials in real-world medical environments. BioMistral 7B may possess inherent risks and biases that have not yet been thoroughly assessed. Additionally, the model's performance has not been evaluated in real-world clinical settings. Consequently, we recommend using BioMistral 7B strictly as a research tool and advise against deploying it in production environments for natural language generation or any professional health and medical purposes.
 
-## Download instructions
+# 1. BioMistral models
 
-With git:
+**BioMistral** is a suite of Mistral-based further pre-trained open source models suited for the medical domains and pre-trained using textual data from PubMed Central Open Access (CC0, CC BY, CC BY-SA, and CC BY-ND). All the models are trained using the CNRS (French National Centre for Scientific Research) [Jean Zay](http://www.idris.fr/jean-zay/) French HPC.
 
-```shell
-git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/BioMistral-7B-exl2 BioMistral-7B-exl2-6_5
-```
+|      Model Name     |             Base Model             |      Model Type     | Sequence Length |                          Download                          |
+|:-------------------:|:----------------------------------:|:-------------------:|:---------------:|:-----------------------------------------------------:|
+|    BioMistral-7B    | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) | Further Pre-trained |       2048      |    [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B)    |
+|  BioMistral-7B-DARE | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) |      Merge DARE     |       2048      |  [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-DARE) |
+|  BioMistral-7B-TIES | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) |      Merge TIES     |       2048      |  [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-TIES) |
+| BioMistral-7B-SLERP | [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) |     Merge SLERP     |       2048      | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-SLERP) |
 
-With huggingface hub (credit to TheBloke for instructions):
+# 2. Quantized Models
 
-```shell
-pip3 install huggingface-hub
-```
+|      Base Model     | Method | q_group_size | w_bit | version | VRAM GB |  Time  | Download |
+|:-------------------:|:------:|:------------:|:-----:|:-------:|:-------:|:------:|:--------:|
+|    BioMistral-7B    | FP16/BF16       |              |   |       |  15.02  |  x1.00 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B)         |
+|    BioMistral-7B    |   AWQ  |      128     |   4   |   GEMM  |   4.68  |  x1.41 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-AWQ-QGS128-W4-GEMM)         |
+|    BioMistral-7B    |   AWQ  |      128     |   4   |   GEMV  |   4.68  | x10.30 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-AWQ-QGS128-W4-GEMV)         |
+|    BioMistral-7B    |  BnB.4 |              |   4   |         |   5.03  |  x3.25 | [HuggingFace](blank)         |
+|    BioMistral-7B    |  BnB.8 |              |   8   |         |   8.04  |  x4.34 | [HuggingFace](blank)         |
+|  BioMistral-7B-DARE |   AWQ  |      128     |   4   |   GEMM  |   4.68  |  x1.41 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-DARE-AWQ-QGS128-W4-GEMM)         |
+|  BioMistral-7B-TIES |   AWQ  |      128     |   4   |   GEMM  |   4.68  |  x1.41 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-TIES-AWQ-QGS128-W4-GEMM)         |
+| BioMistral-7B-SLERP |   AWQ  |      128     |   4   |   GEMM  |   4.68  |  x1.41 | [HuggingFace](https://huggingface.co/BioMistral/BioMistral-7B-SLERP-AWQ-QGS128-W4-GEMM)         |
 
-To download the `main` (only useful if you only care about measurement.json) branch to a folder called `BioMistral-7B-exl2`:
+# 2. Using BioMistral
 
-```shell
-mkdir BioMistral-7B-exl2
-huggingface-cli download bartowski/BioMistral-7B-exl2 --local-dir BioMistral-7B-exl2 --local-dir-use-symlinks False
-```
+You can use BioMistral with [Hugging Face's Transformers library](https://github.com/huggingface/transformers) as follow.
 
-To download from a different branch, add the `--revision` parameter:
+Loading the model and tokenizer :
 
-Linux:
+```python
+from transformers import AutoModel, AutoTokenizer
 
-```shell
-mkdir BioMistral-7B-exl2-6_5
-huggingface-cli download bartowski/BioMistral-7B-exl2 --revision 6_5 --local-dir BioMistral-7B-exl2-6_5 --local-dir-use-symlinks False
+tokenizer = AutoTokenizer.from_pretrained("BioMistral/BioMistral-7B")
+model = AutoModel.from_pretrained("BioMistral/BioMistral-7B")
 ```
 
-Windows (which apparently doesn't like _ in folders sometimes?):
-
-```shell
-mkdir BioMistral-7B-exl2-6.5
-huggingface-cli download bartowski/BioMistral-7B-exl2 --revision 6_5 --local-dir BioMistral-7B-exl2-6.5 --local-dir-use-symlinks False
+# 3. Supervised Fine-tuning Benchmark
+
+| | Clinical KG | Medical Genetics | Anatomy | Pro Medicine | College Biology | College Medicine | MedQA | MedQA 5 opts | PubMedQA | MedMCQA | Avg. |
+|-------------------------------------------|:---------------------------------------------:|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|-----------------------------------------------|------------------|
+| **BioMistral 7B** | 59.9 | 64.0 | 56.5 | 60.4 | 59.0 | 54.7 | 50.6 | 42.8 | 77.5 | 48.1 | 57.3 |
+| **Mistral 7B Instruct** | **62.9** | 57.0 | 55.6 | 59.4 | 62.5 | <u>57.2</u> | 42.0 | 40.9 | 75.7 | 46.1 | 55.9 |
+| | | | | | | | | | | | |
+| **BioMistral 7B Ensemble** | <u>62.8</u> | 62.7 | <u>57.5</u> | **63.5** | 64.3 | 55.7 | 50.6 | 43.6 | 77.5 | **48.8** | 58.7 |
+| **BioMistral 7B DARE** | 62.3 | **67.0** | 55.8 | 61.4 | **66.9** | **58.0** | **51.1** | **45.2** | <u>77.7</u> | <u>48.7</u> | **59.4** |
+| **BioMistral 7B TIES** | 60.1 | <u>65.0</u> | **58.5** | 60.5 | 60.4 | 56.5 | 49.5 | 43.2 | 77.5 | 48.1 | 57.9 |
+| **BioMistral 7B SLERP** | 62.5 | 64.7 | 55.8 | <u>62.7</u> | <u>64.8</u> | 56.3 | <u>50.8</u> | <u>44.3</u> | **77.8** | 48.6 | <u>58.8</u> |
+| | | | | | | | | | | | |
+| **MedAlpaca 7B** | 53.1 | 58.0 | 54.1 | 58.8 | 58.1 | 48.6 | 40.1 | 33.7 | 73.6 | 37.0 | 51.5 |
+| **PMC-LLaMA 7B** | 24.5 | 27.7 | 35.3 | 17.4 | 30.3 | 23.3 | 25.5 | 20.2 | 72.9 | 26.6 | 30.4 |
+| **MediTron-7B** | 41.6 | 50.3 | 46.4 | 27.9 | 44.4 | 30.8 | 41.6 | 28.1 | 74.9 | 41.3 | 42.7 |
+| **BioMedGPT-LM-7B** | 51.4 | 52.0 | 49.4 | 53.3 | 50.7 | 49.1 | 42.5 | 33.9 | 76.8 | 37.6 | 49.7 |
+| | | | | | | | | | | | |
+| **GPT-3.5 Turbo 1106*** | 74.71 | 74.00 | 65.92 | 72.79 | 72.91 | 64.73 | 57.71 | 50.82 | 72.66 | 53.79 | 66.0 |
+
+Supervised Fine-Tuning (SFT) performance of BioMistral 7B models compared to baselines, measured by accuracy (↑) and averaged across 3 random seeds of 3-shot. DARE, TIES, and SLERP are model merging strategies that combine BioMistral 7B and Mistral 7B Instruct. Best model in bold, and second-best underlined. *GPT-3.5 Turbo performances are reported from the 3-shot results without SFT.
+
+# Citation BibTeX
+
+Arxiv : [https://arxiv.org/abs/2402.10373](https://arxiv.org/abs/2402.10373)
+
+```bibtex
+@misc{labrak2024biomistral,
+      title={BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains}, 
+      author={Yanis Labrak and Adrien Bazoge and Emmanuel Morin and Pierre-Antoine Gourraud and Mickael Rouvier and Richard Dufour},
+      year={2024},
+      eprint={2402.10373},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
 ```
 
-Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
+**CAUTION!** Both direct and downstream users need to be informed about the risks, biases, and constraints inherent in the model. While the model can produce natural language text, our exploration of its capabilities and limitations is just beginning. In fields such as medicine, comprehending these limitations is crucial. Hence, we strongly advise against deploying this model for natural language generation in production or for professional tasks in the realm of health and medicine.

config.json

@@ -0,0 +1,24 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.35.0",
+  "use_cache": false,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.35.0"
+}

original_repo_url.txt

@@ -0,0 +1 @@
+https://huggingface.co/BioMistral/BioMistral-7B

output.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:117904190976f810db34b4afb0984422ea99eadff4ee8a570d79c4e329663e73
+size 7371402536

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token + ' ' }}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}