Upload 12 files

copied another project for testing

README.md

@@ -1,204 +1,222 @@
 ---
+language:
+- fr
+- it
+- de
+- es
+- en
 license: apache-2.0
-datasets:
-- HuggingFaceTB/cosmopedia
-metrics:
-- accuracy
-pipeline_tag: question-answering
 tags:
-- medical
+- moe
+model-index:
+- name: Mixtral-8x22B-v0.1
+  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: 70.48
+      name: normalized accuracy
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      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: 88.73
+      name: normalized accuracy
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      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: 77.81
+      name: accuracy
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      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: 51.08
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      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: 84.53
+      name: accuracy
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      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: 74.15
+      name: accuracy
+    source:
+      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=mistral-community/Mixtral-8x22B-v0.1
+      name: Open LLM Leaderboard
 ---
-# Model Card for Model ID
+# Mixtral-8x22B
 
-<!-- Provide a quick summary of what the model is/does. -->
+> [!TIP]
+> MistralAI has uploaded weights to their organization at [mistralai/Mixtral-8x22B-v0.1](https://huggingface.co/mistralai/Mixtral-8x22B-v0.1) and [mistralai/Mixtral-8x22B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1) too. 
 
-This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
+> [!TIP]
+> Kudos to [@v2ray](https://huggingface.co/v2ray) for converting the checkpoints and uploading them in `transformers` compatible format. Go give them a follow!
 
-## Model Details
+Converted to HuggingFace Transformers format using the script [here](https://huggingface.co/v2ray/Mixtral-8x22B-v0.1/blob/main/convert.py).
 
-### Model Description
+The Mixtral-8x22B Large Language Model (LLM) is a pretrained generative Sparse Mixture of Experts.
+## Run the model
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-<!-- Provide a longer summary of what this model is. -->
+model_id = "mistral-community/Mixtral-8x22B-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
 
+model = AutoModelForCausalLM.from_pretrained(model_id)
 
+text = "Hello my name is"
+inputs = tokenizer(text, return_tensors="pt")
 
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+By default, transformers will load the model in full precision. Therefore you might be interested to further reduce down the memory requirements to run the model through the optimizations we offer in HF ecosystem:
+### In half-precision
+Note `float16` precision only works on GPU devices
+<details>
+<summary> Click to expand </summary>
 
-### Model Sources [optional]
+```diff
++ import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-<!-- Provide the basic links for the model. -->
+model_id = "mistral-community/Mixtral-8x22B-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
++ model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16).to(0)
 
-## Uses
+text = "Hello my name is"
++ inputs = tokenizer(text, return_tensors="pt").to(0)
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+</details>
 
-### Direct Use
+### Lower precision using (8-bit & 4-bit) using `bitsandbytes`
+<details>
+<summary> Click to expand </summary>
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+```diff
++ import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-[More Information Needed]
+model_id = "mistral-community/Mixtral-8x22B-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
 
-### Downstream Use [optional]
++ model = AutoModelForCausalLM.from_pretrained(model_id, load_in_4bit=True)
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+text = "Hello my name is"
++ inputs = tokenizer(text, return_tensors="pt").to(0)
 
-[More Information Needed]
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+</details>
 
-### Out-of-Scope Use
+### Load the model with Flash Attention 2
+<details>
+<summary> Click to expand </summary>
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+```diff
++ import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-[More Information Needed]
+model_id = "mistral-community/Mixtral-8x22B-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
 
-## Bias, Risks, and Limitations
++ model = AutoModelForCausalLM.from_pretrained(model_id, use_flash_attention_2=True)
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+text = "Hello my name is"
++ inputs = tokenizer(text, return_tensors="pt").to(0)
 
-[More Information Needed]
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+</details>
 
-### Recommendations
+## Notice
+Mixtral-8x22B-v0.1 is a pretrained base model and therefore does not have any moderation mechanisms.
+# The Mistral AI Team
+Albert Jiang, Alexandre Sablayrolles, Alexis Tacnet, Antoine Roux, Arthur Mensch, Audrey Herblin-Stoop, Baptiste Bout, Baudouin de Monicault,Blanche Savary, Bam4d, Caroline Feldman, Devendra Singh Chaplot, Diego de las Casas, Eleonore Arcelin, Emma Bou Hanna, Etienne Metzger, Gianna Lengyel, Guillaume Bour, Guillaume Lample, Harizo Rajaona, Jean-Malo Delignon, Jia Li, Justus Murke, Louis Martin, Louis Ternon, Lucile Saulnier, Lélio Renard Lavaud, Margaret Jennings, Marie Pellat, Marie Torelli, Marie-Anne Lachaux, Nicolas Schuhl, Patrick von Platen, Pierre Stock, Sandeep Subramanian, Sophia Yang, Szymon Antoniak, Teven Le Scao, Thibaut Lavril, Timothée Lacroix, Théophile Gervet, Thomas Wang, Valera Nemychnikova, William El Sayed, William Marshall.
+# [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_mistral-community__Mixtral-8x22B-v0.1)
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+|             Metric              |Value|
+|---------------------------------|----:|
+|Avg.                             |74.46|
+|AI2 Reasoning Challenge (25-Shot)|70.48|
+|HellaSwag (10-Shot)              |88.73|
+|MMLU (5-Shot)                    |77.81|
+|TruthfulQA (0-shot)              |51.08|
+|Winogrande (5-shot)              |84.53|
+|GSM8k (5-shot)                   |74.15|
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-
-## How to Get Started with the Model
-
-Use the code below to get started with the model.
-
-[More Information Needed]
-
-## Training Details
-
-### Training Data
-
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
-
-### Training Procedure
-
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
-
-## Evaluation
-
-<!-- This section describes the evaluation protocols and provides the results. -->
-
-### Testing Data, Factors & Metrics
-
-#### Testing Data
-
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
-
-#### Factors
-
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
-
-#### Metrics
-
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
-
-### Results
-
-[More Information Needed]
-
-#### Summary
-
-
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
-
-### Compute Infrastructure
-
-[More Information Needed]
-
-#### Hardware
-
-[More Information Needed]
-
-#### Software
-
-[More Information Needed]
-
-## Citation [optional]
-
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-
-**BibTeX:**
-
-[More Information Needed]
-
-**APA:**
-
-[More Information Needed]
-
-## Glossary [optional]
-
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-
-[More Information Needed]
-
-## More Information [optional]
-
-[More Information Needed]
-
-## Model Card Authors [optional]
-
-[More Information Needed]
-
-## Model Card Contact
-
-[More Information Needed]

RELEASE

@@ -0,0 +1,71 @@
+
+                                                                     ▄▄▄░░
+                                                            ▄▄▄▄▄█████████░░░░
+                                                ▄▄▄▄▄▄████████████████████░░░░░
+                                             █████████████████████████████░░░░░
+                        ▄▄▄▄▄▄█████░░░       █████████████████████████████░░░░░
+             ▄▄▄▄▄██████████████████░░░░░░  ██████████████████████████████░░░░░
+      ▄█████████████████████████████░░░░░░░░██████████████████████████████░░░░░
+      ███████████████████████████████░░░░░░░██████████████████████████████░░░░░
+      ███████████████████████████████░░░░░░░██████████████████████████████░░░░░
+      ███████████████████████████████░░░░░░███████████████████████████████░░░░░
+      ████████████████████████████████░░░░░███████████████████████████████░░░░░
+      ████████████████████████████████░░░░████████████████████████████████░░░░░
+      █████████████████████████████████░░░████████████████████████████████░░░░░
+      █████████████████████████████████░░░████████████░███████████████████░░░░░
+      ██████████████████████████████████░█████████████░███████████████████░░░░░
+      ███████████████████░██████████████▄█████████████░███████████████████░░░░░
+      ███████████████████░███████████████████████████░░███████████████████░░░░░
+      ███████████████████░░██████████████████████████░░███████████████████░░░░░
+      ███████████████████░░█████████████████████████░░░███████████████████░░░░░
+      ███████████████████░░░████████████████████████░░░███████████████████░░░░░
+      ███████████████████░░░████████████████████████░░░███████████████████░░░░░
+      ███████████████████░░░░██████████████████████░░░░███████████████████░░░░░
+      ███████████████████░░░░██████████████████████░░░░███████████████████░░░░░
+      ███████████████████░░░░░█████████████████████░░░░███████████████████░░░░░
+      ███████████████████░░░░░██████████████████��█░░░░░███████████████████░░░░░
+      ███████████████████░░░░░░███████████████████░░░░░███████████████████░░░░░
+      ███████████████████░░░░░░██████████████████░░░░░░███████████████████░░░░░
+      ███████████████████░░░░░░░█████████████████░░░░░░███████████████████░░░░░
+      ███████████████████░░░░░░░█████████████████░░░░░░███████████████████░░░░░
+      ███████████████████░░░░░░░░███████████████░░░░░░░██████████░░░░░░░░░░░░░░
+      ███████████████████░░░░░░░░███████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
+      ███████████████████░░░░░░░░███████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
+      ███████████████████░░░░░░░░░██░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
+      ███████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░
+      ██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░  ░░░░░░░
+          ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░      ░░░
+                ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░    ░░░░░░░░░░░░░░░░░░
+                   ░░░░░░░░░░░░░░░░░░░░░░░░░░░░
+                        ░░░░░░░░░░░░░░░░░
+                           ░░░░░
+
+
+ ╓────────────────────────────────────────────────────────────────────────────╖
+ ║                       MIXTRAL 8x22B  ··   24/04/10                         ║
+ ╙────────────────────────────────────────────────────────────────────────────╜
+
+ ╓────────────────────────────────────────────────────────────────────────────╖
+ ║                                                                            ║
+ ║                               ·· md5sum ··                                 ║
+ ║                                                                            ║
+ ║        3816cd2c4f827b4b868bc6481d5d3ba2  consolidated.safetensors          ║
+ ║        37974873eb68a7ab30c4912fc36264ae  tokenizer.model                   ║
+ ╙────────────────────────────────────────────────────────────────────────────╜
+
+ ╓────────────────────────────────────────────────────────────────────────────╖
+ ║                                                                            ║
+ ║                    ·· Released by the Mistral AI team ··                   ║
+ ║     Albert Jiang, Alexandre Sablayrolles, Alexis Tacnet, Antoine Roux,     ║
+ ║ Arthur Mensch, Audrey Herblin-Stoop, Baptiste Bout, Baudouin de Monicault, ║
+ ║      Blanche Savary, Bam4d, Caroline Feldman, Devendra Singh Chaplot,      ║
+ ║   Diego de las Casas, Eleonore Arcelin, Emma Bou Hanna, Etienne Metzger,   ║
+ ║      Gianna Lengyel, Guillaume Bour, Guillaume Lample, Harizo Rajaona,     ║
+ ║    Jean-Malo Delignon, Jia Li, Justus Murke, Louis Martin, Louis Ternon    ║
+ ║    Lucile Saulnier, Lélio Renard Lavaud, Margaret Jennings, Marie Pellat,  ║
+ ║   Marie Torelli, Marie-Anne Lachaux, Nicolas Schuhl, Patrick von Platen,   ║
+ ║      Pierre Stock, Sandeep Subramanian, Sophia Yang, Szymon Antoniak,      ║
+ ║     Teven Le Scao, Thibaut Lavril, Timothée Lacroix, Théophile Gervet,     ║
+ ║    Thomas Wang, Valera Nemychnikova, William El Sayed, William Marshall    ║
+ ║                                                                            ║
+ ╙────────────────────────────────────────────────────────────────────────────╜

app.py

@@ -0,0 +1,83 @@
+import streamlit as st
+from streamlit_chat import message
+# from langchain.llms import OpenAI   #This import has been replaced by the below import please
+from langchain_openai import OpenAI
+from langchain.chains import ConversationChain
+from langchain.chains.conversation.memory import (ConversationBufferMemory, 
+                                                  ConversationSummaryMemory, 
+                                                  ConversationBufferWindowMemory
+               
+                                                  )
+
+if 'conversation' not in st.session_state:
+    st.session_state['conversation'] =None
+if 'messages' not in st.session_state:
+    st.session_state['messages'] =[]
+if 'API_Key' not in st.session_state:
+    st.session_state['API_Key'] =''
+
+# Setting page title and header
+st.set_page_config(page_title="Chat GPT Clone", page_icon=":robot_face:")
+st.markdown("<h1 style='text-align: center;'>How can I assist you? </h1>", unsafe_allow_html=True)
+
+
+st.sidebar.title("😎")
+st.session_state['API_Key']= st.sidebar.text_input("What's your API key?",type="password")
+summarise_button = st.sidebar.button("Summarise the conversation", key="summarise")
+if summarise_button:
+    summarise_placeholder = st.sidebar.write("Nice chatting with you my friend ❤️:\n\n"+st.session_state['conversation'].memory.buffer)
+    #summarise_placeholder.write("Nice chatting with you my friend ❤️:\n\n"+st.session_state['conversation'].memory.buffer)
+
+#import os
+#os.environ["OPENAI_API_KEY"] = "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"
+
+def getresponse(userInput, api_key):
+
+    if st.session_state['conversation'] is None:
+
+        llm = OpenAI(
+            temperature=0,
+            openai_api_key=api_key,
+            model_name='gpt-3.5-turbo-instruct'  # 'text-davinci-003' model is depreciated now, so we are using the openai's recommended model
+        )
+
+        st.session_state['conversation'] = ConversationChain(
+            llm=llm,
+            verbose=True,
+            memory=ConversationSummaryMemory(llm=llm)
+        )
+
+    response=st.session_state['conversation'].predict(input=userInput)
+    print(st.session_state['conversation'].memory.buffer)
+    
+
+    return response
+
+
+
+response_container = st.container()
+# Here we will have a container for user input text box
+container = st.container()
+
+
+with container:
+    with st.form(key='my_form', clear_on_submit=True):
+        user_input = st.text_area("Your question goes here:", key='input', height=100)
+        submit_button = st.form_submit_button(label='Send')
+
+        if submit_button:
+            st.session_state['messages'].append(user_input)
+            model_response=getresponse(user_input,st.session_state['API_Key'])
+            st.session_state['messages'].append(model_response)
+            
+
+            with response_container:
+                for i in range(len(st.session_state['messages'])):
+                        if (i % 2) == 0:
+                            message(st.session_state['messages'][i], is_user=True, key=str(i) + '_user')
+                        else:
+                            message(st.session_state['messages'][i], key=str(i) + '_AI')
+        
+             
+
+        

config.json

@@ -0,0 +1,30 @@
+{
+  "architectures": [
+    "MixtralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 6144,
+  "initializer_range": 0.02,
+  "intermediate_size": 16384,
+  "max_position_embeddings": 65536,
+  "model_type": "mixtral",
+  "num_attention_heads": 48,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 56,
+  "num_key_value_heads": 8,
+  "num_local_experts": 8,
+  "output_router_logits": false,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000,
+  "router_aux_loss_coef": 0.001,
+  "router_jitter_noise": 0.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

convert.py

@@ -0,0 +1,278 @@
+# Copyright 2023 Mistral AI and The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import argparse
+import json
+import os
+
+import torch
+from safetensors.torch import load_file
+
+from transformers import (
+    MixtralConfig,
+    MixtralForCausalLM,
+)
+
+"""
+Sample usage:
+
+```
+python src/transformers/models/mixtral/convert_mixtral_weights_to_hf.py \
+    --input_dir /path/to/downloaded/mixtral/weights --model_size 7B --output_dir /output/path
+```
+
+Thereafter, models can be loaded via:
+
+```py
+from transformers import MixtralForCausalLM
+
+model = MixtralForCausalLM.from_pretrained("/output/path")
+```
+
+Important note: you need to be able to host the whole model in RAM to execute this script (even if the biggest versions
+come in several checkpoints they each contain a part of each weight of the model, so we need to load them all in RAM).
+"""
+
+def compute_intermediate_size(n, ffn_dim_multiplier=1, multiple_of=256):
+    return multiple_of * ((int(ffn_dim_multiplier * int(8 * n / 3)) + multiple_of - 1) // multiple_of)
+
+def read_json(path):
+    with open(path, "r") as f:
+        return json.load(f)
+
+def write_json(text, path):
+    with open(path, "w") as f:
+        json.dump(text, f)
+
+def write_model(model_path, input_base_path, model_size, safe_serialization=True):
+    os.makedirs(model_path, exist_ok=True)
+
+    params = read_json(os.path.join(input_base_path, "params.json"))
+    num_shards = 1
+
+    # For some reason this is a string in the params.json
+    sliding_window = int(params["sliding_window"]) if "sliding_window" in params else None
+    base = params.get("rope_theta", 10000.0)
+    vocab_size = params["vocab_size"]
+
+    if model_size == "7B":
+        dim = params["hidden_size"]
+        max_position_embeddings = 4096 * 8
+        num_local_experts = params["num_local_experts"]
+        ffn_dim = params["intermediate_size"]
+        n_layers = params["num_hidden_layers"]
+        n_heads = params["num_attention_heads"]
+        n_heads_per_shard = n_heads // num_shards
+        dims_per_head = dim // n_heads
+        if "num_key_value_heads" in params:
+            num_key_value_heads = params["num_key_value_heads"] # for GQA / MQA
+            num_local_key_value_heads = num_key_value_heads // num_shards
+            key_value_dim = dims_per_head * num_local_key_value_heads
+        else: # compatibility with other checkpoints
+            num_key_value_heads = n_heads
+            num_local_key_value_heads = n_heads_per_shard
+            key_value_dim = dim
+        rms_norm_eps = params["rms_norm_eps"]
+    elif model_size == "22B":
+        dim = params["dim"]
+        max_position_embeddings = params["max_seq_len"]
+        num_local_experts = params["moe"]["num_experts"]
+        ffn_dim = params["hidden_dim"]
+        n_layers = params["n_layers"]
+        n_heads = params["n_heads"]
+        n_heads_per_shard = n_heads // num_shards
+        dims_per_head = dim // n_heads
+        if "n_kv_heads" in params:
+            num_key_value_heads = params["n_kv_heads"] # for GQA / MQA
+            num_local_key_value_heads = num_key_value_heads // num_shards
+            key_value_dim = dims_per_head * num_local_key_value_heads
+        else: # compatibility with other checkpoints
+            num_key_value_heads = n_heads
+            num_local_key_value_heads = n_heads_per_shard
+            key_value_dim = dim
+        rms_norm_eps = params["norm_eps"]
+    else:
+        raise Exception("Illegal model size:", model_size)
+
+    # permute for sliced rotary
+    def permute(w, n_heads=n_heads, dim1=dim, dim2=dim):
+        return w.view(n_heads, dim1 // n_heads // 2, 2, dim2).transpose(1, 2).reshape(dim1, dim2)
+
+    print(f"Fetching all parameters from the checkpoint at \"{input_base_path}\"...")
+    # Load weights
+    if model_size == "7B":
+        loaded = [
+            torch.load(os.path.join(input_base_path, f"consolidated.{i:02d}.pt"), map_location="cpu") for i in range(8)
+        ]
+        merged_state_dict = {}
+        for state_dict in loaded:
+            merged_state_dict.update(state_dict)
+    elif model_size == "22B":
+        merged_state_dict = load_file(os.path.join(input_base_path, "consolidated.safetensors"))
+    print("Parameters load finished.")
+
+    state_dict = {}
+    for layer_i in range(n_layers):
+        print(f"At layer {layer_i}...")
+        # Sharded
+        # Note that attention.w{q,k,v,o}, feed_fordward.w[1,2,3], attention_norm.weight and ffn_norm.weight share
+        # the same storage object, saving attention_norm and ffn_norm will save other weights too, which is
+        # redundant as other weights will be stitched from multiple shards. To avoid that, they are cloned.
+
+        state_dict.update(
+            {
+                f"model.layers.{layer_i}.input_layernorm.weight": merged_state_dict[
+                    f"layers.{layer_i}.attention_norm.weight"
+                ].clone(),
+                f"model.layers.{layer_i}.post_attention_layernorm.weight": merged_state_dict[
+                    f"layers.{layer_i}.ffn_norm.weight"
+                ].clone(),
+            }
+        )
+
+        state_dict[f"model.layers.{layer_i}.self_attn.q_proj.weight"] = permute(
+            merged_state_dict[f"layers.{layer_i}.attention.wq.weight"]
+            .view(n_heads_per_shard, dims_per_head, dim)
+            .reshape(dim, dim)
+        )
+        state_dict[f"model.layers.{layer_i}.self_attn.k_proj.weight"] = permute(
+            merged_state_dict[f"layers.{layer_i}.attention.wk.weight"]
+            .view(num_local_key_value_heads, dims_per_head, dim)
+            .reshape(key_value_dim, dim),
+            num_key_value_heads,
+            key_value_dim,
+            dim,
+        )
+        state_dict[f"model.layers.{layer_i}.self_attn.v_proj.weight"] = (
+            merged_state_dict[f"layers.{layer_i}.attention.wv.weight"]
+            .view(num_local_key_value_heads, dims_per_head, dim)
+            .reshape(key_value_dim, dim)
+        )
+
+        state_dict[f"model.layers.{layer_i}.self_attn.o_proj.weight"] = merged_state_dict[
+            f"layers.{layer_i}.attention.wo.weight"
+        ]
+
+        if model_size == "7B":
+            w1 = merged_state_dict[f"layers.{layer_i}.block_sparse_moe.w1"]
+            w2 = merged_state_dict[f"layers.{layer_i}.block_sparse_moe.w2"]
+            w3 = merged_state_dict[f"layers.{layer_i}.block_sparse_moe.w3"]
+
+            experts_w1 = [
+                w1[ffn_dim * expert_idx : ffn_dim * (expert_idx + 1), :].contiguous().clone()
+                for expert_idx in range(num_local_experts)
+            ]
+
+            for idx, expert_block in enumerate(experts_w1):
+                expert_key = f"model.layers.{layer_i}.block_sparse_moe.experts.{idx}.w1"
+                state_dict[expert_key + ".weight"] = expert_block.clone()
+
+            experts_w2 = [
+                w2[ffn_dim * expert_idx : ffn_dim * (expert_idx + 1), :].contiguous().clone()
+                for expert_idx in range(num_local_experts)
+            ]
+
+            for idx, expert_block in enumerate(experts_w2):
+                expert_key = f"model.layers.{layer_i}.block_sparse_moe.experts.{idx}.w2"
+                state_dict[expert_key + ".weight"] = expert_block.T.clone().contiguous()
+
+            experts_w3 = [
+                w3[ffn_dim * expert_idx : ffn_dim * (expert_idx + 1), :].contiguous().clone()
+                for expert_idx in range(num_local_experts)
+            ]
+
+            for idx, expert_block in enumerate(experts_w3):
+                expert_key = f"model.layers.{layer_i}.block_sparse_moe.experts.{idx}.w3"
+                state_dict[expert_key + ".weight"] = expert_block.clone()
+
+            state_dict[f"model.layers.{layer_i}.block_sparse_moe.gate.weight"] = merged_state_dict[
+                f"layers.{layer_i}.block_sparse_moe.gate.weight"
+            ]
+        elif model_size == "22B":
+            for expert_i in range(num_local_experts):
+                w1 = merged_state_dict[f"layers.{layer_i}.feed_forward.experts.{expert_i}.w1.weight"]
+                w2 = merged_state_dict[f"layers.{layer_i}.feed_forward.experts.{expert_i}.w2.weight"]
+                w3 = merged_state_dict[f"layers.{layer_i}.feed_forward.experts.{expert_i}.w3.weight"]
+                state_dict[f"model.layers.{layer_i}.block_sparse_moe.experts.{expert_i}.w1.weight"] = w1.contiguous().clone()
+                state_dict[f"model.layers.{layer_i}.block_sparse_moe.experts.{expert_i}.w2.weight"] = w2.contiguous().clone()
+                state_dict[f"model.layers.{layer_i}.block_sparse_moe.experts.{expert_i}.w3.weight"] = w3.contiguous().clone()
+            state_dict[f"model.layers.{layer_i}.block_sparse_moe.gate.weight"] = merged_state_dict[
+                f"layers.{layer_i}.feed_forward.gate.weight"
+            ]
+
+    state_dict.update(
+        {
+            "model.norm.weight": merged_state_dict["norm.weight"],
+            "model.embed_tokens.weight": merged_state_dict["tok_embeddings.weight"],
+            "lm_head.weight": merged_state_dict["output.weight"],
+        }
+    )
+
+    config_additional_kwargs = {}
+    if model_size == "22B":
+        config_additional_kwargs["num_experts_per_tok"] = params["moe"]["num_experts_per_tok"]
+    config = MixtralConfig(
+        hidden_size=dim,
+        intermediate_size=ffn_dim,
+        num_attention_heads=n_heads,
+        num_hidden_layers=n_layers,
+        rms_norm_eps=rms_norm_eps,
+        num_key_value_heads=num_key_value_heads,
+        vocab_size=vocab_size,
+        rope_theta=base,
+        max_position_embeddings=max_position_embeddings,
+        sliding_window=sliding_window,
+        num_local_experts=num_local_experts,
+        **config_additional_kwargs
+    )
+
+    print("Loading the checkpoint in a Mixtral model.")
+    with torch.device("meta"):
+        model = MixtralForCausalLM(config)
+    # Avoid saving this as part of the config.
+    del model.config._name_or_path
+    model.config.torch_dtype = torch.bfloat16
+    print("Saving in the Transformers format.")
+
+    model.load_state_dict(state_dict, strict=True, assign=True)
+
+    for n, p in model.named_parameters():
+        assert p.device.type != "meta", f"{n} has not been loaded!"
+
+    model.save_pretrained(model_path, safe_serialization=safe_serialization)
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--input-dir",
+        help="Location of Mixtral weights, which contains tokenizer.model and model folders",
+        required=True,
+    )
+    parser.add_argument(
+        "--model-size",
+        choices=["7B", "22B"],
+        help="'f' models correspond to the finetuned versions, and are specific to the Mixtral official release. For more details on Mixtral, checkout the original repo: https://huggingface.co/mistral-ai",
+        default="7B",
+    )
+    parser.add_argument("--output-dir", help="Location to write HF model", required=True)
+    parser.add_argument("--safe-serialization", type=bool, default=True, help="Whether or not to save using `safetensors`.")
+    args = parser.parse_args()
+    write_model(
+        model_path=args.output_dir,
+        input_base_path=args.input_dir,
+        model_size=args.model_size,
+        safe_serialization=args.safe_serialization,
+    )
+
+if __name__ == "__main__":
+    main()

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.40.0.dev0"
+}

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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>",
+  "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
+}