Upload README.md with huggingface_hub

README.md

@@ -1,199 +1,236 @@
 ---
-library_name: transformers
-tags: []
+datasets:
+- tiiuae/falcon-refinedweb
+language:
+- en
+inference: false
+license: apache-2.0
 ---
 
-# Model Card for Model ID
+# 🚀 Falcon-7B
 
-<!-- Provide a quick summary of what the model is/does. -->
+**Falcon-7B is a 7B parameters causal decoder-only model built by [TII](https://www.tii.ae) and trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) enhanced with curated corpora. It is made available under the Apache 2.0 license.**
 
+*Paper coming soon* 😊.
 
+🤗 To get started with Falcon (inference, finetuning, quantization, etc.), we recommend reading [this great blogpost fron HF](https://huggingface.co/blog/falcon)!
 
-## Model Details
 
-### Model Description
+## Why use Falcon-7B?
 
-<!-- Provide a longer summary of what this model is. -->
+* **It outperforms comparable open-source models** (e.g., [MPT-7B](https://huggingface.co/mosaicml/mpt-7b), [StableLM](https://github.com/Stability-AI/StableLM), [RedPajama](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1) etc.), thanks to being trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) enhanced with curated corpora. See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
+* **It features an architecture optimized for inference**, with FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)) and multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)). 
+* **It is made available under a permissive Apache 2.0 license allowing for commercial use**, without any royalties or restrictions.
 
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+⚠️ **This is a raw, pretrained model, which should be further finetuned for most usecases.** If you are looking for a version better suited to taking generic instructions in a chat format, we recommend taking a look at [Falcon-7B-Instruct](https://huggingface.co/tiiuae/falcon-7b-instruct). 
 
-- **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]
+🔥 **Looking for an even more powerful model?** [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b) is Falcon-7B's big brother!
 
-### Model Sources [optional]
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import transformers
+import torch
 
-<!-- Provide the basic links for the model. -->
+model = "tiiuae/falcon-7b"
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+tokenizer = AutoTokenizer.from_pretrained(model)
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    torch_dtype=torch.bfloat16,
+    trust_remote_code=True,
+    device_map="auto",
+)
+sequences = pipeline(
+   "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:",
+    max_length=200,
+    do_sample=True,
+    top_k=10,
+    num_return_sequences=1,
+    eos_token_id=tokenizer.eos_token_id,
+)
+for seq in sequences:
+    print(f"Result: {seq['generated_text']}")
 
-## Uses
+```
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+💥 **Falcon LLMs require PyTorch 2.0 for use with `transformers`!**
 
-### Direct Use
+For fast inference with Falcon, check-out [Text Generation Inference](https://github.com/huggingface/text-generation-inference)! Read more in this [blogpost]((https://huggingface.co/blog/falcon). 
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+You will need **at least 16GB of memory** to swiftly run inference with Falcon-7B.
 
-[More Information Needed]
+# Model Card for Falcon-7B
 
-### Downstream Use [optional]
+## Model Details
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+### Model Description
 
-[More Information Needed]
+- **Developed by:** [https://www.tii.ae](https://www.tii.ae);
+- **Model type:** Causal decoder-only;
+- **Language(s) (NLP):** English, German, Spanish, French (and limited capabilities in Italian, Portuguese, Polish, Dutch, Romanian, Czech, Swedish);
+- **License:** Apache 2.0.
 
-### Out-of-Scope Use
+### Model Source
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+- **Paper:** *coming soon*.
 
-[More Information Needed]
+## Uses
 
-## Bias, Risks, and Limitations
+### Direct Use
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+Research on large language models; as a foundation for further specialization and finetuning for specific usecases (e.g., summarization, text generation, chatbot, etc.)
 
-[More Information Needed]
+### Out-of-Scope Use
 
-### Recommendations
+Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful. 
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+## Bias, Risks, and Limitations
+
+Falcon-7B is trained on English and French data only, and will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online.
+
+### Recommendations
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+We recommend users of Falcon-7B to consider finetuning it for the specific set of tasks of interest, and for guardrails and appropriate precautions to be taken for any production use.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
 
-[More Information Needed]
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import transformers
+import torch
+
+model = "tiiuae/falcon-7b"
+
+tokenizer = AutoTokenizer.from_pretrained(model)
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    torch_dtype=torch.bfloat16,
+    trust_remote_code=True,
+    device_map="auto",
+)
+sequences = pipeline(
+   "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:",
+    max_length=200,
+    do_sample=True,
+    top_k=10,
+    num_return_sequences=1,
+    eos_token_id=tokenizer.eos_token_id,
+)
+for seq in sequences:
+    print(f"Result: {seq['generated_text']}")
+
+```
 
 ## 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]
+Falcon-7B was trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb), a high-quality filtered and deduplicated web dataset which we enhanced with curated corpora. Significant components from our curated copora were inspired by The Pile ([Gao et al., 2020](https://arxiv.org/abs/2101.00027)). 
 
-### Training Procedure
+| **Data source**    | **Fraction** | **Tokens** | **Sources**                       |
+|--------------------|--------------|------------|-----------------------------------|
+| [RefinedWeb-English](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) | 79%          | 1,185B     | massive web crawl                 |
+| Books              | 7%           | 110B       |                                   |
+| Conversations      | 6%           | 85B        | Reddit, StackOverflow, HackerNews |
+| Code               | 3%           | 45B        |                                   |
+| RefinedWeb-French  | 3%           | 45B        | massive web crawl                 |
+| Technical          | 2%           | 30B        | arXiv, PubMed, USPTO, etc.        |
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
 
-#### Preprocessing [optional]
+The data was tokenized with the Falcon-[7B](https://huggingface.co/tiiuae/falcon-7b)/[40B](https://huggingface.co/tiiuae/falcon-40b) tokenizer.
 
-[More Information Needed]
+### Training Procedure 
 
+Falcon-7B was trained on 384 A100 40GB GPUs, using a 2D parallelism strategy (PP=2, DP=192) combined with ZeRO.
 
 #### 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. -->
+| **Hyperparameter** | **Value**  | **Comment**                               |
+|--------------------|------------|-------------------------------------------|
+| Precision          | `bfloat16` |                                           |
+| Optimizer          | AdamW      |                                           |
+| Learning rate      | 6e-4       | 4B tokens warm-up, cosine decay to 1.2e-5 |
+| Weight decay       | 1e-1       |                                           |
+| Z-loss       | 1e-4       |                                           |
+| Batch size         | 2304        | 30B tokens ramp-up                         |
 
-### Testing Data, Factors & Metrics
 
-#### Testing Data
+#### Speeds, Sizes, Times
 
-<!-- This should link to a Dataset Card if possible. -->
+Training happened in early March 2023 and took about two weeks. 
 
-[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]
+## Evaluation
 
-<!-- Relevant interpretability work for the model goes here -->
+*Paper coming soon*.
 
-[More Information Needed]
+See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) for early results.
 
-## Environmental Impact
 
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+## Technical Specifications 
 
-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).
+### Model Architecture and Objective
 
-- **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]
+Falcon-7B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token).
 
-## Technical Specifications [optional]
+The architecture is broadly adapted from the GPT-3 paper ([Brown et al., 2020](https://arxiv.org/abs/2005.14165)), with the following differences:
 
-### Model Architecture and Objective
+* **Positionnal embeddings:** rotary ([Su et al., 2021](https://arxiv.org/abs/2104.09864));
+* **Attention:** multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)) and FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135));
+* **Decoder-block:** parallel attention/MLP with a single layer norm.
 
-[More Information Needed]
+| **Hyperparameter** | **Value** | **Comment**                            |
+|--------------------|-----------|----------------------------------------|
+| Layers             | 32        |                                        |
+| `d_model`          | 4544      | Increased to compensate for multiquery                                       |
+| `head_dim`         | 64        | Reduced to optimise for FlashAttention |
+| Vocabulary         | 65024     |                                        |
+| Sequence length    | 2048      |                                        |
 
 ### Compute Infrastructure
 
-[More Information Needed]
-
 #### Hardware
 
-[More Information Needed]
+Falcon-7B was trained on AWS SageMaker, on 384 A100 40GB GPUs in P4d instances. 
 
 #### 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]
+Falcon-7B was trained a custom distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO and high-performance Triton kernels (FlashAttention, etc.)
 
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
 
-[More Information Needed]
+## Citation
 
-## More Information [optional]
+*Paper coming soon* 😊. In the meanwhile, you can use the following information to cite: 
+```
+@article{falcon40b,
+  title={{Falcon-40B}: an open large language model with state-of-the-art performance},
+  author={Almazrouei, Ebtesam and Alobeidli, Hamza and Alshamsi, Abdulaziz and Cappelli, Alessandro and Cojocaru, Ruxandra and Debbah, Merouane and Goffinet, Etienne and Heslow, Daniel and Launay, Julien and Malartic, Quentin and Noune, Badreddine and Pannier, Baptiste and Penedo, Guilherme},
+  year={2023}
+}
+```
 
-[More Information Needed]
+To learn more about the pretraining dataset, see the 📓 [RefinedWeb paper](https://arxiv.org/abs/2306.01116).
 
-## Model Card Authors [optional]
+```
+@article{refinedweb,
+  title={The {R}efined{W}eb dataset for {F}alcon {LLM}: outperforming curated corpora with web data, and web data only},
+  author={Guilherme Penedo and Quentin Malartic and Daniel Hesslow and Ruxandra Cojocaru and Alessandro Cappelli and Hamza Alobeidli and Baptiste Pannier and Ebtesam Almazrouei and Julien Launay},
+  journal={arXiv preprint arXiv:2306.01116},
+  eprint={2306.01116},
+  eprinttype = {arXiv},
+  url={https://arxiv.org/abs/2306.01116},
+  year={2023}
+}
+```
 
-[More Information Needed]
+## License
 
-## Model Card Contact
+Falcon-7B is made available under the Apache 2.0 license.
 
-[More Information Needed]
+## Contact
+falconllm@tii.ae