--- pipeline_tag: text-generation inference: false license: apache-2.0 # datasets: # metrics: # - code_eval library_name: transformers tags: - language - granite-3.0 model-index: - name: granite-3.0-3b-a800m-base results: - task: type: text-generation dataset: type: human-exams name: MMLU metrics: - name: pass@1 type: pass@1 value: 48.64 veriefied: false - task: type: text-generation dataset: type: human-exams name: MMLU-Pro metrics: - name: pass@1 type: pass@1 value: 18.84 veriefied: false - task: type: text-generation dataset: type: human-exams name: AGI-Eval metrics: - name: pass@1 type: pass@1 value: 23.81 veriefied: false - task: type: text-generation dataset: type: commonsense name: WinoGrande metrics: - name: pass@1 type: pass@1 value: 65.67 veriefied: false - task: type: text-generation dataset: type: commonsense name: OBQA metrics: - name: pass@1 type: pass@1 value: 42.20 veriefied: false - task: type: text-generation dataset: type: commonsense name: SIQA metrics: - name: pass@1 type: pass@1 value: 47.39 veriefied: false - task: type: text-generation dataset: type: commonsense name: PIQA metrics: - name: pass@1 type: pass@1 value: 78.29 veriefied: false - task: type: text-generation dataset: type: commonsense name: Hellaswag metrics: - name: pass@1 type: pass@1 value: 72.79 veriefied: false - task: type: text-generation dataset: type: commonsense name: TruthfulQA metrics: - name: pass@1 type: pass@1 value: 41.34 veriefied: false - task: type: text-generation dataset: type: reading-comprehension name: BoolQ metrics: - name: pass@1 type: pass@1 value: 75.75 veriefied: false - task: type: text-generation dataset: type: reading-comprehension name: SQuAD v2 metrics: - name: pass@1 type: pass@1 value: 29.96 veriefied: false - task: type: text-generation dataset: type: reasoning name: ARC-C metrics: - name: pass@1 type: pass@1 value: 46.84 veriefied: false - task: type: text-generation dataset: type: reasoning name: GPQA metrics: - name: pass@1 type: pass@1 value: 24.83 veriefied: false - task: type: text-generation dataset: type: reasoning name: BBH metrics: - name: pass@1 type: pass@1 value: 38.93 veriefied: false - task: type: text-generation dataset: type: code name: HumanEval metrics: - name: pass@1 type: pass@1 value: 26.83 veriefied: false - task: type: text-generation dataset: type: code name: MBPP metrics: - name: pass@1 type: pass@1 value: 34.60 veriefied: false - task: type: text-generation dataset: type: math name: GSM8K metrics: - name: pass@1 type: pass@1 value: 46.02 veriefied: false - task: type: text-generation dataset: type: math name: MATH metrics: - name: pass@1 type: pass@1 value: 17.40 veriefied: false - task: type: text-generation dataset: type: multilingual name: MGSM metrics: - name: pass@1 type: pass@1 value: 25.13 veriefied: false --- # Granite-3.0-3B-A800M-Base ## Model Summary **Granite-3.0-3B-A800M-Base** is an open-source decoder-only language model from IBM Research that supports a variety of text-to-text generation tasks (e.g., question-answering, text-completion). **Granite-3.0-3B-A800M-Base** is trained from scratch and follows a two-phase training strategy. In the first phase, it is trained on 8 trillion tokens sourced from diverse domains, including natural language, math, code, and safety. During the second phase, it is further trained on 2 trillion tokens using a carefully curated mix of high-quality data, aiming to enhance its performance on specific tasks. - **Developers:** IBM Research - **GitHub Repository:** [ibm-granite/granite-language-models](https://github.com/ibm-granite/granite-language-models) - **Paper:** [Granite Language Models](https://) - **Release Date**: October 21st, 2024 - **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0). ## Supported Languages English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, Chinese (Simplified) ## Usage ### Intended use Prominent use cases of LLMs in text-to-text generation include summarization, text classification, extraction, question-answering, and more. All Granite Base models are able to handle these tasks as they were trained on a large amount of data from various domains. Moreover, all Granite language model can serve as baseline to create specialized models for specific application scenarios. ### Generation This is a simple example of how to use **Granite-3.0-3B-A800M-Base** model. Install the following libraries: ```shell pip install torch torchvision torchaudio pip install accelerate pip install transformers ``` Then, copy the code snippet below to run the example. ```python from transformers import AutoModelForCausalLM, AutoTokenizer device = "auto" model_path = "ibm-granite/granite-3.0-3b-a800m-base" tokenizer = AutoTokenizer.from_pretrained(model_path) # drop device_map if running on CPU model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device) model.eval() # change input text as desired input_text = "Where is the MIT-IBM Watson AI Lab located?" # tokenize the text input_tokens = tokenizer(input_text, return_tensors="pt").to(device) # generate output tokens output = model.generate(**input_tokens, max_length=4000) # decode output tokens into text output = tokenizer.batch_decode(output) # print output print(output) ``` ## Model Architeture **Granite-3.0-3B-A800M-Base** is based on a decoder-only sparse Mixture of Experts(MoE) transformer architecture. Core components of this architecture are: Fine-grained Experts, Dropless Token Routing, and Load Balancing Loss. | Model | 2B Dense | 8B Dense | 1B MoE | 3B MoE | | :-------- | :--------| :--------| :--------| :-------- | | Embedding size | 2048 | 4096 | 1024 | **1536** | | Number of layers | 40 | 40 | 24 | **32** | | Attention head size | 64 | 128 | 64 | **64** | | Number of attention heads | 32 | 32 | 16 | **24** | | Number of KV heads | 8 | 8 | 8 | **8** | | MLP hidden size | 8192 | 12800 | 512 | **512** | | MLP activation | SwiGLU | SwiGLU | SwiGLU | **SwiGLU** | | Number of Experts | — | — | 32 | **40** | | MoE TopK | — | — | 8 | **8** | | Initialization std | 0.1 | 0.1 | 0.1 | **0.1** | | Sequence Length | 4096 | 4096 | 4096 | **4096** | | Position Embedding | RoPE | RoPE | RoPE | **RoPE** | | # Paremeters | 2.5B | 8.1B | 1.3B | **3.3B** | | # Active Parameters | 2.5B | 8.1B | 400M | **800M** | | # Training tokens | 12T | 12T | 10T | **10T** | ## Training Data This model is trained on a mix of open-source and proprietary datasets. ## Infrastructure We train the Granite Language models using IBM's super computing cluster, Blue Vela, which is outfitted with NVIDIA H100 GPUs. This cluster provides a scalable and efficient infrastructure for training our models over thousands of GPUs. ## Ethical Considerations and Limitations The use of Large Language Models involves risks and ethical considerations people must be aware of, including but not limited to: bias and fairness, misinformation, and autonomous decision-making. **Granite-3.0-3B-A800M-Base** model is not the exception in this regard. Even though this model is suited for multiple generative AI tasks, it has not undergone any safety alignment, there it may produce problematic outputs. Additionally, it remains uncertain whether smaller models might exhibit increased susceptibility to hallucination in generation scenarios by copying text verbatim from the training dataset due to their reduced sizes and memorization capacities. This aspect is currently an active area of research, and we anticipate more rigorous exploration, comprehension, and mitigations in this domain. Regarding ethics, a latent risk associated with all Large Language Models is their malicious utilization. We urge the community to use **Granite-3.0-3B-A800M-Base** model with ethical intentions and in a responsible way. ## Citation ``` @misc{granite-models, author = {author 1, author2, ...}, title = {}, journal = {}, volume = {}, year = {2024}, url = {https://arxiv.org/abs/0000.00000}, } ```