Update README.md (#1)

- Update README.md (638b530f85213a0cfc13500a39642f806a1d0de7)
- Update README.md (d2a5d22d0811cf1a3ef8af63925e8c498960b8e0)
- Update README.md (d96a69d1a04bfa7ac8e6c4ccdb6511bfee5df31d)
- Update README.md (198973a45d20d8ae448d8a95d21b4f120524c493)
- Update README.md (95655adedb82c3c05ebeba988ae642ed5378628c)
- Update README.md (c70f44d7cf160b863608962c23c7df9cdc44f57b)
- Update README.md (9a0614f3f39337b4336065542cd757f975453a19)


Co-authored-by: Satoru Katsumata <Katsumata420@users.noreply.huggingface.co>
Co-authored-by: big_wing <bigwing@users.noreply.huggingface.co>

README.md

@@ -1,199 +1,148 @@
 ---
 library_name: transformers
-tags: []
+license: apache-2.0
+language:
+- ja
+- en
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
-
-
+# Retrieva BERT Model
+The **RetrievaBERT** is the pre-trained Transformer Encoder using Megatron-LM.
+It is designed for use in Japanese.
 
 ## Model Details
 
 ### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
-
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
+The **RetrievaBERT** is the pre-trained Transformer Encoder using Megatron-LM.
 
-- **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]
+It is designed for use in Japanese.
 
-### Model Sources [optional]
+This model offers several advanced features compared to traditional BERT models:
+- **PreNorm**: Improved stability during training.  
+- **SwiGLU**: Enhanced activation function for better performance.  
+- **Grouped-Query Attention (Multi-Query Attention)**: Efficient attention mechanism.  
+- **Max Sequence Length**: 2048 tokens, allowing for longer context.  
+- **Parameters**: 1.3 billion parameters.  
+- **Pre-training Objective**: Only Masked Language Modeling (MLM), not Next Sentence Prediction (NSP).  
+- **Token Type IDs**: Not used in this model.
 
-<!-- Provide the basic links for the model. -->
+### Model Sources
+- **Developed by:** Retrieva, Inc.
+- **Model type:** Based on MegatronBERT Architecture.
+- **Language(s) (NLP):** Primarily Japanese (optional support for English).
+- **License:** Apache 2.0
 
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
 
 ## 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. -->
+This model can be used as a Masked Language Model (MLM).
+However, it is primarily intended to be fine-tuned on downstream tasks.
+Depending on your use case, follow the appropriate section below.
 
 ### Direct Use
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
-
-### Downstream Use [optional]
-
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
-
-### Out-of-Scope Use
-
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
-
-## Bias, Risks, and Limitations
-
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+This model is pre-trained using Masked Language Modeling.
+The mask token used is `<MASK|LLM-jp>`.
+Note that you need to set `trust_remote_code` to `True` because RetrievaBERT uses a custom model implementation.  
+  
+Example code for direct use: 
 
-[More Information Needed]
+```python
+from transformers import AutoModelForMaskedLM, AutoTokenizer, pipeline
 
-### Recommendations
+model_id = "retrieva-jp/bert-1.3b"
+model = AutoModelForMaskedLM.from_pretrained(model_id, trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+pipe = pipeline("fill-mask", model=model, tokenizer=tokenizer)
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+text = "こんにちは！私の名前は<MASK|LLM-jp>です！"
+print(pipe(text))
+```
 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+### Downstream Use
 
-## How to Get Started with the Model
+RetrievaBERT is compatible with Hugging Face's AutoModels.
+To fine-tune RetrievaBERT for your specific task, use the corresponding AutoModel class.
+For detailed configuration, refer to the config.json file.
 
-Use the code below to get started with the model.
-
-[More Information Needed]
 
 ## Training Details
 
 ### Training Data
+The Retrieva BERT model was pre-trained on the reunion of five datasets:
+- [Japanese CommonCrawl Dataset by LLM-jp](https://gitlab.llm-jp.nii.ac.jp/datasets/llm-jp-corpus-v2).
+- [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb).
+- Chinese Wikipedia dumped on 20240120.
+- Korean Wikipedia dumped on 20240120.
+- [The Stack](https://huggingface.co/datasets/bigcode/the-stack)
 
-<!-- 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]
+The model was trained on 180 billion tokens using the above dataset.
 
 ### Training Procedure
+The model was trained on 4 to 32 H100 GPUs with a batch size of 1,024.
+We adopted the curriculum learning which is similar to the Sequence Length Warmup and training with the following sequence lengths and number of steps.
 
-<!-- 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]
-
+- The sequence length of 128: 31,000 steps.
+- The sequence length of 256: 219,000 steps.
+- The sequence length of 512: 192,000 steps.
+- The sequence length of 2048: 12,000 steps.
 
 #### Training Hyperparameters
+The model was trained on the following 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]
+- Learning rate: 1.5e-4.
+- Learning rate decay style: Linear.
+- Learning rate warmup fraction: 0.01
+- Minimum learning rate: 1e-6
+- Floating point expression: BF16
 
 ## Evaluation
+We fine-tuned the following models and evaluated them on the [JGLUE](https://github.com/yahoojapan/JGLUE) development set. 
+We adjusted the learning rate and training epochs for each model and task in accordance with [the JGLUE paper](https://www.jstage.jst.go.jp/article/jnlp/30/1/30_63/_pdf/-char/ja).
 
-<!-- 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
+| Model                            | MARC-ja/acc | JSTS/pearson | JSTS/spearman | JNLI/acc | JSQuAD/EM | JSQuAD/F1 | JComQA/acc |
+| :--- |---:|---:|---:|---:|---:|---:|---:|
+| tohoku-nlp/bert-base-japanese-v3 | 0.957       | 0.914        | 0.876         | 0.906    | 0.878     | 0.946     | 0.849      |
+| tohoku-nlp/bert-large-japanese-v2| 0.959       | 0.916        | 0.877         | 0.901    | 0.884     | 0.951     | 0.867      |
+| ku-nlp/deberta-v3-base-japanese　　　　| 0.958       | 0.925        | 0.890         | 0.902    | 0.925     | 0.910     | 0.882      |
+| retrieva-jp/bert-1.3b　　　　　　　　　　　　　　　　　　　　　　　　| 0.952       | 0.916        | 0.877         | 0.896    | 0.916     | 0.879     | 0.815      |
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
 
-[More Information Needed]
+## Technical Specifications
 
-#### Metrics
+### Model Architectures
+The Retrieva BERT model is based on BERT with the following hyperparameters:
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+- Number of layers: 48
+- Hidden layer size: 1536
+- FFN hidden layer size: 4096
+- Number of attention heads: 24
+- Maximum length of position embeddings: 2048
 
-[More Information Needed]
+As mentioned earlier, the main differences from the original BERT are:
+- PreNorm: Improved stability during training.  
+- SwiGLU: Enhanced activation function for better performance.  
+- Grouped-Query Attention (Multi-Query Attention): Efficient attention mechanism.  
 
-### 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
+[TSUBAME 4](https://www.t4.gsic.titech.ac.jp/en/hardware)
 
-[More Information Needed]
+This model is based on results obtained from the TSUBAME deep-learning mini-camp.
 
 #### 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]
+The model was trained using [Megatron-LM](https://github.com/NVIDIA/Megatron-LM).
 
-**APA:**
+## More Information
 
-[More Information Needed]
+https://note.com/retrieva/n/n715bea2c2cd1 (in Japanese)
 
-## Glossary [optional]
+## Model Card Authors
 
-<!-- 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]
+Satoru Katsumata, Daisuke Kimura, Jiro Nishitoba
 
 ## Model Card Contact
-
-[More Information Needed]
+pr@retrieva.jp