Update README.md

README.md

@@ -7,96 +7,118 @@ library_name: transformers
 pipeline_tag: text-generation
 ---
 
-# PLaMo-13B
+# PLaMo-13B-Instruct-NC
 
 ## Model Description
-PLaMo-13B is a LLaMA-based 13B model pre-trained on English and Japanese open datasets, developed by Preferred Networks, Inc.
-PLaMo-13B is released under Apache v2.0 license.
+PLaMo-13B-Instruct-NC is a noncommercial instruct fine-tuned model built upon the 8192 context length version of [Plamo-13B](https://huggingface.co/pfnet/plamo-13b) text generation model. PLaMo-13B-Instruct-NC is fine-tuned using multiple publicly available Japanese datasets.
+This model is released under the Apache License 2.0.
 
-[PLaMo-13B Release blog (Japanese)](https://tech.preferred.jp/ja/blog/llm-plamo/)
+[PLaMo-13B-Instruct Release blog (Japanese)](https://tech.preferred.jp/ja/blog/llm-plamo/)
 
-## Usage
 
-### Requirements
+## Usage
+Install the required libraries as follows:
+```sh
+>>> python -m pip install numpy safetensors sentencepiece torch transformers
 
-- numpy
-- safetensors
-- sentencepiece
-- torch
-- transformers
+```
 
-### Use a pipeline as a high-level helper
+Execute the following python code:
 ```python
-import transformers
-pipeline = transformers.pipeline("text-generation", model="pfnet/plamo-13b", trust_remote_code=True)
-print(pipeline("The future of artificial intelligence technology is ", max_new_tokens=32))
+tokenizer = AutoTokenizer.from_pretrained(
+    "pfnet/plamo-13b-instruct-nc",
+    trust_remote_code=True,
+)
+model = AutoModelForCausalLM.from_pretrained(
+    "pfnet/plamo-13b-instruct-nc",
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+def completion(prompt: str, max_new_tokens: int = 128) -> str:
+    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+    generated_ids = model.generate(
+        inputs.input_ids,
+        eos_token_id=2,
+        pad_token_id=3,
+        max_new_tokens=max_new_tokens,
+        temperature=1,
+        top_p=0.95,
+        top_k=50,
+        do_sample=True,
+    )
+    return tokenizer.decode(generated_ids[0], skip_special_tokens=True, clean_up_tokenization_spaces=True)
+
+def generate_prompt(messages: list) -> str:
+    sep = "\n\n### "
+    prompt = [
+        "以下はタスクを説明する指示で、文脈を説明した入力とペアになっています。",
+        "要求を適切に補完するよう応答を書いてください。",
+    ]
+    roles = {"instruction": "指示",  "response": "応答", "input": "入力"}
+    for msg in messages:
+        prompt.append(sep + roles[msg["role"]] + ":\n" + msg['content'])
+    prompt.append(sep + roles["response"] + ":\n")
+    return "".join(prompt)
 ```
 
-### Load model directly
 ```python
-from transformers import AutoTokenizer, AutoModelForCausalLM
-tokenizer = AutoTokenizer.from_pretrained("pfnet/plamo-13b", trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained("pfnet/plamo-13b", trust_remote_code=True)
-text = "これからの人工知能技術は"
-input_ids = tokenizer(text, return_tensors="pt").input_ids
-generated_tokens = model.generate(
-    inputs=input_ids,
-    max_new_tokens=32,
-    do_sample=True,
-    top_k=50,
-    top_p=0.95,
-    temperature=1.0,
-)[0]
-generated_text = tokenizer.decode(generated_tokens)
-print(generated_text)
+prompt = generate_prompt([
+    {"role": "instruction", "content": "日本の首都はどこですか？"},
+    # {"role": "input", "content": "..."}  ## An extra input (optional)
+])
+print(completion(prompt, max_new_tokens=128))
 ```
 
 ## Model Details
 
 - Model size: 13B
 - Trained tokens: 1.5T tokens (English: 1.32T tokens, Japanese: 0.18T tokens)
-- Context length: 4096
-- Developed by: Preferred Networkfs, Inc
+- Tokenizer: sentencepiece tokenizer trained on a subset of the pretraining datasets.
+- Context length: 8192
+- Developed by: Preferred Networks, Inc
 - Model type: Causal decoder-only
-- Language(s): English, Japanese
-- License: Apache v2.0
+- Language(s): Japanese and English
+- License: Apache License 2.0
 
 ## Training Dataset
 
-### English
-
-- C4 - English
-- Project Gutenberg
-- RedPajama - Arxiv
-- RedPajama - CommonCrawl - English
-- RedPajama - Github
-- RedPajama - StackExchange
-- RedPajama - Wikipedia
+- [Stanford Alpaca (Japanese translation)](https://huggingface.co/datasets/fujiki/japanese_alpaca_data)
+- [databricks-dolly-15k (Japanese translation)](https://huggingface.co/datasets/kunishou/databricks-dolly-15k-ja)
+- [Anthropic HH-RLHF (Japanese translation, subset)](https://huggingface.co/datasets/fujiki/japanese_hh-rlhf-49k)
+- [OpenAssistant Conversations Dataset (Japanese translation, oasst1)](https://huggingface.co/datasets/kunishou/oasst1-89k-ja)
+- [Wikinews subset of Izumi-lab llm-japanese-dataset](https://huggingface.co/datasets/izumi-lab/llm-japanese-dataset)
 
-### Japanese
+For the pretraining model, see [Plamo-13B](https://huggingface.co/pfnet/plamo-13b).
 
-- mC4 - Japanese
-- Wikipedia - Japanese
-
-## Tokenizer
-PLaMo-13B uses sentencepiece tokenizer which is trained on a subset of the datasets for model pre-training.
 
 ## Bias, Risks, and Limitations
-PLaMo-13B is a new technology that carries risks with use. Testing conducted to date has been in English and Japanese, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, PLaMo-13B’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of PLaMo-13B, developers should perform safety testing and tuning tailored to their specific applications of the model.
+PLaMo-13B-Instruct-NC is a new technology that carries risks with use. Testing conducted to date has been in English and Japanese, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, PLaMo-13B’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of PLaMo-13B-Instruct-NC, developers should perform safety testing and tuning tailored to their specific applications of the model.
 
 ## How to cite
 ```tex
-@online{PLaMo2023Introducing,
+@online{PLaMoInstruct2023Introducing,
     author    = {Preferred Networks, Inc},
-    title     = {PLaMo-13B},
+    title     = {PLaMo-13B-Instruct-NC},
     year      = {2023},
-    url       = {https://huggingface.co/pfnet/plamo-13b},
-    urldate   = {2023-09-28}
+    url       = {https://huggingface.co/pfnet/plamo-13b-instruct-nc},
+    urldate   = {2023-10-26}
 }
 ```
 
 ## References
 ```tex
+@misc{alpaca,
+  author = {Rohan Taori and Ishaan Gulrajani and Tianyi Zhang and Yann Dubois and Xuechen Li and Carlos Guestrin and Percy Liang and Tatsunori B. Hashimoto },
+  title = {Stanford Alpaca: An Instruction-following LLaMA model},
+  year = {2023},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  howpublished = {\url{https://github.com/tatsu-lab/stanford_alpaca}},
+}
+```
+```tex
 @article{touvron2023llama,
   title={LLaMA: Open and Efficient Foundation Language Models},
   author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},