Quantization made by Richard Erkhov. [Github](https://github.com/RichardErkhov) [Discord](https://discord.gg/pvy7H8DZMG) [Request more models](https://github.com/RichardErkhov/quant_request) sarashina2-13b - GGUF - Model creator: https://huggingface.co/sbintuitions/ - Original model: https://huggingface.co/sbintuitions/sarashina2-13b/ | Name | Quant method | Size | | ---- | ---- | ---- | | [sarashina2-13b.Q2_K.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q2_K.gguf) | Q2_K | 4.91GB | | [sarashina2-13b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.IQ3_XS.gguf) | IQ3_XS | 5.41GB | | [sarashina2-13b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.IQ3_S.gguf) | IQ3_S | 5.69GB | | [sarashina2-13b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q3_K_S.gguf) | Q3_K_S | 5.69GB | | [sarashina2-13b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.IQ3_M.gguf) | IQ3_M | 5.99GB | | [sarashina2-13b.Q3_K.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q3_K.gguf) | Q3_K | 6.32GB | | [sarashina2-13b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q3_K_M.gguf) | Q3_K_M | 6.32GB | | [sarashina2-13b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q3_K_L.gguf) | Q3_K_L | 6.87GB | | [sarashina2-13b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.IQ4_XS.gguf) | IQ4_XS | 6.99GB | | [sarashina2-13b.Q4_0.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q4_0.gguf) | Q4_0 | 7.33GB | | [sarashina2-13b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.IQ4_NL.gguf) | IQ4_NL | 7.37GB | | [sarashina2-13b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q4_K_S.gguf) | Q4_K_S | 7.38GB | | [sarashina2-13b.Q4_K.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q4_K.gguf) | Q4_K | 7.79GB | | [sarashina2-13b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q4_K_M.gguf) | Q4_K_M | 7.79GB | | [sarashina2-13b.Q4_1.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q4_1.gguf) | Q4_1 | 8.09GB | | [sarashina2-13b.Q5_0.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q5_0.gguf) | Q5_0 | 8.86GB | | [sarashina2-13b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q5_K_S.gguf) | Q5_K_S | 8.86GB | | [sarashina2-13b.Q5_K.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q5_K.gguf) | Q5_K | 9.1GB | | [sarashina2-13b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q5_K_M.gguf) | Q5_K_M | 9.1GB | | [sarashina2-13b.Q5_1.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q5_1.gguf) | Q5_1 | 9.63GB | | [sarashina2-13b.Q6_K.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q6_K.gguf) | Q6_K | 10.5GB | | [sarashina2-13b.Q8_0.gguf](https://huggingface.co/RichardErkhov/sbintuitions_-_sarashina2-13b-gguf/blob/main/sarashina2-13b.Q8_0.gguf) | Q8_0 | 13.6GB | Original model description: --- license: mit language: - ja - en --- # Sarashina2-13B This repository provides large language models trained by [SB Intuitions](https://www.sbintuitions.co.jp/). ## How to use ```python import torch from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, set_seed model = AutoModelForCausalLM.from_pretrained("sbintuitions/sarashina2-13b", torch_dtype=torch.bfloat16, device_map="auto") tokenizer = AutoTokenizer.from_pretrained("sbintuitions/sarashina2-13b") # If you want to use slow tokenizer # tokenizer = AutoTokenizer.from_pretrained("sbintuitions/sarashina2-13b", use_fast=False） generator = pipeline("text-generation", model=model, tokenizer=tokenizer) set_seed(123) text = generator( "おはようございます、今日の天気は", max_length=30, do_sample=True, pad_token_id=tokenizer.pad_token_id, num_return_sequences=3, ) for t in text: print(t) ``` ## Configuration | Parameters | Vocab size | Training tokens | Architecture | Position type | Layers | Hidden dim | Attention heads | | :-----: | :-----------: | :-------------: | :------------ | :-----------: | :----: | :--------: | :-------------: | | [7B](https://huggingface.co/sbintuitions/sarashina2-7b) | 102400 | 2.1T | Llama2 | RoPE | 32 | 4096 | 32 | | [13B](https://huggingface.co/sbintuitions/sarashina2-13b) | 102400 | 2.1T | Llama2 | RoPE | 40 | 5120 | 40 | | [70B](https://huggingface.co/sbintuitions/sarashina2-70b) | 102400 | 2.1T | Llama2 | RoPE | 80 | 8192 | 64 | ## Training Corpus For our Japanese training data, we used a Japanese portion of the [Common Crawl corpus](https://commoncrawl.org/), which is the largest Web corpus, as our training dataset. To clean the training corpus, we used [CCNet](https://github.com/facebookresearch/cc_net) and [HojiChar](https://github.com/HojiChar/HojiChar). After cleaning, our Japanese training data contains about 1T tokens. For our English training data, we extracted English documents from [SlimPajama](https://huggingface.co/datasets/cerebras/SlimPajama-627B) but we removed books3 corpus due to copyright infringement. ## Tokenization We use a [sentencepiece](https://github.com/google/sentencepiece) tokenizer with a unigram language model and byte-fallback. We do not apply pre-tokenization with Japanese tokenizer. Thus, a user may directly feed raw sentences into the tokenizer. ## Ethical Considerations and Limitations Sarashina2 has not been tuned to follow an instruction yet. Therefore, sarashina2 might generate some meaningless sequences, some inaccurate instances or biased/objectionable outputs. Before using sarashina2, we would like developers to tune models based on human preferences and safety considerations. ## License [MIT License](https://huggingface.co/sbintuitions/sarashina2-7b/blob/main/LICENSE)