japanese-mistral-300m-base

Overview

Welcome to my model card!

This Model feature is ...

• Suppression of unknown word generation by using byte fallback in SentencePiece tokenizer and conversion to huggingface Tokenizers format
• Pretrained by wikipedia dataset and cc100 dataset
• Use of Mistral 300M

Yukkuri shite ittene!

How to use the model

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
import torch

MODEL_NAME = "ce-lery/japanese-mistral-300m-base"
torch.set_float32_matmul_precision('high')

DEVICE = "cuda"
if torch.cuda.is_available():
    print("cuda")
    DEVICE = "cuda"
else:
    print("cpu")
    DEVICE = "cpu"

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME,use_fast=False)
model = AutoModelForCausalLM.from_pretrained(
    MODEL_NAME,
    trust_remote_code=True,
).to(DEVICE)

# streamer = TextStreamer(tokenizer)

prompt = "大規模言語モデルとは、"

inputs = tokenizer(prompt, add_special_tokens=False,return_tensors="pt").to(model.device)
with torch.no_grad():

    outputs = model.generate(
        inputs["input_ids"],
        max_new_tokens=256,
        do_sample=True,
        early_stopping=False,
        top_p=0.95,
        top_k=50,
        temperature=0.9,
        # streamer=streamer,
        no_repeat_ngram_size=2,
        num_beams=3
    )

print(outputs.tolist()[0])
outputs_txt = tokenizer.decode(outputs[0])
print(outputs_txt)

Receipe

If you want to restruct this model, you can refer this Github repository.

I wrote the receipe for struction this model. For example,

• Preprocess with sentencepiece
• Pretraining with flash attention2 and torch.compile and DeepSpeed
• Fine-tuning with databricks-dolly-15k-ja

If you find my mistake,error,...etc, please create issue. If you create pulreqest, I'm very happy!

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0006
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 64
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.95) and epsilon=0.0001
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.2911	0.12	5000	4.2914
3.9709	0.24	10000	3.9900
3.8229	0.36	15000	3.8388
3.7197	0.47	20000	3.7454
3.652	0.59	25000	3.6739
3.597	0.71	30000	3.6177
3.5554	0.83	35000	3.5770
3.536	0.95	40000	3.5582

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1