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

python

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
import torch

MODEL_NAME = "ce-lery/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 = "自然言語処理とは、"

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=1024,
        do_sample=True,
        early_stopping=False,
        top_p=0.95,
        top_k=50,
        temperature=0.1,
        # 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.

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: 0.0003
train_batch_size: 12
eval_batch_size: 12
seed: 42
gradient_accumulation_steps: 20
total_train_batch_size: 240
optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine_with_min_lr
lr_scheduler_warmup_steps: 1000
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
3.7969	0.2212	10000	3.4418
3.659	0.4424	20000	3.2704
3.5721	0.6635	30000	3.1969
3.5678	0.8847	40000	3.1757

Framework versions

Transformers 4.55.2
Pytorch 2.8.0+cu128
Datasets 4.0.0
Tokenizers 0.21.4