ce-lery/japanese-mistral-300m-base-GGUF

Quantized GGUF model files for japanese-mistral-300m-base from ce-lery

Name	Quant method	Size
japanese-mistral-300m-base.fp16.gguf	fp16	712.33 MB
japanese-mistral-300m-base.q2_k.gguf	q2_k	176.84 MB
japanese-mistral-300m-base.q3_k_m.gguf	q3_k_m	195.04 MB
japanese-mistral-300m-base.q4_k_m.gguf	q4_k_m	234.80 MB
japanese-mistral-300m-base.q5_k_m.gguf	q5_k_m	266.47 MB
japanese-mistral-300m-base.q6_k.gguf	q6_k	307.38 MB
japanese-mistral-300m-base.q8_0.gguf	q8_0	379.17 MB

Original Model Card:

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