Upload README.md with huggingface_hub

README.md

@@ -1,38 +1,39 @@
 ---
-library_name: peft
+license: mit
+language: ja
 ---
-## Training procedure
+Mistral-7B Japanese [LAPT + Random]
+===
 
+## How to use
+```python
+from peft import AutoPeftModelForCausalLM
 
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- _load_in_8bit: True
-- _load_in_4bit: False
-- llm_int8_threshold: 6.0
-- llm_int8_skip_modules: None
-- llm_int8_enable_fp32_cpu_offload: False
-- llm_int8_has_fp16_weight: False
-- bnb_4bit_quant_type: fp4
-- bnb_4bit_use_double_quant: False
-- bnb_4bit_compute_dtype: float32
-- load_in_4bit: False
-- load_in_8bit: True
+model = AutoPeftModelForCausalLM.from_pretrained(
+    "atsuki-yamaguchi/Mistral-7B-v0.1-random-ja"
+)
+tokenizer = AutoTokenizer.from_pretrained(
+    "atsuki-yamaguchi/Mistral-7B-v0.1-random-ja"
+)
 
-The following `bitsandbytes` quantization config was used during training:
-- quant_method: bitsandbytes
-- _load_in_8bit: True
-- _load_in_4bit: False
-- llm_int8_threshold: 6.0
-- llm_int8_skip_modules: None
-- llm_int8_enable_fp32_cpu_offload: False
-- llm_int8_has_fp16_weight: False
-- bnb_4bit_quant_type: fp4
-- bnb_4bit_use_double_quant: False
-- bnb_4bit_compute_dtype: float32
-- load_in_4bit: False
-- load_in_8bit: True
-### Framework versions
+# w/ GPU
+model = AutoPeftModelForCausalLM.from_pretrained(
+    "atsuki-yamaguchi/Mistral-7B-v0.1-random-ja",
+    device_map="auto", 
+    load_in_8bit=True,
+)
+```
+## Citation
+```
+@article{yamaguchi2024empirical,
+  title={An Empirical Study on Cross-lingual Vocabulary Adaptation for Efficient Generative {LLM} Inference}, 
+  author={Atsuki Yamaguchi and Aline Villavicencio and Nikolaos Aletras},
+  journal={ArXiv},
+  year={2024},
+  volume={abs/2402.10712},
+  url={https://arxiv.org/abs/2402.10712}
+}
+```
 
-- PEFT 0.5.0
-
-- PEFT 0.5.0
+## Link
+For more details, please visit https://github.com/gucci-j/llm-cva