Upload 3 files

README.md

@@ -1,3 +1,51 @@
 ---
+library_name: peft
+base_model: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
 license: mit
+language:
+- ko
+- en
+pipeline_tag: translation
 ---
+
+사용 데이터셋: aihub
+
+훈련 환경: RTX3090 x 8
+
+epoch: 1
+
+time: 19시간
+
+``` python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
+from peft import prepare_model_for_kbit_training, PeftModel, PeftConfig
+
+model_path = 'yanolja/EEVE-Korean-Instruct-10.8B-v1.0'
+lora_path = 'qwopqwop/EEVE-ALMA'
+
+bnb_config = BitsAndBytesConfig(load_in_4bit=True,bnb_4bit_quant_type="nf4",bnb_4bit_compute_dtype=torch.float16,)
+model = AutoModelForCausalLM.from_pretrained(model_path, quantization_config=bnb_config, trust_remote_code=True)
+model.config.use_cache = False
+model = PeftModel.from_pretrained(model, lora_path)
+model = prepare_model_for_kbit_training(model)
+tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side='left')
+
+en_text = 'Hi.'
+ko_text = '안녕하세요.'
+
+en_prompt = f"Translate this from English to Korean:\nEnglish: {en_text}\nKorean:"
+ko_prompt = f"Translate this from Korean to English:\nKorean: {ko_text}\nEnglish:"
+
+input_ids = tokenizer(en_prompt, return_tensors="pt", padding=True, max_length=256, truncation=True).input_ids.cuda()
+with torch.no_grad():
+    generated_ids = model.generate(input_ids=input_ids, num_beams=5, max_new_tokens=20, do_sample=True, temperature=0.6, top_p=0.9)
+outputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
+print(outputs)
+
+input_ids = tokenizer(ko_prompt, return_tensors="pt", padding=True, max_length=256, truncation=True).input_ids.cuda()
+with torch.no_grad():
+    generated_ids = model.generate(input_ids=input_ids, num_beams=5, max_new_tokens=20, do_sample=True, temperature=0.6, top_p=0.9)
+outputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
+print(outputs)
+```

adapter_config.json

@@ -0,0 +1,33 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "yanolja/EEVE-Korean-10.8B-v1.0",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "o_proj",
+    "up_proj",
+    "q_proj",
+    "v_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cac94806bf91eb3b57f1f9d7ba427238cb208715e0d73211aced73f5263f4ca9
+size 503407912