Upload folder using huggingface_hub

README.md

@@ -54,7 +54,8 @@ import json
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-model_id = "yanolja/YanoljaNEXT-Rosetta-4B"
+# model_id = "yanolja/YanoljaNEXT-Rosetta-4B"
+model_id = "/data/nas-2/seungduk/eeve2/babel/datasets/gemma-3-4b-rosetta-revision4-stage2"
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
     dtype=torch.bfloat16,
@@ -97,9 +98,20 @@ messages = [
 ]
 
 prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
+print(prompt)
+# <bos><start_of_turn>instruction
+# Translate the user's text to Korean.
+# Context: Simple introduction about a tech company.
+# Tone: Informative and helpful
+# Glossary:
+# - Yanolja NEXT -> 야놀자넥스트
+# - travel industry -> 여행 산업
+# Provide the final translation immediately without any other text.<end_of_turn>
+# <start_of_turn>source
+# {"company_name": "Yanolja NEXT", "description": "Yanolja NEXT is a company that provides cutting-edge technology for the global travel industry."}<end_of_turn>
+# <start_of_turn>translation
 
-# Store input length
+inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
 input_length = inputs["input_ids"].shape[1]
 
 with torch.inference_mode():
@@ -112,7 +124,6 @@ generated_tokens = outputs[0][input_length:]
 translation = tokenizer.decode(generated_tokens, skip_special_tokens=True)
 
 print(json.dumps(json.loads(translation), indent=2, ensure_ascii=False))
-
 # {
 #   "company_name": "야놀자넥스트",
 #   "description": "야놀자넥스트는 글로벌 여행 산업에 최첨단 기술을 제공하는 회사입니다."