(base) ai@User:~$ htop
Command 'htop' not found, but can be installed with:
sudo snap install htop # version 3.4.0, or
sudo apt install htop # version 3.2.2-2
See 'snap info htop' for additional versions.
(base) ai@User:~$ conda activate blackwell_test
(blackwell_test) ai@User:~$ python
Python 3.10.16 (main, Dec 11 2024, 16:24:50) [GCC 11.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from transformers import AutoTokenizer, AutoModelForCausalLM
XAONE/EXAONE-Deep-7.8B"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, device_map="auto")
prompt = "넌 무엇을 할수있어?."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=150)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(result)
>>>
>>> model_id = "LGAI-EXAONE/EXAONE-Deep-7.8B"
>>> tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
>>> model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, device_map="auto")
Loading checkpoint shards: 100%|███████████████████████████████████| 4/4 [00:05<00:00, 1.39s/it]
Some parameters are on the meta device because they were offloaded to the disk and cpu.
>>>
>>> prompt = "넌 무엇을 할수있어?."
>>> inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
>>> outputs = model.generate(**inputs, max_new_tokens=150)
이런식으로 내가 한 질문에 답변을 안하고 따라하기만하는데 내가 코드를 잘못짠건가 지피티가 짜준 코드대로 한거임
회사에서 엑사원쓰라고 ㅈㄹ해서 써봤는데 좃구려서 바로 거름 - dc App
로컬모델은 원래 능지 박살나서 장기 불가능할걸 첫 몇입먹고 새롭게 파야함