Introduction Recently, while reviewing the papers I had previously studied in depth, I found that some notes were still very valuable. I made some minor adjustments and am publishing them for everyone to see. LLama3 is a paper from a few months ago, but each reading still brings new insights. This article discusses key points, … Read more
1. Test using the trained weights from transformers import AutoModelForCausalLM,AutoTokenizer,TextGenerationPipeline import torch tokenizer = AutoTokenizer.from_pretrained(r"E:\大模型AI开发\AI大模型\projects\gpt2\model\models–uer–gpt2-chinese-cluecorpussmall\snapshots\c2c0249d8a2731f269414cc3b22dff021f8e07a3") model = AutoModelForCausalLM.from_pretrained(r"E:\大模型AI开发\AI大模型\projects\gpt2\model\models–uer–gpt2-chinese-cluecorpussmall\snapshots\c2c0249d8a2731f269414cc3b22dff021f8e07a3") # Load our own trained weights (Chinese poetry) model.load_state_dict(torch.load("net.pt")) # Use the system's built-in pipeline tool to generate content pipline = TextGenerationPipeline(model,tokenizer,device=0) print(pipline("天高", max_length=24)) The performance is actually not good: 2. Post-process the AI-generated results # Customized … Read more
On April 18th, local time in the US, Meta released the highly anticipated open-source large language model Llama3, which offers two model sizes: 8B and 70B parameters, with a 400B version expected to be released in the future. Meta mentioned in their blog that thanks to improvements in training techniques, the Llama3 models, both 8B … Read more
1. Chatting If there is one industry that feels the most threatened by the arrival of AI, I believe e-commerce customer service should be on the list. Currently, platforms like Taobao and JD.com primarily use AI customer service for replies, only resorting to human agents if customers are dissatisfied, thus achieving cost reduction and efficiency … Read more