Reproduction / 复现过程 import torch from transformers import AutoModelForCausalLM, AutoTokenizer device = "cuda" tokenizer = AutoTokenizer.from_pretrained("/opt/glm-4-9b-chat",trust_remote_code=True) query = "你好" inputs = tokenizer.apply_chat_template([{"role": "user", "content": query...
1、在\fastchat\model\model_chatglm.py文件的91行添加一个判断: elif "glm-4" in model_type: message_list = recover_message_list(prompt) inputs = tokenizer.apply_chat_template( message_list, tokenize=True, return_tensors="pt", return_dict=True, add_generation_prompt=True ).to(model.device...
File "C:\Users\hello.cache\huggingface\modules\transformers_modules\glm4\modeling_chatglm.py", line 260, in forward value_layer = value_layer.view(output_size[0] * output_size[1], value_layer.size(2), -1) RuntimeError: shape '[224, 7, -1]' is invalid for input of size 28672 如...
在语义、数学、推理、代码和知识等多方面的数据集测评中,GLM-4-9B及其人类偏好对齐的版本GLM-4-9B-Chat均表现出超越 Llama-3-8B 的卓越性能。 开源地址:【https://github.com/THUDM/GLM-4】 在线体验:【https://modelscope.cn/studios/dash-infer/GLM-4-Chat-DashInfer-Demo/summary】魔搭社区提供 GLM-4...
能否支持GLM-4-9B-Chat#4858 Forevery1opened this issueJun 6, 2024· 1 comment Forevery1commentedJun 6, 2024 希望支持GLM-4-9B-Chat The text was updated successfully, but these errors were encountered: Forevery1added themodel requestModel requestslabelJun 6, 2024 ...
Please can we get an official release of THUDM/glm-4-9b-chat It is doing really well in chat and coding for a 9b parameter model. 👍 6 ekolawole added the model request label Jun 6, 2024 Member mchiang0610 commented Jun 6, 2024 Thank you for submitting this... merging this ...
siegrainwongcommentedAug 30, 2024 开过flash attention,不起作用 junruizh2021mentioned this issueOct 14, 2024 ollama generate incorrect answer when it run glm-4-9b-chat modelintel-analytics/ipex-llm#12193 Open sixsixcoderclosed this ascompletedOct 21, 2024...
return await loop.run_in_executor(pool, self._chat, *input_args) File "miniconda3/envs/new_llama_factory_py3.10/lib/python3.10/concurrent/futures/thread.py", line 58, in run result = self.fn(*self.args, **self.kwargs) File "miniconda3/envs/new_llama_factory_py3.10/lib/python3.10/...
现在还没没有支持glm-4-9b-chat模型吗,模型配置文件里没有找到glm-4-9b-chat相关配置? https://github.com/THUDM/GLM-4 解决的问题 / Problem Solved 实现建议 / Implementation Suggestions 替代方案 / Alternative Solutions 其他信息 / Additional Information...
ChatGLM3-6B-Base61.469.026.872.325.758.5 GLM-4-9B74.777.134.384.030.470.1 由于GLM-4-9B在预训练过程中加入了部分数学、推理、代码相关的 instruction 数据,所以将 Llama-3-8B-Instruct 也列入比较范围。 长文本 在1M 的上下文长度下进行大海捞针实验,结果如下: ...