ruanshudong

I have a submodule with another token, how can i set build-push-action?

3

``` #1 [internal] load git source https://github.com/TarsCloudMarket/CountServer.git#4c66a29c33814cc24e93ed470d4a1ccc3ed98b05 #0 0.041 hint: Using 'master' as the name for the initial branch. This default branch name #0 0.041 hint: is subject to change....

使用预训练脚本, 输入目录下是txt文本, 为啥有这个报错, 貌似用json来解析?可能是哪里传参不对?

20

Traceback (most recent call last): File "/home/data/LLM/llama/data/run_clm_pt_with_peft.py", line 461, in main processed_dataset = datasets.load_from_disk(cache_path, keep_in_memory=False) File "/home/data/anaconda3/lib/python3.10/site-packages/datasets/load.py", line 1886, in load_from_disk return DatasetDict.load_from_disk(dataset_path, keep_in_memory=keep_in_memory, storage_options=storage_options) File "/home/data/anaconda3/lib/python3.10/site-packages/datasets/dataset_dict.py", line 1308, in...

合并完以后, 推理都有问题? 原始模型推理怎没有问题

1

合并: python ../Chinese-LLaMA-Alpaca/scripts/merge_llama_with_chinese_lora.py --base_model=../pyllama_data/7B_hf --lora_model=../chinese-llama-plus-lora-7b/,../chinese-alpaca-plus-lora-7b --output_type=huggingface --output_dir=merge_llama 执行推理: python inference_hf.py --base_model=merge_llama/ --interactive load: merge_llama/ Loading checkpoint shards: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:06

合并了生成了alpaca模型, 基于这个模型, 做预训练, 然后用inference_hf, 出现错误, 可能是什么原因?还是说不能这么干?

15

python inference_hf.py --base_model=merge_alpaca_plus/ --lora_model=lora-llama-7b/ --interactive --with_prompt load: merge_alpaca_plus/ Loading checkpoint shards: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:21

请教一下, 预训练节点, eos和bos是怎么添加到训练文本里面的?

2

具体是谁来做的? 是文本预处理的时候就添加好(在每篇文章开头和结尾增加eos/bos), 还是代码里面自动做的?这里一直没太弄明白, 多谢!

Execution failed: terminated

1

### Describe the bug 2025-10-30T06:59:11.360Z [ERROR] Command threw error: terminated TypeError: terminated at tHe.G (file:///Users/jarod/.nvm/versions/node/v22.21.0/lib/node_modules/@github/copilot/index.js:775:43922) at tHe.emit (node:events:519:28) at tHe.terminate (file:///Users/jarod/.nvm/versions/node/v22.21.0/lib/node_modules/@github/copilot/index.js:775:30640) at Object.onError (file:///Users/jarod/.nvm/versions/node/v22.21.0/lib/node_modules/@github/copilot/index.js:775:46236) at Nrt.onError (file:///Users/jarod/.nvm/versions/node/v22.21.0/lib/node_modules/@github/copilot/index.js:648:77963) at Object.v$n [as...