-
Try to load model, but failed. Below is the output.
```
D:\AI\ChatRWKV\rwkv-cpp-cuda> chat.exe
D:\AI\ChatRWKV\rwkv-cpp-cuda/model.bin
n_layers: 431231561062
n_embed: 17592186094693
…
-
https://github.com/BlinkDL/ChatRWKV
-
(JittorLLMs) C:\o\JittorLLMs>python cli_demo.py [chatglm|pangualpha|llama|chatrwkv]
'pangualpha' 不是内部或外部命令，也不是可运行的程序
或批处理文件。
(JittorLLMs) C:\o\JittorLLMs>python cli_demo.py [chatglm|pangualpha|ll…
wsmdn updated
9 months ago
-
(gh_ChatRWKV_py11) amd00@MZ32-00:~/llm_dev/ChatRWKV$ python v2/chat.py
ChatRWKV v2 https://github.com/BlinkDL/ChatRWKV
Chinese - cuda fp16 - /home/amd00/llm_dev/ChatRWKV/v2/prompt/default/Ch…
-
Hi I am the dev of https://github.com/BlinkDL/ChatRWKV and it is a RNN (so faster and saves VRAM) that can match transformer performance (and already scaled to 14B params. more to come).
Let me kno…
-
### 🐛 Describe the bug
Run [ChatRWKV](https://github.com/BlinkDL/ChatRWKV) using 'mps', returna a very big number, looks like overflow.
MBP(intel CPU, not M1/M2), with eGPU[rx6800 16G]
pytorch==2…
-
when i run the RWKV-LM-Lora,i meet that error,
the model i use:
RWKV-4-Raven-1B5-v9-Eng99%-Other1%-20230411-ctx4096.pth
my insruction run in wsl2：
python3 train.py --load_model /home/wubo/chatRWKV…
-
希望后续支持chatRWKV并行训练，这个现在用的人那么火，不搞一下。 ~~
-
运行Readme文档中的这句部署命令时python cli_demo.py [chatglm|pangualpha|llama|chatrwkv]
报错如下：
pangualpha : The term 'pangualpha' is not recognized as the name of a cmdlet, function, script file, or operable pro…
-
2023.6.12 发布 PaddleNLP v2.6rc 预览版
🔨 大模型全流程范例：全面支持主流开源大模型Bloom, ChatGLM, GLM, Llama, OPT的训练和推理；Trainer API新增张量训练能力, 简单配置即可开启分布式训练；新增低参数微调能力PEFT, 助力大模型高效微调
------------------------- 提个建议 希望支持chatR…