vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

desc

Browse source
This commit is contained in:
Concedo 2023-06-26 16:58:59 +08:00
parent e4c9aea840
commit 1fdf9d1131
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
koboldcpp.py
View file

@ -848,7 +848,7 @@ if __name__ == '__main__':
parser.add_argument("--highpriority", help="Experimental flag. If set, increases the process CPU priority, potentially speeding up generation. Use caution.", action='store_true') parser.add_argument("--highpriority", help="Experimental flag. If set, increases the process CPU priority, potentially speeding up generation. Use caution.", action='store_true')
parser.add_argument("--contextsize", help="Controls the memory allocated for maximum context size, only change if you need more RAM for big contexts. (default 2048)", type=int,choices=[512,1024,2048,4096,8192], default=2048) parser.add_argument("--contextsize", help="Controls the memory allocated for maximum context size, only change if you need more RAM for big contexts. (default 2048)", type=int,choices=[512,1024,2048,4096,8192], default=2048)
parser.add_argument("--blasbatchsize", help="Sets the batch size used in BLAS processing (default 512). Setting it to -1 disables BLAS mode, but keeps other benefits like GPU offload.", type=int,choices=[-1,32,64,128,256,512,1024], default=512) parser.add_argument("--blasbatchsize", help="Sets the batch size used in BLAS processing (default 512). Setting it to -1 disables BLAS mode, but keeps other benefits like GPU offload.", type=int,choices=[-1,32,64,128,256,512,1024], default=512)
parser.add_argument("--stream", help="Uses pseudo streaming when generating tokens. Only for the Kobold Lite UI.", action='store_true') parser.add_argument("--stream", help="Uses streaming when generating tokens. Only for the Kobold Lite UI.", action='store_true')
parser.add_argument("--smartcontext", help="Reserving a portion of context to try processing less frequently.", action='store_true') parser.add_argument("--smartcontext", help="Reserving a portion of context to try processing less frequently.", action='store_true')
parser.add_argument("--unbantokens", help="Normally, KoboldAI prevents certain tokens such as EOS and Square Brackets. This flag unbans them.", action='store_true') parser.add_argument("--unbantokens", help="Normally, KoboldAI prevents certain tokens such as EOS and Square Brackets. This flag unbans them.", action='store_true')
parser.add_argument("--usemirostat", help="Experimental! Replaces your samplers with mirostat. Takes 3 params = [type(0/1/2), tau(5.0), eta(0.1)].",metavar=('[type]', '[tau]', '[eta]'), type=float, nargs=3) parser.add_argument("--usemirostat", help="Experimental! Replaces your samplers with mirostat. Takes 3 params = [type(0/1/2), tau(5.0), eta(0.1)].",metavar=('[type]', '[tau]', '[eta]'), type=float, nargs=3)