ChatGPT Optimierung / Korrektur
This commit is contained in:
parent
c117a5561d
commit
d080f54087
|
|
@ -25,7 +25,7 @@ services:
|
||||||
- --port=8005
|
- --port=8005
|
||||||
- --quantization=awq
|
- --quantization=awq
|
||||||
- --tensor-parallel-size=2
|
- --tensor-parallel-size=2
|
||||||
- --gpu-memory-utilization=0.84
|
- --gpu-memory-utilization=0.86
|
||||||
- --max-model-len=65536
|
- --max-model-len=65536
|
||||||
- --hf-overrides={"rope_scaling":{"rope_type":"yarn","factor":2.0,"original_max_position_embeddings":32768}}
|
- --hf-overrides={"rope_scaling":{"rope_type":"yarn","factor":2.0,"original_max_position_embeddings":32768}}
|
||||||
- --max-num-seqs=4
|
- --max-num-seqs=4
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue