25 1 year ago

deepseek预训练1.5B

ollama run weixuan/pre_train_deepseek1.5B

Details

1 year ago

57efb37951bf · 2.3GB ·

llama
·
3.82B
·
Q4_K_M
{{ if .System }}<|system|> {{ .System }}<|end|> {{ end }}{{ if .Prompt }}<|user|> {{ .Prompt }}<|end
{ "num_ctx": 4096, "num_keep": 4, "stop": [ "<|user|>", "<|assistant|>",

Readme

No readme