From 1a9141b6c3101ce8a7f54a2466816fee6bdfa810 Mon Sep 17 00:00:00 2001 From: Randall Fitzgerald Date: Fri, 9 Jun 2023 16:29:10 -0400 Subject: [PATCH] Remove model assign in main(). Clarified stop in README. The model will now load the default from gptparams ("models/7B/ggml-model.bin") --- examples/server/README.md | 2 +- examples/server/server.cpp | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/examples/server/README.md b/examples/server/README.md index 364291ee3..4ff4f5f6d 100644 --- a/examples/server/README.md +++ b/examples/server/README.md @@ -110,7 +110,7 @@ node . `prompt`: Provide a prompt. Internally, the prompt is compared, and it detects if a part has already been evaluated, and the remaining part will be evaluate. - `stop`: Specify the strings that indicate a stop. + `stop`: Specify a JSON array of stopping strings. These words will not be included in the completion, so make sure to add them to the prompt for the next iteration (default: []). `tfs_z`: Enable tail free sampling with parameter z (default: 1.0, 1.0 = disabled). diff --git a/examples/server/server.cpp b/examples/server/server.cpp index afe0cc03c..ee645c143 100644 --- a/examples/server/server.cpp +++ b/examples/server/server.cpp @@ -774,7 +774,6 @@ int main(int argc, char ** argv) // struct that contains llama context and inference llama_server_context llama; - params.model = "ggml-model.bin"; server_params_parse(argc, argv, sparams, params);