From 7a9b6c3a8bdc1cb75fefc826dfaa7331eb63695d Mon Sep 17 00:00:00 2001
From: Georgi Gerganov <ggerganov@gmail.com>
Date: Fri, 24 Mar 2023 23:17:37 +0200
Subject: Reduce memory usage and allocate enough memory for largest context
 (#473)

* Reduce memory usage and allocate enough memory for large contexts

* Simpler scratch buffer usage

* Reenable BLAS for quantized mul_mat

* Fix number of layers in 30B and 65B

* Fix KV cache size for F32
---
 utils.cpp | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'utils.cpp')

diff --git a/utils.cpp b/utils.cpp
index 10673fb..2f995c1 100644
--- a/utils.cpp
+++ b/utils.cpp
@@ -79,8 +79,8 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
                 break;
             }
             params.n_ctx = std::stoi(argv[i]);
-        } else if (arg == "--memory_f16") {
-            params.memory_f16 = true;
+        } else if (arg == "--memory_f32") {
+            params.memory_f16 = false;
         } else if (arg == "--top_p") {
             if (++i >= argc) {
                 invalid_param = true;
@@ -111,6 +111,7 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
                 break;
             }
             params.n_batch = std::stoi(argv[i]);
+            params.n_batch = std::min(512, params.n_batch);
         } else if (arg == "-m" || arg == "--model") {
             if (++i >= argc) {
                 invalid_param = true;
@@ -131,6 +132,8 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
             params.use_color = true;
         } else if (arg == "--mlock") {
             params.use_mlock = true;
+        } else if (arg == "--mtest") {
+            params.mem_test = true;
         } else if (arg == "-r" || arg == "--reverse-prompt") {
             if (++i >= argc) {
                 invalid_param = true;
@@ -193,7 +196,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     fprintf(stderr, "  --repeat_penalty N    penalize repeat sequence of tokens (default: %.1f)\n", params.repeat_penalty);
     fprintf(stderr, "  -c N, --ctx_size N    size of the prompt context (default: %d)\n", params.n_ctx);
     fprintf(stderr, "  --ignore-eos          ignore end of stream token and continue generating\n");
-    fprintf(stderr, "  --memory_f16          use f16 instead of f32 for memory key+value\n");
+    fprintf(stderr, "  --memory_f32          use f32 instead of f16 for memory key+value\n");
     fprintf(stderr, "  --temp N              temperature (default: %.1f)\n", params.temp);
     fprintf(stderr, "  --n_parts N           number of model parts (default: -1 = determine from dimensions)\n");
     fprintf(stderr, "  -b N, --batch_size N  batch size for prompt processing (default: %d)\n", params.n_batch);
@@ -201,6 +204,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     if (ggml_mlock_supported()) {
         fprintf(stderr, "  --mlock               force system to keep model in RAM rather than swapping or compressing\n");
     }
+    fprintf(stderr, "  --mtest               compute maximum memory usage\n");
     fprintf(stderr, "  -m FNAME, --model FNAME\n");
     fprintf(stderr, "                        model path (default: %s)\n", params.model.c_str());
     fprintf(stderr, "\n");
-- 
cgit v1.2.3