llama : fix handling of "future" tokens when loading sessions

This commit is contained in:
Georgi Gerganov 2023-10-03 18:29:22 +03:00
parent 0f332a9104
commit 337120cc0d
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
6 changed files with 41 additions and 40 deletions

View file

@ -543,6 +543,9 @@ int main(int argc, char ** argv) {
if (i > 0) {
embd.erase(embd.begin(), embd.begin() + i);
}
// remove any "future" tokens that we might have inherited from the session from the KV cache
llama_kv_cache_tokens_rm(ctx, n_past, -1);
}
// evaluate tokens in batches

View file

@ -332,7 +332,7 @@ int main(int argc, char ** argv) {
}
// delete only the generated part of the sequence, i.e. keep the system prompt in the cache
llama_kv_cache_seq_rm(ctx, client.id, n_tokens_system, n_ctx);
llama_kv_cache_seq_rm(ctx, client.id, n_tokens_system, -1);
const auto t_main_end = ggml_time_us();

View file

@ -448,7 +448,7 @@ struct llama_server_context
n_past = common_part(embd, prompt_tokens);
// since #3228 we now have to manually manage the KV cache
llama_kv_cache_seq_rm(ctx, 0, n_past, params.n_ctx);
llama_kv_cache_seq_rm(ctx, 0, n_past, -1);
embd = prompt_tokens;
if (n_past == num_prompt_tokens)

View file

@ -172,7 +172,7 @@ int main(int argc, char ** argv) {
LOG("out of drafted tokens\n");
}
llama_kv_cache_seq_rm(ctx_dft, 0, n_past_dft, n_ctx);
llama_kv_cache_seq_rm(ctx_dft, 0, n_past_dft, -1);
llama_decode(ctx_dft, llama_batch_get_one(&id, 1, n_past_dft, 0));
++n_past_dft;
@ -257,7 +257,7 @@ int main(int argc, char ** argv) {
}
// evaluate the drafted token on the draft model
llama_kv_cache_seq_rm(ctx_dft, 0, n_past_cur, n_ctx);
llama_kv_cache_seq_rm(ctx_dft, 0, n_past_cur, -1);
llama_decode(ctx_dft, llama_batch_get_one(&drafted.back(), 1, n_past_cur, 0));
++n_past_cur;
@ -267,7 +267,7 @@ int main(int argc, char ** argv) {
}
// evaluate the target model on the drafted tokens
llama_kv_cache_seq_rm(ctx_tgt, 0, n_past_tgt, n_ctx);
llama_kv_cache_seq_rm(ctx_tgt, 0, n_past_tgt, -1);
llama_decode(ctx_tgt, llama_batch_get_one(drafted.data(), drafted.size(), n_past_tgt, 0));
++n_past_tgt;