Skip to content

Some improvements to loading the session with --prompt-cache #1550

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
May 26, 2023
Merged

Some improvements to loading the session with --prompt-cache #1550

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
de5bf5b
Some improvements to loading the session with --prompt-cache
KerfuffleV2 May 21, 2023
2d79928
Apply clang suggestions.
KerfuffleV2 May 21, 2023
e1ec489
Use existing session behavior when in instruct or interact first mode
KerfuffleV2 May 23, 2023
76c7398
Use the initial value of params.seed to determine if user supplied seed
KerfuffleV2 May 24, 2023
156d70b
Always set RNG seed when restoring cached prompt in main example.
KerfuffleV2 May 25, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion examples/main/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -272,7 +272,7 @@ These options help improve the performance and memory usage of the LLaMA models.

### Prompt Caching

- `--prompt-cache FNAME`: Specify a file to cache the model state after the initial prompt. This can significantly speed up the startup time when you're using longer prompts. The file is created during the first run and is reused and updated in subsequent runs.
- `--prompt-cache FNAME`: Specify a file to cache the model state after the initial prompt. This can significantly speed up the startup time when you're using longer prompts. The file is created during the first run and is reused and updated in subsequent runs. **Note**: Restoring a cached prompt does not imply restoring the exact state of the session at the point it was saved. So even when specifying a specific seed, you are not guaranteed to get the same sequence of tokens as the original generation.

### Quantization

Expand Down
18 changes: 14 additions & 4 deletions examples/main/main.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,8 +134,6 @@ int main(int argc, char ** argv) {
return 0;
}

// Add a space in front of the first character to match OG llama tokenizer behavior
params.prompt.insert(0, 1, ' ');

std::string path_session = params.path_prompt_cache;
std::vector<llama_token> session_tokens;
Expand All @@ -155,6 +153,7 @@ int main(int argc, char ** argv) {
return 1;
}
session_tokens.resize(n_token_count_out);
llama_set_rng_seed(ctx, params.seed);

fprintf(stderr, "%s: loaded a session with prompt size of %d tokens\n", __func__, (int) session_tokens.size());
} else {
Expand All @@ -163,7 +162,16 @@ int main(int argc, char ** argv) {
}

// tokenize the prompt
auto embd_inp = ::llama_tokenize(ctx, params.prompt, true);
std::vector<llama_token> embd_inp;

if (params.interactive_first || params.instruct || !params.prompt.empty() || session_tokens.empty()) {
// Add a space in front of the first character to match OG llama tokenizer behavior
params.prompt.insert(0, 1, ' ');

embd_inp = ::llama_tokenize(ctx, params.prompt, true);
} else {
embd_inp = session_tokens;
}

const int n_ctx = llama_n_ctx(ctx);

Expand All @@ -181,7 +189,9 @@ int main(int argc, char ** argv) {
}
n_matching_session_tokens++;
}
if (n_matching_session_tokens >= embd_inp.size()) {
if (params.prompt.empty() && n_matching_session_tokens == embd_inp.size()) {
fprintf(stderr, "%s: using full prompt from session file\n", __func__);
} else if (n_matching_session_tokens >= embd_inp.size()) {
fprintf(stderr, "%s: session file has exact match for prompt!\n", __func__);
} else if (n_matching_session_tokens < (embd_inp.size() / 2)) {
fprintf(stderr, "%s: warning: session file has low similarity to prompt (%zu / %zu tokens); will mostly be reevaluated\n",
Expand Down