sampling : refactor + optimize penalties sampler (#10803)

* sampling : refactor + optimize penalties sampler

ggml-ci

* common : apply ignore_eos as logit bias

ggml-ci

* batched : remove penalties sampler

* params : allow penalty_last_n == -1 to be equal to context size

ggml-ci

* common : by default, move the penalties at the end of the sampling chain

ggml-ci

* common : ignore all EOG tokens

Co-authored-by: Diego Devesa <slarengh@gmail.com>

* common : move back the penalties at the front of the sampling chain

ggml-ci

* readme : restore hint about --ignore-eos flag [no ci]

* llama : minor

ggml-ci

* webui : update

---------

Co-authored-by: Diego Devesa <slarengh@gmail.com>
This commit is contained in:
Georgi Gerganov 2024-12-16 12:31:14 +02:00 committed by GitHub
parent 4ddd199f6f
commit 644fd71b44
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
17 changed files with 111 additions and 152 deletions

View file

@ -940,6 +940,25 @@ struct common_init_result common_init_from_params(common_params & params) {
params.sampling.ignore_eos = false;
}
if (params.sampling.ignore_eos) {
for (llama_token i = 0; i < llama_n_vocab(model); i++) {
if (llama_token_is_eog(model, i)) {
LOG_INF("%s: added %s logit bias = %f\n", __func__, common_token_to_piece(lctx, i).c_str(), -INFINITY);
params.sampling.logit_bias.push_back({i, -INFINITY});
}
}
}
if (params.sampling.penalty_last_n == -1) {
LOG_INF("%s: setting penalty_last_n to ctx_size = %d\n", __func__, llama_n_ctx(lctx));
params.sampling.penalty_last_n = llama_n_ctx(lctx);
}
if (params.sampling.dry_penalty_last_n == -1) {
LOG_INF("%s: setting dry_penalty_last_n to ctx_size = %d\n", __func__, llama_n_ctx(lctx));
params.sampling.dry_penalty_last_n = llama_n_ctx(lctx);
}
if (params.warmup) {
LOG_WRN("%s: warming up the model with an empty run - please wait ... (--no-warmup to disable)\n", __func__);