talk-llama : sync llama.cpp

Author: ggerganov

examples/talk-llama/llama.cpp

@@ -67,6 +67,7 @@ extern "C" {
         LLAMA_VOCAB_TYPE_SPM  = 1, // LLaMA tokenizer based on byte-level BPE with byte fallback
         LLAMA_VOCAB_TYPE_BPE  = 2, // GPT-2 tokenizer based on byte-level BPE
         LLAMA_VOCAB_TYPE_WPM  = 3, // BERT tokenizer based on WordPiece
+        LLAMA_VOCAB_TYPE_UGM  = 4, // T5 tokenizer based on Unigram
     };
 
     // pre-tokenization types
@@ -87,6 +88,10 @@ extern "C" {
         LLAMA_VOCAB_PRE_TYPE_DBRX           = 13,
         LLAMA_VOCAB_PRE_TYPE_SMAUG          = 14,
         LLAMA_VOCAB_PRE_TYPE_PORO           = 15,
+        LLAMA_VOCAB_PRE_TYPE_CHATGLM3       = 16,
+        LLAMA_VOCAB_PRE_TYPE_CHATGLM4       = 17,
+        LLAMA_VOCAB_PRE_TYPE_VIKING         = 18,
+        LLAMA_VOCAB_PRE_TYPE_JAIS           = 19,
     };
 
     // note: these values should be synchronized with ggml_rope
@@ -177,6 +182,12 @@ extern "C" {
         LLAMA_POOLING_TYPE_LAST = 3,
     };
 
+    enum llama_attention_type {
+        LLAMA_ATTENTION_TYPE_UNSPECIFIED = -1,
+        LLAMA_ATTENTION_TYPE_CAUSAL      = 0,
+        LLAMA_ATTENTION_TYPE_NON_CAUSAL  = 1,
+    };
+
     enum llama_split_mode {
         LLAMA_SPLIT_MODE_NONE    = 0, // single GPU
         LLAMA_SPLIT_MODE_LAYER   = 1, // split layers and KV across GPUs
@@ -294,6 +305,7 @@ extern "C" {
 
         enum llama_rope_scaling_type rope_scaling_type; // RoPE scaling type, from `enum llama_rope_scaling_type`
         enum llama_pooling_type      pooling_type;      // whether to pool (sum) embedding results by sequence id
+        enum llama_attention_type    attention_type;    // attention type to use for embeddings
 
         // ref: https://github.com/ggerganov/llama.cpp/pull/2054
         float    rope_freq_base;   // RoPE base frequency, 0 = from model
@@ -482,6 +494,13 @@ extern "C" {
     // Get a llama model tensor
     LLAMA_API struct ggml_tensor * llama_get_model_tensor(struct llama_model * model, const char * name);
 
+    // Returns true if the model contains an encoder that requires llama_encode() call
+    LLAMA_API bool llama_model_has_encoder(const struct llama_model * model);
+
+    // For encoder-decoder models, this function returns id of the token that must be provided
+    // to the decoder to start generating output sequence. For other models, it returns -1.
+    LLAMA_API llama_token llama_model_decoder_start_token(const struct llama_model * model);
+
     // Returns 0 on success
     LLAMA_API uint32_t llama_model_quantize(
             const char * fname_inp,
@@ -767,6 +786,14 @@ extern "C" {
     // Frees a batch of tokens allocated with llama_batch_init()
     LLAMA_API void llama_batch_free(struct llama_batch batch);
 
+    // Processes a batch of tokens with the ecoder part of the encoder-decoder model.
+    // Stores the encoder output internally for later use by the decoder cross-attention layers.
+    //   0 - success
+    // < 0 - error
+    LLAMA_API int32_t llama_encode(
+            struct llama_context * ctx,
+              struct llama_batch   batch);
+
     // Positive return values does not mean a fatal error, but rather a warning.
     //   0 - success
     //   1 - could not find a KV slot for the batch (try reducing the size of the batch or increase the context)
@@ -857,6 +884,7 @@ extern "C" {
     LLAMA_API llama_token llama_token_cls(const struct llama_model * model); // classification
     LLAMA_API llama_token llama_token_sep(const struct llama_model * model); // sentence separator
     LLAMA_API llama_token llama_token_nl (const struct llama_model * model); // next-line
+    LLAMA_API llama_token llama_token_pad(const struct llama_model * model); // padding
 
     // Returns -1 if unknown, 1 for true or 0 for false.
     LLAMA_API int32_t         llama_add_bos_token(const struct llama_model * model);
@@ -878,6 +906,7 @@ extern "C" {
     /// @param tokens The tokens pointer must be large enough to hold the resulting tokens.
     /// @return Returns the number of tokens on success, no more than n_tokens_max
     /// @return Returns a negative number on failure - the number of tokens that would have been returned
+    /// @param add_special Allow to add BOS and EOS tokens if model is configured to do so.
     /// @param parse_special Allow tokenizing special and/or control tokens which otherwise are not exposed and treated
     ///                      as plaintext. Does not insert a leading space.
     LLAMA_API int32_t llama_tokenize(
@@ -892,15 +921,31 @@ extern "C" {
     // Token Id -> Piece.
     // Uses the vocabulary in the provided context.
     // Does not write null terminator to the buffer.
-    // User code is responsible to remove the leading whitespace of the first non-BOS token when decoding multiple tokens.
+    // User can skip up to 'lstrip' leading spaces before copying (useful when encoding/decoding multiple tokens with 'add_space_prefix')
     // @param special If true, special tokens are rendered in the output.
     LLAMA_API int32_t llama_token_to_piece(
               const struct llama_model * model,
                            llama_token   token,
                                   char * buf,
                                int32_t   length,
+                               int32_t   lstrip,
                                   bool   special);
 
+    /// @details Convert the provided tokens into text (inverse of llama_tokenize()).
+    /// @param text The char pointer must be large enough to hold the resulting text.
+    /// @return Returns the number of chars/bytes on success, no more than text_len_max.
+    /// @return Returns a negative number on failure - the number of chars/bytes that would have been returned.
+    /// @param remove_special Allow to remove BOS and EOS tokens if model is configured to do so.
+    /// @param unparse_special If true, special tokens are rendered in the output.
+    LLAMA_API int32_t llama_detokenize(
+        const struct llama_model * model,
+               const llama_token * tokens,
+                         int32_t   n_tokens,
+                            char * text,
+                         int32_t   text_len_max,
+                            bool   remove_special,
+                            bool   unparse_special);
+
     /// Apply chat template. Inspired by hf apply_chat_template() on python.
     /// Both "model" and "custom_template" are optional, but at least one is required. "custom_template" has higher precedence than "model"
     /// NOTE: This function does not use a jinja parser. It only support a pre-defined list of template. See more: https://github.com/ggerganov/llama.cpp/wiki/Templates-supported-by-llama_chat_apply_template
@@ -924,6 +969,12 @@ extern "C" {
     // Grammar
     //
 
+    /// Initialize a llama_grammar.
+    ///
+    /// @param rules The rule elements of the grammar to initialize.
+    /// @param n_rules The number of rules.
+    /// @param start_rule_index The index of the root rule (the starting point of the grammar).
+    /// @return The initialized llama_grammar or nullptr if initialization failed.
     LLAMA_API struct llama_grammar * llama_grammar_init(
             const llama_grammar_element ** rules,
                                  size_t    n_rules,

examples/talk-llama/llama.h

@@ -35,10 +35,10 @@ static std::vector<llama_token> llama_tokenize(struct llama_context * ctx, const
 
 static std::string llama_token_to_piece(const struct llama_context * ctx, llama_token token) {
     std::vector<char> result(8, 0);
-    const int n_tokens = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), false);
+    const int n_tokens = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), 0, false);
     if (n_tokens < 0) {
         result.resize(-n_tokens);
-        int check = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), false);
+        int check = llama_token_to_piece(llama_get_model(ctx), token, result.data(), result.size(), 0, false);
         GGML_ASSERT(check == -n_tokens);
     } else {
         result.resize(n_tokens);

examples/talk-llama/talk-llama.cpp

@@ -7030,4 +7030,3 @@ const std::vector<range_nfd> unicode_ranges_nfd = {  // start, last, nfd
 {0x02FA1C, 0x02FA1C, 0x009F3B},
 {0x02FA1D, 0x02FA1D, 0x02A600},
 };
-

examples/talk-llama/unicode-data.cpp

@@ -23,7 +23,7 @@ static std::string unicode_cpts_to_utf8(const std::vector<uint32_t> & cps) {
     return result;
 }
 
-static uint32_t unicode_cpt_from_utf8(const std::string & utf8, size_t & offset) {
+uint32_t unicode_cpt_from_utf8(const std::string & utf8, size_t & offset) {
     assert(offset < utf8.size());
     if (!(utf8[offset + 0] & 0x80)) {
         auto result = utf8[offset + 0];
@@ -232,8 +232,7 @@ static std::vector<size_t> unicode_regex_split_custom_gpt2(const std::string & t
         };
 
         auto _get_flags = [&] (const size_t pos) -> codepoint_flags {
-            static const codepoint_flags undef(codepoint_flags::UNDEFINED);
-            return (offset_ini <= pos && pos < offset_end) ? unicode_cpt_flags(cpts[pos]) : undef;
+            return (offset_ini <= pos && pos < offset_end) ? unicode_cpt_flags(cpts[pos]) : codepoint_flags{};
         };
 
         size_t _prev_end = offset_ini;
@@ -295,9 +294,9 @@ static std::vector<size_t> unicode_regex_split_custom_gpt2(const std::string & t
                 continue;
             }
             // regex: <space>?[^\s\p{L}\p{N}]+
-            if (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number || flags2.is_undefined)) {
+            if (!(flags2.is_whitespace | flags2.is_letter | flags2.is_number) && flags2.as_uint()) {
                 pos += (cpt == ' ');
-                while (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number || flags2.is_undefined)) {
+                while (!(flags2.is_whitespace | flags2.is_letter | flags2.is_number) && flags2.as_uint()) {
                     flags2 = _get_flags(++pos);
                 }
                 _add_token(pos);
@@ -351,8 +350,7 @@ static std::vector<size_t> unicode_regex_split_custom_llama3(const std::string &
         };
 
         auto _get_flags = [&] (const size_t pos) -> codepoint_flags {
-            static const codepoint_flags undef(codepoint_flags::UNDEFINED);
-            return (offset_ini <= pos && pos < offset_end) ? unicode_cpt_flags(cpts[pos]) : undef;
+            return (offset_ini <= pos && pos < offset_end) ? unicode_cpt_flags(cpts[pos]) : codepoint_flags{};
         };
 
         size_t _prev_end = offset_ini;
@@ -394,8 +392,8 @@ static std::vector<size_t> unicode_regex_split_custom_llama3(const std::string &
                 }
             }
 
-            // regex: [^\r\n\p{L}\p{N}]?\p{L}+  //####FIXME: the first \p{L} is correct?
-            if (!(cpt == '\r' || cpt == '\n' || /*flags.is_letter |*/ flags.is_number)) {
+            // regex: [^\r\n\p{L}\p{N}]?\p{L}+
+            if (!(cpt == '\r' || cpt == '\n' || flags.is_number)) {
                 if (flags.is_letter || _get_flags(pos+1).is_letter) {  // one or more letters
                     pos++;
                     while (_get_flags(pos).is_letter) {
@@ -421,9 +419,9 @@ static std::vector<size_t> unicode_regex_split_custom_llama3(const std::string &
 
             // regex: <space>?[^\s\p{L}\p{N}]+[\r\n]*
             auto flags2 = (cpt == ' ' ? _get_flags(pos+1) : flags);
-            if (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number || flags2.is_undefined)) {
+            if (!(flags2.is_whitespace | flags2.is_letter | flags2.is_number) && flags.as_uint()) {
                 pos += (cpt == ' ');
-                while (!(flags2.is_whitespace || flags2.is_letter || flags2.is_number || flags2.is_undefined)) {
+                while (!(flags2.is_whitespace | flags2.is_letter | flags2.is_number) && flags2.as_uint()) {
                     flags2 = _get_flags(++pos);
                 }
                 uint32_t cpt2 = _get_cpt(pos);

examples/talk-llama/unicode.cpp

@@ -48,6 +48,7 @@ struct codepoint_flags {
 
 
 std::string unicode_cpt_to_utf8(uint32_t cp);
+uint32_t unicode_cpt_from_utf8(const std::string & utf8, size_t & offset);
 std::vector<uint32_t> unicode_cpts_from_utf8(const std::string & utf8);
 
 std::vector<uint32_t> unicode_cpts_normalize_nfd(const std::vector<uint32_t> & cpts);

examples/talk-llama/unicode.h

@@ -2949,7 +2949,7 @@ struct whisper_global_cache {
 // Mel spectrogram
 
 void whisper_mel_init(whisper_mel & mel, ggml_backend_t backend, int n_len, int n_len_org, int n_mel) {
-    WHISPER_LOG_INFO("%s: n_len = %d, n_len_org = %d, n_mel = %d\n", __func__, n_len, n_len_org, n_mel);
+    //WHISPER_LOG_INFO("%s: n_len = %d, n_len_org = %d, n_mel = %d\n", __func__, n_len, n_len_org, n_mel);
     mel.n_len_org = n_len_org;
     assert(!mel.ctx);
     mel.ctx = ggml_init({ggml_tensor_overhead(), nullptr, true});