gguf : new file format with flexible meta data (beta) (#2398)

* gguf : first API pass

* gguf : read header + meta data

* gguf : read tensor info

* gguf : initial model loading - not tested

* gguf : add gguf_get_tensor_name()

* gguf : do not support passing existing ggml_context to gguf_init

* gguf : simplify gguf_get_val

* gguf : gguf.c is now part of ggml.c

* gguf : read / write sample models

* gguf : add comments

* refactor : reduce code duplication and better API (#2415)

* gguf : expose the gguf_type enum through the API for now

* gguf : add array support

* gguf.py : some code style changes

* convert.py : start a new simplified implementation by removing old stuff

* convert.py : remove GGML vocab + other obsolete stuff

* GGUF : write tensor (#2426)

* WIP: Write tensor

* GGUF : Support writing tensors in Python

* refactor : rm unused import and upd todos

* fix : fix errors upd writing example

* rm example.gguf

* gitignore *.gguf

* undo formatting

* gguf : add gguf_find_key (#2438)

* gguf.cpp : find key example

* ggml.h : add gguf_find_key

* ggml.c : add gguf_find_key

* gguf : fix writing tensors

* gguf : do not hardcode tensor names to read

* gguf : write sample tensors to read

* gguf : add tokenization constants

* quick and dirty conversion example

* gguf : fix writing gguf arrays

* gguf : write tensors one by one and code reuse

* gguf : fix writing gguf arrays

* gguf : write tensors one by one

* gguf : write tensors one by one

* gguf : write tokenizer data

* gguf : upd gguf conversion script

* Update convert-llama-h5-to-gguf.py

* gguf : handle already encoded string

* ggml.h : get array str and f32

* ggml.c : get arr str and f32

* gguf.py : support any type

* Update convert-llama-h5-to-gguf.py

* gguf : fix set is not subscriptable

* gguf : update convert-llama-h5-to-gguf.py

* constants.py : add layer norm eps

* gguf.py : add layer norm eps and merges

* ggml.h : increase GGML_MAX_NAME to 64

* ggml.c : add gguf_get_arr_n

* Update convert-llama-h5-to-gguf.py

* add gptneox gguf example

* Makefile : add gptneox gguf example

* Update convert-llama-h5-to-gguf.py

* add gptneox gguf example

* Update convert-llama-h5-to-gguf.py

* Update convert-gptneox-h5-to-gguf.py

* Update convert-gptneox-h5-to-gguf.py

* Update convert-llama-h5-to-gguf.py

* gguf : support custom alignment value

* gguf : fix typo in function call

* gguf : mmap tensor data example

* fix : update convert-llama-h5-to-gguf.py

* Update convert-llama-h5-to-gguf.py

* convert-gptneox-h5-to-gguf.py : Special tokens

* gptneox-main.cpp : special tokens

* Update gptneox-main.cpp

* constants.py : special tokens

* gguf.py : accumulate kv and tensor info data + special tokens

* convert-gptneox-h5-to-gguf.py : accumulate kv and ti + special tokens

* gguf : gguf counterpart of llama-util.h

* gguf-util.h : update note

* convert-llama-h5-to-gguf.py : accumulate kv / ti + special tokens

* convert-llama-h5-to-gguf.py : special tokens

* Delete gptneox-common.cpp

* Delete gptneox-common.h

* convert-gptneox-h5-to-gguf.py : gpt2bpe tokenizer

* gptneox-main.cpp : gpt2 bpe tokenizer

* gpt2 bpe tokenizer (handles merges and unicode)

* Makefile : remove gptneox-common

* gguf.py : bytesarray for gpt2bpe tokenizer

* cmpnct_gpt2bpe.hpp : comments

* gguf.py : use custom alignment if present

* gguf : minor stuff

* Update gptneox-main.cpp

* map tensor names

* convert-gptneox-h5-to-gguf.py : map tensor names

* convert-llama-h5-to-gguf.py : map tensor names

* gptneox-main.cpp : map tensor names

* gguf : start implementing libllama in GGUF (WIP)

* gguf : start implementing libllama in GGUF (WIP)

* rm binary commited by mistake

* upd .gitignore

* gguf : calculate n_mult

* gguf :  inference with 7B model working (WIP)

* gguf : rm deprecated function

* gguf : start implementing gguf_file_saver (WIP)

* gguf : start implementing gguf_file_saver (WIP)

* gguf : start implementing gguf_file_saver (WIP)

* gguf : add gguf_get_kv_type

* gguf : add gguf_get_kv_type

* gguf : write metadata in gguf_file_saver (WIP)

* gguf : write metadata in gguf_file_saver (WIP)

* gguf : write metadata in gguf_file_saver

* gguf : rm references to old file formats

* gguf : shorter name for member variable

* gguf : rm redundant method

* gguf : get rid of n_mult, read n_ff from file

* Update gguf_tensor_map.py

* Update gptneox-main.cpp

* gguf : rm references to old file magics

* gguf : start implementing quantization (WIP)

* gguf : start implementing quantization (WIP)

* gguf : start implementing quantization (WIP)

* gguf : start implementing quantization (WIP)

* gguf : start implementing quantization (WIP)

* gguf : start implementing quantization (WIP)

* gguf : quantization is working

* gguf : roper closing of file

* gguf.py : no need to convert tensors twice

* convert-gptneox-h5-to-gguf.py : no need to convert tensors twice

* convert-llama-h5-to-gguf.py : no need to convert tensors twice

* convert-gptneox-h5-to-gguf.py : simplify nbytes

* convert-llama-h5-to-gguf.py : simplify nbytes

* gptneox-main.cpp : n_layer --> n_block

* constants.py : n_layer --> n_block

* gguf.py : n_layer --> n_block

* convert-gptneox-h5-to-gguf.py : n_layer --> n_block

* convert-llama-h5-to-gguf.py : n_layer --> n_block

* gptneox-main.cpp : n_layer --> n_block

* Update gguf_tensor_map.py

* convert-gptneox-h5-to-gguf.py : load model in parts to save memory

* convert-llama-h5-to-gguf.py : load model in parts to save memory

* convert : write more metadata for LLaMA

* convert : rm quantization version

* convert-gptneox-h5-to-gguf.py : add file_type key

* gptneox-main.cpp : add file_type key

* fix conflicts

* gguf : add todos and comments

* convert-gptneox-h5-to-gguf.py : tensor name map changes

* Create gguf_namemap.py : tensor name map changes

* Delete gguf_tensor_map.py

* gptneox-main.cpp : tensor name map changes

* convert-llama-h5-to-gguf.py : fixes

* gguf.py : dont add empty strings

* simple : minor style changes

* gguf : use UNIX line ending

* Create convert-llama-7b-pth-to-gguf.py

* llama : sync gguf-llama.cpp with latest llama.cpp (#2608)

* llama : sync gguf-llama.cpp with latest llama.cpp

* minor : indentation + assert

* llama : refactor gguf_buffer and gguf_ctx_buffer

* llama : minor

* gitignore : add gptneox-main

* llama : tokenizer fixes (#2549)

* Merge tokenizer fixes into the gguf branch.

* Add test vocabularies

* convert : update convert-new.py with tokenizer fixes (#2614)

* Merge tokenizer fixes into the gguf branch.

* Add test vocabularies

* Adapt convert-new.py (and fix a clang-cl compiler error on windows)

* llama : sync gguf-llama with llama (#2613)

* llama : sync gguf-llama with llama

* tests : fix build + warnings (test-tokenizer-1 still fails)

* tests : fix wstring_convert

* convert : fix layer names

* llama : sync gguf-llama.cpp

* convert : update HF converter to new tokenizer voodoo magics

* llama : update tokenizer style

* convert-llama-h5-to-gguf.py : add token types

* constants.py : add token types

* gguf.py : add token types

* convert-llama-7b-pth-to-gguf.py : add token types

* gguf-llama.cpp :  fix n_head_kv

* convert-llama-h5-to-gguf.py : add 70b gqa support

* gguf.py : add tensor data layout

* convert-llama-h5-to-gguf.py : add tensor data layout

* convert-llama-7b-pth-to-gguf.py : add tensor data layout

* gptneox-main.cpp : add tensor data layout

* convert-llama-h5-to-gguf.py : clarify the reverse permute

* llama : refactor model loading code (#2620)

* llama : style formatting + remove helper methods

* llama : fix quantization using gguf tool

* llama : simplify gguf_file_saver

* llama : fix method names

* llama : simplify write_header()

* llama : no need to pass full file loader to the file saver

just gguf_ctx

* llama : gguf_file_saver write I32

* llama : refactor tensor names (#2622)

* gguf: update tensor names searched in quantization

* gguf : define tensor names as constants

* gguf : initial write API (not tested yet)

* gguf : write to file API (not tested)

* gguf : initial write API ready + example

* gguf : fix header write

* gguf : fixes + simplify example + add ggml_nbytes_pad()

* gguf : minor

* llama : replace gguf_file_saver with new gguf write API

* gguf : streaming support when writing files

* gguf : remove oboslete write methods

* gguf : remove obosolete gguf_get_arr_xxx API

* llama : simplify gguf_file_loader

* llama : move hparams and vocab from gguf_file_loader to llama_model_loader

* llama : merge gguf-util.h in llama.cpp

* llama : reorder definitions in .cpp to match .h

* llama : minor simplifications

* llama : refactor llama_model_loader (WIP)

wip : remove ggml_ctx from llama_model_loader

wip : merge gguf_file_loader in llama_model_loader

* llama : fix shape prints

* llama : fix Windows build + fix norm_rms_eps key

* llama : throw error on missing KV paris in model meta data

* llama : improve printing + log meta data

* llama : switch print order of meta data

---------

Co-authored-by: M. Yusuf Sarıgöz <yusufsarigoz@gmail.com>

* gguf : deduplicate (#2629)

* gguf : better type names

* dedup : CPU + Metal is working

* ggml : fix warnings about unused results

* llama.cpp : fix line feed and compiler warning

* llama : fix strncpy warning + note token_to_str does not write null

* llama : restore the original load/save session implementation

Will migrate this to GGUF in the future

* convert-llama-h5-to-gguf.py : support alt ctx param name

* ggml : assert when using ggml_mul with non-F32 src1

* examples : dedup simple

---------

Co-authored-by: klosax <131523366+klosax@users.noreply.github.com>

* gguf.py : merge all files in gguf.py

* convert-new.py : pick #2427 for HF 70B support

* examples/gguf : no need to keep q option for quantization any more

* llama.cpp : print actual model size

* llama.cpp : use ggml_elements()

* convert-new.py : output gguf (#2635)

* convert-new.py : output gguf (WIP)

* convert-new.py : add gguf key-value pairs

* llama : add hparams.ctx_train + no longer print ftype

* convert-new.py : minor fixes

* convert-new.py : vocab-only option should work now

* llama : fix tokenizer to use llama_char_to_byte

* tests : add new ggml-vocab-llama.gguf

* convert-new.py : tensor name mapping

* convert-new.py : add map for skipping tensor serialization

* convert-new.py : convert script now works

* gguf.py : pick some of the refactoring from #2644

* convert-new.py : minor fixes

* convert.py : update to support GGUF output

* Revert "ci : disable CI temporary to not waste energy"

This reverts commit 7e82d25f40.

* convert.py : n_head_kv optional and .gguf file extension

* convert.py : better always have n_head_kv and default it to n_head

* llama : sync with recent PRs on master

* editorconfig : ignore models folder

ggml-ci

* ci : update ".bin" to ".gguf" extension

ggml-ci

* llama : fix llama_model_loader memory leak

* gptneox : move as a WIP example

* llama : fix lambda capture

ggml-ci

* ggml : fix bug in gguf_set_kv

ggml-ci

* common.h : .bin --> .gguf

* quantize-stats.cpp : .bin --> .gguf

* convert.py : fix HF tensor permuting / unpacking

ggml-ci

* llama.cpp : typo

* llama : throw error if gguf fails to init from file

ggml-ci

* llama : fix tensor name grepping during quantization

ggml-ci

* gguf.py : write tensors in a single pass (#2644)

* gguf : single pass for writing tensors + refactoring writer

* gguf : single pass for writing tensors + refactoring writer

* gguf : single pass for writing tensors + refactoring writer

* gguf : style fixes in simple conversion script

* gguf : refactor gptneox conversion script

* gguf : rename h5 to hf (for HuggingFace)

* gguf : refactor pth to gguf conversion script

* gguf : rm file_type key and method

* gguf.py : fix vertical alignment

* gguf.py : indentation

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* convert-gptneox-hf-to-gguf.py : fixes

* gguf.py : gptneox mapping

* convert-llama-hf-to-gguf.py : fixes

* convert-llama-7b-pth-to-gguf.py : fixes

* ggml.h : reverse GGUF_MAGIC

* gguf.py : reverse GGUF_MAGIC

* test-tokenizer-0.cpp : fix warning

* llama.cpp : print kv general.name

* llama.cpp : get special token kv and linefeed token id

* llama : print number of tensors per type + print arch + style

* tests : update vocab file with new magic

* editorconfig : fix whitespaces

* llama : re-order functions

* llama : remove C++ API + reorganize common source in /common dir

* llama : minor API updates

* llama : avoid hardcoded special tokens

* llama : fix MPI build

ggml-ci

* llama : introduce enum llama_vocab_type + remove hardcoded string constants

* convert-falcon-hf-to-gguf.py : falcon HF --> gguf conversion, not tested

* falcon-main.cpp : falcon inference example

* convert-falcon-hf-to-gguf.py : remove extra kv

* convert-gptneox-hf-to-gguf.py : remove extra kv

* convert-llama-7b-pth-to-gguf.py : remove extra kv

* convert-llama-hf-to-gguf.py : remove extra kv

* gguf.py : fix for falcon 40b

* falcon-main.cpp : fix for falcon 40b

* convert-falcon-hf-to-gguf.py : update ref

* convert-falcon-hf-to-gguf.py : add tensor data layout

* cmpnct_gpt2bpe.hpp : fixes

* falcon-main.cpp : fixes

* gptneox-main.cpp : fixes

* cmpnct_gpt2bpe.hpp : remove non-general stuff

* Update examples/server/README.md

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

* cmpnct_gpt2bpe.hpp : cleanup

* convert-llama-hf-to-gguf.py : special tokens

* convert-llama-7b-pth-to-gguf.py : special tokens

* convert-permute-debug.py : permute debug print

* convert-permute-debug-master.py : permute debug for master

* convert-permute-debug.py : change permute type of attn_q

* convert.py : 70b model working (change attn_q permute)

* Delete convert-permute-debug-master.py

* Delete convert-permute-debug.py

* convert-llama-hf-to-gguf.py : fix attn_q permute

* gguf.py : fix rope scale kv

* convert-llama-hf-to-gguf.py : rope scale and added tokens

* convert-llama-7b-pth-to-gguf.py : rope scale and added tokens

* llama.cpp : use rope scale kv

* convert-llama-7b-pth-to-gguf.py : rope scale fix

* convert-llama-hf-to-gguf.py : rope scale fix

* py : fix whitespace

* gguf : add Python script to convert GGMLv3 LLaMA models to GGUF (#2682)

* First pass at converting GGMLv3 LLaMA models to GGUF

* Cleanups, better output during conversion

* Fix vocab space conversion logic

* More vocab conversion fixes

* Add description to converted GGUF files

* Improve help text, expand warning

* Allow specifying name and description for output GGUF

* Allow overriding vocab and hyperparams from original model metadata

* Use correct params override var name

* Fix wrong type size for Q8_K

Better handling of original style metadata

* Set default value for gguf add_tensor raw_shape KW arg

* llama : improve token type support (#2668)

* Merge tokenizer fixes into the gguf branch.

* Add test vocabularies

* Adapt convert-new.py (and fix a clang-cl compiler error on windows)

* Improved tokenizer test

But does it work on MacOS?

* Improve token type support

- Added @klosax code to convert.py
- Improved token type support in vocabulary

* Exclude platform dependent tests

* More sentencepiece compatibility by eliminating magic numbers

* Restored accidentally removed comment

* llama : add API for token type

ggml-ci

* tests : use new tokenizer type API (#2692)

* Merge tokenizer fixes into the gguf branch.

* Add test vocabularies

* Adapt convert-new.py (and fix a clang-cl compiler error on windows)

* Improved tokenizer test

But does it work on MacOS?

* Improve token type support

- Added @klosax code to convert.py
- Improved token type support in vocabulary

* Exclude platform dependent tests

* More sentencepiece compatibility by eliminating magic numbers

* Restored accidentally removed comment

* Improve commentary

* Use token type API in test-tokenizer-1.cpp

* py : cosmetics

* readme : add notice about new file format

ggml-ci

---------

Co-authored-by: M. Yusuf Sarıgöz <yusufsarigoz@gmail.com>
Co-authored-by: klosax <131523366+klosax@users.noreply.github.com>
Co-authored-by: goerch <jhr.walter@t-online.de>
Co-authored-by: slaren <slarengh@gmail.com>
Co-authored-by: Kerfuffle <44031344+KerfuffleV2@users.noreply.github.com>

examples/train-text-from-scratch/train-text-from-scratch.cpp

@@ -1,4 +1,5 @@
 #include "ggml.h"
+#include "common.h"
 #include "llama.h"
 #include <unordered_map>
 #include <vector>
@@ -16,7 +17,7 @@
 #pragma warning(disable: 4244 4267) // possible loss of data
 #endif
 
-static const float rms_norm_eps = LLAMA_DEFAULT_RMS_EPS;
+static const float rms_norm_eps = 1e-5f;
 
 struct random_normal_distribution {
     std::mt19937 gen;
@@ -169,14 +170,16 @@ struct ggml_tensor * randomize_tensor_uniform(struct ggml_tensor * tensor, struc
 struct llama_vocab {
     using id    = int32_t;
     using token = std::string;
+    using ttype = llama_token_type;
 
-    struct token_score {
-        token tok;
+    struct token_data {
+        token text;
         float score;
+        ttype type;
     };
 
     std::unordered_map<token, id> token_to_id;
-    std::vector<token_score> id_to_token;
+    std::vector<token_data> id_to_token;
 };
 
 struct my_llama_hparams {
@@ -1961,7 +1964,7 @@ void print_matrix(struct ggml_tensor * probs) {
 
 
 void print_token(struct llama_context * ctx, llama_token token) {
-    printf("%s", llama_token_to_str(ctx, token));
+    printf("%s", llama_token_to_str(ctx, token).c_str());
 }
 
 void print_tokens(struct llama_context* ctx, struct ggml_tensor * tokens) {
@@ -1995,7 +1998,7 @@ void print_tokens_batch(struct llama_context* ctx, struct ggml_tensor * tokens)
     }
 }
 
-void get_example_targets(const int * train_samples, size_t n_train_samples, const llama_token * train_data, size_t n_train_data, int example_id, struct ggml_tensor * tokens_input, struct ggml_tensor * target_logits, struct ggml_tensor * target_probs) {
+void get_example_targets(struct llama_context * lctx, const int * train_samples, size_t n_train_samples, const llama_token * train_data, size_t n_train_data, int example_id, struct ggml_tensor * tokens_input, struct ggml_tensor * target_logits, struct ggml_tensor * target_probs) {
     int n_tokens = tokens_input->ne[0];
     int n_vocab  = target_logits->ne[0];
 
@@ -2004,7 +2007,7 @@ void get_example_targets(const int * train_samples, size_t n_train_samples, cons
 
     ggml_set_f32(target_logits, -1.0f/n_vocab);
     ggml_set_f32(target_probs, 0.0f);
-    ggml_set_i32_1d(tokens_input, 0, llama_token_bos());
+    ggml_set_i32_1d(tokens_input, 0, llama_token_bos(lctx));
     for (int i=1; i<n_tokens+1; ++i) {
         int token = clamp(train_data[sample+i-1], 0, n_vocab-1);
         set_f32_2d(target_logits, token, i-1, +1.0f);
@@ -2015,7 +2018,7 @@ void get_example_targets(const int * train_samples, size_t n_train_samples, cons
     }
 }
 
-void get_example_targets_batch(struct llama_context * /*lctx*/, const int * train_samples, size_t n_train_samples, const llama_token * train_data, size_t n_train_data, int example_id, struct ggml_tensor * tokens_input, struct ggml_tensor * target_logits, struct ggml_tensor * target_probs) {
+void get_example_targets_batch(struct llama_context * lctx, const int * train_samples, size_t n_train_samples, const llama_token * train_data, size_t n_train_data, int example_id, struct ggml_tensor * tokens_input, struct ggml_tensor * target_logits, struct ggml_tensor * target_probs) {
     GGML_ASSERT(tokens_input->n_dims  == 2);
     GGML_ASSERT(target_logits->n_dims == 3);
     GGML_ASSERT(target_probs->n_dims  == 3);
@@ -2035,7 +2038,7 @@ void get_example_targets_batch(struct llama_context * /*lctx*/, const int * trai
         size_t sample = train_samples[(example_id*n_batch + k) % n_train_samples];
         GGML_ASSERT(sample+n_tokens-1 < n_train_data);
 
-        set_i32_2d(tokens_input, 0, k, llama_token_bos());
+        set_i32_2d(tokens_input, 0, k, llama_token_bos(lctx));
         for (int i=1; i<n_tokens+1; ++i) {
             int token = clamp(train_data[sample+i-1], 0, n_vocab-1);
             // print_token(lctx, token);
@@ -2188,11 +2191,10 @@ int tokenize_file(struct llama_context * lctx, const char * filename, std::vecto
     f.read_raw(buf.data(), f.size);
     buf[f.size] = '\0';
 
-    out.resize(buf.size());
-
-    int n_tokens = llama_tokenize(lctx, buf.data(), out.data(), buf.size(), false);
-    if (n_tokens >= 0) {
-        out.resize(n_tokens);
+    int n_tokens = llama_tokenize(lctx, buf.data(), out.data(), out.size(), false);
+    if (n_tokens < 0) {
+        out.resize(-n_tokens);
+        llama_tokenize(lctx, buf.data(), out.data(), out.size(), false);
     }
 
     bool verify = false;
@@ -2200,17 +2202,17 @@ int tokenize_file(struct llama_context * lctx, const char * filename, std::vecto
         const char * in  = buf.data();
         const char * end = buf.data() + buf.size();
         for (int i = 0; i < (int) out.size(); ++i) {
-            const char * s = llama_token_to_str(lctx, out[i]);
-            int len = strlen(s);
+            std::string s = llama_token_to_str(lctx, out[i]);
+            int len = s.length();
             if (in >= end) {
                 printf("%s: unexpected end of original text.\n", __func__);
                 break;
             }
-            const bool matches = (strncmp(in, s, len) == 0);
+            const bool matches = (strncmp(in, s.c_str(), len) == 0);
             if (matches) {
                 in += len;
             } else {
-                printf("%s: mismatch: expected '%s', but got '%s'\n", __func__, std::string(in, len).c_str(), s);
+                printf("%s: mismatch: expected '%s', but got '%s'\n", __func__, std::string(in, len).c_str(), s.c_str());
             }
         }
     }
@@ -2294,7 +2296,7 @@ llama_token sample(struct my_llama_sampler * sampler, float * logits, const llam
     const auto params = sampler->params;
 
     // Apply penalties
-    const float nl_logit = logits[llama_token_nl()];
+    const float nl_logit = logits[llama_token_nl(ctx)];
 
     const int n_last = std::min(std::min(n_last_tokens, params.repeat_last_n), sampler->n_ctx);
 
@@ -2313,7 +2315,7 @@ llama_token sample(struct my_llama_sampler * sampler, float * logits, const llam
         params.alpha_presence);
 
     if (!params.penalize_nl) {
-        logits[llama_token_nl()] = nl_logit;
+        logits[llama_token_nl(ctx)] = nl_logit;
     }
 
     llama_token token = 0;
@@ -2612,42 +2614,45 @@ void save_as_llama_model(struct llama_vocab * vocab, struct my_llama_model * mod
         return;
     }
 
-    // write_magic
-    file.write_u32(LLAMA_FILE_MAGIC);   // magic
-    file.write_u32(LLAMA_FILE_VERSION); // version
-    // write_hparams
-    file.write_u32(model->hparams.n_vocab);
-    file.write_u32(model->hparams.n_embd);
-    file.write_u32(model->hparams.n_mult);
-    file.write_u32(model->hparams.n_head);
-    file.write_u32(model->hparams.n_layer);
-    file.write_u32(model->hparams.n_rot);
-    file.write_u32(LLAMA_FTYPE_ALL_F32);
-    // write_vocab
-    uint32_t n_vocab = model->hparams.n_vocab;
-    for (uint32_t i = 0; i < n_vocab; i++) {
-        const auto & token_score = vocab->id_to_token.at(i);
-        file.write_u32((uint32_t) token_score.tok.size());
-        file.write_raw(token_score.tok.data(), token_score.tok.size());
-        file.write_raw(&token_score.score, sizeof(token_score.score));
-    }
-    // write tensors
-    write_tensor(&file, model->tok_embeddings);
-    write_tensor(&file, model->norm);
-    write_tensor(&file, model->output);
-    for (uint32_t i = 0; i < model->hparams.n_layer; ++i) {
-        auto & layer = model->layers[i];
-
-        write_tensor(&file, layer.attention_norm);
-        write_tensor(&file, layer.wq);
-        write_tensor(&file, layer.wk);
-        write_tensor(&file, layer.wv);
-        write_tensor(&file, layer.wo);
-        write_tensor(&file, layer.ffn_norm);
-        write_tensor(&file, layer.w1);
-        write_tensor(&file, layer.w2);
-        write_tensor(&file, layer.w3);
-    }
+#pragma message("TODO: implement file saving using gguf")
+    (void) vocab;
+    (void) model;
+//    // write_magic
+//    file.write_u32(LLAMA_FILE_MAGIC);   // magic
+//    file.write_u32(LLAMA_FILE_VERSION); // version
+//    // write_hparams
+//    file.write_u32(model->hparams.n_vocab);
+//    file.write_u32(model->hparams.n_embd);
+//    file.write_u32(model->hparams.n_mult);
+//    file.write_u32(model->hparams.n_head);
+//    file.write_u32(model->hparams.n_layer);
+//    file.write_u32(model->hparams.n_rot);
+//    file.write_u32(LLAMA_FTYPE_ALL_F32);
+//    // write_vocab
+//    uint32_t n_vocab = model->hparams.n_vocab;
+//    for (uint32_t i = 0; i < n_vocab; i++) {
+//        const auto & token_data = vocab->id_to_token.at(i);
+//        file.write_u32((uint32_t) token_data.tok.size());
+//        file.write_raw(token_data.tok.data(), token_data.tok.size());
+//        file.write_raw(&token_data.score, sizeof(token_data.score));
+//    }
+//    // write tensors
+//    write_tensor(&file, model->tok_embeddings);
+//    write_tensor(&file, model->norm);
+//    write_tensor(&file, model->output);
+//    for (uint32_t i = 0; i < model->hparams.n_layer; ++i) {
+//        auto & layer = model->layers[i];
+//
+//        write_tensor(&file, layer.attention_norm);
+//        write_tensor(&file, layer.wq);
+//        write_tensor(&file, layer.wk);
+//        write_tensor(&file, layer.wv);
+//        write_tensor(&file, layer.wo);
+//        write_tensor(&file, layer.ffn_norm);
+//        write_tensor(&file, layer.w1);
+//        write_tensor(&file, layer.w2);
+//        write_tensor(&file, layer.w3);
+//    }
 }
 
 float cosine_decay(const int decay_steps, const float alpha, int step) {
@@ -3052,20 +3057,13 @@ int main(int argc, char ** argv) {
 
     struct llama_vocab vocab;
     {
-        std::vector<const char *> strings;
-        std::vector<float> scores;
-        int n_vocab = llama_n_vocab(lctx);
-        strings.resize(n_vocab, NULL);
-        scores.resize(n_vocab, 0);
-        n_vocab = llama_get_vocab(lctx, strings.data(), scores.data(), n_vocab);
-        GGML_ASSERT(n_vocab == llama_n_vocab(lctx));
+        const int n_vocab = llama_n_vocab(lctx);
         vocab.id_to_token.resize(n_vocab);
         for (int i=0; i<n_vocab; ++i) {
-            std::string tok   = std::string(strings[i]);
-            float       score = scores[i];
-            vocab.id_to_token[i].tok   = tok;
-            vocab.id_to_token[i].score = score;
-            vocab.token_to_id.emplace(tok, i);
+            vocab.id_to_token[i].text  = llama_token_get_text(lctx, i);
+            vocab.id_to_token[i].score = llama_token_get_score(lctx, i);
+            vocab.id_to_token[i].type  = llama_token_get_type(lctx, i);
+            vocab.token_to_id.emplace(vocab.id_to_token[i].text, i);
         }
     }
 
@@ -3178,7 +3176,7 @@ int main(int argc, char ** argv) {
     std::vector<int> train_samples;
     train_samples.push_back(0);
     for (int i = 1; i < (int) train_tokens.size() - n_tokens; ++i) {
-        if (!params.samples_start_after_nl || (train_tokens[i-1] == llama_token_nl())) {
+        if (!params.samples_start_after_nl || (train_tokens[i-1] == llama_token_nl(lctx))) {
             train_samples.push_back(i);
         }
     }
@@ -3338,7 +3336,7 @@ int main(int argc, char ** argv) {
         struct ggml_tensor * target_logits = ggml_new_tensor_2d(model.ctx, GGML_TYPE_F32, n_vocab,  n_tokens);
         struct ggml_tensor * target_probs  = ggml_new_tensor_2d(model.ctx, GGML_TYPE_F32, n_vocab,  n_tokens);
 
-        get_example_targets(train_samples.data(), train_samples.size(), train_tokens.data(), train_tokens.size(), rand()%train_samples.size(), tokens_input, target_logits, target_probs);
+        get_example_targets(lctx, train_samples.data(), train_samples.size(), train_tokens.data(), train_tokens.size(), rand()%train_samples.size(), tokens_input, target_logits, target_probs);
         for (int i=sample_ctx; i<n_tokens; ++i) {
             ggml_set_i32_1d(tokens_input, i, n_vocab/2);
         }