Commit graph

1334 commits

Author SHA1 Message Date
pudepiedj a8777ad84e
parallel : add option to load external prompt file (#3416)
* Enable external file and add datestamp

* Add name of external file at end

* Upload ToK2024

* Delete ToK2024.txt

* Experiments with jeopardy

* Move ParallelQuestions to /proimpts and rename

* Interim commit

* Interim commit

* Final revision

* Remove trailing whitespace

* remove cmake_all.sh

* Remove cmake_all.sh

* Changed .gitignore

* Improved reporting and new question files.

* Corrected typo

* More LLM questions

* Update LLM-questions.txt

* Yet more LLM-questions

* Remove jeopardy results file

* Reinstate original jeopardy.sh

* Update examples/parallel/parallel.cpp

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-06 16:16:38 +03:00
Jhen-Jie Hong 97af49fa39
server : reuse llama_sample_token common util (#3494)
* server : reuse llama_sample_token common function

* common : use n_probs for temperature sampling
2023-10-06 15:44:24 +03:00
l3utterfly 16820a5a0d
llama : correct hparams comparison (#3446)
* fixed floating point comparison issues

* updated implementation for hparam comparison to handle inf and NaN

* fixed code review comments

* minor simplification

* rename is_float_eq -> is_float_close

---------

Co-authored-by: Cebtenzzre <cebtenzzre@gmail.com>
2023-10-06 13:47:59 +03:00
Jhen-Jie Hong 04b2f4386e
ci : fix xcodebuild destinations (#3491)
* ci : fix xcodebuild destinations

* ci : add .swift to paths
2023-10-06 13:36:43 +03:00
cebtenzzre 48edda30ee
convert : update Falcon script for new HF config (#3448)
Also adds Falcon-180B support.
Closes #3049

Co-authored-by: jb <jonathan.t.barnard@gmail.com>
2023-10-05 15:00:34 -04:00
Kenvix ⭐ 45eba9369f
build : use std::make_tuple() for compatibility with older GCC versions (#3488) 2023-10-05 20:16:39 +03:00
staviq acec9eaaa9
common : process escape sequences in reverse prompts (#3461) 2023-10-05 19:17:29 +03:00
shibe2 e2583cbc29 CLBlast: Fix handling of on-device tensor data
Fix uploading tensor data to device, including 3D, 4D, and non-contiguous tensors.
Use correct offsets into data that is already in VRAM.
Correct handling of OpenCL events when multiple commands are queued.
2023-10-05 18:25:23 +04:00
Jhen-Jie Hong e8b8d32e86
server : fix incorrect num_tokens_predicted (#3480) 2023-10-05 17:02:55 +03:00
Jhen-Jie Hong 8f3a642ec1
swift : disable ACCELERATE_NEW_LAPACK (#3481) 2023-10-05 17:00:07 +03:00
Jhen-Jie Hong 0745384449
ci : add swift build via xcodebuild (#3482) 2023-10-05 16:56:21 +03:00
Kerfuffle 019ba1dcd0
convert : fix Baichuan2 models by using vocab size in config.json (#3299)
Use local GGUF package when possible in Baichuan converter
2023-10-04 17:20:28 +03:00
Georgi Gerganov beabc8cfb0
readme : add project status link 2023-10-04 16:50:44 +03:00
Georgi Gerganov 0d152b37fe
ggml : fix build after #3329 2023-10-04 16:25:41 +03:00
ds5t5 f8c90cdbaa
llm : add Refact model (#3329)
* add refact model

* resolve comments

* rebase to the latest

* solve alibi cpu error

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-04 16:23:39 +03:00
Georgi Gerganov f93af02488
sync : ggml (conv 1d + 2d updates, UB fixes) (#3468)
* sync : ggml (conv 1d + 2d updates)

ggml-ci

* ggml : fix UB in q5_0 and q5_1 quantize code

ggml.c:1033:39: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior

ggml.c:1081:39: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior

ggml-ci

* tests : fix UB in test-quantize-perf
2023-10-04 15:29:58 +03:00
Merrick Christensen f72f8f22c9
finetune : readme fix typo (#3465)
Fix small typo
2023-10-04 09:33:13 +03:00
Tameem 79f34abddb
ggml : add RISC-V Vector Support for K-Quants and improved the existing intrinsics (#3453)
* Added RVV intrinsics support for Q8 quantize row and also improved the existing dot product function for risc-v.

The RVV intrinsics is added for the following quantize row functions
   quantize_row_q8_0
   quantize_row_q8_1

The following dot product functions have also been optimized by using LMUL = 1/2 instead of LMUL = 1
   ggml_vec_dot_q4_0_q8_0
   ggml_vec_dot_q4_1_q8_1
   ggml_vec_dot_q5_0_q8_0
   ggml_vec_dot_q5_1_q8_1

And vector initialization in Q5 by temporary array is also replaced by the vid intrinsics

Signed-off-by: Ahmad Tameem <ahmad.tameem@10xengineers.ai>

* Added RVV intrinsics support for k_quants

This adds RISC-V Vector intrinsics support for the following K_quants functions for both QKK = 256 and QKK = 64
   ggml_vec_dot_q2_K_q8_K
   ggml_vec_dot_q3_K_q8_K
   ggml_vec_dot_q4_K_q8_K
   ggml_vec_dot_q5_K_q8_K
   ggml_vec_dot_q6_K_q8_K

Signed-off-by: Ahmad Tameem <ahmad.tameem@10xengineers.ai>

---------

Signed-off-by: Ahmad Tameem <ahmad.tameem@10xengineers.ai>
2023-10-03 21:38:19 +03:00
h-h-h-h 8186242b6d
main : consistent prefix/suffix coloring (#3425)
* Typo

* No `--in-prefix` coloring

The `--in-prefix` text was inconsistently colored. Now, it's never colored, just like the `--in-suffix` text.
2023-10-03 21:16:15 +03:00
Georgi Gerganov ac2219fef3
llama : fix session saving/loading (#3400)
* llama : fix session saving/loading

* llama : temp fix for clearing "future" tokens from the KV cache

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

* llama : fix comments for llama_kv_cache API
2023-10-03 21:04:01 +03:00
Alex Klinkhamer 48be797ffb
llama : expose model's rope_freq_scale in the API (#3418)
so it can be scaled further before creating a context.
2023-10-03 20:09:28 +03:00
Jiahao Li f56e1baec3
metal : alibi for arbitrary number of heads (#3426) 2023-10-03 19:55:21 +03:00
Eve 017efe899d
cmake : make LLAMA_NATIVE flag actually use the instructions supported by the processor (#3273)
* fix LLAMA_NATIVE

* syntax

* alternate implementation

* my eyes must be getting bad...

* set cmake LLAMA_NATIVE=ON by default

* march=native doesn't work for ios/tvos, so disable for those targets. also see what happens if we use it on msvc

* revert 8283237 and only allow LLAMA_NATIVE on x86 like the Makefile

* remove -DLLAMA_MPI=ON

---------

Co-authored-by: netrunnereve <netrunnereve@users.noreply.github.com>
2023-10-03 19:53:15 +03:00
goerch ff5a3f0c09
Work on the BPE tokenizer (#3252)
* Work on the BPE tokenizer

Tokenizer tests work for Falcon-7B

* Try to fix build problem

* Fix debug assertion failure

* Fix MSVC Unicode BOM problem

* Cleanup and an improvement

* Fix compiler warning

* Cleanup

* Test doesn't work over the full range of Unicodes

* Update .gitignore and Makefile

* Another Makefile rule

* Testing Aquila

* Moving byte decoding back to `token_to_piece` ...

... because everyone is using it.

* Guarding some unusable code pathes

* Streamlining code and adding some more assertions

Important change: I'm classifying added tokens as control tokens now for BPE.

* Adding a comment

* Adding another assertion

* Fixed vocabulary guarding assertions

* Fix PR for recent change

* Fix PR for recent change

* Fix for compiler warning

* Fix PR for recent change

* Fix PR for recent change

* Fix PR for recent change

* Fix for compiler warning

* Fixes for more compiler warnings

* Remove unused code

* Fix initialization of static maps

* Add scores and token types back, adapt gptneox

* Update llama.cpp

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

* Update unicode.h

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

* Update unicode.h

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

* Ported Starcoder and added some assertions

* Fix coding style

* Apply @jploski 's fix for missing tokens

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-10-03 09:16:26 +02:00
cebtenzzre 1c84003c08
convert : fix vocab size when not defined in hparams (#3421) 2023-10-02 18:07:24 -04:00
cebtenzzre e78f0b0d05
cmake : increase minimum version for add_link_options (#3444) 2023-10-02 22:38:43 +03:00
shibe2 665018c749
CLBlast: Add broadcast support for matrix multiplication (#3402)
Broadcast src0 into src1 across dimensions 2 and 3 when needed.
This is required for models that use GQA.
2023-10-02 21:26:15 +02:00
cebtenzzre 29a404a951
gguf : add BERT, MPT, and GPT-J arch info (#3408) 2023-10-02 15:20:28 -04:00
cebtenzzre 0fe321031a
gguf : general usability improvements (#3409) 2023-10-02 14:58:46 -04:00
cebtenzzre 9476b01226
cmake : make CUDA flags more similar to the Makefile (#3420)
* cmake : fix misuse of cxx_flags

* cmake : make CUDA flags more similar to the Makefile

* cmake : fix MSVC build
2023-10-02 16:16:50 +03:00
xaedes a03ce38455
finetune : fix #3404 (#3437)
the shapes for init model of gqa models was wrong
2023-10-02 16:15:45 +03:00
Adrian a847676984
metal : set log callback before initializing (#3427) 2023-10-02 13:49:59 +03:00
bandoti 095231dfd3
cmake : fix transient definitions in find pkg (#3411) 2023-10-02 12:51:49 +03:00
Kevin Ji ea55295a74
docker : ignore Git files (#3314) 2023-10-02 11:53:53 +03:00
vvhg1 c97f01c362
infill : add new example + extend server API (#3296)
* vvhg-code-infill (#1)

* infill in separate example (#2)

* reverted changes to main and added infill example

* cleanup

* naming improvement

* make : add missing blank line

* fix missing semicolon

* brought infill up to current main code

* cleanup

---------

Co-authored-by: Cebtenzzre <cebtenzzre@gmail.com>
2023-10-02 10:42:02 +03:00
slaren f5ef5cfb18
ggml-cuda : perform cublas mat mul of quantized types as f16 (#3412)
* ggml-cuda : perform cublas matrix multiplication of quantized types as fp16

* rename CC_TURING to CC_VOLTA

* disable fp16 mat mul completely with multi GPU
2023-09-30 18:12:57 +02:00
slaren 40e07a60f9
llama.cpp : add documentation about rope_freq_base and scale values (#3401)
* llama.cpp : add documentation about rope_freq_base and scale values

* add notice to hot topics
2023-09-29 18:42:32 +02:00
Georgi Gerganov bc34dd4f5b
train : fix KQ_pos allocation (#3392)
* train : fix KQ_pos allocation

* make sure KQ_pos is not reallocated in finetune

---------

Co-authored-by: xaedes <xaedes@gmail.com>
2023-09-29 19:05:18 +03:00
Cebtenzzre 2777a84be4
llama : quantize up to 31% faster on Linux and Windows with mmap (#3206)
* llama : enable mmap in quantize on Linux -> 31% faster

* also enable mmap on Windows

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-09-29 16:48:45 +03:00
BarfingLemurs 0a4a4a0982
readme : update hot topics + model links (#3399) 2023-09-29 15:50:35 +03:00
Andrew Duffy 569550df20
readme : add link to grammars app (#3388)
* Add link to grammars app per @ggernagov suggestion

Adding a sentence in the Grammars section of README to point to grammar app, per https://github.com/ggerganov/llama.cpp/discussions/2494#discussioncomment-7138211

* Update README.md
2023-09-29 14:15:57 +03:00
Jhen-Jie Hong c71bf2c45c
swift : fix build on xcode 15 (#3387) 2023-09-29 08:25:13 +03:00
Cebtenzzre bc39553c90
build : enable more non-default compiler warnings (#3200) 2023-09-28 17:41:44 -04:00
Hua Jiang 0ccfc62a96
ggml_tensor: update the structure comments. (#3283)
* ggml_tensor: update the structure comments.

* remove semicolon

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

* Update ggml.h

---------

Co-authored-by: Cebtenzzre <cebtenzzre@gmail.com>
Co-authored-by: slaren <slarengh@gmail.com>
2023-09-28 23:06:18 +03:00
Qu Zongfu 7f1a0fe709
ggml : release the requested thread pool resource (#3292)
* Release the requested thread pool resource

* Release the requested thread pool resource 2

---------

Co-authored-by: Zongfu ZF3 Qu <quzf3@Lenovo.com>
2023-09-28 22:51:52 +03:00
slaren 16bc66d947
llama.cpp : split llama_context_params into model and context params (#3301)
* llama.cpp : split llama_context_params into model and context params

ggml-ci

* fix metal build

* fix freq_base/scale default to model value

* llama-bench : keep the same model between tests when possible

* move n_threads to llama_context_params, add n_threads_batch

* fix mpi build

* remove kv_size(), cuda scratch fixes

* remove low-vram option

* add n_threads_batch to system info, refactor to get_system_info()

* add documentation about --threads-batch to the READMEs

* llama-bench fix

* main : fix rope freq/scale warning

* llama.cpp : add llama_get_model
common : add llama_tokenize from model

* remove duplicated ctx/model functions

ggml-ci

* cuda : print total VRAM used
2023-09-28 22:42:38 +03:00
Eve 0512d66670
ci : multithreaded builds (#3311)
* mac and linux threads

* windows

* Update build.yml

* Update build.yml

* Update build.yml

* automatically get thread count

* windows syntax

* try to fix freebsd

* Update build.yml

* Update build.yml

* Update build.yml
2023-09-28 22:31:04 +03:00
xaedes 0e76a8992c
train : finetune LORA (#2632)
* fix track_max_mem in forward_batch_wo_cache_flash_attn_train

* remove unnecessary Adam(W) optimizer tensors.

reduces optimizer memory overhead from 7*modelsize to 2*modelsize.

additionally allows to optimize models with more than 2^31 parameters by replacing int with int64_t.

bumps training checkpoint file version, but old checkpoints can still be read.
new version with less tensors is saved.

* add gradient clipping to AdamW

* Fix reset of unused g->nodes and g->grads to NULL

* implement gradient checkpointing for training

reduces memory overhead from O(n_layer) to O(sqrt(n_layer))

as explained in readme of https://github.com/cybertronai/gradient-checkpointing

* remove unused compute buffer 3

* add and use function ggml_build_backward_expand to avoid stack overflows with large maximum number of nodes

GGML_API void ggml_build_backward_expand(struct ggml_context * ctx, struct ggml_cgraph * gf, struct ggml_cgraph * gb, bool keep);

* change AdamW decay parameter to work like the torch AdamW decay parameter

It is now relative to Adam learning rate `alpha*sched`.
Before that it was relative to `sched` only.

`alpha` being the maximum learning rate and `sched` being a scaling parameter in [0..1]

* change default AdamW weight decay parameter used in training to 0.1 as used in nanoGPT

* change default AdamW weight decay parameter defined in ggml to 0.0, making Adam default instead of AdamW

btw: the default weight decay parameter for torch.optim.AdamW is 0.01

* bug fixes for cross entropy loss

ggml_cross_entropy_loss: sums where not correctly added in workload of each thread
ggml_cross_entropy_loss_back: simplify backward process, reducing numerical issues

guard usage of exp f16 lookup in cross entropy by #define GGML_CROSS_ENTROPY_EXP_FP16

cross entropy loss is only used once during training, but it is quite sensitive to numerical errors introduced by exp-f16-lookup.
so exp-f16-lookup for cross entropy loss is disabled by default, trading better gradients for very slightly worse runtime performance.

* fix test-grad0 for cross_entropy_loss

the second argument to cross_entropy_loss must sum up to 1 for each row

* fix test-grad0 for soft_max

dont use only sum as aggregation, because sum of softmax is always 1 -> finite differences should not work
instead use sum(log(soft_max()*(1-eps)+eps)); use eps to avoid log(0)

* improve finite differences of test-grad0 by using double instead of float

* change cross_entropy_loss to output average over all rows

this helps keeping the loss and gradients in a sane range

* improve gradient checkpointing

sqrt(n_layers) is only the best checkpoint step when mem size of checkpoints and mem size of layers are equal.
since layers require more memory than the single-tensor-checkpoint we use, the optimal values are compute different:

```
  given: n, u, v
  objective: minimize(a*u+b*v) where a*b=n, a>0, b>0
  b=n/a
  minimize(a*u+v*n/a)
  diff(a*u+v*n/a, a) = u - (v*n/a)/a
  diff(a*u+v*n/a, a) == 0
  u - (v*n/a)/a == 0
  u == v*n/(a*a)
  u*a*a = v*n
  a*a = v*n/u
  a = sqrt(n*v/u)
```

this change results in more checkpoints, requiring less layers to store between checkpoints, overall improving memory usage.

* disable gradient checkpointing debug output

* llama : fix rope usage in train-text-from-scratch after ChatGLM change

* add more training parameters:

--enable-restart N         Only for Adam optimizer. Enable restarts of cos-decay
--disable-restart N        Only for Adam optimizer. Disable restarts of cos-decay
--opt-past N               Number of optimization iterations to track for delta convergence test. Disabled when zero.
--opt-delta N              Maximum delta for delta convergence test. Disabled when <= zero.
--opt-max-no-improvement N Maximum number of optimization iterations with no improvement. Disabled when <= zero.
--adam-epsf N              AdamW epsilon for convergence test. Disabled when <= zero.
--adam-min-alpha N         Adam minimum learning rate alpha, usually 0.1 * alpha

* replace memcpy with reshape operation so that the graph is not cut at the input

this makes it possible to store other values into the input tensor and then simply recompute the graph without rebuilding it

* remove unused function argument from get_example_targets_batch

* measure and print total training time

* add optimization callback to ggml_opt_resume_g

this callback is called before each iteration with custom data and pointer to learning schedule parameter (only used in Adam(W)).

can be used for dynamic learning schedule and setting input data for batches before each iteration

* use optimization callback in training

allows dynamic learning schedule and different batch data for each iteration without relying on low n_iter and high n_examples parameters

reduces runtime by avoiding restart of optimization function and improves training convergence by providing a different batch for each iteration

* add minimum number of tensor dimensions to apply weight decay (default 2)

this allows to not apply weight decay to bias parameters

* rename training parameter cos-decay-alpha to cos-decay-min and clarify that adam-min-alpha also applies to warmup

* fix increase of model.train_samples and model.train_tokens

now that each optimizer iteration gets its own batch we need to multiply by number of opt iterations

* change sampling parameters for prediction after training to defaults of common.h

and clarify what is context for prediction and what are generated tokens

* tighten abs error bounds for cross_entropy_loss in test-grad0

* add conditional compilation of using F16 exp in flash attention

uncomment `// #define GGML_FLASH_ATTN_EXP_FP16` to enable usage of f16 exp in flash attention

* tighten abs error bounds for flash_attn in test-grad0

* tighten abs error bounds for sqrt in test-grad0

* remove out-commented vectorized code of opt_adam

the vectorized code might be bit faster for low number of parameters, but it had a big memory usage overhead

* ggml : update ggml_rms_norm_back with configurable eps

* llama training : fix ggml_rms_norm_back calls to pass configurable eps

* remove trailing whitespace

* add train function using automatic gradient checkpointing backward pass and allocator

* in train function replace add_inplace by regular add

because using add_inplace seems to result in different gradients

* don't use allocate hash_map on context

because the context has no_alloc=True when using memory allocator resulting in NULL data pointers

* correctly clone reshape and permute operations by also cloning tensor->nb values

* fix variable name and add missing type cast

* terminate recursive tensor cloning when reaching tensor without src tensors

* correctly clone view tensors by setting data pointers

without this the checkpointing would only work when being used together with memory allocator

* fix variable names

* swap arguments to commutative ops to be the same as in `forward_batch_wo_cache_flash_attn`

* add input tensors as checkpoints

so that recursive tensor cloning of gradient checkpointing terminates on input tensors

* fix variable name and add missing boolean negation

* make sure some tensors are not reallocated by inserting new temporary nodes depending on them:

output and parameter gradient tensors need to be available at the end of the graph execution

parameter gradient tensors also need to be available before the graph execution because they are set to zero before each optimizer iteration

checkpoint tensors are allocated all together to reduce memory allocator fragmentation

afterwards, in addition to the temporary nodes, we also need to reset the temporary leafs

* fix ASSERT to work with zero layers

* add training options whether to use allocator and/or unified training function

* integrate unified training function which may use memory allocator

the unified training function also supports arguments whether to use flash attention and/or gradient checkpointing

* format name of cloned tensors with " (clone)" suffix

* set names for tensors in unified train function for easier debugging

* allocate graph on context using ggml_new_graph

* remove handwritten training functions

* remove unused training parameters "use_scratch" and "use_unified"

* remove trailing whitespace

* remove unused train params: mem_compute1_gb & mem_compute2_gb

mem_compute_gb is used for compute when automatic memory allocator is not enabled, otherwise it can be very small to only hold the tensor definitions
mem_compute0_gb is used for automatic memory allocator (as long as measurement of max required size is not implemented)

* remove unused forward_batch function

* add debug asserts in ggml_allocr_alloc to some common pitfalls when using this function directly

* only use ggml_allocr_alloc when tensor has NULL data and is no view

* fix test when to create temporary backward graph

temporary backward graph is only necessary when using checkpointing

* fix memory "leak" in optimizers

each iteration a new cplan with new memory for work data was allocated.
now cplan creation only happens at the start of optimization, with each iteration reusing the cplan and its work data.

* reverse order of for loop in ggml_build_backward_expand to save memory when using gradient checkpointing and allocator

with this loop order gradient checkpointing with allocator on 16 layer model saves 13% memory; 2 layer memory it saves 2% memory.

the computation results are the same

* add API functions to access llama model tensors

* add stub example for finetuning, based on train-text-from-scratch

* move and remove code

* add API functions to access remaining model parameters:

mult, head and rot

* first draft for LORA finetune training

* remove const model and layer arguments in API functions for accessing model tensors

* bug fixes to make finetune compile

automatic allocator does not work yet

* add debug prints for training memory improvements

* fix names of lora tensors

* avoid stack overflow resulting from big ggml_cgraph

replace stack allocation and ggml_build_forward by ggml_new_graph in combination with ggml_build_forward_expand

* replace llama API functions to get model tensors by one function to get model tensor by name

LLAMA_API struct ggml_tensor * llama_get_model_tensor(struct llama_model * model, const char * name);

* remove unused call to not existing llama_get_layer_from_model

* implement ggml_compute_forward_out_prod_q_f32

* remove trailing whitespace

* add lora finetune support on quantized base model tensors

* add ggml_add_cast API function

this function works like ggml_add, but accepts a data type for the resulting tensor.
only supported for quantized src0 input.

* use ggml_add_cast in finetuning

lora-applied weights will now have data type F32, which improves gradients when finetuning quantized base models

* bug fix: actually use result type passed to ggml_add_cast

* make sure base model tensors data cannot be used in viewable operations

memory allocator would try to make lora application inplace on base model tensors.
since those are memory mapped this will result in memory access violations

* fix bug in ggml_out_prod which resulted in wrong n_dims of result tensors

* avoid keeping in memory ALL of the gradients

The problem here stems from ggml_graph_reset. This function is called in the optimization function, before each graph computation, to reset the gradients to zero. This required a unique memory slot for each gradient: allocating memory from a previosly freed memory location might lead to non-zero input gradients.

During ggml_compute_backward the gradients are build stepwise by adding or substracting new values, starting from a OP_NONE tensor which needs to contain zero-values. This requires the graph reset.

To avoid this I now remember in ggml_build_backward_expand the original OP_NONE gradient tensors in a hash table, which is passed to ggml_compute_backward. There instead of using add (or sub or similar) I test whether the existing gradient to be changed is a zero-valued-tensor by looking up its existence in the hash table. When it is such a zero-tensor it will not be modified, but replaced by the value to be added, otherwise the regular add (not inplace, allocator will take care of this) will be used. This way none of those zero-tensor values will be necessary in the final backward graph and more importantly they won't need a unique memory slot, just to make them zero.

* remove trailing whitespace

* remove debug prints and function to compute tensor data hash

* improve optimization iteration prints

* adjust maximal values to support finetuning 3B models

* change default finetune params lora_r and lora_alpha to match the n_rank parameters of 4

* bug fix: make sure finetune input gradient is allocated at begin and kept until end

* remove unnecessary src tensor from ggml_get_rows_back

we don't need data of src[2] for computation, only to setup the correct output shape.
remove dependency on src[2], so that allocator can work more freely.

the computational graph is still completely determined, because the output shape is naturally included.
this is similar to how ggml_reshape does it.

* remove unnecessary src tensor from ggml_repeat & ggml_repeat_back

we don't need data of src[1] for computation, only to setup the correct output shape.
remove dependency on src[1], so that allocator can work more freely.

the computational graph is still completely determined, because the output shape is naturally included

* resolve todo

allocator will only make it inplace when they are of the same type

* mixing multiple LORA adapters is now possible

pass more than one '--lora FNAME' argument to apply more than one LORA.
use '--lora-scaled FNAME S' when you want to specify a user-defined scale for an adapter.

* add option to save finetune output every N iterations

* also save latest finetune output with ITERATION="LATEST" and print where files are saved

saving with LATEST makes it easier to resume training from the latest checkpoint
the string "LATEST" can be configured with command line option "--fn-latest STR"

* update checkpoint train stats before saving via "--save-every"

* add command line option `--rank-wo N` for rank of wo tensor

* update finetune README

* fix dump_non_result_info_yaml to output multiple lora adapters

* bug fix: replace GGML_TYPE_SIZE[t] by ggml_type_size(t)

* replace llama_n_mult by llama_n_ff

* finetune bug fixes to compile with merged in code from master

* remove prediction related code to reduce duplicated code with main

use main instead

* reduce large memory overhead in train-text-from-scratch

all gradients had to be pinned so that graph_reset works correctly.
this is no longer necessary with the changes to ggml_compute_backward introduced in this PR.

* add comment explaining why finetune checkpoints are allocated in one block

* make default value of float member a float literal

* handle rms_norm and rope parameters the same as in train-text-from-scratch

* remove unused code

* remove vocab related code as it is unnecessary

* add LLM_KV_TRAINING_TYPE to train-text-from-scratch checkpoints

so that they can be differentiated from lora finetune checkpoints

* add gguf constants and load/save functions from train-text-from-scratch

* add load & save lora finetune checkpoints via gguf

* add python script to convert old finetune checkpoint files to gguf

* remove old checkpoint save & load code

* remove code to print data checksums which was used to verify correctness of new gguf code

* omit tokenization when training is disabled, only save llama lora adapter

training can be disabled by passing '-n 0' to finetune

* remove trailing whitespace

* update README.md

* implement ggml_compute_forward_repeat_f16

* avoid stack overflow of large cgraphs in test-grad0

* add ggml API functions ggml_unravel_index, ggml_get_i32_nd and its analogs for set and for f32

ggml_get_i32_1d, ggml_set_i32_1d, ggml_get_f32_1d, ggml_set_f32_1d now support non-contiguous tensors.
in case of non-contiguous tensor, the 1d index is unraveled into a multi index using ggml_unravel_index to be passed to '_nd' function equivalent.

this fixes a bug in test-grad0 which happens due to ggml_build_backward not building purely contiguous tensors anymore

* increase test-grad0 context mem size to accommodate for bigger cgraph

* add sanity check to ggml_compute_backward, asserting the correct shape of gradients

* fix ggml_acc_or_set to return tensor of correct shape

* remove unused 'inplace' argument from ggml_compute_backward function

inplace operations to add gradients are no longer created by ggml_compute_backward
use allocator to automatically make inplace operations

* add missing argument 'int i0' to ggml_get_i32_nd & ggml_set_i32_nd header declarations

* fix error message in ggml_allocr_alloc to display actual max_avail

* fix check_gradient

ggml_build_backward_expand was previously replaced by ggml_build_backward, but the assignment of forward graph to backward graph missing

* use tensor->view_src instead of ggml_is_view and get_view_source

* move gradient checkpointing code into ggml, new API function:

// build gradient checkpointing backward graph gb for gf using provided checkpoints
// gb_tmp will contain original backward graph with rewritten backward process nodes,
// but without the second forward pass nodes.
GGML_API void ggml_build_backward_gradient_checkpointing(
        struct ggml_context   * ctx,
        struct ggml_cgraph    * gf,
        struct ggml_cgraph    * gb,
        struct ggml_cgraph    * gb_tmp,
        struct ggml_tensor  * * checkpoints,
        int                     n_checkpoints);

* replace custom data getters and setters by ggml functions

* train-text-from-scratch can train (full finetune) gguf models

just pass the gguf model via `--checkpoint-in FN`.
after this, to continue training, pass the generated checkpoint instead of the original gguf model.

tested with smaller models, bigger models may exceed available memory.
use (LORA) finetune for those.

* remove trailing whitespace

* add option to save train-text-from-scratch output every N iterations

* update README.md

* fix warnings

* fix warnings

* remove finetune option to disable allocator

the allocator should always be used.
by making sure that it is always used it gets easier to implement automatic memory requirements computation

* add tensor checkpoints only when gradient checkpointing is enabled

* initialize opt ggml context if none was provided

* add ggml-alloc API function 'ggml_allocr_max_size' to get max size of alloc

GGML_API size_t ggml_allocr_max_size(struct ggml_allocr * alloc);

* finetune: automatically allocate all memory and changes to command line options

remove '--n_examples N' parameter, as it no longer makes sense to call optimization process multiple times in a loop.
add '--only_write_lora' command line option: will skip tokenization and training, to only write a llama.cpp comptabile LORA adapter.
remove memory buffer related command line options.
improve iteration console output.

* add finetune to Makefile

* update README.md

* print time per iteration and estimate remaining time

* increase measured alloc size by tensor_alignment

ggml_allocr_reset will reduce the given size by up to tensor_alignment-1

* fix README.md

* add some more allocator debug prints

* bug fix, probably solves the 'ggml_allocr_alloc: not enough space in the buffer' issue

* revert last commit

"bug fix, probably solves the 'ggml_allocr_alloc: not enough space in the buffer' issue"

"alloc was freeing an externally allocated tensor, because it calculated the end of allocator memory as alloc->data + alloc->max_size instead of alloc->data + alloc->size."

This is intentional to reduce the risk of freeing external tensors when measuring. Unless max_size is not properly calculated, I don't see why this is an issue.

* remove unnecessary "0x" before "%p" output

* move measurement memory segment to upper region of the address space

* update README.md

* fix printf format warnings

* add missing gguf_free in load_checkpoint_lora_file

* load default rms_norm and rope parameters from base model

* add gradient accumulation

specify number accumulation steps with '--grad-acc N'.
this will simulate a bigger batch size of grad_acc*batch.

* fix tracking of train_samples and train_tokens

* build : fix compile warnings

* ggml : fix L-BFGS linesearch loop

* improve finetune time measurement

fix printf warnings on system where int64_t is (long int).
change time datatypes to double because values get big with long training times.
exclude file saving from time measurement.
converge faster to actual time per iteration by removing very small first duration before first iteration was performed.
fix bug in output of total training time, the reported value was 1000 times to small.

* specify default lora rank with '--lora-r N'

'--lora-r N' will specify default rank for all tensors
'--rank-wq N', etc. will override this default rank for specific tensor types.

* fix gradient accumulation bug where the same batch was used for each microstep

* fix gradient accumulation bug where the same batch was used for each microstep

* support grouped-query-attention in ggml_flash_attn and ggml_flash_attn_back

k and v can now be repeated in q along ne[2]

in forward pass just use modulo to compute k and v indices, like ik2 = iq2 % nek2.

in backard pass this won't work as easy, because multiple threads will compete to accumulate to the same k->grad[:,ik1,ik2,ik3] and v->grad[:,iv1,iv2,iv3].
so we change the parallelization over q rows to be over k rows. this ensures non-overlapping (ik2,ik3) across threads.
in each thread we then iterate over the number of repetitions of k/v in q to compute iq2 as iq2 = ik2 + irep*nek2.

since ne2 is not the same for q,k and v we also change how the gradients are concatenated into the result tensor.
additionally the offsets of gradq, gradk and gradv in the result tensor are now memory aligned.

we also simplify the compute_backward part of flash_attn to use ggml_reshape instead of switching over the number of dimensions.
this needs a small change to ggml_reshape, removing the assertion of second argument to be contiguous.
since only the shape (ne) of the second reshape argument is of relevance, its memory layout (nb) is irrelevant -> it can very well be non-contiguous.

change test-grad0 to also test for repeated k/v in q.

this changes the rng and now results in small gradient differences in softmax. these solely come from using f16 exp table lookup in forward softmax: when temporarily changing softmax to use actual exp function, the reported gradient differences go away. gradient differences coming solely from f16 table lookup are acceptable.
added a note to explain this.

* add llama API functions to get grouped-query-attention n_head parameter 'n_head_kv'.

* fix finetune to support grouped-query-attention (using flash-attention)

note: ggml changes to ggml_out_prod are necessary to support grouped-query-attention without flash-attention.

* support broadcastable a in out_prod(a, b) and backward pass of broadcasting mul_mat(a, b)

* test broadcasting mul_mat backward pass

* decouple random number generator of each operation test

when changing one test the rng of others tests is not influenced anymore

* add comment briefly describing what ggml_repeat_back does

* simplify broadcasting mul_mat backward using ggml_repeat_back

* add cgraph evaluation order member and corresponding enum type

this controls in which order ggml_build_forward visits source nodes.
by default the nodes are visited left to right, i.e. src[0] first.
in some cases it is beneficial for ggml-alloc to visit in a different order.
two possible orders are supported: left-to-right (src[0] first) and right-to-left (src[0] last).

* measure max compute size for each cgraph eval order and use best order

this can bring huge memory savings:
e.g. codellama-34b with n_ctx=64, n_batch=1 goes from 92927.8mb down to 4627.6 MB

* remove unused command line options

* add sample start patterns and options to force new or by default resume last shuffling

* update shuffle rng state on reshuffle

* exclude known zero values from computations in flash_attn_f32 & flash_attn_back_f32

* remove probably unnecessary exception type flags from stringstream

* pass correct max number of tokens to llama_tokenize

* account for possible leading whitespace that will be added by tokenizer
e.g. '\t' will be tokenized by llama spm tokenizer to [29871, 12]

* use unrolled vec_mad in out_prod

y is vec_mad result vec.
x is vec_mad input vec.
v is vec_mad input scalar.

ggml_vec_mad_f32_unroll will internally loop over x and v with same y.

GGML_VEC_MAD_UNROLL is by default defined to 32.

This value is empirical optimized using performance test runs of out-prod in openllama-3b finetune with 256 context length and batch size 1. It gives 23% performance boost for out_prod.

Full measurements of out-prod runtime in ms:
	unroll_xv	unroll_yv
1	67014.643	87826.469
2	77117.552	89077.656
4	72091.311	109121.657
8	61077.543	88678.334
16	56914.67	79514.947
24	59024.595	84350.254
28	55952.446	83368.73
32	51476.658	85177.745
36	55973.792	84659.92
40	55139.616	93844.738
48	60736.392	93330.267
64	99856.878	116994.99

Second column is when unrollying yv instead of xv

* set lora_alpha to value of lora_r if it is not set via command line

otherwise only changing lora_r will change scaling of lora adapter used in prediction

* reshuffle original sample order instead of the previous shuffled order

otherwise resumed reshuffle will not result in same sample order

* block tiling for out-prod inspired by mul-mat

block sizes are empirically optimized

roughly doubles the flops of out-prod

* exclude some more known zero values from computations in flash_attn_f32 & flash_attn_back_f32

* add static keywords

* remove outcommented old code

* update train-text-from-scratch with tokenization, sample selection and shuffling from finetune

* remove lbfgs related train parameters

* move common train functions into common/train.[h|cpp]

* move train state into struct train_state

* move train data saving code into callback to unify code of opt_callback

train_params are still different in finetune and train-text-from-scratch, so it can't yet be moved to train.h|cpp

* move common train params into common/train

* move common opt_callback into common/train

* fix consume_common_train_arg

* save and load head_count_kv in lora checkpoints

* increase train_samples by used_samples instead of number of batches

on batch can contain more than one sample when option "fill_with_next_samples" is used

* fix usage of llama_tokenize

* remove static from process_escape since we need it exposed in header

* fix code formating of long function declarations

* fix condition in load_train_state_gguf

* use die("msg") instead of replace GGML_ASSERT(!"msg") or throw std::runtime_error("msg")

* fix saving and loading of training type

* remove terminating '\0' from tokenization

(llama_tokenize is now passed the string length instead of relying on terminating '\0')

* fix compile warnings

* fix compile warnings

* use new/delete for train_state instead of malloc/free

using malloc may result in seg faults when trying to assign string fields

* assert that sample_count > 0, avoiding division by zero

* fix frand to return value in interval [0,1)

* add train option "--sample-random-offsets"

Use samples beginning at random offsets.
The offset is only applied to the first sample in each batch context window.
Together with "--fill-with-next-samples" this may help for training endless text generation.

For example given a dataset containing samples "abcd", "ABCD", "0123".
With context size of 8 and options "--fill-with-next-samples", "--no-separate-with-eos", "--no-separate-with-bos",
the context windows of batches could only be filled with "abcdABCD", "ABCDabcd", "0123abcd", etc.

With "--sample-random-offsets" it can also be filled with "23abcdAB", "bcd0123A", etc.

* deduplicate code into function

* remove n_rot hparam, as it must always be hparam.n_embd_head()

* align code

* assert correct base model tensor shapes

* move some params from lora hparams into model hparams and load model params from gguf

this equalizes the model definition in finetune and text-from-scratch and removes the need for additional llama api functions to get model parameters

* remove now unnecessary llama API functions to get model params that where added by this PR

* train-text-from-scratch: automatically allocate model tensors, remove option '--mem-model N'

* train-text-from-scratch: automatically allocate opt context

* train-text-from-scratch: automatically allocate input tensors

* train-text-from-scratch: automatically allocate compute memory

* remove unused options and equalize train-text-from-scratch with finetune

* initialize opt->loss_after with zero

* add export-lora program

* remove trailing whitespace

* add export-lora build in Makefile

* remove unused struct tensor_info from export-lora

* add export-lora build dependency to llama

because it depends on common, which depends on llama

* update finetune README.md

* cancel optimization when specified number of epochs is completed

* improve handling of export-lora arguments

print errors and warnings when files could not be read or created

* Fix export-lora.cpp "not enough space in the context's memory pool" (#1)

* Fix export-lora.cpp "not enough space in the context's memory pool"

Without this patch, export-lora would sometimes error with "not enough space in the context's memory pool (needed 656784, available 656800)".

* increase required context size by 5*GGML_MEM_ALIGN instead of plain 16

---------

Co-authored-by: xaedes <xaedes@gmail.com>

* improve handling of not yet supported tensor types

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: meatbag-18a <145869052+meatbag-18a@users.noreply.github.com>
2023-09-28 21:40:11 +03:00
Cebtenzzre 2db94d98ed
gguf : basic type checking in gguf_get_* (#3346) 2023-09-28 14:30:31 -04:00
Cebtenzzre ecf90b1a51
gguf : make token scores and types optional (#3347) 2023-09-28 14:30:15 -04:00