Commit graph

40 commits

Author SHA1 Message Date
Georgi Gerganov
13c351ad72
ggml : various fixes (#1450)
- `ggml_rope()`
- `ggml_diag_mask_inf()` multi-threaded
- compatibility with scratch buffers
2023-05-14 18:22:50 +03:00
Georgi Gerganov
601a033475
ggml : add GGML_QNT_VERSION to track quantization format changes
https://github.com/ggerganov/ggml/issues/150#issuecomment-1546625668
2023-05-14 10:20:19 +03:00
Johannes Gäßler
905d87b70a
ggml : GPU-accelerated token generation (#1412)
* CUDA kernel for q4_0 dequant. + mat. vec. mult.

* Added q4_1 via template

* Added missing __syncthreads();

* --gpu_layers -> --gpu-layers

* Shorter dequantize_mul_mat_vec line

* q5_0 dequantize_mul_mat kernel

* More readable dequantize_mul_mat_vec logic

* dequantize_mul_mat_vec kernels for q5_1, q8_0, f16

* llama : offload "output" tensor to GPU too + coding style fixes

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-05-13 16:38:36 +03:00
xaedes
f954edda93
ggml : implement backward pass for llama + small training-llama-from-scratch example (#1360)
* implement 8 of 14 missing backward pass operations used by llama

- GGML_OP_ADD_AT
- GGML_OP_CPY
- GGML_OP_MUL_MAT (src0.grad)
- GGML_OP_PERMUTE
- GGML_OP_RESHAPE
- GGML_OP_SCALE
- GGML_OP_TRANSPOSE
- GGML_OP_VIEW

implement additional ggml operation GGML_OP_ADD_AT, which is necessary for backward pass of GGML_OP_VIEW.

this operation adds src1 to src0 with data offset, i.e. to view(src0, ..., offset).
the values are return in a tensor size of src0. values outside of [data+offset:data+offset+nbytes(src1)] are just the original values from src0.

still missing backward passes for llama:

- GGML_OP_DIAG_MASK_INF
- GGML_OP_GET_ROWS
- GGML_OP_RMS_NORM
- GGML_OP_ROPE
- GGML_OP_SILU
- GGML_OP_SOFT_MAX

* implement 5 of 6 missing backward pass operations used by llama

- GGML_OP_DIAG_MASK_INF
- GGML_OP_GET_ROWS
- GGML_OP_RMS_NORM
- GGML_OP_SILU
- GGML_OP_SOFT_MAX

add necessary ggml operations GGML_OP_ADD1, GGML_OP_SILU_BACK, GGML_OP_RMS_NORM_BACK, GGML_OP_DIAG_MASK_ZERO, and GGML_OP_ROPE_BACK

GGML_OP_ADD1 is necessary to add a scalar value in the backward pass of GGML_OP_SOFT_MAX
GGML_OP_ADD1 could also be replaced by using GGML_OP_ADD and GGML_OP_REPEAT, but the performance would be worse. additionally GGML_OP_REPEAT will return unexpected value when the the input to GGML_OP_SOFT_MAX contains only a single scalar. in this case GGML_OP_REPEAT will not return the value that should be repeated (src1) but the value which shape the result should take (src0). So in this case it can not replace GGML_OP_ADD1.

GGML_OP_SILU_BACK, GGML_OP_RMS_NORM_BACK and GGML_OP_ROPE_BACK are necessary for backward pass of GGML_OP_SILU, GGML_OP_RMS_NORM and GGML_OP_ROPE. The backward pass for these functions cannot be easily composed of existing operations. Since the backward pass builds a computation graph we need operations forward pass implementations of the the required backward passes. Sounds a bit confusing at first, I know...

GGML_OP_DIAG_MASK_ZERO is necessary for backward pass of GGML_OP_DIAG_MASK_INF.

Some operations where previously inplace-only. for backward pass there needs to be non-inplace variants.
staying consistent with other operations that have non-inplace and inplace variants, the operations are changed to non-inplace and
functions with "_inplace" are added which are inplace.
in llama we need to call the inplace variants so that it is implemented as before.
for llama backward pass we need to use the non-inplace variants.

still not completely implemented backward passes for llama:

- GGML_OP_ROPE: needs forward pass for GGML_OP_ROPE_BACK
- GGML_OP_GET_ROWS: only necessary for tokenizer

* norm & rms_norm can not be threaded:

after investigation rms norm for quite some time I come to the conclusion that neither norm, nor rms_norm can be threaded, because we need mean over all items, not just of the slices each thread sees.

* remove already resolved TODO

* implement backward pass of ggml_rope and ggml_rope_back

* implement backward pass for ggml_get_rows and for new operation ggml_get_rows_back

* add test-grad0.c

* use GGML_PRINT_DEBUG for debug messages which will otherwise flood the console

* test both gradients of mul_mat

* disable graph dot export as it floods console

* bug fixes for silu_back

* successfully test silu backward

* bug fix for scale backward pass

use sum instead of mean for gradient of scalar scale parameter

* successfully test scale backward

* improve performance of sum backward pass

use add1(x,y) instead of add(x,repeat(y,x))

* improve performance of sqr backward pass

use scale(x,y) instead of mul(x,repeat(y,x))

* successfully test rope backward

* bug fix for cpy backward pass

* successfully test cpy backward

* bug fix for reshape backward pass

* successfully test reshape backward

* add test-opt.c

this uses ggml_opt to train a,b for minimal e=sum(sqr(c - a*b)) for random initial a,b,c

* correctly implement softmax backward pass using new operation ggml_diag

ggml_diag constructs diagonal matrices with entries.
ggml_diag(shape[a,1,c,d]) -> shape[a,a,c,d]

* successfully test soft_max backward

* align shape annotations

* add shape annotations for llama

* de-duplicate ggml_forward_dup code taking care of contiguous tensors of same type.

with this we can duplicate tensor of any typ as long as they are contiguous.

* fix ggml_compute_forward_dup_same_cont for when nelements < nthreads

when more threads are used than elements exist ie1 was less than ie0, resulting in invalid negative byte count argument in memcpy

* bug fix for add_at forward

required for view backward pass

src0 values must be copied to dst, because during addition we don't touch all dst elements in contrast to the normal add function.

* successfully test view backward

* minor code format improvement

* fix ggml_forward_add functions to work correctly with transposed tensors

uses the same logic as in ggml_compute_forward_add_q_f32, but make it consistent across all ggml_compute_forward_add_... functions.
this also slightly changes the mem access pattern of the different threads to works as in ggml_compute_forward_add_q_f32.

* fix ggml_forward_add1 functions to work correctly with transposed tensors

uses the same logic as in ggml_compute_forward_add1_q_f32, but make it consistent across all ggml_compute_forward_add1_... functions.
this also slightly changes the mem access pattern of the different threads to works as in ggml_compute_forward_add1_q_f32.

* test-grad0.c : add print_elements to help with debugging

* successfully test permute backward

* some minor test-grad0 fixes

* fix sub, mul and div functions to work correctly with transposed tensors

uses the same logic as in add

* implement ggml_cont backward pass

* successfully test transpose backward and permute for all permutations

also test sub, mul and div up to max n_dims

* test-grad0.c add TODO for view_2d and view_3d

add_at (required for view backward pass) is a bit tricky for n_dims > 1.

* fix comments

* successfully test diag_mask_inf and diag_mask_zero backward

* test-grad0 : fix test for div

nargs and ndims was swapped, corrupting the stack

* fix diag_mask to work with non-inplace input

* move dup call into the actual add_at functions

* fix get rows backward pass

* successfully test get_rows backward

* fix view backward pass

add nb parameters to add_at like in view.
together with offset they define how to view dst and src0 during the add_at operation.

* successfully test backward pass of view_1d, view_2d and view_3d

* fix backward pass for rms_norm

I would have used formulas from other frameworks, but they differed so I could not decide which is correct.
Instead it was derived here in comment using manual forward-backward automatic differention of rms_norm and simplification.

* successfully test backward pass of rms_norm

some tests may fail when gradients are large.
could not find a satisfying configuration to check for abs error and relative error that passes all tests while still actually testing the results with tight enough error bounds.
when looking at the values the "failed" tests look actually ok. for example:

rms_norm: ndims=2, i=0, k=2, x0=0.000153, xm=0.000053, xp=0.000253, f0=0.278594, f1=0.086213, g0=961.905457, g1=966.064941, eps=0.000100, error_abs=4.159485, error_rel=0.004324

it is due to the test logic in check_gradients that they fail.

* add todos for llama backward pass

- implementation for ADD1 backward pass should probably use sum instead of mean (but this backward pass is not required)
- repeat is not yet tested and looks like it only works for single element src0 inputs.

* add operation ggml_sum_rows

ggml_sum_rows(shape[a,b,c,d]) -> shape[1,b,c,d]

* add missing GGML_OP_SUM_ROWS

* fix backward pass for repeat

requires ggml_sum_rows

* successfully test backward pass of repeat

* update quantization types in switch-case of add_at and add1

* add baby-llama example training a very small llama model from scratch to output a sinusoidal wave.

had to increase maximum number of optimization parameters to train from scratch.

* fix softmax in baby-llama example

* switching from training with adam to lbfgs produces much better results in the baby-llama example

* train with two examples, creating new tensors each time..

* fix bug when using ggml_opt to optimize params in one context and use a renewable context for eval and opt

when not keeping gradients of model parameters they are overwritten by tensors created by opt, which may be invalid after opt context is renewed.
so we need to keep the original gradients and make dups for opt

* train on multiple examples, generate & print tokens with trained model afterwards

ctx0 for evaluation and optimization is renewed for each sample

* add ggml_reshape_1d, ggml_reshape_4d and ggml_view_4d

* fix soft_max backward pass for input->ne[1] != 1

* add ggml_log operation necessary for cross entropy loss

* add test for ggml_log gradients

* implement backward pass for ggml_sum_rows, necessary for cross entropy loss

* implement ggml_repeat support for rank > 2 tensors

* add test for ggml_sum_rows gradients

* fix training get_example_targets

predict the next token, not the current token!

* add square_error_loss and cross_entropy_loss functions

* optimize loss over multiple samples

this increases computation graph, need parallel batched forward for more efficiency.

* fix backward pass for add_at and change arguments to have same order as in view

* add ggml_set(ctx, a, b) to set b in view of a and return modified a

necessary to set values into kv_self cache and properly propagate the gradients

* fix kv_self gradients for training

use ggml_set instead of ggml_cpy to set kv_self cache with properly propagating gradients

* replace inplace operations for training with copying operations to allow gradient propagation

* add GGML_ASSERT to catch ggml_rope and back value errors

* add trainable lora-only model with all big matrices C split into A,B with A*B=C

this is not a lora-finetune, but the whole model changed to have only low-rank "lora" matrices.

training this instead of the normal model resulted in much worse results though...

* vastly improve training results

instead of logit targets 0 and 1 use -1 and +1.

* shorten code using a variable

* change name of GGML_OP_ADD_AT to GGML_OP_ACC

* smaller default values for baby llama model parameters

* update static assert of GGML_OP_COUNT

* remove shape annotations in llama_eval_internal

* revert disabling of threading for rms_norm and norm

* rename print functions in baby-llama example

* fix call to ggml_set_name

* add missing include for strcmp, etc

* remove trailing whitespace

* reduce number of test-grad0 iterations

avoid exceeding timeout of automated tests

* remove busy loop that was used as sleep for slower sinus wave generation

* disable slow tests grad0 and opt to avoid exceeding timeouts

* c++ in baby-llama example

use c++ includes instead of c includes
use std::min, std::max instead of MIN, MAX macros

* c++ in baby-llama example

use c++ includes instead of c includes
use std::min, std::max instead of MIN, MAX macros

* ggml : fix compiler warnings + cosmetic changes

* ggml : fix nullptr derefs in GGML_OP_CONT and GGML_OP_RESHAPE back

* swap arguments to vDSP_vdiv call

documentation for vDSP_vdiv states: "Note that B comes before A!"

* swap arguments to vDSP_vdiv call

documentation for vDSP_vdiv states: "Note that B comes before A!"

* ggml : swap vDSP_vsub args as per documentation

* add parallel batched forward function for baby-llama training

* cleanup code for batched training

* remove trailing whitespace

* minor : fix compiler warnings + indentation style

* ggml : fix null ptr deref in backward pass

* ggml : remove Q4_2 remnants

* ggml : fix clang-tidy warnings

* baby-llama : couple of clang-tidy warnings

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-05-13 15:56:40 +03:00
Georgi Gerganov
b9fd7eee57
ggml : remove bit shuffling (#1405)
* ggml : remove Q4_0 bit shufling (ARM NEON)

* ggml : remove Q4_1 bit shuffling (ARM NEON + reference)

* ggml : nibbles_from_floats() + bytes_from_nibbles() (ARM NEON)

* ggml : remove Q4_2 bit shuffling (WIP, BROKEN)

* ggml : remove Q5_0 bit shuffling (ARM NEON)

* ggml : 2x faster scalar implementations

* ggml : remove Q5_1 bit shuffling (ARM NEON + scalar)

* ggml : simplify scalar dot

* ggml : remove WASM SIMD bit shuffling + remove vzip for ARM 32-bit

* ggml : fix Q4_1 quantization

* ggml : update cuBLAS + normalize variable names

* ggml : remove Q4_2 mode

* ggml : minor formatting

* ggml : fix Q5_0 quantization

* scripts : add script for measuring the time per token

* AVX implementations (#1370)

* ggml : uniform 5th bit extraction

* llama : produce error upon loading old model files

* llama : fix model magic/version write

* ggml : speed-up Q5_0 + Q5_1 at 4 threads

* ggml : preserve old Q4 and Q5 formats

* ggml : simplify Q8_1 - no need for low / high sums anymore

* ggml : fix Q8_0 and Q8_1 rounding

* Revert "AVX implementations (#1370)"

This reverts commit 948d124837f9d287d8490f41338e0e4cceb0814f.

* ggml : fix AVX2 implementation

* sha : update hashes for 7B and 13B

* readme : update timings + remove warning banner

* llama : update v2 PR number to 1405

* ggml : fix WASM comments

* ggml : back to original bit order

* readme : add note that Q4 and Q5 have been changed

* llama : fix return for unknown version

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
2023-05-12 00:23:08 +03:00
slaren
2d099e5193
ggml: add names to tensors (#1268)
* ggml: add names to tensors

* minor improvements to dot file formatting
2023-05-02 16:03:00 +02:00
slaren
58b367c2d7
cuBLAS: refactor and optimize f16 mat mul performance (#1259)
* cuBLAS: refactor, convert fp16 to fp32 on device

* cuBLAS: use multiple streams, choose smartly between mul_mat_q and mul_mat_f16

* fix build

* cuBLAS: update block_q5_1
2023-05-01 18:11:07 +02:00
Georgi Gerganov
6bc4400e67
ggml : add Q5 WASM SIMD + GGML_FTYPE 2023-04-30 19:07:43 +03:00
Georgi Gerganov
0b5a935099
ggml : fix visibility and unused warnings 2023-04-29 19:28:36 +03:00
Stephan Walter
36d19a603b
Remove Q4_3 which is no better than Q5 (#1218) 2023-04-28 23:10:43 +00:00
Georgi Gerganov
55390bcaf2
ggml : sync ggml (ggml_alibi) 2023-04-28 20:51:05 +03:00
0cc4m
7296c961d9
ggml : add CLBlast support (#1164)
* Allow use of OpenCL GPU-based BLAS using ClBlast instead of OpenBLAS for context processing

* Improve ClBlast implementation, avoid recreating buffers, remove redundant transfers

* Finish merge of ClBlast support

* Move CLBlast implementation to separate file

Add buffer reuse code (adapted from slaren's cuda implementation)

* Add q4_2 and q4_3 CLBlast support, improve code

* Double CLBlast speed by disabling OpenBLAS thread workaround

Co-authored-by: Concedo <39025047+LostRuins@users.noreply.github.com>
Co-authored-by: slaren <2141330+slaren@users.noreply.github.com>

* Fix device selection env variable names

* Fix cast in opencl kernels

* Add CLBlast to CMakeLists.txt

* Replace buffer pool with static buffers a, b, qb, c

Fix compile warnings

* Fix typos, use GGML_TYPE defines, improve code

* Improve btype dequant kernel selection code, add error if type is unsupported

* Improve code quality

* Move internal stuff out of header
* Use internal enums instead of CLBlast enums
* Remove leftover C++ includes and defines
* Make event use easier to read

Co-authored-by: Henri Vasserman <henv@hot.ee>

* Use c compiler for opencl files

* Simplify code, fix include

* First check error, then release event

* Make globals static, fix indentation

* Rename dequant kernels file to conform with other file names

* Fix import cl file name

---------

Co-authored-by: Concedo <39025047+LostRuins@users.noreply.github.com>
Co-authored-by: slaren <2141330+slaren@users.noreply.github.com>
Co-authored-by: Henri Vasserman <henv@hot.ee>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-04-28 17:57:16 +03:00
Georgi Gerganov
574406dc7e
ggml : add Q5_0 and Q5_1 quantization (#1187)
* ggml : add Q5_0 quantization (cuBLAS only)

* ggml : fix Q5_0 qh -> uint32_t

* ggml : fix q5_0 histogram stats

* ggml : q5_0 scalar dot product

* ggml : q5_0 ARM NEON dot

* ggml : q5_0 more efficient ARM NEON using uint64_t masks

* ggml : rename Q5_0 -> Q5_1

* ggml : adding Q5_0 mode

* quantize : add Q5_0 and Q5_1 to map

* ggml : AVX2 optimizations for Q5_0, Q5_1 (#1195)

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
2023-04-26 23:14:13 +03:00
Georgi Gerganov
7a32fcb3b2
ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179)
* ggml : add Q8_0 quantization format (rename the old one to Q8_1)

* tests : fix test-quantize-fns

* ggml : finalize Q8_0 implementation

* ggml : use q4_0_q8_0 and q4_2_q8_0

* ggml : fix Q8_0 dot product bug (ARM)

* ggml : Q8_0 unroll x2

* ggml : fix bug - using wrong block type

* ggml : extend quantize_fns_t with "vec_dot_type"

* ggml : fix Q8_0 to use 255 values out of 256

* ggml : fix assert using wrong QK4_2 instead of QK4_3
2023-04-25 23:40:51 +03:00
Georgi Gerganov
8a0f8673ba
ggml : export symbols (#1155) 2023-04-24 22:18:25 +03:00
Georgi Gerganov
12b5900dbc
ggml : sync ggml (add GPT-NeoX RoPE implementation) 2023-04-20 23:32:59 +03:00
Kawrakow
38de86a711
llama : multi-threaded quantization (#1075)
* Multi-threading quantization.

Not much gain for simple quantizations, bit it will be important
for quantizations that require more CPU cycles.

* Multi-threading for quantize-stats

It now does the job in ~14 seconds on my Mac for
Q4_0, Q4_1 and Q4_2. Single-threaded it was taking
more than 2 minutes after adding the more elaborate
version of Q4_2.

* Reviewer comments

* Avoiding compiler confusion

After changing chunk_size to const int as suggested by
@ggerganov, clang and GCC starting to warn me that I don't
need to capture it in the lambda. So, I removed it from the
capture list. But that makes the MSVC build fail. So,
making it a constexpr to make every compiler happy.

* Still fighting with lambda captures in MSVC

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-04-20 20:42:27 +03:00
Georgi Gerganov
e0305ead3a
ggml : add Q4_3 quantization (#1082) 2023-04-20 20:35:53 +03:00
slaren
8944a13296
Add NVIDIA cuBLAS support (#1044) 2023-04-19 11:22:45 +02:00
Georgi Gerganov
77a73403ca
ggml : add new Q4_2 quantization (ARM only) (#1046)
* ggml : Q4_2 ARM

* ggml : add ggml_is_quantized()

* llama : update llama_type_name() with Q4_2 entry

* ggml : speed-up q4_2

- 4 threads: ~100ms -> ~90ms
- 8 threads:  ~55ms -> ~50ms

* ggml : optimize q4_2 using vmlaq_n_f32 + vmulq_n_f32
2023-04-18 23:54:57 +03:00
slaren
315a95a4d3
Add LoRA support (#820) 2023-04-17 17:28:55 +02:00
Ivan Komarov
f266259ad9
Speedup the AVX-512 implementation of ggml_vec_dot_q4_0() (#933) 2023-04-17 15:10:57 +02:00
Georgi Gerganov
e95b6554b4
ggml : add Q8_0 quantization for intermediate results (#951)
* ggml : add Q8_0 quantization for intermediate results

* quantize-stats : fix test + add it to Makefile default

* Q8: use int8_t, AVX/AVX2 optimizations

* ggml : fix quantize_row_q8_0() ARM_NEON rounding

* minor : updates after rebase to latest master

* quantize-stats : delete obsolete strings

* ggml : fix q4_1 dot func

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
2023-04-15 17:53:22 +03:00
Pavol Rusnak
c56b715269
Expose type name from ggml (#970)
Avoid duplication of type names in utils

Co-authored-by: Håkon H. Hitland <haakon@likedan.net>
2023-04-14 20:05:37 +02:00
Kerfuffle
c9a59b70a5
ggml : add unary and binary map operations (#874)
* GGML map ops proof of concept.

* Various cleanups.

Add handling for task setting.

Add handling for ggml_compute_backward.

Rename functions to ggml_map_unary_f32 and ggml_map_binary_f32

Fix compiler warnings related to casting function pointers and `void *`

Reorder functions and definitions based on the GGML op number.

Use typedefs for map op function pointer types.

* Fix position of map ops cases in ggml_compute_forward
2023-04-14 17:43:55 +03:00
Georgi Gerganov
a3a2a0eda8
ggml : add GGML_DEFAULT_N_THREADS 2023-04-13 18:36:48 +03:00
Stephan Walter
3e6e70d8e8
Add enum llama_ftype, sync ggml_type to model files (#709) 2023-04-11 15:03:51 +00:00
Georgi Gerganov
c3ac702e5e
ggml : add ggml_cont() + optimize ggml_cpy() for contiguous dst 2023-04-10 22:42:28 +03:00
comex
f963b63afa Rewrite loading code to try to satisfy everyone:
- Support all three formats (ggml, ggmf, ggjt).  (However, I didn't
  include the hack needed to support GPT4All files without conversion.
  Those can still be used after converting them with convert.py from my
  other PR.)

- Support both mmap and read (mmap is used by default, but can be
  disabled with `--no-mmap`, and is automatically disabled for pre-ggjt
  files or on platforms where mmap is not supported).

- Support multi-file models like before, but automatically determine the
  number of parts rather than requiring `--n_parts`.

- Improve validation and error checking.

- Stop using the per-file type field (f16) entirely in favor of just
  relying on the per-tensor type/size fields.  This has no immediate
  benefit, but makes it easier to experiment with different formats, and
  should make it easier to support the new GPTQ-for-LLaMa models in the
  future (I have some work in progress on that front).

- Support VirtualLock on Windows (using the same `--mlock` option as on
  Unix).

    - Indicate loading progress when using mmap + mlock.  (Which led me
      to the interesting observation that on my Linux machine, with a
      warm file cache, mlock actually takes some time, whereas mmap
      without mlock starts almost instantly...)

      - To help implement this, move mlock support from ggml to the
        loading code.

- madvise/PrefetchVirtualMemory support (based on #740)

- Switch from ifstream to the `fopen` family of functions to avoid
  unnecessary copying and, when mmap is enabled, allow reusing the same
  file descriptor for both metadata reads and mmap (whereas the existing
  implementation opens the file a second time to mmap).

- Quantization now produces a single-file output even with multi-file
  inputs (not really a feature as much as 'it was easier this way').

Implementation notes:

I tried to factor the code into more discrete pieces than before.

Regarding code style: I tried to follow the code style, but I'm naughty
and used a few advanced C++ features repeatedly:

- Destructors to make it easier to ensure everything gets cleaned up.

- Exceptions.  I don't even usually use exceptions when writing C++, and
  I can remove them if desired... but here they make the loading code
  much more succinct while still properly handling a variety of errors,
  ranging from API calls failing to integer overflow and allocation
  failure.  The exceptions are converted to error codes at the
  API boundary.)

Co-authored-by: Pavol Rusnak <pavol@rusnak.io> (for the bit I copied from #740)
2023-04-10 01:10:46 +02:00
unbounded
62cfc54f77
Add quantize-stats command for testing quantization (#728)
Command that calculates some statistics over the errors introduced by
quantization, like mean square error, max error and some percentile errors for layer
weights. Should be useful for testing quantization improvements.

Exposes some internal state from ggml and llama for testing
2023-04-08 00:09:18 +02:00
Georgi Gerganov
986b6ce9f9
ggml, llama : avoid heavy V transpose + improvements (#775)
ggml :

- added ggml_view_3d()
- ggml_view_tensor() now inherits the stride too
- reimplement ggml_cpy() to account for dst stride
- no longer require tensor->data to be memory aligned

llama :

- compute RoPE on 32-bit tensors (should be more accurate)
- store RoPE-ed K in the KV cache
- store transposed V in the KV cache (significant speed-up)
- avoid unnecessary Q copy
2023-04-05 22:07:33 +03:00
Marian Cepok
c0bb1d3ce2
ggml : change ne to int64_t (#626) 2023-04-02 13:21:31 +03:00
Justine Tunney
6f23ba5ee2 Ensure --mlock works properly with mmap() support 2023-03-30 12:28:25 -07:00
Slaren
c03ae8dca1 Add mmap support for model files 2023-03-30 12:28:25 -07:00
Stephan Walter
c1f885067c
ggml : introduce structs for the q4 data blocks (#356)
* Introduce structs for the q4 data blocks

* ggml : rename quant struct variables + fix ARM_NEON

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-28 18:56:03 +03:00
comex
563cdc391d
Support calling mlock() on loaded model data on Linux and macOS (#453)
* Support calling mlock() on loaded model data on Linux and macOS

This is enabled by a new --mlock command line option.

Using mlock() disables swapping and memory compression for the model
data.  Doing so can be useful on systems where the model takes up a
large fraction of system RAM.  In my experience, macOS is quite eager to
start compressing llama.cpp's memory, which then makes it halt for a few
seconds while it decompresses, even with a model that uses "only" 25GB
out of 32GB.

Of course, this comes at the cost of forcing the system to swap or
compress other processes' memory instead, so it needs to be used with
care and shouldn't be enabled by default.

In theory it should be possible to support this on Windows as well using
VirtualLock(), but I'm not much of a Windows user.

* Update llama.cpp

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-24 17:19:05 +02:00
Stephan Walter
69c92298a9
Deduplicate q4 quantization functions (#383)
* Deduplicate q4 quantization functions

* Use const; add basic test

* Re-enable quantization test

* Disable AVX2 flags in CI

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-22 19:29:06 +02:00
Georgi Gerganov
f5a77a629b
Introduce C-style API (#370)
* Major refactoring - introduce C-style API

* Clean up

* Add <cassert>

* Add <iterator>

* Add <algorithm> ....

* Fix timing reporting and accumulation

* Measure eval time only for single-token calls

* Change llama_tokenize return meaning
2023-03-22 07:32:36 +02:00
hoangmit
6eac39ba95
Add RMS norm and use it (#187)
* add ggml_rms_norm

* update op num
2023-03-16 00:41:38 +02:00
Georgi Gerganov
26c0846629
Initial release 2023-03-10 20:56:40 +02:00