Commit Graph

1013 Commits

Author SHA1 Message Date
Steward Garcia
7e4ea5beff
examples : add server example with REST API (#1443)
* Added httplib support

* Added readme for server example

* fixed some bugs

* Fix the build error on Macbook

* changed json11 to nlohmann-json

* removed some whitespaces

* remove trailing whitespace

* added support custom prompts and more functions

* some corrections and added as cmake option
2023-05-21 20:51:18 +03:00
Georgi Gerganov
ec2e10c444
llama : add llama_init_backend() API (close #1527) 2023-05-20 11:06:37 +03:00
DannyDaemonic
d2c59b8ba4
Fix for mingw (#1462) 2023-05-20 00:40:02 -07:00
Evan Jones
943e6081cc
examples : add persistent chat (#1495)
* examples : add persistent chat

* examples : fix whitespace

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-05-19 20:39:51 +03:00
Jason McCartney
7694b52b9a
main : make reverse prompt option act as a stop token in non-interactive mode (#1032)
* Make reverse prompt option act as a stop token in non-interactive scenarios

* Making requested review changes

* Update gpt_params_parse and fix a merge error

* Revert "Update gpt_params_parse and fix a merge error"

This reverts commit 2bb2ff1748.

* Update gpt_params_parse and fix a merge error take 2
2023-05-19 20:24:59 +03:00
Georgi Gerganov
4b7e245adf
minor : fix compile warnings 2023-05-19 20:14:51 +03:00
DannyDaemonic
ee9654138a
Fixes #1511 lambda issue for w64devkit (mingw) (#1513)
* Fix for w64devkit and mingw
2023-05-18 19:30:40 +02:00
Stephan Walter
dc271c52ed
Remove unused n_parts parameter (#1509) 2023-05-17 22:12:01 +00:00
rankaiyx
c238b5873a
benchmark-matmul: Print the average of the test results (#1490) 2023-05-17 16:47:58 +02:00
András Salamon
9560655409
define default model path once, sync path with readme (#1366) 2023-05-16 17:46:34 +02:00
zrm
63d20469b8
fix get_num_physical_cores() (#1436)
* fix get_num_physical_cores()
had been broken on complex topologies because "cpu cores" in /proc/cpuinfo is per-"physical id"

* Add spaces to maintain consistent formatting

---------

Co-authored-by: slaren <ddevesa@gmail.com>
2023-05-15 04:25:42 +02:00
slaren
b5c9295eef
benchmark-matmul: fix clang-tidy issues, report results in GFLOPS (#1458)
* benchmark-matmul: fix command line parsing, replace macros with functions, report results in GFLOPS
2023-05-14 22:46:00 +02:00
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
Rinne
6456a4eb9f
embedding : remove unused code (#1426) 2023-05-13 10:24:20 +03:00
Georgi Gerganov
fb62f92433
llama : fix --mtest option (close #1414) 2023-05-12 21:44:20 +03:00
Johannes Gäßler
773ee249fb
CLI args use - instead of _, backwards compatible (#1416) 2023-05-12 14:34:55 +00:00
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 948d124837.

* 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
Evan Jones
cf348a60e0
main : add option to save full output to session (#1338)
* main : add option to save full output to session

* split behavior into --session and --prompt-cache

* restore original implementation with new names

* PR comments

* move the check for incompatible parameters to gpt_params_parse

* Fix whitespace

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

---------

Co-authored-by: DannyDaemonic <DannyDaemonic@gmail.com>
2023-05-10 11:37:14 -04:00
DannyDaemonic
e6a46b0ed1
Locale fix for Windows (#1379) 2023-05-09 19:53:28 +02:00
DannyDaemonic
41654efea8
Interface improvements and --multiline-input (previously --author-mode) (#1040)
* Interface improvements
* Multiline input
* Track character width
* Works with all characters and control codes + Windows console fixes
2023-05-08 19:45:48 -07:00
Georgi Gerganov
f9a6364912
llama : require first token to be BOS (#1303)
* llama : require first token to be BOS

* scripts : add ppl-run-all.sh

* perplexity : add BOS for each chunk

* readme : update perplexity values after BOS fix

* perplexity : add clarifying comments
2023-05-08 17:41:54 +03:00
Johannes Gäßler
1f48b0abcf
Documented CUDA reproducibility, added warning (#1346) 2023-05-08 02:42:01 +02:00
Jed Fox
3924088512
Remove default arguments from sampling functions (#1343) 2023-05-06 17:01:47 -04:00
slaren
94c5652fc0
quantize: make output filename optional, default to ggml-model-<ftype>.bin (#1301) 2023-05-05 00:58:56 +02:00
44670
2edbdb0f99
main : add --in-suffix option (#1318)
* adding --in-suffix option

* print input suffix before generation
2023-05-04 18:41:12 +03:00
DannyDaemonic
db1080876a
Only escape prompts when used with -e (#1311) 2023-05-04 05:08:25 -07:00
DannyDaemonic
c65a7fbfa9
Update main's README.md with new features (#1296) 2023-05-04 03:02:59 -07:00
Tomas
f647ce040f
fix #1224 reverse prompt and multi line (#1297)
* fix reverse prompt and multi line

* Code Formatting

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

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-05-04 03:02:30 -07:00
khimaros
6daa09d879
examples : read chat prompts from a template file (#1196) 2023-05-03 20:58:11 +03:00
CRD716
a8a2efdc81
examples : various prompt and example fixes (#1298)
* fix dan.txt

* miku prompt improvements

* use common characters
2023-05-03 18:26:47 +03:00
DannyDaemonic
2485d7a4d3
Process escape sequences given in prompts (#1173) 2023-05-02 18:46:20 -07:00
DannyDaemonic
13b0c68ed7
Handle signals properly on Windows (#1123) 2023-05-02 18:01:57 -07:00
slaren
bf4b22ffe4
fix missing parameters in llama_init_from_gpt_params (#1293) 2023-05-03 01:36:45 +02:00
Ron Evans
67c77799e0
examples : add llama_init_from_gpt_params() common function (#1290)
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-05-02 23:39:51 +03:00
Georgi Gerganov
0e6cbff1b7
llama : fix compile warnings 2023-05-02 23:09:08 +03:00
Ron Evans
8c9be35ff9
examples : improve vertical alignment of a few variables (#1286)
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-05-02 20:53:52 +03:00
Robert Brisita
2bb992f034
llama : allow 0 as a seed number. (#1275) 2023-05-02 19:23:44 +03:00
Ron Evans
e2cd506999
main : switch input_noecho to input_echo to remove negation (#979)
Signed-off-by: deadprogram <ron@hybridgroup.com>
2023-05-02 19:13:26 +03:00
DannyDaemonic
f4cef87edf
Add git-based build information for better issue tracking (#1232)
* Add git-based build information for better issue tracking

* macOS fix

* "build (hash)" and "CMAKE_SOURCE_DIR" changes

* Redo "CMAKE_CURRENT_SOURCE_DIR" and clearer build messages

* Fix conditional dependency on missing target

* Broke out build-info.cmake, added find_package fallback, and added build into to all examples, added dependencies to Makefile

* 4 space indenting for cmake, attempt to clean up my mess in Makefile

* Short hash, less fancy Makefile, and don't modify build-info.h if it wouldn't change it
2023-05-01 18:23:47 +02:00
Georgi Gerganov
70269cae37
llama : fix session load / save (#1263) 2023-05-01 14:54:59 +03:00
jon-chuang
a5d30b1f53
common : better default number of threads (#934)
* commit

* fix

* try-catch

* apply code review

* improve

* improve

* add macos headers

* done

* remove color

* fix windows

* minor

* fix

* Apply suggestions from code review

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

* remove

* minor

* minor

---------

Co-authored-by: jon-chuang <jon-chuang@users.noreply.github.com>
Co-authored-by: DannyDaemonic <DannyDaemonic@gmail.com>
2023-04-30 21:41:35 +03:00
Stephan Walter
f0d70f147d
Various fixes to mat_mul benchmark (#1253) 2023-04-30 12:32:37 +00:00
Georgi Gerganov
305eb5afd5
build : fix reference to old llama_util.h 2023-04-29 13:53:12 +03:00
Georgi Gerganov
84ca9c2ecf
examples : fix save-load-state + rename llama-util.h 2023-04-29 13:48:11 +03:00
Georgi Gerganov
334637e43e
common : change default parameters to pre-#1126 (#1223) 2023-04-29 09:51:06 +03:00
Ivan Stepanov
dd7eff57d8
llama : new sampling algorithms (#1126)
* Sample interface, new samplers.

New samplers:
- locally typical sampling
- tail free sampling
- frequency and presence penalty
- mirostat

Ignore EOS fix: -inf should be used.

* mirostat

* Added --logit-bias and --no-penalize-nl, removed std::span

* Use C++11, clarify llama API documentation, rename Mirostat parameters to --mirostat_lr and --mirostat_ent, add temperature sampling for Mirostat, simplify Mirostat sampling API parameters (removed N and *k)

Use C++11, clarify llama API documentation, rename Mirostat parameters to --mirostat_lr and --mirostat_ent, add temperature sampling for Mirostat, simplify Mirostat sampling API parameters (removed N and *k)

* Save and load example adjust

* Tests

* Windows build fix

* Windows test fix
2023-04-29 08:34:41 +03:00
Stephan Walter
36d19a603b
Remove Q4_3 which is no better than Q5 (#1218) 2023-04-28 23:10:43 +00:00
CRD716
5fba3c016b
examples : add Jeopardy example (#1168)
* Basic Setup

* Prevent Results.txt from coming up

* Prefixes, Line separators, etc

* editorcheck

* introduction to give more consistent results

* Basic graph thing

* Grading, ready for testing!

* Y'all ready to get funky?

* fix column removal stuff

* missed a few
2023-04-28 19:13:33 +03:00
Evan Jones
1481a9cf25
llama : add session file format and saved sessions in main (#1169) 2023-04-28 18:59:37 +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
Pavol Rusnak
859fee6dfb
quantize : use map to assign quantization type from string (#1191)
instead of `int` (while `int` option still being supported)

This allows the following usage:

`./quantize ggml-model-f16.bin ggml-model-q4_0.bin q4_0`

instead of:

`./quantize ggml-model-f16.bin ggml-model-q4_0.bin 2`
2023-04-26 18:43:27 +02:00
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
xaedes
0c5692345d
examples : add save_load_state example (#1150)
* add save_load_state example

* use <cstdio> instead of <iostream> and fprintf / printf instead of cout

* renamed save-load-state example files replacing underscores by dashes
2023-04-24 19:23:31 +03:00
mgroeber9110
9b0a4d4214
examples/main README improvements and some light refactoring (#1131) 2023-04-24 15:45:32 +00:00
slaren
1d78fecdab
Fix LoRA acronym (#1145) 2023-04-23 23:03:44 +02:00
DannyDaemonic
edce63baa9
Added README.md for main with examples and explanations (#1139) 2023-04-23 15:37:02 +00:00
Stephan Walter
c50b628810
Fix CI: ARM NEON, quantization unit tests, editorconfig (#1122) 2023-04-22 10:54:13 +00:00
wbpxre150
36b4f7e064
llama : print timings on ctrl+c exit (#1021)
* print timings on ctrl+c exit

* remove redundant free memory call.

* add global pointer to ctx.
2023-04-22 11:56:35 +03:00
eiery
10f19c1121
llama : have n_batch default to 512 (#1091)
* set default n_batch to 512 when using BLAS

* spacing

* alternate implementation of setting different n_batch for BLAS

* set n_batch to 512 for all cases
2023-04-22 11:27:05 +03:00
Clint Herron
e9a9cb0c54
examples : Improve Alpaca Default Repeat Penalty: Better Match Alpaca.cpp Experience (#1107)
* Moving parameters to separate lines for readability.

* Increasing repeate_penalty to 1.1 to make alpaca more usable by default.

* Adding trailing newline.
2023-04-22 09:54:33 +03:00
Alex Klinkhamer
9411288271
main : evaluate tokens in batches after swapping context (#1014)
* examples : evaluate tokens in batches after swapping context

* Update examples/main/main.cpp

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-04-21 21:18:09 +03:00
slaren
3d59769c3b
Show perplexity ETA in hours and minutes (#1096) 2023-04-21 14:57:57 +02:00
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
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
Georgi Gerganov
eb17a026fd
quantize-stats : fix bug in --type argument 2023-04-17 17:31:06 +03:00
Pavol Rusnak
489537e6cf
examples: add missing <ctime> include for time() (#1011) 2023-04-16 10:13:00 +00:00
Ivan Komarov
c12b14b77f
benchmark : fix result validation in benchmark-q4_0-matmult (#987) 2023-04-15 08:51:54 +03:00
Pavol Rusnak
c85e03d12e
Revert "main : alternative instruct mode (Vicuna support, etc.) (#863)" (#982)
This reverts commit f4d277ae17.
2023-04-14 22:58:43 +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
Tomáš Pazdiora
f4d277ae17
main : alternative instruct mode (Vicuna support, etc.) (#863)
* Add support for configs, add configurable prefixes / suffixes, deprecate instruct mode, add stop prompt

* Add multiline mode, update text input.

* bugfix

* update implementation

* typos

* Change --multiline implementation to be toggled by EOF.

* bugfix

* default multiline mode

* add more configs

* update formating

* update formatting

* apply suggestions
2023-04-14 18:19:17 +03:00
Gary Linscott
be87b6ed20
perplexity : add support for batch size to --perplexity (#407)
* Add support to batch size for perplexity

* Revert "Fix memory allocation issues and seg faults"

This reverts commit 4870e455b3.

* update from merge

* Remove perplexity from main

* updates

* Update batch size for efficiency
2023-04-14 00:50:42 +03:00
CRD716
0e07e6a839
common : remove unnecessary includes (#947) 2023-04-13 18:39:25 +03:00
Georgi Gerganov
9190e8eac8
llama : merge llama_internal.h into llama.h
Hide it behind an #ifdef
2023-04-13 18:04:45 +03:00
CRD716
8cda5c981d
fix whitespace (#944) 2023-04-13 16:03:57 +02:00
niansa/tuxifan
107980d970
examples : add -n to alpaca and gpt4all scripts (#706) 2023-04-13 16:03:39 +03:00
SebastianApel
95ea26f6e9
benchmark : add tool for timing q4_0 matrix multiplication (#653)
* Initial version of q4_0 matrix multiplication benchmark

* Bugfix: Added dependency to ggml.o to benchmark

* Reviewer requests: added parameter for threads, switched to ggml_time_us()

* Reviewer input: removed rtsc, use epsilon for check

* Review comment: Removed set_locale

* Feature: Param for numer of iterations, Bugfix for use of parameter threads

* Reviewer suggestion: Moved to examples

* Reviewer feedback: Updated clean: and benchmark: sections

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-04-13 15:46:23 +03:00
Pavol Rusnak
8b679987cd
Fix whitespace, add .editorconfig, add GitHub workflow (#883) 2023-04-11 19:45:44 +00:00
Stephan Walter
3e6e70d8e8
Add enum llama_ftype, sync ggml_type to model files (#709) 2023-04-11 15:03:51 +00:00
comex
2663d2c678
Windows fixes (#890)
Mostly for msys2 and mingw64 builds, which are different from each other
and different from standard Visual Studio builds.  Isn't Windows fun?

- Define _GNU_SOURCE in more files (it's already used in ggml.c for
  Linux's sake).

- Don't use PrefetchVirtualMemory if not building for Windows 8 or later
  (mingw64 doesn't by default).  But warn the user about this situation
  since it's probably not intended.

- Check for NOMINMAX already being defined, which it is on mingw64.

- Actually use the `increment` variable (bug in my `pizza` PR).

- Suppress unused variable warnings in the fake pthread_create and
  pthread_join implementations for Windows.

- (not Windows-related) Remove mention of `asprintf` from comment;
  `asprintf` is no longer used.

Fixes #871.
2023-04-11 15:19:54 +02:00
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
Tomáš Pazdiora
aaf3b23deb
fix for windows utf-8 input (#840)
Use UTF-16 as input on Windows, since UTF-8 does not work and reads multibyte characters as zeros
2023-04-08 17:49:39 +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
Sergey Alirzaev
cc9cee8e9e
Do not crash when it has nothing to say. (#796)
Otherwise observing this in the interactive mode:
/usr/lib/gcc/x86_64-pc-linux-gnu/12/include/g++-v12/bits/stl_vector.h:1230: reference std::vector<int>::back() [_Tp = int, _Alloc = std::allocator<int>]: Assertion '!this->empty()' failed.
2023-04-06 17:59:11 +02:00
at8u
ff05d05c96
miku.sh : add executable bit (#780) 2023-04-05 18:59:13 +03:00
at8u
88ed5761b8
examples : add Miku.sh (#724)
* Add Miku.sh to examples

* Add missing line to prompt in Miku.sh

* Add --keep param to Miku.sh

* Remove '[end_of_conversation]' line from Miku.sh

No longer is necessary.
2023-04-05 17:32:42 +03:00
mgroeber9110
53dbba7695
Windows: reactive sigint handler after each Ctrl-C (#736) 2023-04-03 18:00:55 +02:00
Leonardo Neumann
6e7801d08d
examples : add gpt4all script (#658) 2023-04-02 10:56:20 +03:00
Murilo Santana
5b70e7de4c
fix default params for examples/main (#697) 2023-04-02 04:41:12 +02:00
Slaren
0d054e292e Show error message when -f fails 2023-04-01 16:08:40 +02:00
Slaren
64bde3ffd4 Fix ggml_init_params in quantize 2023-03-30 12:28:25 -07:00
Thérence
d9ad104440
Create chat-13B.bat (#592)
* Create chat-13B.bat

Same script than chat-13B.sh, but for windows users.
Tested and working on windows 10/11 v 22H2

* Apply suggestions from code review

---------

Co-authored-by: anzz1 <anzz1@live.com>
2023-03-29 20:21:09 +03:00
Tobias Lütke
a6956b25a1
add example of re-act pattern (#583)
* add example of re-act pattern

* spelling...

* fixed whitespace in reverse prompt issue
2023-03-29 10:10:24 -05:00
anzz1
7f4c5c6651
llama : fix linkage with mingw (#551)
* Revert 7e53955 (#542)

Still needs to be fixed properly

* Fix linking on mingw32
2023-03-28 21:23:09 +03:00
Stephan Walter
436e561931
all : be more strict about converting float to double (#458)
* Be more strict about converting float to double

* Test equivalence of round, SILU implementations

Test module is commented out in CMakeLists.txt because the tests may
take a long time, depending on how much the compiler optimizes.

* Fix softmax in perplexity.cpp

* all : prefer float over double where appropriate

* perplexity : add <cmath>

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-28 19:48:20 +03:00
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
anzz1
7b8dbcb78b
main.cpp fixes, refactoring (#571)
- main: entering empty line passes back control without new input in interactive/instruct modes
- instruct mode: keep prompt fix
- instruct mode: duplicate instruct prompt fix
- refactor: move common console code from main->common
2023-03-28 17:09:55 +03:00
Marco Matthies
7e5395575a
Fix missing ggml link in cmake for examples/* on w64-mingw32 (#542) 2023-03-27 07:55:26 +03:00
Stephan Walter
b391579db9
Update README and comments for standalone perplexity tool (#525) 2023-03-26 16:14:01 +03:00
anzz1
7a87d31f4f
[main] fix infinite generation (-n == -1) (#523) 2023-03-26 16:06:10 +03:00
Harald Fernengel
33e35b8fe8
Exit from interactive mode if input stream is bad (#491)
Allow exiting the interactive prompt also with CTRL-D on Unix and CTRL-Z
on Windows.
2023-03-26 08:25:46 +03:00
anzz1
34ab526843
(Windows) Set console to UTF-8 on init (#420)
Sets console codepage to 65001 (CP_UTF8) on start for both input and output, should fix problems with UTF-8 characters.
2023-03-25 22:29:22 +02:00
Georgi Gerganov
c2b25b6912
Fix colors enabling on WIN32 2023-03-25 21:53:39 +02:00
Georgi Gerganov
79b2b266db
If n_predict == -1, generate forever 2023-03-25 21:51:41 +02:00
Georgi Gerganov
e2d490dafd
Inifinite generation via context swapping (#71) 2023-03-25 21:36:22 +02:00
Georgi Gerganov
03f7e33560
Cleanup STL headers + fix embedding examples + minor stuff 2023-03-25 20:51:14 +02:00
Georgi Gerganov
55ad42af84
Move chat scripts into "./examples" 2023-03-25 20:37:09 +02:00
Georgi Gerganov
a316a425d0
Overhaul the examples structure
- main -> examples
- utils -> examples (renamed to "common")
- quantize -> examples
- separate tools for "perplexity" and "embedding"

Hope I didn't break something !
2023-03-25 20:26:40 +02:00
Georgi Gerganov
04c6f5ed6f
Immediately start processing the prompt before user input has been provided (#476) 2023-03-24 23:17:58 +02:00
Mathieu Nayrolles
3f9c6135e4
fix typo in chatLLaMa (#368)
The prompt contains a typo where 'alound' is used instead of 'aloud'.
2023-03-21 22:52:27 +02:00
Jean-Christophe Hoelt
3ab3e6582f
Add chatLLaMa script (#198)
* Add chatLLaMa script

* Fix shellcheck errors and do some cleanup

* Move chatLLaMa script to `examples` directory

* Reduce chatLLaMa context size to 2048

Ref d7def1a752

* Include n_predict to 2048 in examples/chatLLaMa
2023-03-21 18:23:15 +02:00