Commit Graph

7 Commits

Author SHA1 Message Date
Wing Lian
2bb0b78975 Attention mask and position id fixes for packing (#285)
* fix attetion mask with packing

* set position ids and use block diagonal attn mask

* fix expand mask for multiple batch items, make sure we pad position_ids

* don't move masks to cpu

* use multi pack dataloader w random sampler

* add position_ids back

* more fixes for dataloader integration

* est total tokens, fix field loop

* more fixes, position_ids seems broken

* more fixes for sample packing

* use distributed sampler, avoid accelerate prepare

* use accelerator prepare for dataloader

* fix for position_ids w packing

* Update src/axolotl/utils/dataloader.py

* validation for sample packing and doc

* more fixes for 4k and optimizations

* optimized expand mask fn

* better handling of variance in multipack dataloader length and trainer hanging when it runs out of data

* fix rounding of len of batches to int

* better handling so that all devices have the same dataloader len

* fix step calc for packing

* pass sample packing efficiency to training args

* add a test for the mask expansion for sequence packing

* only process eval dataset for packing if not None

* don't split batches when packing

* weighted CE losses

* weighted CEL fixes

* limit packing to sequences of max seq len

* seq_len_multiple for packing

* make sure the chunk size is an int

* sample_packing_seq_len_multiplier config

* use cumulative seq len with var len flash attn v2 w packing

* properly calculate max len

* fix flash-attn, xformers, packing, support chatml

* fix chatml system prompt for openorca, legacy tokenizer opts

* add chatml

* add unit tests for cum seq lens, add ability to build cu_seq_lens from positional ids, fix prompt test

* fix test and pylint checks

* more packing and dataset optimizations and fixes

* filter w multiple cpus

* more fixes and optimizations

* fixes and go back to distributed sampler since batch sampler won't work

* fix counts by accounting for num devices

* fix steps calculation

* previous accelerate is still most performant

* add numba to requirements.

* use custom distributed checks

* fix sampler to prevent overfit w new epochs

* let's not cleanup the cached datasets

* calculate cum seq lens with pos_ids instead of mask, simplify packing params, fix distributed barrier

* speed optimizations and set accelerate fsdp env vars

* optimize dataset concatenation?

* more optimizations for dataset handling

* fix import for annotation

* manual pre-commit fixes

* another sum optimization and bug fix for calc steps

* fix packing estimations

* fix formatting

* pylint problems

* add back flash attention branch for handling unpacked sequences seperately

* Address PR feedback

* add optional sample packing config params to readme
2023-08-12 15:14:56 -04:00
Wing Lian
3d4984b9a5 update prompts for open orca to match the paper (#317)
fix the test for the updated system tokenizer
2023-07-22 13:49:11 -04:00
Wing Lian
3a38271276 add tests and supoort for loader for sys prompt data 2023-06-25 22:28:07 -04:00
Wing Lian
8d20e0a3d3 initial wip to get sys prompt from dataset 2023-06-25 22:28:07 -04:00
NanoCode012
7eb33a77dd Lint test_prompters 2023-05-31 02:53:53 +09:00
NanoCode012
392dfd9b07 Lint and format 2023-05-31 02:53:22 +09:00
Wing Lian
d199d6c261 automated testing in github actions 2023-05-27 11:51:01 -04:00