Wing Lian ff5d6393c8 replace legacy attention boolean flags with capability properties ...

Replace checks with capability-based properties derived from attn_implementation

This separates three concerns that were conflated under flash_attention:
1. Backend selection -> attn_implementation enum
2. Packing capability -> attn_supports_packing property
3. Flash-attn library dependency -> attn_uses_flash_lib property

2026-04-23 22:27:01 +00:00

..

cli

feat: support dot-notation CLI args for nested config options (#3419)

2026-02-23 10:10:06 -05:00

core

DPO support loss types (#3566)

2026-04-23 00:25:28 -04:00

e2e

DPO support loss types (#3566)

2026-04-23 00:25:28 -04:00

fixtures

Respect sequence_len in config for type: llama2_chat (#926)

2023-12-12 09:39:22 -08:00

integrations

fix async prefetch with nemogym (#3606)

2026-04-22 09:05:46 -04:00

kernels

fix async prefetch with nemogym (#3606)

2026-04-22 09:05:46 -04:00

monkeypatch

fix async prefetch with nemogym (#3606)

2026-04-22 09:05:46 -04:00

patched

allow bf16 flag but warn (#3563) [skip ci]

2026-04-01 09:54:01 -04:00

prompt_strategies

feat: move to uv first (#3545)

2026-04-21 10:16:03 -04:00

telemetry

Fix DO_NOT_TRACK not being correctly handled (#3580)

2026-04-04 05:16:58 -04:00

utils

fix async prefetch with nemogym (#3606)

2026-04-22 09:05:46 -04:00

__init__.py

fix: minor patches for multimodal (#2441)

2025-03-31 13:40:12 +07:00

conftest.py

feat: move to uv first (#3545)

2026-04-21 10:16:03 -04:00

constants.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

hf_offline_utils.py

transformers v5 upgrade (#3272)

2026-01-27 17:08:24 -05:00

test_attn_implementation.py

replace legacy attention boolean flags with capability properties

2026-04-23 22:27:01 +00:00

test_chunked_xentropy.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

test_context_parallel_batch_size.py

fix: correct total_num_steps and batch_size calculation with context parallelism (#3444)

2026-03-05 12:33:28 -05:00

test_convert.py

fix: fix CONTRIBUTING.md placeholders, bare except clauses, and add convert.py tests (#3485) [skip ci]

2026-03-16 00:12:40 -04:00

test_data.py

Fix: excess_length_strategy truncation method (#3401)

2026-02-25 11:31:11 +07:00

test_datasets.py

Feat: add support for datasets with str saved messages field (#3607)

2026-04-23 00:25:48 -04:00

test_dict.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

test_ebft_kernels.py

EBFT: Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models (#3527) [skip ci]

2026-03-24 18:43:46 -04:00

test_ebft_strided_structured.py

EBFT: Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models (#3527) [skip ci]

2026-03-24 18:43:46 -04:00

test_exact_deduplication.py

feat:add support dataset_num_processes (#3129) [skip ci]

2025-10-13 17:18:12 +07:00

test_freeze.py

Train parameters exclusively in specific ranges (#1390)

2024-03-14 11:05:42 -04:00

test_http_weight_sync.py

EBFT: Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models (#3527) [skip ci]

2026-03-24 18:43:46 -04:00

test_loaders.py

fix: transformers deprecate load_in_Xbit in model_kwargs (#3205)

2025-10-16 16:07:27 +07:00

test_logging_config_file_capture.py

Debug log, logging improvements (#3159)

2025-09-17 13:27:03 -04:00

test_lora.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

test_normalize_config.py

transformers v5 upgrade (#3272)

2026-01-27 17:08:24 -05:00

test_opentelemetry_callback.py

Feat/opentelemetry (#3215)

2025-10-22 19:16:55 -07:00

test_packed_batch_sampler.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

test_packed_dataset.py

feat:add support dataset_num_processes (#3129) [skip ci]

2025-10-13 17:18:12 +07:00

test_packed_pretraining.py

Streaming SFT support (#3101)

2025-09-02 12:08:44 -04:00

test_perplexity.py

transformers v5 upgrade (#3272)

2026-01-27 17:08:24 -05:00

test_prompt_tokenizers.py

DPO transformers v0.29 fixes (#3560) [skip ci]

2026-03-31 19:04:53 -04:00

test_prompters.py

fix: prompt phi (#1845) [skip ci]

2024-08-22 11:46:57 -04:00

test_revision_parameter.py

feat: add processor_kwargs YAML field forwarded to from_pretrained (#3612)

2026-04-23 00:26:34 -04:00

test_save_deduplicated.py

fix: Save de-duplicated dataset during pre-processing (#3427)

2026-03-02 12:55:59 -05:00

test_schedulers.py

Add ruff, remove black, isort, flake8, pylint (#3092)

2025-08-23 23:37:33 -04:00

test_streaming.py

text diffusion training plugin (#3067)

2025-09-10 20:27:00 -04:00

test_tensor_parallel_batch_size.py

use new tf32 APIs for torch 2.9+ (#3467) [skip ci]

2026-03-06 11:40:32 -05:00

test_tokenizers.py

upgrade transformers==5.3.0 trl==0.29.0 kernels (#3459)

2026-03-06 09:11:20 -05:00

test_train.py

refactor dupes from merge/rebase (#2919) [skip ci]

2025-07-14 10:05:26 -04:00

test_triton_kernels.py

use custom triton kernels for entropy from logits and selective softmax (#3510)

2026-03-19 02:02:43 -04:00

test_utils_tee.py

Debug log, logging improvements (#3159)

2025-09-17 13:27:03 -04:00

test_validation_dataset.py

automatically enable tf32 if supported (#3473) [skip ci]

2026-03-16 23:47:00 -04:00