Default Branch

activeblue/main
Some checks failed
Tests Nightly against upstream main / pre-commit (push) Has been cancelled
Tests Nightly against upstream main / Prefetch S3 once to prime the CDN cache (push) Has been cancelled
Tests Nightly against upstream main / PyTest (3.12, 2.10.0) (push) Has been cancelled
Tests Nightly against upstream main / PyTest (3.12, 2.9.1) (push) Has been cancelled
Tests Nightly against upstream main / docker-e2e-tests (<nil>, 128, 12.8.1, 1, 3.11, 2.10.0) (push) Has been cancelled
Tests Nightly against upstream main / docker-e2e-tests (<nil>, 128, 12.8.1, true, 1, 3.11, 2.9.1) (push) Has been cancelled
Tests Nightly against upstream main / docker-e2e-tests (<nil>, 130, 13.0.0, true, 1, 3.12, 2.9.1) (push) Has been cancelled
Tests Nightly against upstream main / docker-e2e-multigpu-tests (<nil>, 128, 12.8.1, true, 2, 3.11, 2.9.1) (push) Has been cancelled
docker-nightlies / build-axolotl (<nil>, 128, 12.8.1, 3.11, 2.9.1) (push) Has been cancelled
docker-nightlies / build-axolotl-cloud (<nil>, 128, 12.8.1, 3.11, 2.9.1) (push) Has been cancelled
docker-multigpu-tests-biweekly / test-axolotl-multigpu (<nil>, 130, 13.0.0, 2, 3.11, 2.9.1) (push) Has been cancelled
docker-multigpu-tests-biweekly / test-axolotl-multigpu (fbgemm-gpu, 128, 12.8.1, 2, 3.11, 2.10.0) (push) Has been cancelled

c6da9b9e92 · Update SETUP_MIAAI.md: add bare Ubuntu rebuild section (driver, packages, Ollama) · Updated 2026-05-13 21:33:02 +00:00

Branches

9c221a6761 · code review feedback · Updated 2024-03-15 21:10:22 +00:00    tocmo0nlord

1404
7

34eb4e1677 · fix handling of ddp_find_unused_parameters · Updated 2024-03-14 21:45:42 +00:00    tocmo0nlord

1404
1

8c171aadb4 · drop unused padding_mask in llama patch · Updated 2024-03-14 21:26:30 +00:00    tocmo0nlord

1404
1

b7fe46579d · make the conversations/messages field configurable for sharegpt · Updated 2024-03-08 13:08:29 +00:00    tocmo0nlord

1418
1

3b432346e3 · WIP · Updated 2024-03-07 13:30:13 +00:00    tocmo0nlord

1419
1

718a8f4153 · update flash attention to 2.5.5 for gemma · Updated 2024-02-22 04:32:44 +00:00    tocmo0nlord

1455
1

d465b9fd98 · wip, jagged restarts · Updated 2024-02-16 19:34:08 +00:00    tocmo0nlord

1466
1

e08df47584 · wip load remote data from postgres · Updated 2024-02-12 14:55:24 +00:00    tocmo0nlord

1466
1

39ad38a1fb · update address and port for spaces · Updated 2024-02-08 22:55:44 +00:00    tocmo0nlord

1477
4

d46d7dfe30 · wip · Updated 2024-02-01 05:28:16 +00:00    tocmo0nlord

1489
3

1a538be9c2 · add a prelim test for expading the 4d mask · Updated 2024-01-26 05:41:24 +00:00    tocmo0nlord

1501
1

34de5b3bd5 · extras for the various flash attn subdirs and build those in the base module as it is a slow step · Updated 2024-01-26 05:40:39 +00:00    tocmo0nlord

1502
2

1b33588f09 · use low_cpu_mem_usage with ds zero 1 or 2 · Updated 2024-01-17 00:33:44 +00:00    tocmo0nlord

1549
2

eea6e8303a · Disable datasets caching when preparing dataset for packing · Updated 2024-01-15 22:48:24 +00:00    tocmo0nlord

1552
1

7ecc3a408c · Fix(debug): Use space delimiter for debug_text_only also · Updated 2024-01-07 03:45:19 +00:00    tocmo0nlord

1593
1

272bced137 · cpu offloading · Updated 2023-12-31 21:17:43 +00:00    tocmo0nlord

1609
11

856f5f6115 · Update README.md · Updated 2023-12-20 01:15:58 +00:00    tocmo0nlord

1631
1

450e04d3c4 · fix: remove excessive newlines in system prompt(s) for alpaca (#936) · Updated 2023-12-13 07:40:02 +00:00    tocmo0nlord

1641
0
Included

5bb4a782ce · dataloader defaults · Updated 2023-12-12 22:33:31 +00:00    tocmo0nlord

1647
1

a58a9e5f6c · Only fuse if flash_attn_fuse_mlp is True · Updated 2023-12-10 18:17:12 +00:00    tocmo0nlord

1647
5