tocmo0nlord
  • Joined on 2026-02-11
tocmo0nlord created branch streaming-remote-dataset in tocmo0nlord/axolotl 2026-04-26 15:08:03 +00:00
tocmo0nlord pushed to streaming-remote-dataset at tocmo0nlord/axolotl 2026-04-26 15:08:03 +00:00
e08df47584 wip load remote data from postgres
tocmo0nlord created branch streaming-v2 in tocmo0nlord/axolotl 2026-04-26 15:08:03 +00:00
tocmo0nlord pushed to streaming-v2 at tocmo0nlord/axolotl 2026-04-26 15:08:03 +00:00
d3bea3a2eb broken
2e2302aae3 remove unused
3a35076513 seems to be working?
Compare 3 commits »
tocmo0nlord created branch swe-rebench-rl-rebase in tocmo0nlord/axolotl 2026-04-26 15:08:03 +00:00
tocmo0nlord created branch stable in tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
tocmo0nlord pushed to stable at tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
tocmo0nlord created branch streaming in tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
tocmo0nlord pushed to streaming at tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
78a039e1be add depr warning for preprocess --iterable
53bbca2591 bugfix for sample packing
49bd6ece4a remove unused
42b38a718a remove eval streaming (not HF supported)
Compare 10 commits »
tocmo0nlord created branch streaming-on-the-fly-preprocess in tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
tocmo0nlord pushed to streaming-on-the-fly-preprocess at tocmo0nlord/axolotl 2026-04-26 15:08:02 +00:00
4870638734 initial impl of streaming preprocessing
ba681125d7 separate streaming and pretraining
Compare 3 commits »
tocmo0nlord created branch squash_position_ids in tocmo0nlord/axolotl 2026-04-26 15:08:01 +00:00
tocmo0nlord pushed to squash_position_ids at tocmo0nlord/axolotl 2026-04-26 15:08:01 +00:00
21ba1cd3f1 wire up squash_position_ids
tocmo0nlord created branch ssmi-main in tocmo0nlord/axolotl 2026-04-26 15:08:01 +00:00
tocmo0nlord pushed to ssmi-main at tocmo0nlord/axolotl 2026-04-26 15:08:01 +00:00
9793faf6dc pre-commit formatting fixes
64852ae15a Whitespace bug fix
1fed74b1d9 Catch configs without pretraining_tp
a300a4db1d Fix XFormers attention for Llama-2 70B (GQA)
Compare 4 commits »
tocmo0nlord pushed to sp-rl-v3 at tocmo0nlord/axolotl 2026-04-26 15:08:00 +00:00
926dc4af90 updates
6810f0ee19 minimize diffs to GRPO trainer
6c65eeaaf7 finalizing SP + GRPO trainer
7f4e4076e1 progress
4f2d092216 subclassing constructor
Compare 10 commits »
tocmo0nlord created branch split-batches-sizes in tocmo0nlord/axolotl 2026-04-26 15:08:00 +00:00
tocmo0nlord pushed to split-batches-sizes at tocmo0nlord/axolotl 2026-04-26 15:08:00 +00:00
bb65157dcf fix conditional for None values
7fd3d8abc4 handle batch size correchtly when using split and dispatch batches
Compare 2 commits »
tocmo0nlord created branch sppo in tocmo0nlord/axolotl 2026-04-26 15:08:00 +00:00
tocmo0nlord pushed to sppo at tocmo0nlord/axolotl 2026-04-26 15:08:00 +00:00