tocmo0nlord
0 Followers · 0 Following
  • Joined on 2026-02-11
Repositories
11
 Projects Packages Public Activity Starred Repositories
tocmo0nlord created branch kd-trainer-rebased in tocmo0nlord/axolotl 2026-04-26 15:07:23 +00:00
tocmo0nlord pushed to kd-trainer-rebased at tocmo0nlord/axolotl 2026-04-26 15:07:23 +00:00
ab491804e0 chore: lint
f7334a1719 make sure to use tensorboard to capture loss for checks
c45ab03487 fix adapter model check
0da0cd02e5 make sure to use the correct tokenizer
dd48ce7365 make sure to set tokenizer from l3 70b and save safetensors
Compare 10 commits »
tocmo0nlord created branch kd-trainer-v2 in tocmo0nlord/axolotl 2026-04-26 15:07:23 +00:00
tocmo0nlord created branch kd-logprob-data in tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord pushed to kd-logprob-data at tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
8fc4c420a4 Add kd coefficient scheduler
tocmo0nlord created branch kd-trainer in tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord pushed to kd-trainer at tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord created branch kd-trainer-2 in tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord pushed to kd-trainer-2 at tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
4a0ab11fcf chore: lint
73b6b0a580 chore: lint
9db5072407 make sure to use tensorboard to capture loss for checks
42d3e36a6f fix adapter model check
b12d93bedf make sure to use the correct tokenizer
Compare 10 commits »
tocmo0nlord created branch kd-trainer-pre in tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord pushed to kd-trainer-pre at tocmo0nlord/axolotl 2026-04-26 15:07:22 +00:00
tocmo0nlord created branch jagged-restart-lr-scheduler-v3 in tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
tocmo0nlord pushed to jagged-restart-lr-scheduler-v3 at tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
fe12aa79c8 jagged lr restart scheduler
tocmo0nlord created branch kd-fix-20250519-v2 in tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
tocmo0nlord pushed to kd-fix-20250519-v2 at tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
2491303c46 improve handling of train len
2c66483a47 default to dropping last batch in multipack batch sampler
01382b9a79 fix rebase issues
cfcd69df0d rename vars for consistency
2302b14a84 fix to remove attention_mask
Compare 10 commits »
tocmo0nlord created branch kd-logits-view in tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
tocmo0nlord pushed to kd-logits-view at tocmo0nlord/axolotl 2026-04-26 15:07:21 +00:00
ca379405c1 use narrow as a view on the student logits instead of slicing
tocmo0nlord pushed to ia3-peft at tocmo0nlord/axolotl 2026-04-26 15:07:20 +00:00
d0b534292f Add e2e test for ia3 ft
0bd89b38c6 migrate lora_ to peft_
481ef187a5 update README for IA3 peft
d645b19fcf include task type for ia3 config
203369411e consolidate as peft_model_dir
Compare 10 commits »
tocmo0nlord created branch iterable-optional in tocmo0nlord/axolotl 2026-04-26 15:07:20 +00:00
tocmo0nlord pushed to iterable-optional at tocmo0nlord/axolotl 2026-04-26 15:07:20 +00:00
1cfb8feb2d add iterable argument to preprocess-cli