tocmo0nlord
  • Joined on 2026-02-11
tocmo0nlord created branch lora_bf16 in tocmo0nlord/axolotl 2026-04-26 15:07:31 +00:00
tocmo0nlord pushed to lora_bf16 at tocmo0nlord/axolotl 2026-04-26 15:07:31 +00:00
a7676af44d hmmm
52e37077fc Merge branch 'main' into lora_bf16
850489405b working?
6874d32e0c more lora handling
Compare 4 commits »
tocmo0nlord created branch lora_kernels_fsdp in tocmo0nlord/axolotl 2026-04-26 15:07:31 +00:00
tocmo0nlord pushed to lora_kernels_fsdp at tocmo0nlord/axolotl 2026-04-26 15:07:31 +00:00
ede973b76c nits
tocmo0nlord created branch main-base in tocmo0nlord/axolotl 2026-04-26 15:07:31 +00:00
tocmo0nlord created branch llmcompressor-sft-wing in tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
tocmo0nlord pushed to llmcompressor-sft-wing at tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
c9880977be split llmcompressor from vllm checks
f196941315 additional fixes for docker and saving compressed
5be047ac46 Fix: Test
758115b8c6 Apply patch from @winglian
0dc1da5876 Add: line about further optimizations using llmcompressor
Compare 10 commits »
tocmo0nlord created branch lora-fsdp2-doc in tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
tocmo0nlord pushed to lora-fsdp2-doc at tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
3299f182ba ungate lora with bias
2fc430d365 update lora optims doc
Compare 2 commits »
tocmo0nlord created branch lora-kernels-deepspeed in tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
tocmo0nlord pushed to lora-kernels-deepspeed at tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
700409be6f removing deepspeed guard for LoRA Triton kernels
tocmo0nlord created branch lora-kernels-doc-fix in tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
tocmo0nlord pushed to lora-kernels-doc-fix at tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
02efd7e83d quick formatting fix for LoRA optims doc
tocmo0nlord created branch lora-quant-state-offset in tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
tocmo0nlord pushed to lora-quant-state-offset at tocmo0nlord/axolotl 2026-04-26 15:07:30 +00:00
1a22d16842 handle empty offset for quant state
tocmo0nlord created branch llmcompressor-sft in tocmo0nlord/axolotl 2026-04-26 15:07:29 +00:00
tocmo0nlord pushed to llmcompressor-sft at tocmo0nlord/axolotl 2026-04-26 15:07:29 +00:00
b708a1cc45 validate config to set defaults
daa9a58f83 Add: line about further optimizations using llmcompressor
ae7069e15b Merge branch 'main' into llmcompressor-sft
20d48cd617 Address Review Comments:
Compare 4 commits »
tocmo0nlord created branch llmcompressor-sft-v2 in tocmo0nlord/axolotl 2026-04-26 15:07:29 +00:00
tocmo0nlord pushed to llmcompressor-sft-v2 at tocmo0nlord/axolotl 2026-04-26 15:07:29 +00:00
6affbb1f85 move import of llmcompressor to reset session inside test
0ed4b4c310 make sure to reset the session after each test
f4a0f496a0 move decorator to test method instead of class
82b16bd040 split llmcompressor from vllm checks
fd5c985038 additional fixes for docker and saving compressed
Compare 10 commits »
tocmo0nlord pushed to llava-train at tocmo0nlord/axolotl 2026-04-26 15:07:28 +00:00