Skip to content

Actions: hiyouga/LLaMA-Factory

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
3,148 workflow runs
3,148 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

qwen2vl-72B全参微调时4张A800会从每张30多G的显存占用突然升到80G后报错 label_issue #1762: Issue #6270 opened by tdtgi
December 6, 2024 11:33 10s
December 6, 2024 11:33 10s
WorkNCCL label_issue #1761: Issue #6269 opened by bisque-qwe
December 6, 2024 11:17 16s
December 6, 2024 11:17 16s
qwen全量微调的时候报错 label_issue #1760: Issue #6268 opened by iwanttosignunp
December 6, 2024 11:03 11s
December 6, 2024 11:03 11s
评估预测问题 label_issue #1759: Issue #6267 opened by alixiaodi
December 6, 2024 09:27 14s
December 6, 2024 09:27 14s
If I want to dynamically adjust the loss coefficient in KTO, and need to pass a new feature to the loss computation part in the data, how should I modify the code? label_issue #1758: Issue #6266 opened by Jeryi-Sun
December 6, 2024 07:52 10s
December 6, 2024 07:52 10s
多卡运行报错 label_issue #1757: Issue #6264 opened by belikesunshine
December 6, 2024 04:35 13s
December 6, 2024 04:35 13s
是否支持14b 参数模型PPO label_issue #1756: Issue #6263 opened by hljjjmssyh
December 6, 2024 03:33 11s
December 6, 2024 03:33 11s
NPU 910B 获取版本报错 label_issue #1755: Issue #6261 opened by ninth99
December 6, 2024 02:50 15s
December 6, 2024 02:50 15s
react问题，重新描述 label_issue #1754: Issue #6260 opened by ccp123456789
December 6, 2024 02:36 10s
December 6, 2024 02:36 10s
微调后原始模型没有react功能 label_issue #1753: Issue #6258 opened by ccp123456789
December 6, 2024 01:53 11s
December 6, 2024 01:53 11s
请问单机多卡怎么训练 label_issue #1752: Issue #6257 opened by iwanttosignunp
December 5, 2024 17:26 15s
December 5, 2024 17:26 15s
Model Stops Generating Prematurely with llamafactory-cli chat label_issue #1751: Issue #6256 opened by Sherlock1956
December 5, 2024 15:09 20s
December 5, 2024 15:09 20s
Llama-Factory vs Axolotl? label_issue #1750: Issue #6254 opened by fzyzcjy
December 5, 2024 13:49 20s
December 5, 2024 13:49 20s
Merge pull request #6253 from hiyouga/hiyouga/qwen2vl_mm_proj tests #1632: Commit bcb4fb3 pushed by hiyouga
December 5, 2024 12:25 8m 8s main
main
December 5, 2024 12:25 8m 8s
[model] support qwen2vl train proj only tests #1631: Pull request #6253 opened by hiyouga
December 5, 2024 11:36 6m 52s hiyouga/qwen2vl_mm_proj
hiyouga/qwen2vl_mm_proj
December 5, 2024 11:36 6m 52s
单机多卡速度降低 label_issue #1749: Issue #6252 opened by rzhao-zhsq
December 5, 2024 10:36 9s
December 5, 2024 10:36 9s
Merge pull request #6251 from hiyouga/hiyouga/vllm_qwen2vl_infer tests #1630: Commit 561a8e5 pushed by hiyouga
December 5, 2024 10:26 7m 19s main
main
December 5, 2024 10:26 7m 19s
[infer] support qwen2vl vllm infer tests #1629: Pull request #6251 opened by hiyouga
December 5, 2024 10:18 7m 33s hiyouga/vllm_qwen2vl_infer
hiyouga/vllm_qwen2vl_infer
December 5, 2024 10:18 7m 33s
启动微调命令报类似错误 label_issue #1748: Issue #6250 opened by 2792715512
December 5, 2024 09:55 16s
December 5, 2024 09:55 16s
grad norm too large，梯度爆炸 label_issue #1747: Issue #6249 opened by Zbaoli
December 5, 2024 09:31 10s
December 5, 2024 09:31 10s
qwen2vl微调后怎么加载模型 label_issue #1746: Issue #6247 opened by Youmo1
December 5, 2024 08:51 11s
December 5, 2024 08:51 11s
kto训练有计划支持desirable weight和undesirable weight参数吗 label_issue #1745: Issue #6245 opened by xjtulien
December 5, 2024 08:06 14s
December 5, 2024 08:06 14s
在进行过一次全参SFT之后的Qwen2.5-7b-instruct上继续第二次全参SFT，最后Checkpoint从原本的4个变为了7个，total_size大了一倍 label_issue #1744: Issue #6244 opened by mori173
December 5, 2024 05:29 11s
December 5, 2024 05:29 11s
训练qwen2-vl出现问题 label_issue #1743: Issue #6243 opened by GasolSun36
December 5, 2024 05:16 12s
December 5, 2024 05:16 12s
Merge pull request #6242 from hiyouga/hiyouga/fix_script tests #1628: Commit c42890b pushed by hiyouga
December 5, 2024 03:54 8m 48s main
main
December 5, 2024 03:54 8m 48s