issues
search
InternLM
/
InternEvo
InternEvo is an open-sourced lightweight training framework aims to support model pre-training without the need for extensive dependencies.
https://internevo.readthedocs.io/zh-cn/latest/?badge=latest
Apache License 2.0
310
stars
52
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
fix(linear.py): linear module uneven split is forbidden
#374
huangting4201
opened
2 days ago
0
fix(monitor): send exception when feishu alert is enable && remove light monitoring address
#373
JiaoPL
opened
3 days ago
0
[QA] Does internEvo support loongtrain selective checkpoint++?
#372
wplf
opened
3 days ago
1
fix(gmm): change communicator.grad_hook to async
#371
blankde
opened
3 days ago
0
fix(mha.py): fix evaluation argu key err
#370
huangting4201
closed
5 days ago
0
feat(fp8): [Work In Progress] enable FP8 training
#369
zigzagcai
opened
2 weeks ago
1
remove unused moe changes , modify _q_kv_without_cu_seqlens and _SplitForwardGatherBackward
#368
KkHu-Kistch
closed
2 weeks ago
0
Add hetero feat
#367
fumihwh
closed
2 weeks ago
0
fix(isp.py): fix isp overlap backward allgather twice when activation ckpt 0.x
#366
huangting4201
opened
3 weeks ago
0
Add z loss to PipelineSchedule
#365
zhhsplendid
closed
3 weeks ago
0
fix lumina model and add lumina ckpt support
#364
SHshenhao
closed
3 weeks ago
0
fix lumina model and add lumina ckpt support
#363
SHshenhao
closed
3 weeks ago
0
fix lumina model and add lumina ckpt support
#362
SHshenhao
closed
3 weeks ago
0
A PR Provides Multi Machine MPI scripts
#361
zhhsplendid
closed
3 weeks ago
0
fix(mlp.py): fix mlp w1w2w3 init order to w1w3w2
#360
huangting4201
opened
3 weeks ago
0
fix llava model device bugs
#359
hellozmz
opened
3 weeks ago
0
Feat/refactor process group
#358
mwiacx
opened
3 weeks ago
0
feat(pipeline): Zero Bubble V Shape Memory Efficient Editon
#357
li126com
closed
3 weeks ago
0
Tmp fix QK norm bug
#356
zhhsplendid
closed
3 weeks ago
0
Feat/heterogeneous x pu training
#355
KkHu-Kistch
closed
3 weeks ago
0
[QA] 如何进行单卡微调的，需要调整那些设置
#354
OkGuai
opened
4 weeks ago
0
[Feature] Add Lumina Model to InternEvo. Tested on MUXI single card
#353
zhhsplendid
closed
1 month ago
0
feat(moe): add gshard token rearrange optim
#352
blankde
opened
1 month ago
0
fix(checkpoint/components.py): fix lr scheduler resume step count
#351
huangting4201
closed
1 month ago
0
feat(moe): support moe zero1 setting
#350
blankde
opened
1 month ago
0
feat(model): support kv head copy
#349
yingtongxiong
closed
3 weeks ago
0
fix(moe): dropless moe loss
#348
blankde
closed
1 month ago
0
doc(2d): docs for 2d-attention
#347
yingtongxiong
closed
1 month ago
0
[QA] loong train 支持packed_sample_into_one=false吗
#346
Lzhang-hub
opened
1 month ago
1
feat(moe): support group mlp for moe
#345
blankde
closed
1 month ago
0
feat(dataloader): refine implementation of mocked and megatron dataloader
#344
zigzagcai
opened
2 months ago
0
feat(zero bubble): update zbh1
#343
li126com
opened
2 months ago
0
[Bug] There will be timeout in some cases.
#342
kkscilife
closed
1 month ago
1
fix inject model and add multimodal dataloader
#341
sallyjunjun
closed
1 month ago
0
fix(enable_qkv_fusion): minor fix for qkv fusion
#340
zigzagcai
closed
2 months ago
0
fix dispatch model
#339
sallyjunjun
closed
2 months ago
0
fix(enable_qkv_fusion): refine wqkv fusion
#338
zigzagcai
closed
2 months ago
0
fix wqkv fusion
#337
zigzagcai
closed
2 months ago
0
fix wqkv fusion
#336
zigzagcai
closed
2 months ago
0
fix wqkv dim when enable qkv fusion
#335
sallyjunjun
closed
2 months ago
0
fix(pipeline): fix zero bubble pipeline parallelism
#334
li126com
closed
2 months ago
0
Feat(adam): support apex FusedAdam
#333
li126com
closed
2 months ago
0
feat(moe): add moe async param handler
#332
blankde
opened
2 months ago
0
feat(usability): Refine model inject helper to support huggingface models
#331
zigzagcai
closed
2 months ago
0
remove isp memory pool
#330
mwiacx
closed
2 months ago
0
update test loss
#329
li126com
opened
2 months ago
0
fix(isp): fix unnecessary module gather for isp
#328
blankde
closed
2 months ago
2
add qwen2moe and mixtral
#327
sallyjunjun
closed
2 months ago
1
feat(model: impl gpt 567 b
#326
blankde
closed
1 month ago
0
[Feature] MoE模型里稠密层和专家层zero和并行的解耦
#325
sunpengsdu
opened
2 months ago
0
Next