alibaba / Pai-Megatron-Patch
File Age & Freshness

File age measurements show the distribution of file ages (days since the first commit) and the file freshness (days since the latest commit).

Summary
File Change History Overall
File Age Distribution Overall
Days since first update
  • There are 280 files with 72,165 lines of code in files.
    • 165 files that are 366+ days old (50,611 lines of code)
    • 41 files that are 181-365 days old (9,083 lines of code)
    • 41 files that are 91-180 days old (6,708 lines of code)
    • 19 files that are 31-90 days old (4,013 lines of code)
    • 14 files that are 1-30 days old (1,750 lines of code)
70% | 12% | 9% | 5% | 2%
Legend:
366+
181-365
91-180
31-90
1-30

explore: grouped by folders | grouped by age
File Freshness Distribution Overall
Days since last update
  • There are 280 files with 72,165 lines of code in files.
    • 130 files have been last changed 366+ days ago (37,838 lines of code)
    • 43 files have been last changed 181-365 days ago (11,875 lines of code)
    • 57 files have been last changed 91-180 days ago (10,092 lines of code)
    • 30 files have been last changed 31-90 days ago (8,813 lines of code)
    • 20 files have been last changed 1-30 days ago (3,547 lines of code)
52% | 16% | 13% | 12% | 4%
Legend:
366+
181-365
91-180
31-90
1-30

explore: grouped by folders | grouped by freshness
File Change History per File Extension
py, sh, md, json, patch, txt, gitignore, gitmodules
File Age Distribution per Extension
Days since first update
366+
181-365
91-180
31-90
1-30
py70% | 12% | 9% | 5% | 2%
File Freshness Distribution per Extension
Days since last update
366+
181-365
91-180
31-90
1-30
py52% | 16% | 13% | 12% | 4%
File Change History per Logical Decomposition
primary
primary (file age distribution)
Days since first update
366+
181-365
91-180
31-90
1-30
megatron_patch70% | 13% | 10% | 3% | 1%
toolkits67% | 11% | 6% | 9% | 4%
rlhf100% | 0% | 0% | 0% | 0%
primary (file freshness distribution)
Days since last update
366+
181-365
91-180
31-90
1-30
megatron_patch59% | 12% | 18% | 6% | 3%
toolkits35% | 26% | 5% | 24% | 8%
rlhf100% | 0% | 0% | 0% | 0%
Oldest Files (Top 50)
File# lines# unitscreatedlast modified# changes
(days)
# contributorsfirst
contributor
latest
contributor
transformer.py
in megatron_patch/model/llama2
1296 35 2023-09-04 2024-02-28 13 4 jerryli1981@users.noreply.g... 38210876+lwmlyy@users.norep...
transformer.py
in megatron_patch/model/qwen
1243 35 2023-09-04 2024-02-02 5 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/baichuan
1179 32 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/glm130b
875 25 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/starcoder
848 31 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/falcon
845 31 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/bloom
811 27 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/llama
715 26 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/falcon40b
683 28 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/baichuan
649 14 2023-09-04 2024-02-02 8 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/falcon
612 12 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
training.py
in megatron_patch
612 8 2023-09-04 2024-02-02 8 3 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/chatglm
604 20 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/falcon40b
583 12 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/starcoder
583 12 2023-09-04 2024-02-02 3 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
reward_model_to_megatron.py
in toolkits/model_checkpoints_convertor/bloom
573 12 2023-09-04 2024-02-02 3 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/bloom
572 12 2023-09-04 2024-02-02 3 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer.py
in megatron_patch/model/galactica
570 21 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/baichuan
515 20 2023-09-04 2023-10-12 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/galactica
501 19 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/llama
501 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/falcon40b
491 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/falcon
491 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/chatglm
473 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/galactica
454 13 2023-09-04 2024-02-02 4 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/llama2
454 15 2023-09-04 2024-03-05 8 3 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
arguments.py
in megatron_patch
449 2 2023-09-04 2025-04-28 35 5 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
language_model.py
in megatron_patch/model/qwen
440 15 2023-09-04 2023-10-12 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/glm130b
434 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/bloom
411 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/chatglm
396 11 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/starcoder
387 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
hf2te.py
in toolkits/model_checkpoints_convertor/baichuan
378 8 2023-09-04 2024-02-02 5 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/glm
378 10 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
checkpoint_reshaping_and_interoperability.py
in toolkits/model_checkpoints_convertor/glm130b
363 10 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
rm_main.py
in rlhf/deepspeed-chat
319 2 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
generation.py
in megatron_patch/generation
317 5 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
icetk_glm130b_tokenizer.py
in megatron_patch/tokenizer
273 39 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
finetune_utils.py
in megatron_patch
202 8 2023-09-04 2024-06-13 6 4 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
preprocess_data.py
in toolkits/pretrain_data_preprocessing
198 6 2023-09-04 2024-12-05 6 3 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
trlx_bloom_rlhf.py
in rlhf/trlx
178 5 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
api.py
in megatron_patch/generation
170 4 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
151 7 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
deepspeed_to_megatron_ori.py
in toolkits/model_checkpoints_convertor/bloom
149 9 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
deepspeed_to_megatron.py
in toolkits/model_checkpoints_convertor/bloom
149 9 2023-09-04 2024-02-02 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
tokenization_baichuan.py
in megatron_patch/tokenizer
139 13 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
positional_embeddings.py
in megatron_patch/model/bloom
122 10 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/baichuan
106 7 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
configuration_RW.py
in toolkits/model_checkpoints_convertor/falcon
101 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
merge_130b_ckpts.py
in toolkits/model_checkpoints_convertor/glm130b
96 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
Files Not Recently Changed (Top 50)
File# lines# unitscreatedlast modified# changes
(days)
# contributorsfirst
contributor
latest
contributor
__init__.py
in megatron_patch/model/starcoder
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/qwen
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/llama2
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/falcon
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/llama
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/bloom
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/baichuan
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/chatglm
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/glm130b
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in megatron_patch/model/falcon40b
1 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
convert_json_to_list.py
in toolkits/pretrain_data_preprocessing
10 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
enums.py
in megatron_patch/model/starcoder
19 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
glu_activations.py
in megatron_patch/model/starcoder
32 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
configuration_RW.py
in toolkits/model_checkpoints_convertor/falcon40b
50 3 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
jiebabpe_tokenizer.py
in megatron_patch/tokenizer
53 8 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
positional_embeddings.py
in megatron_patch/model/llama
54 4 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
positional_embeddings.py
in megatron_patch/model/chatglm
60 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
clean_raw_text.py
in toolkits/pretrain_data_preprocessing
69 3 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
tokenization.py
in megatron_patch/generation
76 3 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
preprocess_wudao2.py
in toolkits/pretrain_data_preprocessing
76 3 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/glm130b
80 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/bloom
81 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
81 2 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/chatglm
82 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/starcoder
83 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
layers.py
in megatron_patch/model/bloom
87 2 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/qwen
88 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/llama
92 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/falcon
94 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/galactica
94 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/falcon40b
94 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
merge_130b_ckpts.py
in toolkits/model_checkpoints_convertor/glm130b
96 - 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
configuration_RW.py
in toolkits/model_checkpoints_convertor/falcon
101 6 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
gpt_model.py
in megatron_patch/model/baichuan
106 7 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
positional_embeddings.py
in megatron_patch/model/bloom
122 10 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
tokenization_baichuan.py
in megatron_patch/tokenizer
139 13 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
deepspeed_to_megatron_ori.py
in toolkits/model_checkpoints_convertor/bloom
149 9 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
151 7 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
api.py
in megatron_patch/generation
170 4 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
trlx_bloom_rlhf.py
in rlhf/trlx
178 5 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
icetk_glm130b_tokenizer.py
in megatron_patch/tokenizer
273 39 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
generation.py
in megatron_patch/generation
317 5 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
rm_main.py
in rlhf/deepspeed-chat
319 2 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/starcoder
387 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/bloom
411 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/glm130b
434 15 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/chatglm
473 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/falcon
491 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model.py
in megatron_patch/model/falcon40b
491 17 2023-09-04 2023-09-04 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
Most Recently Created Files (Top 50)
File# lines# unitscreatedlast modified# changes
(days)
# contributorsfirst
contributor
latest
contributor
gpt_layer_specs.py
in megatron_patch/model/qwen3_moe
347 6 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
router.py
in megatron_patch/model/qwen3_moe/moe
111 4 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_utils.py
in megatron_patch/model/qwen3_moe/moe
79 2 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_layer.py
in megatron_patch/model/qwen3_moe/moe
70 2 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_module_specs.py
in megatron_patch/model/qwen3_moe
61 1 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
m2h_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
499 17 2025-04-15 2025-05-12 2 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
h2m_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
259 13 2025-04-15 2025-05-12 2 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
119 9 2025-04-15 2025-05-12 3 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
convert.py
in toolkits/distributed_checkpoints_convertor/impl
80 2 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
h2m_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
46 3 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
m2h_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
42 3 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
patch.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
25 2 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
__init__.py
in toolkits/distributed_checkpoints_convertor/impl/general
6 - 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
__init__.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
6 - 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
hf2mcore_qwen2_moe.py
in toolkits/model_checkpoints_convertor/qwen
555 11 2025-03-25 2025-04-29 3 3 wanqian5@tal.com 46404040+lostkevin@users.no...
hf2mcore_qwen2.5_vl.py
in toolkits/model_checkpoints_convertor/qwen
607 10 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
transformer_block.py
in megatron_patch/model/qwen2_5_vl
434 12 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
dataset_helpers.py
in megatron_patch/data
381 9 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
visionmodel.py
in megatron_patch/model/qwen2_5_vl
224 10 2025-03-21 2025-04-10 2 2 46404040+lostkevin@users.no... wzuck.wang@gmail.com
model.py
in megatron_patch/model/qwen2_5_vl
191 6 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
build_llava_frame_dataset.py
in toolkits/multimodal_data_preprocessing
123 5 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
convert_custom_dataset_to_wds_chatml.py
in toolkits/multimodal_data_preprocessing
98 2 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
image_processing.py
in megatron_patch/data
67 6 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
transformer_config.py
in megatron_patch/model/qwen2_5_vl
55 2 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
replace_llava_image_key.py
in toolkits/multimodal_data_preprocessing
29 1 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
convert_llava_pretrain_to_wds.py
in toolkits/multimodal_data_preprocessing
25 1 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
layer_specs.py
in megatron_patch/model/qwen2_moe
281 6 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
transformer_config.py
in megatron_patch/model/qwen2_moe
55 1 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
__init__.py
in megatron_patch/model/qwen2_moe
1 - 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
fp8_cast_bf16.py
in toolkits/model_checkpoints_convertor/deepseek
88 3 2025-02-25 2025-02-25 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
hf2mcore_deepseek_v3_moe.py
in toolkits/model_checkpoints_convertor/deepseek
578 12 2025-02-21 2025-04-03 13 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
helper.py
in megatron_patch/template
115 3 2025-02-21 2025-05-09 6 2 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
json_sft.py
in megatron_patch/data
106 7 2025-02-21 2025-02-21 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in toolkits/model_checkpoints_convertor/utils
146 5 2025-01-17 2025-04-03 6 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...
multi_latent_attention.py
in megatron_patch/model/deepseek_v2
276 4 2025-01-16 2025-04-28 2 2 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
mlp.py
in megatron_patch/model/deepseek_v2
196 4 2025-01-16 2025-01-16 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
shared_experts.py
in megatron_patch/model/deepseek_v2/moe
180 9 2025-01-16 2025-01-16 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
language_model_embedding.py
in megatron_patch/model/qwen2_vl
98 3 2025-01-15 2025-01-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
tensor_parallel.py
in megatron_patch
66 3 2024-12-27 2024-12-27 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
chatml.py
in megatron_patch/data/energon
46 3 2024-12-27 2024-12-27 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
attention.py
in megatron_patch/model/mixtral_bak/transformer
322 11 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
token_dispatcher.py
in megatron_patch/model/mixtral_bak/moe
172 7 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
model.py
in megatron_patch/model/mixtral_bak
162 4 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
transformer_config.py
in megatron_patch/model/mixtral_bak
142 1 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
experts.py
in megatron_patch/model/mixtral_bak/moe
136 4 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
mlp.py
in megatron_patch/model/mixtral_bak/transformer
131 6 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
router.py
in megatron_patch/model/mixtral_bak/moe
113 11 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
layer_specs.py
in megatron_patch/model/mixtral_bak
86 3 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
moe_layer.py
in megatron_patch/model/mixtral_bak/moe
57 4 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
moe_utils.py
in megatron_patch/model/mixtral_bak/moe
39 6 2024-12-24 2024-12-24 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
Most Recently Changed Files (Top 50)
File# lines# unitscreatedlast modified# changes
(days)
# contributorsfirst
contributor
latest
contributor
m2h_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
499 17 2025-04-15 2025-05-12 2 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
h2m_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
259 13 2025-04-15 2025-05-12 2 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/general
119 9 2025-04-15 2025-05-12 3 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
helper.py
in megatron_patch/template
115 3 2025-02-21 2025-05-09 6 2 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
hf2mcore_qwen2_moe.py
in toolkits/model_checkpoints_convertor/qwen
555 11 2025-03-25 2025-04-29 3 3 wanqian5@tal.com 46404040+lostkevin@users.no...
preprocess_data_megatron.py
in toolkits/pretrain_data_preprocessing
360 13 2024-04-15 2025-04-29 10 4 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
gpt_layer_specs.py
in megatron_patch/model/qwen3_moe
347 6 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
router.py
in megatron_patch/model/qwen3_moe/moe
111 4 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_utils.py
in megatron_patch/model/qwen3_moe/moe
79 2 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_layer.py
in megatron_patch/model/qwen3_moe/moe
70 2 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
moe_module_specs.py
in megatron_patch/model/qwen3_moe
61 1 2025-04-29 2025-04-29 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
arguments.py
in megatron_patch
449 2 2023-09-04 2025-04-28 35 5 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
multi_latent_attention.py
in megatron_patch/model/deepseek_v2
276 4 2025-01-16 2025-04-28 2 2 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
transformer_config.py
in megatron_patch/model/deepseek_v2
42 1 2024-05-27 2025-04-28 5 3 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
convert.py
in toolkits/distributed_checkpoints_convertor/impl
80 2 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
h2m_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
46 3 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
m2h_synchronizer.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
42 3 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
patch.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
25 2 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
__init__.py
in toolkits/distributed_checkpoints_convertor/impl/general
6 - 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
__init__.py
in toolkits/distributed_checkpoints_convertor/impl/deepseek_v3
6 - 2025-04-15 2025-04-15 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
visionmodel.py
in megatron_patch/model/qwen2_5_vl
224 10 2025-03-21 2025-04-10 2 2 46404040+lostkevin@users.no... wzuck.wang@gmail.com
hf2mcore_deepseek_v3_moe.py
in toolkits/model_checkpoints_convertor/deepseek
578 12 2025-02-21 2025-04-03 13 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
__init__.py
in toolkits/model_checkpoints_convertor/utils
146 5 2025-01-17 2025-04-03 6 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...
sample_stats.py
in toolkits/sft_data_preprocessing
23 - 2024-07-24 2025-04-03 2 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
utils.py
in megatron_patch/data
318 5 2024-01-28 2025-03-31 20 3 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
hf2mcore_qwen2.5_vl.py
in toolkits/model_checkpoints_convertor/qwen
607 10 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
transformer_block.py
in megatron_patch/model/qwen2_5_vl
434 12 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
dataset_helpers.py
in megatron_patch/data
381 9 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
model.py
in megatron_patch/model/qwen2_5_vl
191 6 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
build_llava_frame_dataset.py
in toolkits/multimodal_data_preprocessing
123 5 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
convert_custom_dataset_to_wds_chatml.py
in toolkits/multimodal_data_preprocessing
98 2 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
image_processing.py
in megatron_patch/data
67 6 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
transformer_config.py
in megatron_patch/model/qwen2_5_vl
55 2 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
replace_llava_image_key.py
in toolkits/multimodal_data_preprocessing
29 1 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
convert_llava_pretrain_to_wds.py
in toolkits/multimodal_data_preprocessing
25 1 2025-03-21 2025-03-21 1 1 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
hf2mcore_qwen2_vl.py
in toolkits/model_checkpoints_convertor/qwen
616 10 2024-11-27 2025-03-19 9 2 46404040+lostkevin@users.no... 46404040+lostkevin@users.no...
layer_specs.py
in megatron_patch/model/qwen2_moe
281 6 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
transformer_config.py
in megatron_patch/model/qwen2_moe
55 1 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
__init__.py
in megatron_patch/model/qwen2_moe
1 - 2025-03-12 2025-03-12 1 1 qianwan@ymail.com qianwan@ymail.com
hf2mcore_deepseek_v2_moe.py
in toolkits/model_checkpoints_convertor/deepseek
454 8 2024-05-27 2025-03-05 13 3 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
__init__.py
in megatron_patch/data
88 7 2023-11-10 2025-02-26 12 3 jerryli1981@users.noreply.g... 46404040+lostkevin@users.no...
fp8_cast_bf16.py
in toolkits/model_checkpoints_convertor/deepseek
88 3 2025-02-25 2025-02-25 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
hf2mcore_qwen2_dense_and_moe_gqa.py
in toolkits/model_checkpoints_convertor/qwen
821 10 2024-06-19 2025-02-21 12 5 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
hf2mcore_llama3_1.py
in toolkits/model_checkpoints_convertor/llama
710 11 2024-08-23 2025-02-21 3 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...
hf2mcore_mixtral.py
in toolkits/model_checkpoints_convertor/mistral
672 10 2024-04-22 2025-02-21 4 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
attention.py
in megatron_patch/model/qwen2_vl
530 13 2024-11-27 2025-02-21 2 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...
attention_vision.py
in megatron_patch/model/qwen2_vl
529 13 2024-11-27 2025-02-21 2 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...
hf2mcore.py
in toolkits/model_checkpoints_convertor/mistral
468 14 2024-04-21 2025-02-21 5 2 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
json_sft.py
in megatron_patch/data
106 7 2025-02-21 2025-02-21 1 1 jerryli1981@users.noreply.g... jerryli1981@users.noreply.g...
layer_specs.py
in megatron_patch/model/qwen2_vl
95 3 2024-11-27 2025-02-21 2 2 46404040+lostkevin@users.no... jerryli1981@users.noreply.g...