diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..a1c4145b9efb4c676757d723b2233ccf8629a36b --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ec706b3d0b2c294647d558c141620a5a22be0efdfcc182674355d783b410710 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..0f1ba2470670ac63dc93371da7b28a3d1ed7688f --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19042acb2ad7f43d269fe6e6e2bdf2bec3fef1708bb8608206fddc4e1be9ec07 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..9f1651547a1652bb0f03ed1b258ec7b69d2bf2f0 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:028d29e998e9f22cd278e19b0a6729e02bbf7e4c45b666b957346a4b50add891 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..74019eac3ac0df6ea4186c959c36592caa49d94b --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b337fa2838504e202c535c1e807d254104830ae763eebeb6f57e4313f777b75 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..d0829317c2aafd92be9c4cbed4fae31bf0ea35f4 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15f04e623030a1c516c650e6830cd5aa816f1974cf4cd79cd79bfd83d77fffc7 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..58332c2cc102672c143522697413580d89187c9a --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b2be2b926e89278b395da3a787df4b1a494ee5fb49a47cbaf2d5b510ec63b29e +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..4868af83ec349392105f0241f8b90b31574e71c9 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6eeb7c6cbebfa625051f0fbc6167f5e788571ba8d6e9f4b80dde42180814e015 +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..66c64e8c575bbeadeeb3764b42849603f3405986 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/extra_state_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb47fe292e2bd42fbbcd95e15c34f2de787be26033ce5d72dd005c9780ca33f3 +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..8699f07fc2a12b8717156ca20ec7303f4d2aeef0 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90f06363ec7b8bb60d3a22f13ac1d1fd356b9cd91d06f5cb25bef7288060af00 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..2898c71bb37374c2c0a148717318505c4e143899 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4074e1f2b92f23d92b46665e62e9d15009cb186ef85a19d8267c4d49c17af87 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..38b314b2b88cee56a286e3ae371c84355dad1fca --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad53f481bf47b6b2b86544c4872caf8df923dae9c61ca74729d71c21e9562e39 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..8932d0a8c4a8ab186ff6e6e15d31ec72b89e554a --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1473fb0a56f35b2c958a335e24bac0b9400fdd46aaebe2e37b041dcb1fd5fb43 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..68dd24068f76c98b5b5a2e3717a413e7e01f55f6 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02bc74f81b394ca2516e23ad19c78d0830ef7d6f7e8dd2aabf6917b8e056105b +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..2265c50e3bf78a1042f33978b26f71bf061ba2d2 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bec17da5388f3756cf89c624709d3c0f505ebc422d0861838b41ac94803cb5d +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..c211b42c39b360166fb590a9839eb3716a8553b1 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3dc5fc3c7bf422cfd5b59ce576d62f51d5e5252a8554be9ed77dcd5a83c44f41 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..17e52e796f919741096760b81ff946d5832b9072 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/model_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eaf1e4f94e587791db7feab7bc7c224a5682cb280d5462ed983560dba40da989 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..c287f5b32d0e5cafebd116838c23898480532769 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64a3d45ae24c724e205e214f885e50b43f8eec83a76f9863914cbb83bd903a40 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..0857a805e39c166ee5bac85415e385fae9bf0854 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:345693c574ba75057d728e0eb1b9a85647c7d2f071736614a12c91401fef09d3 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..54fdf3f151363fee766ff9f06e87c405a207b2a6 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c0efa8e707f4151e02724200450c1bcc55677aa0949f475562a0708ccc3d1e0 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..8cdd1ecc8d03bb4083dcbe0606fe3d4624d52820 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42339bc29e8dee8b1ae7dcbf4617c8b8c6b1eacd49cfb9dc21da27409b0195f7 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..024654bfa20bd6048fe95478e00ba78ca912cdda --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5160676f1ae9855b69469be0a2705c25e8009c694ff090fdcaa18d3f72612408 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..bc7f3cee465b4af6bce948c5c0da26564475fea4 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02182a0829cd29af011137180527291a22fe5e5b3106eb98eb8c28e2226b3225 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..1498cfafe256dd484553706b6c4b953abb1c716a --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac89994cf3c2cbfa5147ff8d93594fcdf4370c9cbec68ce876f347085df87c25 +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..a6b9faa4a30933a688927a28cc5cbdcf2c7f0072 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/actor/optim_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cdbcba0aca7d26a237cd57dc21b37f95417b56bcfb41962a9819a69a742e5648 +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/data.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..381e23b47da45b981346bf914cd28a962997160d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_10/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:65f96f3b613a1e07c7133914cacc9043779e35ccd7968a6f1d49a1c1bf2e9777 +size 6948429 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..6305398ae32d61da35a0a4f31b6c99ba10293dc0 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a322865d55115565d33d2c8901b869e0ab548579eb4052785c3b5c1df03c164e +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..8c1d713c0dab73f076d28a6e4498911ed0a50af2 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:077a7f22cf56438a14e90f21bfd16e2b6ce776737ca8acdd7db12d83629dd8ff +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..bcd0a9c05dae556cda7c8988d2d09c2cf81b56e7 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:67d7242529563cb2715f15dcb57c6a699cdc7b78710a896bac0b834f7d89f387 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..25b47e7f6436d9c2a53c9ea8023065eafdc1dd7c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf07697cebe4c12997265aa01282cdcef23abe5c3b74d428f54689067334b534 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..18b8bc5f1b845ffb1ee14c12b1719d3c322456c8 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35ff14051e464e0d36b5b1a70772c205cf8dd669b3043a778266953331020edc +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..fb2b796cc87b706ee05a898175ed0419b9f40720 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8fe1c466178201bc0a31a52c3b613652785f981b4355ad3374db18f96e8d1f85 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..ce37cfee5ec53c44467bc3b83e0bc24f75efed50 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c9f5cdcc60de6900a9bfb5c40590a15605af1c55733a6ec9cbe3680d80f1e2b +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..642bc82a4cab92f7cbd5d7ca55d4b4d4499fcbc2 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/extra_state_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5f16db076258b2224b935db1d82eebc1f53742d747081ba165a2779437989e5 +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..673838a19a178a1351698d7d28483345d84ca395 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ead6d38b4574b6bdb51a57ab010c537a6d4bf96f6f91db38120c75ccc26527d2 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..e7eeccda71bb4e47a5f6cf9928bc322db171177d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7b60cb231ac322f9208b0d36b6d7448c262c3dec72618c0ca349f85cc5d8ecb +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..ca9be2e16fad82793f798c0ff2544e3255e5b596 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35af83606e6099294b39e5ec8da909d7265d59b8cf0daaf376780f39ea075498 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..5f3689afa983d3d9f38c5b34a823227fc08888e2 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2d57a0a26c885cc8aa92d068db3b094cfb81f74edf6448a7da8b815399c209c +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..6e1ad290b1f807be954ba9756fc2fbf1d4417c47 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2cfb0f3cdc5d13ab78760c1436edc347567efbc728b9b5f2d2d2b93d8430b0e7 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..5b4143a1e9e4a72ca2d8adacc89a033ba56a1087 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab5898bcc26fa0ee1b5fa6d320430ba8fcc00982e4bb13189518a37018c72903 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..8e86326c0cdccab5dd57639287988d2b22705761 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:411b7568707eebcd15117069ea902f1a54fed5bb29d3a1222e0cd57d887c4258 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..216a8953c04116b929b7385a52b7c6ddccf182f2 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/model_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bd8769076e556a6539a9b3954f4a393f40164e44f99cbb1dd21e2423db7ca8f +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..1925280281ea205a4bc942a13a58c34b8769627f --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae40ad77748b635c9219b967fc725f0004cef4cfbed19dfd721d7b9358fbf22a +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..9d18100317ae4e2faf3994a18b07c416d1739bd3 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:743a973aae9340119452e1e7d837cdf96733932c4af936dec3c416db85579013 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..c1d2a3454cf83ddbbe2920545143192f3aa440ec --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e295dc7dfdfba1400e7c84acfc6f8c17a38638972e7a3f58bf6a359acc6026eb +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..e1370080765834a4fa1307e3d11428309172e38f --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:624493460a3eaa7a5df3953575aa8c263b918446fcab215a739449af64c1c454 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..91bd3e3992a1554a5fe0cc55f0581fb6a59b68a9 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f083d0c7204d8670b9bae7687f5360e1b4d7c85baab8fdc93b25aee4fea2bce6 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..09392bc5f8062ff406924c1ee12ca57e4f05411b --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af4a8d9dccefd37b51e614610fa3cd385d924d82fb121bc2cfccec39f85f0d42 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..24b97d73b36c2d306d2ba24867485d3b692e3bf7 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0164128fdbd72049b43f6195b041454a49a5797c5e196dc0a78527b5bc855d0 +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..b7f66ba849ed570554322772611b99c8513d1f95 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/actor/optim_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4886ae984f51f2bbaf710eb9c89a50c64c74c11b3831df375ff83761fcad7290 +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/data.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..381e23b47da45b981346bf914cd28a962997160d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_15/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:65f96f3b613a1e07c7133914cacc9043779e35ccd7968a6f1d49a1c1bf2e9777 +size 6948429 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..97100776d2192b809022424e44dfd236e9c12b7e --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b70e27e9b0963d30d26599fb1c3ca09e901b8cbedace047fb7b3947fcdbca49 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..3941aa8e51d151fe1c612e9b6584454b36e48408 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5f643661aad08a58a1c562f3aa99870e4bbf008b52d81113fef19c4590b8709 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..0b63ee10aecfa260af119ee76daa091c45fad419 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b29a0586caf56d3a3b90a6b8454086d2f165525fbcd0593dba33c388afe03381 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..3efe7ad7cdcea23ad1be15e2751d67761242a57c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7cac41ad7129d4663822b118f03367e4b13428cd8d1681393e6753f987730ff6 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..ca0b2b80e314d1a3b1c9f541beeb02bef5f0c011 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09f503766dce38ea202f91023a7422354b8a174b23cd79fc5374c3e13cf45f3a +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..fd736c36d81c9148284285f2cdac7368cde34a6a --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e7ea5aebc38f3f238e93d58be57272be2deaf915252e980d7575f659e847434 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..ec58a377cd47f49a89e42ecee88e631ca75d4b0c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1de87a783430ce03ca37ea0eab0b0324a17a98f181098991ca76f674ae2ad841 +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..5e6516309a068ce57202dc158d4117689748fd51 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/extra_state_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:03093427c6e97818df731eb6d8154a5e510690d4761c24fc2581bcaf5bfa162c +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..e14544380834a01a0cd70f85a8d5264e286c7881 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:981b39d95370729d6fa53a9fd3657fcd066c0fc5030f381690803c38fd8ab0e9 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..2c61c912f340e583342c03aa876d6fa61586c87b --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6713f5b1238e3a8c9148a7ab38f97dd459aeb239108e4080519f3332adb6a0cd +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..2797ef57be1690ab2fc943b28f144833bb765f62 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4dc892df785eb3e8045a81dc31ce55b821bf123f7f1c722b5e4e93f721ff2f03 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..a848b7c35e226da69fb43c566f9824176fe452f6 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b632b3c75300b3991548c2fc583f64f29dc0d5502b7d23a5f4f509ce50ddb92 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..b86108485f704003f11aabb315864c30f699d444 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30d4c7fc106269d26bf8c74eb78898a97ce5515472540de9e9cb5b0db4ff16ed +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..f792b3fc36ae0ccc3aad250713424aa039b90176 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b401cab4a180d6a0e087667d8902293dedbce1d98aeda0c2ec089cbaea60610 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..4b60f0d5c7cc34257a09a1f4e4b83a108d1f651c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59fe1b85917267ffaebf8094aea24deb7682423c965c1f73a30a03d277820f18 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..142bdcb6d1ba946610fb86d9940f6a7d42fb55b0 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/model_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3c4eda12c954f67abd5b162f3724ddf29b34d7f6b8ea93c510a38cacb871ca3 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..eec575015b1131927dd87ca1256c78bdf73351dc --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb341b0bb85573d1b2b2d37d59e93e9e3bd6e758145373d9edf8aff4634978d7 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..4f37fb2129d31f67df0f43c784f900fef4ada027 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66da04bcc77f6624d55ecb7bbe9c312e996ed4350792ff51c6c480b4483ec44d +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..041fce0d4bf34311a3985ecec985941680dc85fe --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52719dd55cdc29b1a3fc5acb180e79f62eb142c4945ebda7df5762acd427b3e7 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..4134740086bd2709bf224bcd81ab55d437689f7d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b86d8d5d2fdaacf59ae39c909fd6bbde5a1e7414ed7ca431a885552d87293f99 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..d74d7310196f156a58f61ee1613a624b7cfe1e19 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1d60115ea2f93eaab14e18aef809cda8e2b5577c2b1439a132fe40b1e646cf1 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..b0edcc182b3590f61f02f0e0f02d3dfeebfe7423 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:004ed4065087f2c7b488b6e7d0fccb00fbe67a95f6aa9ae91425f28f44590a03 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..2acd500fcce2df5d04d192bf4ee2b80fbfbfb41c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc7344b715e085c0a1e92962c07ff47cf7f32f9ded53a052066074f82868921a +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..6425ff19a77d953aca0eb14d1c61e8dea85cd027 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_20/actor/optim_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12a9307c3144b943373b1fe24a61a75898dbde7a980550e05593303ccd8a648f +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..a16198e6e1dee8a857af2b34bb422a4cf530ec07 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d1c2169b53a493bd7987e91a14f5113fbc37790d48473196d2f0be9a980d99a +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..7dccfe0fcece942f8a9afba737038d40618d90cd --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ce404f56affec9ce56a9c6524d9904072ea1eb3cfd8d73513b2dab969f1a0f8 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..0a9b5d8e97cbc5fb498353a36d0ab84348d42d18 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc69e21cf4adf7e40f095970dd376ef297fe849b7d8f121aa4db005f30e30ae0 +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..785b4c7133ab5e9e91d5895b0296cb2024a6a264 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:22891e3677cef987c738ddba431ce5a9b100c614322e52c6c499145c45c6f0ed +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..872616fc65614822e26cf507ec8fbf9c6c748bde --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88fa0d7f15c83dfe10b9d86cce69f1ad4e3e074423f929a5626be42a8df7680e +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..e80f1a1351d95d4496e182af0936069a124398f7 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9f81c93bcd82c6a4b25a8ee34d26a3e192b4b62044e2a6440632e16e3e8b11c +size 14644 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..dd7890dd615f798d33c0d2e82219c86d567e09f9 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1149355cc86894787cb4b7eee539af25351392b3126005475f69b776ebd3a7cb +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..6357f9cc9485e45af518eab99a68bbd06edb0311 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/extra_state_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d94e9f46093e2e876ffd341ca7db80e5a89795be58f1520528efeed5ea649940 +size 14638 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..4519f1b67ea7dd0030e953dc3b7407035a7abdc8 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9aaa2bc4a2da9eb9121f8edde5cdee4c2c6dd3d3b819dbea6dfd16bb7f2746c +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..ca933e823137a91e43c5b4e62e30f7bdaed71d46 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e46f1ce9d5e9d70f47ce91daf41ad761ee7effa6ffde190bcc0be775555214a4 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..b2ce1a5c40f13198310c088c8ba85482690234c3 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:87395f23a6ced1285601943641655bea688082d25491c4c110ab538fb228d7c0 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..4d3766c3d596c52ba6061201b57b8c4c38222177 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea21f55c91a6a815f92d7c42fb45bf45aa197d48c62a75f95dcabbb4adc41c58 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..6e0674105f6809a7d2ad6b02b51b0dea4df0eb3d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b64c39882fd0b1f48df096891fb9bbbe40686616cd70d84dc67b8bca365a4cb5 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..e613d8b99b5dbc87d98fce6981992f7bdcf25107 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db1d9714339888eeafef2bdf35e9d82ef64886694bec1abea6875c2d591a3d93 +size 1904094442 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..33b3fdf0d8c1d60bc38a9092b825902c907e6d85 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:089c3e52d908037dc8a5d4d4140068ef91b4cbcc29dba8cdc6c156145f637f2b +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..68cc1a7d9343ad74b88ec3f0d095d98d2f47a0a9 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/model_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:051780d967c647a137b3a9e237b0e83d903d90c870c82105ba3c51ff18987b21 +size 1904094098 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_10.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_10.pt new file mode 100644 index 0000000000000000000000000000000000000000..d55d9c91bff590bfe8d3715c8b41733947fc90b5 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_10.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7fabcbb5433af03cd4c4801878a4b5b9f2ac779fc793a51a54f02cce9bd68ea +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_11.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_11.pt new file mode 100644 index 0000000000000000000000000000000000000000..2ddaae617efc984e105f275246b3d443c75c2bd8 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_11.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89a1816611584f01ac17e15b68b81c30960f60da5ff00b90cfc63aa122d6d2fe +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_12.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_12.pt new file mode 100644 index 0000000000000000000000000000000000000000..753f2578a6e588a05b50af9a8051b693dc5c538d --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_12.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:187dde7bfa64ac68fe61f4d693990106646a4bc845f72854b0189f21be38770b +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_13.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_13.pt new file mode 100644 index 0000000000000000000000000000000000000000..49b16d73170d8c6071b5a87f170363c120ff7a7c --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_13.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2edfde2b2b6f5bf401029ed06f9752ce0682462b75c3b24efbe1e5133813f42 +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_14.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_14.pt new file mode 100644 index 0000000000000000000000000000000000000000..89c5b930200c7b661abc4ba6306f61a34d247bed --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_14.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e42b726befcb195afc9b15fada9c9bf3a2134af2e0ce62a114751ec2cc9888ff +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_15.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_15.pt new file mode 100644 index 0000000000000000000000000000000000000000..54068a4028d3290ab73b8f67cfc69ba2278b5896 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_15.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:abc04e77e585ad408d3b0b0bb2f43acbd474cf2da76b0abc3e534bc20c7957fb +size 3807834605 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_8.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_8.pt new file mode 100644 index 0000000000000000000000000000000000000000..4eb1e90ff2bdde65274328b3d9fe6e23a6f51746 --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_8.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9fc818d5993c2babd9f9101ec4b6dcffc3c88125619e49080a5b76151239564 +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_9.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_9.pt new file mode 100644 index 0000000000000000000000000000000000000000..00e1fb38ea0fe440d5c9691ac3f89fed76a26e7f --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/actor/optim_world_size_16_rank_9.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cec405c454a85083c1b00d907b492f8cb5f218e33bb615847bf26b2691779d1f +size 3807834514 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/data.pt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..0e9c5a80416fb554d9322777c1792d138ddd5d0b --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/global_step_5/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9c678af306e14b6c9da1e6983d05c193700eb8814b4dee2994633a1be76e280 +size 6948429 diff --git a/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/latest_checkpointed_iteration.txt b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/latest_checkpointed_iteration.txt new file mode 100644 index 0000000000000000000000000000000000000000..3f10ffe7a4c473619c926cfb1e8d95e726e5a0ec --- /dev/null +++ b/3_Qwen_Qwen2.5-Math-7B_batch1024_rollout8_klcoef0.0001_entcoef0.001_simplelr_math_35/latest_checkpointed_iteration.txt @@ -0,0 +1 @@ +15 \ No newline at end of file