diff --git a/.gitattributes b/.gitattributes index 1cd8b379dec5a48cfae8ed93751613e053b32f9a..94bc834eeee8852f65a7648efb06c2ecd502ba38 100644 --- a/.gitattributes +++ b/.gitattributes @@ -41,3 +41,7 @@ pilecc_cont2-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rouge7-temperature-1.0/c pilecc_cont2-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rouge7-temperature-1.0/actor/global_step_300/tokenizer.json filter=lfs diff=lfs merge=lfs -text mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_300/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_250/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..3f465c8835c0a301f5832e67f2b3879e80abf190 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bda746bab51495c1fd50c0a53c381ddf10d75429a0297c183f6f8c4a6ed372b9 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..39b0096c4930ff448effb7b2a70a47f4aa55dd0e --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e7276b975a6ffc96397e410e554bca17350dd3c2600a24e74b5b578efa2af6e5 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..8a4fc427441089f2f4a5e3aa14d81b67ff5412b6 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab1811d0066a64a8c1b8f3363d93b48821e0fe09e742779f068e22318fb26c9c +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..618fc8f862e0726cdaba756786932f841e356d29 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:22757ffb05725478b1c1dbeee4e78ae6f669111c6f91c613698170530fa56ce4 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..a9e966d8b35d491cdc9ba20997f8a4bc971a70db --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:59b5f9fe02c2c43f9d4a8245eccd5bc9c548dff3646929ba39d201a2d89b90b9 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..df3812d189812d1ff5bcd47cf4ee7f89f9f8a8f9 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d227687d673a6d4cddcce2eadf63e665740371ea76ed050757c0eed4293b654 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..e63793978cf16ed143ee9e3db70dd0b6ca3d42c3 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2279557e6c239e3075d6478b60f1d1800daaba9caa478b96fed8533e21c76a18 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..7480db27a3d9dc5f39c538582e8559983f105259 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/extra_state_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a1390deeccdee907893134bdab17cccbbfe5f2a3a3afebf94da6d2b2af263993 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fc65e2f815e5e098b24a21847aeb87001e981223 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ec842ae48b943f09d19bda9f63b33d9e1fdc4cdef1593ef74cde9f17ae8c28b +size 4998767360 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e9ffeae97cb1b59a27e9fb17618fbc058d9f847d --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee3ba43a8a43233c94cb213a46092b52f3dfc04e1eeda1b8a673c977d20fe62a +size 4932808968 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a8457a21070d68e839d58b664e7e20a83deef236 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa8d95a55079c5bc11fe90f3664d216db24e49b71fc9170ba5cd9a6de276f3b4 +size 4495462048 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/huggingface/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..efd21fc2b34ba81f7d044998141b6803e8995b3d --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fca5f21ad2fa64864bdd73f250f24ab488575646cfd5c22cbbd06098ebd40324 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..e370fe60ffb5eea70bade6d4987070edf6f11173 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6859aace96b3fce39f18906a82ef19185628eed81525c5b24aa0aa965bd338bf +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..c6a64bf0145061b1f6f0eae00a6325277fadf402 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2e44209deb3fabb668c0cbf17d596d244678cce68f5fe6e703e0ff6fd2123cc +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..0b79c78766c0cdd70f01f1d79cbc5cf6a3195023 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6679899f49cdb4869b4acfd4d77fb915f578bc9f1a791a7ff9a64c8528825644 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..e273b5e9977f61e40464ccc65998b40d88f395da --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d7e52926f1bacc948a26c695d5b37748a7e7eded2dce2e72b4730de774d0c22 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..917a91b7681b9b938b6d2ece0df80faa4934c342 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97432940cc3ce780040397de634f494471b2a9612a6cd116a780d3c2820abaa5 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..e9621651bc986ac25688b9da040cdfa69999b34c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b4fa93cac4b0145b65619801dc7e854517cd781d43d83e92a2549bfe59d6fcb +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..845c33245c75af2fe29f0a29cf9a053c16d9a467 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/actor/model_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:398cdadbc08509505e2622e803826723690d7d9e02d965525ff24af89c175686 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/data.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..94c99d0dd38436742b536a50ec28247e70109bc4 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_100/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b64ddf6dce72bb8a8f54bb44d9e987375e973f8d6c4d6095af9d1f64f7be30c5 +size 1492 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..551a70aa7829a68955696888dced90e35bfab8af --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09438a0f5a631f830b690bca793c0f28df2c586bc2c833f343c7e24f8088f673 +size 4998767360 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..31f0839106a175705ab7cd81c9c2e04bcad0ef83 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05c85656564025722d77e8144d4b6ad20b411cd76beeb4151896d34a8737f09b +size 4932808968 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8c6dcfa2bc61650db62eff11d1fd6f505b6f265d --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71b4996012450d861121e2bc24f1bd03b12f5414448b5d1a92ce49e35adef604 +size 4495462048 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/actor/huggingface/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/data.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..194dce230258f9c7623548382a76126bbde161ca --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_200/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e367aa4e4921b1598c0b062d8d887d4797836b6b0be8b7f6bd8bf99a39edb5c5 +size 1492 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..50d0422a2a4df0e0ff019534855476e0e218bd61 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d4fab691cdfa4d6eb38580815e64c37bf4dc5252d8310f12fbfbd0c94be2ddb +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..2f98bdb4987f6521ef0d6eabc2a4ec8ef4d9e251 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fec2b2d82bc1d805abee2e501422874d269d1f615103e9ee9423335b54ecd7f2 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..c912cf2cd2c68d15f4cc24a14a47eb95e9a84161 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3629bfd10723cbdf0b98b704ac0dcb42fec1cb705b4f832b74dc6d3bbbb21f50 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..fda256b1eaf13355214d1f03dd3616495d0b2b09 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:88eefb71372d9824891a294659b776f4d5a5dd043269ebce180818a7ee176a0f +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..563aee0540a89a75dffcce9fe13806e922a84514 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/extra_state_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4d8dcfab32b481bec0d7e30f47041adc700bcfc68f3657c9a94c0d74e78261b6 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2852453d5ea42802fb604bdce6011f612d0c0515 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7114c732b04a4a6e6adb25ced9c1f88c08b952ee4f8316158729312f980dadab +size 4998767360 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..467958658226b1bbcd09bf6f236eac9d67bcfb98 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5565261aca4dfd41cb59aa5c61101964649ae78f24b5722c18143c6d144559c +size 4932808968 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..39df7ae9c00a9e4d4a913c87819f615637a8058d --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7dc19388c83663136e8b7f925cf83eac8632718aa2a57a5ffaadff68ea99c98 +size 4495462048 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/huggingface/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..7b1eac4c6cba39b5b0484bba4697143d362614a6 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad39bb288a7a18789636d697338ef830a67907fb69973062c4b6eea463061f2e +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..cd58dc3aeec4c773f5dd7be150c85ef186704bba --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5d13dbc3220c84b43f3f53ee17727af5c3b54e87e5edf7245377176da37432c5 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..acd4e97f14508456d8b8518e8d34cfa7679608c3 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:661c51a8aad9c9eeb2505d4fdef8a1185fd2bdc8f9e7d4c56f84ce3ee4e51abb +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..a06c21dbc458124718cbb704ec2c8b5009e17d4c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5ca4eb78ea880666fb2ceba1da80601b0e933a384080f8c53b291dd0f7c81b0 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..186130bd2c1e6afa3c12e5418e168d1ab1016548 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7aede43b5d8623710b4fc252881ac0f6d7a0770873778a7db624bfe6b2ecf78c +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..3d329548086bc61fd1802ae2832255e02be0f536 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/model_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7875231710cca2abfee985b8853bb1e18cf933a1ac3777d43a1afaedfca9dbd1 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..7556f3ece93c4b99d4aa541ae640f8c968ae0a21 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a0c4b436659ca449ad94016dd718d42137f546ca0753ae794d4c6a392617979a +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..747802a4f4ec5f766393ab6bbd26774d2296cefb --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0dfb6efbf51755a996b2c92a665ed87fa673d5e989c02beeabe1b3076b48e70f +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..16e3e0765009fa1ffd5e6e1cfe04d7196de1df23 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d301e26b08b29b3205010bd4441cd5c4e1e28009c956fba634a1117d2290de6 +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..902ade185dc652c0cf520e37db6a9cac78ca209c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f58f2dce372ab11ca58cbe0fbe09a5d5e5f754f70854d7bc135a65b4efb38bf +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..7e08397f06808858ef40e7b1cfa2791467f06ac6 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfc75ccf429bc18d8348d17953c31cdc92b1951c3012c5a2b35177d6e0e80c8d +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..d810fa03d2cf23b0aeaa19582e64d31f2660db5b --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.8/global_step_301/actor/optim_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2980341f34c1da32d64c89e65b074b17402a6714634a0f8f5f8ce124a0421d82 +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/model-00001-of-00002.safetensors b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..10422e2fb67ad5f7033eb5219fd36908c6009821 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d58c7b596f3cede5b114a84a2348659decc6c20ead506be71adb29b276180679 +size 4996670464 diff --git a/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/tokenizer.json b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..51ebb3ba93988c73bbd83a3c163a14343fbc5b06 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/actor/global_step_100/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa +size 11421896 diff --git a/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/random_state.pkl b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/random_state.pkl new file mode 100644 index 0000000000000000000000000000000000000000..0bc9ff4b5f7b06ac687a7468084a7aca8733bf86 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-qwen2.5-math-1.5b-oldreward-corrupt_delete-0.8/global_step_100/random_state.pkl @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14aac75551337e17d9c0639c85dd03da73449cab6b8dcf461ab647dfd9d63a6a +size 12846