diff --git a/.gitattributes b/.gitattributes index f32f1e39a17d3192d92b7b0b0e99c70322291da0..8910a0ba4c5eb6391dcfb7c784cfac69db9d4bb8 100644 --- a/.gitattributes +++ b/.gitattributes @@ -68,3 +68,7 @@ pilecc_cont1-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rouge5-temperature-1.0/c pilecc_cont1-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rouge5-temperature-1.0/actor/global_step_300/tokenizer.json filter=lfs diff=lfs merge=lfs -text mathhard2-mutualposclip0.0-qwen3-1.7b-oldreward-4k-corrupt-0.2-step-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text mathhard2-mutualposclip0.2-qwen3-1.7b-oldreward-4k-corrupt-0.2-step-250/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathhard2-grpo-qwen3-1.7b-4k-step-50/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/mathhard2-grpo-qwen3-1.7b-4k-step-50/model.safetensors b/mathhard2-grpo-qwen3-1.7b-4k-step-50/model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..70fb73592ce919ee8c13679ce53b2b5c7af538c4 --- /dev/null +++ b/mathhard2-grpo-qwen3-1.7b-4k-step-50/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:201d2bec3dc923f60caaec94fe334383e895f8c5ab011c3bf88090ca05d6dceb +size 4063515640 diff --git a/mathhard2-grpo-qwen3-1.7b-4k-step-50/tokenizer.json b/mathhard2-grpo-qwen3-1.7b-4k-step-50/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..cd71f61a15a522601badb3dc960d800d9cb3766c --- /dev/null +++ b/mathhard2-grpo-qwen3-1.7b-4k-step-50/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4 +size 11422654 diff --git a/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00001-of-00002.safetensors b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4d22f1e4e315fbc967b1832a2a3f32cfefd2c8c3 --- /dev/null +++ b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bdd113738213bdd1c9618a4add96abc3fe9023b650239eea0e73331874de541 +size 4969539560 diff --git a/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00002-of-00002.safetensors b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b98f8c6b23ce69e9905688de8df9398a9a39f540 --- /dev/null +++ b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cee00cb3293986e460fdd18604d1bc37a9e34a83097ecf531563895688f95db +size 3157455608 diff --git a/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/tokenizer.json b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..cd71f61a15a522601badb3dc960d800d9cb3766c --- /dev/null +++ b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/huggingface/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4 +size 11422654 diff --git a/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/model_world_size_8_rank_0.pt b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/model_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..bec26e095b145f070e55e6fe63860e5c0b13a437 --- /dev/null +++ b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/model_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:661b7c6ed04bab3b10756192e11ca61dd600b126480212ec410f36513142d342 +size 1016051898 diff --git a/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/optim_world_size_8_rank_0.pt b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/optim_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..8dbc4e92e4dc003badf77e42ece44fb4c8e326b7 --- /dev/null +++ b/mathmedium2-grpo-qwen3-1.7b-4k/global_step_300/actor/optim_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64f62437460a18fae7c22d276ce9994a636635e28e50b0e5b16b2ae72c73fa43 +size 1720600311 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..a28af77cad975fd4314cd8f3a9161e2b1c1c8c07 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de3ee5de802e18a6c464c668cbdea573f3a0762277f1da014fc4de56061b19ae +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_1.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..258c0ae6d4e8fba5428310ad48eae0b44db0fa10 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:919a2fa7656698779da0467195200081bea59e9188d28ba480928d2c640b236c +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_2.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..bbae616e3272108a9d16ba4a0faf8364a7fd1e6a --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba3af5db6866f4159722e3292cdd17a0eaf592558b03b04d5b8d1a4519eb9c33 +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_3.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..f823033c7fb391dd83658e199f03f600ca5f5513 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af942cccf3e655acb5044223ed1e67480b3ea44abc0dcec8e766a10f10bf0e4a +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_4.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..71c959990762c4095b2f013313bf1cdf39dbf2b1 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:542616d296554beeb26e2c0e3ca00bf39476ab873ce590bd527f6259c67eddd5 +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..445ff6c64e8527969a073f3f5e205d64406d3e15 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb4e97adb81cdd7a88915021119adbca638cb20fb5354c3b7e38da09ebe3011f +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_6.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..50e071d9fd89572f302cfdc7e7fed0bdf76afeb4 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5142697a3fac0fad6a9ce7a2385c9ae7643c8c5fbb7d9dc767c93589cf184b99 +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..36233988c5c94b4bf083d30f483ef129599e8bc7 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77a6569381d581fbc1f645ce86beb08209de562214cf5167c2a7f864edd16b7f +size 14696 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..29a60b8f9e8a4dbaeedbcfffd281a0f8696a6881 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c7edb3bd529ed3cc91699fc1911492ee661cfc8d403c2ee5b34aac7cfe9f9e2 +size 4998767360 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6c1b9088ba15c18b3e7aa4d86356aeeea6507140 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99292ee13e5ce09eb8c68625fbeeb0aceef1e4e099be808c37ca85a9df71a26d +size 4932808968 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9dd9600be0371ba2b9277689e7890d2ab7787d9f --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba8ca9cf9a3aac680cd265ee5db91e8123ef3c5269c701e552969894f4cd12c5 +size 4495462048 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/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-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_0.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..c87604751c8f4f10f3b53e5ec90c8997e6c0d8a7 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:776e3b83eec99fc34e9ffa051db20ea8cf43e81f2fcdf60b071780ee7b6ea0f9 +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_1.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..84cd3fea6edfc7b2e016960397e40c57185b30e5 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb80609cd4649a51fd6abbb9b33342554c3ccdb62fd3a288f136e199956d4358 +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..db83dcbbbd6129646457ea9d59bf2d68741337bf --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9164126dcdddd1f12db8c276306f51ac50f6f817f032c39da72218227bad712b +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_3.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..c3dd623f745da54c9035dc6ebfd65164b34e973c --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76266f3980106a3b4c8a0f979a9f05279c609e8fd71d77943133d2b8ddc5162a +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_4.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..71e0ff78c47009e50f4f50d1f6ed146c73325a9a --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe94a244ad4b96e7520bd9179730874387554c5a089aee7279ff5ff86d68e59c +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..4445a57e2121216e3d24cdd9e9d2397705cc8ba1 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff3e07897a2d7b441470e128e692e600836c4711acfba4e641ee827ae4fa58ad +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_6.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..e53ced9eabd8659caccaceb0c15ac87051eccf34 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85e2d59e00a2357a44806d289bf6735e93325daebb3eb5e83cede0a5f49cf09d +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_7.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..d2c95403e70a22bdf28c0aa824cac48931e30b11 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9ad4377e255291ba3e2421547ac1570aa4005ca155fbf636cdbfcc3d77fe8ea +size 1803528186 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_0.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..ac5624b3f3325ee9ee527977974ec47426a203df --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:df773d6ac160dc515224ad0a6a187ab2d5724912c2b9b1aad255d66070128c75 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_1.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..845b99b96fa4b913eebb99ee1c508cfd25cddca1 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e794d73c57ed52a05e41c8eb029e2cac1cca5573ff6fd4f816a17ab9b3320c61 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_2.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..3005a176b20871a73f1b682b57d237cbfa8621db --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f713aa48f6c41790255bff1f68c4e9cad1b6d2138a21fb90c5f826b16de7bf38 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_3.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..6f57e19b7d72381e7eb9fc8ac5cdfc6c0f3fbdae --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:379b52b90e6987680e507caeb6c55ebd5d38e7505956ab374e7a70cd5dfd38f6 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_4.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..a1ccbf3f207dd293b8edeee2d6dd7eec8862d449 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd818ec95ce43bda97e4caa6787b22af6ac0a7e9b1ae36d213cc9188ac462af1 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_5.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..310f25148e1457f501136edeb58106abd84ef58f --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a887a315dc209983280795925b5bae6891e2d7855c15192e8713f59d87d1c44 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_6.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..2a12277c0052e6f5f842210e0b4e20a9d0e16749 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:275296354734ab1b1e62950747a7fbb48780fb7f82af4c6f0922df72f7675dc3 +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_7.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..47bbb4cc7fc55d7646ab65e99888f8371e45e135 --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81e31b687c6e1ee4c73ab3bff001886f37a987e30cdea71d85aa52d89f5bd62c +size 3212775159 diff --git a/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/data.pt b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..266b8fab40f43ba2367d43cccdd4191fd972156b --- /dev/null +++ b/mathmedium2-mutualposclip-0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e3304d4f734e7f6ea306229510c87cdf58c788bd687887718acec32e8b6e4303 +size 1492 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..c3d79f7cd4a311ff5edc797014b42cd55ede052a --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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 sha256:3cc4ef2e6ee8f7092508c4792b5bd4090cea7760a7568a0aaf7c5ca806aa1bd4 +size 14696 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..79157d19a0339c72e15160154a7d215123b91886 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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 sha256:8e6765e06801fe60b69e69b6327e50f0d3c492ecb2ba9630507afc8dff28ba7a +size 14632 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..88c03f7990b99334047390c7418503765b8982f6 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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:ea0d0064f02965b08cde553ccc3bed62e6a14f4e1fe80221569bf55b78184b1a +size 1803528186 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..a2202c30de800424d7a5239beb7ec2e656ab09b4 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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:aea4272ed9b7cb127f75ebfed022e62247270fdbff1bfe2a837406ef70a0bb09 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/optim_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/optim_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..d8aa88a4c5cc7287b84e3f5b6d5456d72ad7a1c5 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/optim_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:651861073270d2d71e6e1e58b61abbd44a96ba42fa1e05c289581cc4c3d51859 +size 3212775159 diff --git a/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/data.pt b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..06a54b3049d008ab5cfe1c4be9006d4a5fc4bb8b --- /dev/null +++ b/mathmedium2-mutualposclip0.0-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06a0f167e052fb597b6af95e4ecfebbe88f35785da9c87cc7f29d19611a690a6 +size 1492 diff --git a/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0aef050726b76ac901ab0b1dbebd13d89e5d54a4 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1af812c5ddca6be843496c3b89acff79eaecfdee98e1fdbf01bccc4f22f28af2 +size 4982131536 diff --git a/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e6397ce210b9ede0029209f4926d967bbff7f27f --- /dev/null +++ b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:831f7f81c6ca14bd93c1a2f36c818cf20f0989a0ac71463e79840f14d62f86a2 +size 4932949336 diff --git a/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..587d3fe10cbbf1099d742c4de5dfd5459086bad8 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:003c2b9476d9c8fdf38c6a797d883a7e774bb0b082bea8627e33723435952089 +size 3673383040 diff --git a/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/tokenizer.json b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..51ebb3ba93988c73bbd83a3c163a14343fbc5b06 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/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-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl new file mode 100644 index 0000000000000000000000000000000000000000..1a0ccfe23fc8a6b37efeb306ccb00eb176b0c4d5 --- /dev/null +++ b/mathmedium2-mutualposclip0.0-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e913a22d2ca9468b044b25ecc554f48e1d83cb32a3c18a1cb17c449ffeec734 +size 12846 diff --git a/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..780c4e3a975daa25463a6816da9d08a60053688a --- /dev/null +++ b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bae2eb5bb3bfef872fba0056668901b27728e72a33f9b98e8690609f103bcaf5 +size 4982131536 diff --git a/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..eef7c80ae70d3b512e4bae1fdb9b032c86e68247 --- /dev/null +++ b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/actor/global_step_100/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18b1d3d07f0955603ea2dc9f8700b583a634fc4103a39273a732cddc8408abcd +size 4932949336 diff --git a/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl new file mode 100644 index 0000000000000000000000000000000000000000..51c14f7a717e58252dae39a0113534e6a411ebd4 --- /dev/null +++ b/mathmedium2-mutualposclip0.2-qwen2.5-3b-oldreward-corrupt-0.1/global_step_100/random_state.pkl @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:47ec7cd51818a3a1fe43c1160bfb0f3269d27d42b42be7c72e0def87128affe1 +size 12846