diff --git a/.gitattributes b/.gitattributes index caa358f08503468f2415ec1949ad197b44df80d5..f5429f6aa4ccbefb9f905d1f2ffa3f96e75093c1 100644 --- a/.gitattributes +++ b/.gitattributes @@ -53,3 +53,5 @@ pilecc_cont1-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rougeL-temperature-1.0/a pilecc_cont2-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rougeL-temperature-1.0/critic/global_step_300/tokenizer.json filter=lfs diff=lfs merge=lfs -text pilecc_cont2-r1-ppo-llama3.2-3b-it-em-warmup-0.05-rouge-rougeL-temperature-1.0/actor/global_step_300/tokenizer.json filter=lfs diff=lfs merge=lfs -text mathhard2-grpo-qwen3-1.7b-4k-step-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text +mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..04b907a44249bd2ea3b3b844b1b59a98745e86a8 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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:e2d30f21600212ee83b2b70f1d53d4ea0329b308e968c5e3d474ae168fe1b110 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-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..38f26a4e17c98299cf2512c89dd191a856cbb891 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:44fd864055554b8151147e4facc8641dcf80fb074b24d19ca104e73caebe9fb5 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..5d92057434531cca16555f73d86ab996be400b91 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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:f83cd7bd00e11a1570bf3c218ba1dc8a80bc3354b6021cfd12ac31c67a0e9d30 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..360cec1030abb0574d45b9bb01735bc7fba62cf2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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 sha256:26beb4354c2df0103478e3a66abe8096dd6d252e9507c6239c13aa79f16e3590 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..0b6032a4d26992d63233e1b3d9813465946e91ce --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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 sha256:1c21bf59d7e36ec429873a36769cfd635e724bdf6ada47e900fc8f215e429d98 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-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..1d15c0ac2f739c962554e2a1f027836c8eb27588 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:b74178c31413293c2787b7586846d0e699ab2d5311f4a99dae9889167afe6fa2 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/extra_state_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..5535d40f73646bf591c3b833507ffd862a7a1fd7 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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 sha256:4d6e920b8ce2c3e4504f63eca2804ee9ec238847214eef1da449360843bb25f4 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..58cd3e90c0fdf22c254c0985b02f492c2982a56f --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/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:ff4faaf30e49a1294bae608c018167bc65e25c4fc3c27ca8a127e449769dfbf8 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..a689a043b3360cf52105cdecfaf2b63d43ca0048 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a462383f48edaf859b857bafa343068b9949b830e1407f4a95ee685a454a92f +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-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..fc99a6137fe5a72dbf3cc62aa7065b71d44097fd --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:f74a956c73b98aae2fac1407369868e379638e5a26ae5a060cd3fc557380fde5 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-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..7b9dfa079cd3ba2754d38c89dfdac943997a28d6 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:183b97af9173700dc463e56c6938d0cd40985a0a822d61b4498465d2342e7def +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..2cf0fdbe56b13e9095b00d3fe67afbe67ea3a393 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6139b2022b84f20e7760f2083213a3541a9191072c9b043719378ff02c49651a +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..facc99d6ae975a77f67ad2aff8515b392efd06f9 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_100/actor/model_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:830e7ec119124f707c69a25e50a878a544d86a07267f5b6979a7b500bdde9dba +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..1a8b5c79912aabd4308b163bbfa91fbe0151ce7a --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f695ec4b7f6ccfd51fc2bbb698d1a2cd19f849f7dcd39ae6b668ae1978ffeca5 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..fbae2d7d7dfa4a8342b789a782ab4af306caab9b --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97516852f8ef40fb591ba1d3c0f81bbfd7d694607bfcceea5316c5ab0c992412 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..6b09a3e8a2552cb222904da333c38447d08b3b50 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31dd1d9fd6c8046e6a82ff9131a22e4c2d618633c41c69f0b1a65296cb95b4aa +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..96f1ddf5ac4f68e526d00fa486b258a46523f976 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5a1787304f31770a2751c1ff9c74d733f0aa03ac77e8b088e881d607853ea6a +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..840341d8fcfdb47291f618b12af5966a863654c7 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f28045f00f8c3e7f4f2bcc2c6fe190b438972795d32c251ccc248cee7a248fad +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..50fd11c520c80b815757a190cf194e415a6eba1c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4c644002d6baea3749d3d8e4bcbde517fc8ad07c61357e0bc03f69f828fbc61c +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..e4f0e103a76a273063c46bf97f4bb131b7835d4f --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ef92db8692639c16094aae16f6b398ca49096f41e201ea6bc4672cbde4738f8 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..1a4d2fa68892b4068a5f04e968acf59007743036 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/extra_state_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4973d210d9e1e6f0ba7b71bf763a17571666bfbe35c15609786718d367b00768 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00001-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d27d7b54ecb2ef3c890cfff72c880a16e0ae8239 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00001-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26b8cc644be5482d71b28ca7bc1b16170268460d967c37036a8b58294c9d00a7 +size 4998767360 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00002-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1c0ebb007a3a6a65e2738552ccd926b0a2bb1256 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00002-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e13626e5780f73c5130354aabf8cace7a47e7bcb8275a8c9fa22edb6ca031e54 +size 4932808968 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00003-of-00003.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..43a7881f416a6f6ce3cd51b95605b6182665eb41 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/model-00003-of-00003.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3916e2ae5394168aa90b67e1347ea49a70043d77bd7819066c13d3f29feb9641 +size 4495462048 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/huggingface/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/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-0.2/global_step_250/actor/model_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_0.pt new file mode 100644 index 0000000000000000000000000000000000000000..e8b99a46663d0ce7126a711152fa593878694db4 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_0.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4cc00e0d7a76ecba39a8efe4bf4814059c1ed34c48f1a776e7ec7ac04c474ad6 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_1.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_1.pt new file mode 100644 index 0000000000000000000000000000000000000000..8eda4e7de2526d9b6039cdbfd5a6c969ec0dcc37 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_1.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75edb3f81a2b39707715c7c5aae37fde81d8c65955bcc3c575f8061706a6f773 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_2.pt new file mode 100644 index 0000000000000000000000000000000000000000..c554d9336634303a97273ef769747608960c3ffd --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_2.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ceaaf6d6416a6ba5eee066949513d0a16c4397299655451b776b16441a4ddc51 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_3.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_3.pt new file mode 100644 index 0000000000000000000000000000000000000000..3a36775381a97dd8aec31174edbdd5300c0ed744 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_3.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7332bc6f6b7f4060b3c145c16d962cc3db59735f880ef92f3375782f68989ddd +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_4.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_4.pt new file mode 100644 index 0000000000000000000000000000000000000000..b9190096fcdd517a5ed3732b582c36c3a8e216b2 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_4.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09e8997c6873f9a7ccaa5b1dc4e87496cb0f7840ff619c44dd4f3103a120023e +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_5.pt new file mode 100644 index 0000000000000000000000000000000000000000..dd5bccf1c5169da67114302dd22431391471139f --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_5.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9a8343ee116c72652a3d91085737e8113b87889902e40fa50d769f733d80c3c +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_6.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_6.pt new file mode 100644 index 0000000000000000000000000000000000000000..7bdd1a0bebc1acb2765a177265dda15fd2d70dc7 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_6.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ac703e48598d88676c7f08a987ae79c724c046dba140a1ee49038913712eb63 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_7.pt new file mode 100644 index 0000000000000000000000000000000000000000..a5765ca0d044205a85cc06e61baa30ec57b52ee1 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/actor/model_world_size_8_rank_7.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eb8759ef81dee3be9cf7d5a6cc12edb98adfb40f85d70f2b4d004175c3266241 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/data.pt b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/data.pt new file mode 100644 index 0000000000000000000000000000000000000000..e4f4af5314abdabb53d14ad693f1e1b123e07ca0 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_250/data.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52d7f2fd75e837024c0566961a8d1ab6f5387587358a4cd1075c3451aa882907 +size 1492 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-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..4387a8a682e56179e6d62b961eefccb116e2b173 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:ef573cf20896c8d518f977f92b6da08dee015590bb6fced1de21c0b0e33e6918 +size 14760 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-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..e593603256bb88e9053212aabdfeb5b3b25003c3 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:adadec2eb769d8dfd0cdbd8e604f83cae6639193eb850a29a35d78da380e0d31 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-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..80281a038d98f170674afd7acc4eb544b1675bfc --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:90ccc12a04c90a4e7d5c5eb6ead6e6f2817fb462f19223252514438e45bea677 +size 14632 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/extra_state_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-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..0db904e031114078d6b43501f29810ca760ba62c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:21e3daf95b2bcb0b97b965a07452731444348da186c3125749488ea52ba7a973 +size 14696 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00001-of-00003.safetensors b/mathmedium2-mutualposclip0.5-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..913ac53a1067216dae75f977d832189d742cf04f --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:d684a967c6296625c563bd732071b0ca0ca99cf0bceddbb18c18ff0479570386 +size 4998767360 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00002-of-00003.safetensors b/mathmedium2-mutualposclip0.5-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..ee30f14f71e7960bc1e37440ee318da0fdb154bc --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:d5b4348d441f5f4bedeb67174ba539eac26623b94af04e85139942d4c9a3b373 +size 4932808968 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/model-00003-of-00003.safetensors b/mathmedium2-mutualposclip0.5-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..012f4e5396631e90ab441664a1c374fbff8658e4 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:f796074753d0b5b7f53cb713d2517370f26126abf34abe6297ca078f11cf328a +size 4495462048 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/huggingface/tokenizer.json b/mathmedium2-mutualposclip0.5-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-mutualposclip0.5-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-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-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..39065296362f73862d13e5506b4acfc729f7f29b --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:7d3c9b71169d7712a36beda701ab9d26beefcce8fde7874c2c7926e0002ac7ce +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-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..570700affec50061be30954c9ccf6124fa97da30 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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:b5e90c2782d698cf3d9808acefd8a3727690f119720391c8c511407f36907bb7 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-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..24d81da2c8f143cf7f33e0df003b135f2d71f84c --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:7a41e8a73fe1ad270a5e951f07e39a6b8902777ebad4cbfc9eeee2fe204d87d0 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/model_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-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..4e1f8ae572ff51c24151f1ef9e6ca8ae686a9293 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:89f3e1dbc8572222950d3013670857cf6f383f42fc94cd63504c8c7b5222b8f2 +size 1803528186 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_0.pt b/mathmedium2-mutualposclip0.5-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..43401f12bf0ab47b1918e2d8dc772fba556c5946 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:2be6d01006217087bc20deb8c2a5e210c74dffbb38bd59e5c4c3d262e88f1891 +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_2.pt b/mathmedium2-mutualposclip0.5-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..23a6a85f364554671cf0b68b25901ea3db6f975e --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:653657bff34ccf8c13e520b4289b60ac302c774413d9577f1189bdbb225ca19e +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_5.pt b/mathmedium2-mutualposclip0.5-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..b4ea27b0e2ced61f7573d9b79b62a7c2ac6af64d --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:5557ea687582a0a8d5741968ed1de6c5581b90649a7bf10f8dd96ab25659f84f +size 3212775159 diff --git a/mathmedium2-mutualposclip0.5-llama3.2-3b-it-oldreward-4k-corrupt-0.2/global_step_300/actor/optim_world_size_8_rank_7.pt b/mathmedium2-mutualposclip0.5-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..49e6b4805793e21c2826b97e06a6a6db67f536b9 --- /dev/null +++ b/mathmedium2-mutualposclip0.5-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 sha256:2806b836ec9c5af9f9dc06e07117a1be1f81cdf599d45a9a32a191cf401396e1 +size 3212775159