Training in progress, epoch 2, checkpoint

last-checkpoint/global_step592/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f823e2ca6867f0940b154d733ea57d99d76abb2f7c148f7c9956125cf1db0c7
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6660fc4586008310e185ba684465ed436f87337a26346c4efe2ef5094ef92679
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2c408b3ba0f63af28599944f0a6805cc6440b87170357a9d434098400ee9fda
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:690784657770564d591a9e5cb6e8ad35f3d7002af2b64e7078946e011258c894
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57c969dedb50d82f909f47521e331aceb85d776327de5dbf6d0956e1654b6572
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b882316e4281fbfda9d0ff5ce250d5cfe326358a677748454a48c976a583f54
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c74f2aa7f7695d9acb0d57b88649005d9dfcad6469d4863482a5d70baf77a89f
+size 2506176112

last-checkpoint/global_step592/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:345dc3266b64e1fca3db8d85fc99261e6e7ea124325441fc3396496f21da2b34
+size 2506176112

last-checkpoint/global_step592/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca2f7161f6f6bf481448018527d783d2f436082ef4838d583fc47ac8f0528106
+size 85570

last-checkpoint/global_step592/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35fde5b74e6a068a2abe279701a3e030c55dcd90262e0c83c5d20aec0a8a9d75
+size 85506

last-checkpoint/global_step592/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17b4a45cb2859c5d01326a75bba0eb6fd5b2783c2d40f2a1947b3199a1ad0a87
+size 85506

last-checkpoint/global_step592/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a806163c623783356ab0581d3018f18e7a665a04338cb79f796e56bfb939e4c
+size 85506

last-checkpoint/global_step592/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dcc133f5a1f48b83e527c5defdb690efab3d250dd49745885d1661ef4107161
+size 85506

last-checkpoint/global_step592/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71bc5650e5833833667a713cf4f0e0ffdfe57a6c81ba6319dab9e75d1f41f561
+size 85506

last-checkpoint/global_step592/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c06c1cc811682c0f45f6e7a105fd5c90b6b40aa3d17f99cfcd767ecba5c10516
+size 85506

last-checkpoint/global_step592/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e55c0d10b78a967ec7a240d011de40558f2e5146617067c2847d5d830ccb6077
+size 85506

last-checkpoint/latest

@@ -1 +1 @@
-global_step395
+global_step592

last-checkpoint/model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:731d149712afa98318e30f48083df4b1e2a718b770eb7e709dd7592a1503e28f
+oid sha256:c3134047f2f0f5b6606ba7f39ebac7011215d2546becd91f868f44e1a3a2a36f
 size 4945242264

last-checkpoint/model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ccbe171405888a5d7edb0b1d77178ea2e7f377564c0c2a9d150a60457fc4e88
+oid sha256:a3d94a5b51729f4e9144b68f58645cf9e4208ff217517d5c899baafc7b82b472
 size 67121608

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad8a35afd8967cbb748405387e44426e43ad127028e826eddc9b67d2ca873c85
+oid sha256:08282b46825aa78d10fe10e3fea89555c5b5a691b261a3ddfd58fcb58370edff
 size 15984

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f338ce80d7c441076bfc8c53b84067a0181f5a14e80c13d5acb8150b659f4d73
+oid sha256:dbab71d98a3a9a92df82a6bba463947327c3a1bcf35cd9f4f46114641fc42dd9
 size 15984

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9fbc9fa428939be10b46779f0eb5cd833e0da426b1cbdee77b3a55b6952235b
+oid sha256:caac82d57d878d30219a4f9ec289a97ff90c53afc160b968f251b3fd3454b8d8
 size 15984

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac55dba0b79d5fa4699d239da2f966d52040d576d31234ac8d4632e6956481bc
+oid sha256:19762d2d370222b01817da11bbaa6665d542293373186d66f754e7246bb861ed
 size 15984

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af2d0c015100768ffa23faf3b6c2d54ea89eb045603e30e55cd211e06ff34972
+oid sha256:00c7508b346a7d3c5c23392845f1d013331114ade778794b76e919cb3ed5d33e
 size 15984

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c60a1b40608e34bc801c8231f97b81c53b5290dfaed1b9cd0ccbeca29574a991
+oid sha256:b89de7d14dd20a191f56b74c816ef8b7fe5c171e31efbeadbf321c4539ed68c3
 size 15984

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ad6a142a403eb9aafc4a3a9a856bca648fe31fd22d796867baca31fb13656aa
+oid sha256:1c71152053553e6e22d670fbc4fd7550bf8a046b54cad7b71869787986a6a42c
 size 15984

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38bc23a138cc800b22881742c0f3f9a71731a9a7111c6058a0077e6274d21773
+oid sha256:4b67db12a26a26ffe03d9afc84a43857eb2e5b2fec2dd189653b415f74208190
 size 15984

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4d84901e6619ac64838b9206ce48d7b495f836174ff70c8bf3fe0cbb24ca9e1
+oid sha256:83e8d6cb171d4563e280d4ca4761e24d8366d2865f6f1dda2bd2597b587d30c2
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
+  "epoch": 2.9974683544303797,
   "eval_steps": 100,
-  "global_step": 395,
+  "global_step": 592,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -607,6 +607,306 @@
       "rewards/margins": 0.011001082137227058,
       "rewards/rejected": -0.007340874522924423,
       "step": 390
+    },
+    {
+      "epoch": 2.0253164556962027,
+      "grad_norm": 767339.6192091529,
+      "learning_rate": 4.929489188342212e-07,
+      "logits/chosen": -12.052891731262207,
+      "logits/rejected": -11.94625473022461,
+      "logps/chosen": -225.0377197265625,
+      "logps/rejected": -243.81039428710938,
+      "loss": 119737.85,
+      "rewards/accuracies": 0.7749999761581421,
+      "rewards/chosen": 0.006956162396818399,
+      "rewards/margins": 0.01727995090186596,
+      "rewards/rejected": -0.010323788039386272,
+      "step": 400
+    },
+    {
+      "epoch": 2.0759493670886076,
+      "grad_norm": 936793.207320047,
+      "learning_rate": 4.913820119084926e-07,
+      "logits/chosen": -11.38767147064209,
+      "logits/rejected": -11.339715957641602,
+      "logps/chosen": -219.8796844482422,
+      "logps/rejected": -252.80581665039062,
+      "loss": 114021.05,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/chosen": 0.008199459873139858,
+      "rewards/margins": 0.031510110944509506,
+      "rewards/rejected": -0.023310650140047073,
+      "step": 410
+    },
+    {
+      "epoch": 2.1265822784810124,
+      "grad_norm": 1035986.8564166825,
+      "learning_rate": 4.89815104982764e-07,
+      "logits/chosen": -10.819408416748047,
+      "logits/rejected": -10.774351119995117,
+      "logps/chosen": -231.78854370117188,
+      "logps/rejected": -260.20355224609375,
+      "loss": 116051.6,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.0056950985454022884,
+      "rewards/margins": 0.027868490666151047,
+      "rewards/rejected": -0.02217339165508747,
+      "step": 420
+    },
+    {
+      "epoch": 2.1772151898734178,
+      "grad_norm": 1036991.7861177241,
+      "learning_rate": 4.882481980570354e-07,
+      "logits/chosen": -10.84526252746582,
+      "logits/rejected": -10.708145141601562,
+      "logps/chosen": -221.5430908203125,
+      "logps/rejected": -257.36114501953125,
+      "loss": 113501.175,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.005057200789451599,
+      "rewards/margins": 0.038923002779483795,
+      "rewards/rejected": -0.033865805715322495,
+      "step": 430
+    },
+    {
+      "epoch": 2.2278481012658227,
+      "grad_norm": 1227488.243303788,
+      "learning_rate": 4.866812911313068e-07,
+      "logits/chosen": -10.5010986328125,
+      "logits/rejected": -10.63232135772705,
+      "logps/chosen": -233.42373657226562,
+      "logps/rejected": -276.0982666015625,
+      "loss": 112100.4,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.004779786802828312,
+      "rewards/margins": 0.040522992610931396,
+      "rewards/rejected": -0.03574320673942566,
+      "step": 440
+    },
+    {
+      "epoch": 2.278481012658228,
+      "grad_norm": 1079397.6974786038,
+      "learning_rate": 4.851143842055782e-07,
+      "logits/chosen": -10.104026794433594,
+      "logits/rejected": -10.142271995544434,
+      "logps/chosen": -216.66940307617188,
+      "logps/rejected": -258.98858642578125,
+      "loss": 112483.4,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/chosen": 0.0053299954161047935,
+      "rewards/margins": 0.03484385460615158,
+      "rewards/rejected": -0.029513856396079063,
+      "step": 450
+    },
+    {
+      "epoch": 2.329113924050633,
+      "grad_norm": 1367054.8438774655,
+      "learning_rate": 4.835474772798496e-07,
+      "logits/chosen": -10.148681640625,
+      "logits/rejected": -10.183786392211914,
+      "logps/chosen": -233.730224609375,
+      "logps/rejected": -278.64349365234375,
+      "loss": 111561.6625,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.011530257761478424,
+      "rewards/margins": 0.04578756168484688,
+      "rewards/rejected": -0.034257303923368454,
+      "step": 460
+    },
+    {
+      "epoch": 2.379746835443038,
+      "grad_norm": 1298484.9349088285,
+      "learning_rate": 4.819805703541209e-07,
+      "logits/chosen": -10.018949508666992,
+      "logits/rejected": -10.097805976867676,
+      "logps/chosen": -224.6026153564453,
+      "logps/rejected": -270.0591735839844,
+      "loss": 112710.1875,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.011801879853010178,
+      "rewards/margins": 0.040784891694784164,
+      "rewards/rejected": -0.028983011841773987,
+      "step": 470
+    },
+    {
+      "epoch": 2.430379746835443,
+      "grad_norm": 1428524.6930006845,
+      "learning_rate": 4.804136634283923e-07,
+      "logits/chosen": -9.595979690551758,
+      "logits/rejected": -9.634994506835938,
+      "logps/chosen": -265.3009338378906,
+      "logps/rejected": -315.98541259765625,
+      "loss": 110031.3,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.0027348275762051344,
+      "rewards/margins": 0.05151837319135666,
+      "rewards/rejected": -0.048783544450998306,
+      "step": 480
+    },
+    {
+      "epoch": 2.481012658227848,
+      "grad_norm": 1467649.8441612076,
+      "learning_rate": 4.788467565026637e-07,
+      "logits/chosen": -8.871723175048828,
+      "logits/rejected": -8.764354705810547,
+      "logps/chosen": -203.2312774658203,
+      "logps/rejected": -241.612548828125,
+      "loss": 110534.325,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.01302252896130085,
+      "rewards/margins": 0.03907207027077675,
+      "rewards/rejected": -0.02604953944683075,
+      "step": 490
+    },
+    {
+      "epoch": 2.5316455696202533,
+      "grad_norm": 1382959.9591988046,
+      "learning_rate": 4.772798495769351e-07,
+      "logits/chosen": -8.468270301818848,
+      "logits/rejected": -8.384966850280762,
+      "logps/chosen": -226.46237182617188,
+      "logps/rejected": -269.6461181640625,
+      "loss": 110480.175,
+      "rewards/accuracies": 0.824999988079071,
+      "rewards/chosen": 0.015191495418548584,
+      "rewards/margins": 0.0456535741686821,
+      "rewards/rejected": -0.030462080612778664,
+      "step": 500
+    },
+    {
+      "epoch": 2.5822784810126582,
+      "grad_norm": 1369494.2190603705,
+      "learning_rate": 4.757129426512065e-07,
+      "logits/chosen": -8.634099006652832,
+      "logits/rejected": -8.640868186950684,
+      "logps/chosen": -232.20022583007812,
+      "logps/rejected": -304.80352783203125,
+      "loss": 109921.975,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/chosen": 0.013228721916675568,
+      "rewards/margins": 0.07378505170345306,
+      "rewards/rejected": -0.060556329786777496,
+      "step": 510
+    },
+    {
+      "epoch": 2.632911392405063,
+      "grad_norm": 1750255.0550240122,
+      "learning_rate": 4.741460357254779e-07,
+      "logits/chosen": -7.8379316329956055,
+      "logits/rejected": -7.4784440994262695,
+      "logps/chosen": -213.3401641845703,
+      "logps/rejected": -258.43743896484375,
+      "loss": 111730.3875,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.016342563554644585,
+      "rewards/margins": 0.048144370317459106,
+      "rewards/rejected": -0.03180180490016937,
+      "step": 520
+    },
+    {
+      "epoch": 2.6835443037974684,
+      "grad_norm": 1447093.2174814222,
+      "learning_rate": 4.7257912879974927e-07,
+      "logits/chosen": -8.354089736938477,
+      "logits/rejected": -7.889782905578613,
+      "logps/chosen": -225.5243682861328,
+      "logps/rejected": -276.7877502441406,
+      "loss": 109226.9625,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/chosen": 0.013667022809386253,
+      "rewards/margins": 0.05627403408288956,
+      "rewards/rejected": -0.042607005685567856,
+      "step": 530
+    },
+    {
+      "epoch": 2.7341772151898733,
+      "grad_norm": 1477083.7533012358,
+      "learning_rate": 4.710122218740207e-07,
+      "logits/chosen": -7.921019077301025,
+      "logits/rejected": -7.979846000671387,
+      "logps/chosen": -237.23715209960938,
+      "logps/rejected": -285.4289855957031,
+      "loss": 109592.125,
+      "rewards/accuracies": 0.9125000238418579,
+      "rewards/chosen": 0.014952963218092918,
+      "rewards/margins": 0.055934417992830276,
+      "rewards/rejected": -0.040981464087963104,
+      "step": 540
+    },
+    {
+      "epoch": 2.7848101265822782,
+      "grad_norm": 1486366.6324330876,
+      "learning_rate": 4.6944531494829204e-07,
+      "logits/chosen": -7.12634801864624,
+      "logits/rejected": -7.396058082580566,
+      "logps/chosen": -226.1304168701172,
+      "logps/rejected": -276.8672790527344,
+      "loss": 108245.925,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.00994019117206335,
+      "rewards/margins": 0.0537477545440197,
+      "rewards/rejected": -0.04380756989121437,
+      "step": 550
+    },
+    {
+      "epoch": 2.8354430379746836,
+      "grad_norm": 1560304.698196799,
+      "learning_rate": 4.6787840802256345e-07,
+      "logits/chosen": -7.268878936767578,
+      "logits/rejected": -7.414219856262207,
+      "logps/chosen": -215.24661254882812,
+      "logps/rejected": -276.79437255859375,
+      "loss": 110187.5125,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.016926631331443787,
+      "rewards/margins": 0.05572710186243057,
+      "rewards/rejected": -0.03880046680569649,
+      "step": 560
+    },
+    {
+      "epoch": 2.8860759493670884,
+      "grad_norm": 1647695.8714812996,
+      "learning_rate": 4.663115010968348e-07,
+      "logits/chosen": -8.584083557128906,
+      "logits/rejected": -8.43793773651123,
+      "logps/chosen": -239.3496856689453,
+      "logps/rejected": -301.948974609375,
+      "loss": 108493.15,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.007640582975000143,
+      "rewards/margins": 0.06335236132144928,
+      "rewards/rejected": -0.0557117760181427,
+      "step": 570
+    },
+    {
+      "epoch": 2.9367088607594938,
+      "grad_norm": 1523200.3846012072,
+      "learning_rate": 4.647445941711062e-07,
+      "logits/chosen": -8.875934600830078,
+      "logits/rejected": -8.860316276550293,
+      "logps/chosen": -234.2982635498047,
+      "logps/rejected": -293.39727783203125,
+      "loss": 107204.65,
+      "rewards/accuracies": 0.9375,
+      "rewards/chosen": 0.0077833631075918674,
+      "rewards/margins": 0.061719853430986404,
+      "rewards/rejected": -0.05393648147583008,
+      "step": 580
+    },
+    {
+      "epoch": 2.9873417721518987,
+      "grad_norm": 1605115.356703113,
+      "learning_rate": 4.631776872453776e-07,
+      "logits/chosen": -8.788633346557617,
+      "logits/rejected": -8.637460708618164,
+      "logps/chosen": -257.7025146484375,
+      "logps/rejected": -303.82147216796875,
+      "loss": 108959.225,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.0057020229287445545,
+      "rewards/margins": 0.053022872656583786,
+      "rewards/rejected": -0.04732084274291992,
+      "step": 590
     }
   ],
   "logging_steps": 10,