Training in progress, epoch 14, checkpoint

last-checkpoint/global_step2962/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:e3471e29a40024ecfc28a8eea3bb1eb8c173c0609f355df8a706d782fd72a6f2
+size 2506176112

last-checkpoint/global_step2962/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f3ab56eb73e7d800b1ce4430ccac77be0b6e08cde748e3316b3d44702f578fb
+size 2506176112

last-checkpoint/global_step2962/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:af0097a70a7247c363820d83118beb965b4feaa08ce4536fc3c02a548ca6b380
+size 2506176112

last-checkpoint/global_step2962/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84e7119e680ce910838814ddf94b554d606d5e7476354a860c1a8d55374d1439
+size 2506176112

last-checkpoint/global_step2962/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ea4f33951cbb414f84b9a39e35b2a8be1a9c097ad68119aacd329beb669d4db
+size 2506176112

last-checkpoint/global_step2962/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:b9e1284901d00cb10e9afe3b47732279ca766e0d6683ca3402c6f06872de5e4e
+size 2506176112

last-checkpoint/global_step2962/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:a3e426eb952650947dc7a89121f573fd79dc8d3051854c02d2171607d393ff4f
+size 2506176112

last-checkpoint/global_step2962/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ec593fe6aee5d849160b36774007bf4c2a4f123200ad6bb931cbdf26823786c
+size 2506176112

last-checkpoint/global_step2962/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8894ca4b20727ca9a3b8d9b04e6070339729765a18dda5530f295e94d25e4bf
+size 85570

last-checkpoint/global_step2962/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c9eea5b5486c31f10c150a7429b6b01fcec5184835f886ed7ac9ff9bc9676e9
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c520a6b9475cbcd8422c5021f1acb91e83f01f855c5bc86271aaadd93e8e321a
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f22b61ce07985a14d7dd5d28da2a4c67e70cc37037742abd6945b69df55ba82
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:828a5dd2c1f30bdd3116927dd889f8388e7a86fab99eb91e177b163c838854e4
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ef7d47d77deed99c61384314c2aa706dc6305ef596b77eda2482f1e5a49a2cb
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:947db6371c681606d6fe68f817cc28c7de90e30a193cafef7dee08dfe850711f
+size 85506

last-checkpoint/global_step2962/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:106a9948b226f2c26e9bd883317e2721530b2b7a29d30d49c7b252a8f554fe92
+size 85506

last-checkpoint/latest

@@ -1 +1 @@
-global_step2765
+global_step2962

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51554b1b141cb737fe919c7a7d90b054dc54778551653b359782ad641e140b5d
+oid sha256:d6ab34130d90aebe29868ecf9b47a15403e74bd3aa5e09f06dc3ea9032f8089b
 size 4945242264

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:39ccf095d82d77e6ad3b265d9fdc49a86ab3fad18daa92e729265d0f5f3bbb71
+oid sha256:3cf91018b2355dca95814934115beac2e49e42607748ab6a28986a106363bbcd
 size 67121608

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7891ffa7c7dae99113aa986d67278b52b8c57db55001dc3547a61f24569a34ee
+oid sha256:07d994b317c4df888a1a1aabc0c532e81f1fa34c18c8313cb2feadca3bb37194
 size 15984

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8b92875cb04deec367605433847d1bda444b178b643d2da7ed9aaf738d232b4
+oid sha256:4f5b05860618aa49c7f5d8c366d6ee73cf8b3b0d0adc17d9313b72621630d0aa
 size 15984

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9f5f3338a05e325b5408a1cd0b6f5e5b10fad05fe479d63f44bec4cf18107d6
+oid sha256:7262faf861e984775b4fd85bc76a11b0b8b04037690e8a08a58cf9ff5328a042
 size 15984

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1be749fea477a3867d44010631937e0d8f071ca5f9614f9795c92c7fa68833a6
+oid sha256:9479cad91150e2e266d17eb95fe678579a770f6df6b53496cf72067b186b094d
 size 15984

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbc4a5ea4532c621f4c8e9891117b2e597a7f005001e8b4f2a1b4da8c82bf964
+oid sha256:435cb6cf559e0ce3fe0d4582cac16ea40b48b7a64589952402a4c399cafbfc00
 size 15984

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:480f9fe7dd71b54d915b46162e34b780ba2467d5542115cc809dbca60b394c0e
+oid sha256:f51001b0d8dc5792180c3a9705ccbfa66b61d46d7639afb6f7abf409629ed74f
 size 15984

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c11d982dcd813e82c2d97a5491ce9624cff2dd22e8655ea617ccef1fc1474470
+oid sha256:e1e87084f11088fdce293e1fbbb05e35f5c7385b00e2f9ba195bf61cb36f757d
 size 15984

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73494fac3a001cba7cedd097b97f028d4c1d136ee6709214b0a7fe305e5b9089
+oid sha256:3d32e9bdd65145ae509e6c6ef4f6ea9d842f94a34c34a0d7d2ab6c248d3f2121
 size 15984

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:136693ea21ecf0b59fde813d184b14a037ef0bca92ae910a6f73169e6198ccb5
+oid sha256:3ef29c4eabe559fffbf188b61164c94ef6c3807ccd683770ebd49ca46d0f6823
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 14.0,
+  "epoch": 14.99746835443038,
   "eval_steps": 100,
-  "global_step": 2765,
+  "global_step": 2962,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4162,6 +4162,306 @@
       "rewards/margins": 0.5153056383132935,
       "rewards/rejected": -0.3264002799987793,
       "step": 2760
+    },
+    {
+      "epoch": 14.025316455696203,
+      "grad_norm": 454762.9481647176,
+      "learning_rate": 1.2159197743654026e-07,
+      "logits/chosen": 2.4223504066467285,
+      "logits/rejected": 3.487738847732544,
+      "logps/chosen": -44.93278503417969,
+      "logps/rejected": -561.3870849609375,
+      "loss": 16557.4125,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.1806286722421646,
+      "rewards/margins": 0.5195534229278564,
+      "rewards/rejected": -0.33892473578453064,
+      "step": 2770
+    },
+    {
+      "epoch": 14.075949367088608,
+      "grad_norm": 487680.4985531969,
+      "learning_rate": 1.2002507051081164e-07,
+      "logits/chosen": 1.9585473537445068,
+      "logits/rejected": 2.446890354156494,
+      "logps/chosen": -39.52117919921875,
+      "logps/rejected": -561.9512939453125,
+      "loss": 15203.5906,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.1834731251001358,
+      "rewards/margins": 0.5265246629714966,
+      "rewards/rejected": -0.343051552772522,
+      "step": 2780
+    },
+    {
+      "epoch": 14.126582278481013,
+      "grad_norm": 335633.29006652284,
+      "learning_rate": 1.1845816358508304e-07,
+      "logits/chosen": -0.2361418753862381,
+      "logits/rejected": 0.4229121804237366,
+      "logps/chosen": -56.944580078125,
+      "logps/rejected": -581.4995727539062,
+      "loss": 14980.4906,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18962779641151428,
+      "rewards/margins": 0.5268105268478394,
+      "rewards/rejected": -0.3371827304363251,
+      "step": 2790
+    },
+    {
+      "epoch": 14.177215189873417,
+      "grad_norm": 433336.52566667914,
+      "learning_rate": 1.1689125665935443e-07,
+      "logits/chosen": -0.8853734135627747,
+      "logits/rejected": 0.24162235856056213,
+      "logps/chosen": -49.96304702758789,
+      "logps/rejected": -587.9956665039062,
+      "loss": 15952.2594,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.1883043497800827,
+      "rewards/margins": 0.5334208607673645,
+      "rewards/rejected": -0.345116525888443,
+      "step": 2800
+    },
+    {
+      "epoch": 14.227848101265822,
+      "grad_norm": 352832.2810093542,
+      "learning_rate": 1.1532434973362581e-07,
+      "logits/chosen": -0.9270970225334167,
+      "logits/rejected": -0.8106321096420288,
+      "logps/chosen": -50.61150360107422,
+      "logps/rejected": -579.4258422851562,
+      "loss": 15482.3031,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18148374557495117,
+      "rewards/margins": 0.5241626501083374,
+      "rewards/rejected": -0.34267887473106384,
+      "step": 2810
+    },
+    {
+      "epoch": 14.278481012658228,
+      "grad_norm": 518734.4787371263,
+      "learning_rate": 1.137574428078972e-07,
+      "logits/chosen": 2.115744113922119,
+      "logits/rejected": 2.9750027656555176,
+      "logps/chosen": -41.601097106933594,
+      "logps/rejected": -573.6159057617188,
+      "loss": 15787.4719,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.18657180666923523,
+      "rewards/margins": 0.5376033186912537,
+      "rewards/rejected": -0.35103151202201843,
+      "step": 2820
+    },
+    {
+      "epoch": 14.329113924050633,
+      "grad_norm": 637771.2756103254,
+      "learning_rate": 1.1219053588216858e-07,
+      "logits/chosen": -0.09557388722896576,
+      "logits/rejected": -0.5708149671554565,
+      "logps/chosen": -44.071807861328125,
+      "logps/rejected": -585.6417236328125,
+      "loss": 15660.4813,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18580812215805054,
+      "rewards/margins": 0.5347784757614136,
+      "rewards/rejected": -0.34897032380104065,
+      "step": 2830
+    },
+    {
+      "epoch": 14.379746835443038,
+      "grad_norm": 469592.5817335632,
+      "learning_rate": 1.1062362895643998e-07,
+      "logits/chosen": 0.14405778050422668,
+      "logits/rejected": 0.6720622181892395,
+      "logps/chosen": -45.77620315551758,
+      "logps/rejected": -562.7021484375,
+      "loss": 15265.0797,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.18696969747543335,
+      "rewards/margins": 0.5200961828231812,
+      "rewards/rejected": -0.3331265151500702,
+      "step": 2840
+    },
+    {
+      "epoch": 14.430379746835444,
+      "grad_norm": 381405.89470487926,
+      "learning_rate": 1.0905672203071137e-07,
+      "logits/chosen": -0.46474942564964294,
+      "logits/rejected": -0.6803582906723022,
+      "logps/chosen": -43.475257873535156,
+      "logps/rejected": -578.9302978515625,
+      "loss": 15502.7,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18612739443778992,
+      "rewards/margins": 0.5332227945327759,
+      "rewards/rejected": -0.34709542989730835,
+      "step": 2850
+    },
+    {
+      "epoch": 14.481012658227849,
+      "grad_norm": 389034.05049605225,
+      "learning_rate": 1.0748981510498275e-07,
+      "logits/chosen": 0.192867711186409,
+      "logits/rejected": 0.04235720634460449,
+      "logps/chosen": -45.57283020019531,
+      "logps/rejected": -573.7398071289062,
+      "loss": 16059.1625,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18987932801246643,
+      "rewards/margins": 0.5239830613136292,
+      "rewards/rejected": -0.33410370349884033,
+      "step": 2860
+    },
+    {
+      "epoch": 14.531645569620252,
+      "grad_norm": 1027736.0673764712,
+      "learning_rate": 1.0592290817925414e-07,
+      "logits/chosen": -0.14229407906532288,
+      "logits/rejected": 0.4352554380893707,
+      "logps/chosen": -52.69159698486328,
+      "logps/rejected": -584.4544067382812,
+      "loss": 15405.6859,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.19550864398479462,
+      "rewards/margins": 0.5430904626846313,
+      "rewards/rejected": -0.34758180379867554,
+      "step": 2870
+    },
+    {
+      "epoch": 14.582278481012658,
+      "grad_norm": 384385.74028987245,
+      "learning_rate": 1.0435600125352554e-07,
+      "logits/chosen": -2.178337335586548,
+      "logits/rejected": -0.7508569955825806,
+      "logps/chosen": -59.098426818847656,
+      "logps/rejected": -576.6027221679688,
+      "loss": 14664.3844,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18826426565647125,
+      "rewards/margins": 0.5217211842536926,
+      "rewards/rejected": -0.33345693349838257,
+      "step": 2880
+    },
+    {
+      "epoch": 14.632911392405063,
+      "grad_norm": 329341.72262227273,
+      "learning_rate": 1.0278909432779692e-07,
+      "logits/chosen": -0.5238679647445679,
+      "logits/rejected": 0.5422592163085938,
+      "logps/chosen": -45.037288665771484,
+      "logps/rejected": -568.7276000976562,
+      "loss": 15557.125,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.18480226397514343,
+      "rewards/margins": 0.5315712094306946,
+      "rewards/rejected": -0.34676894545555115,
+      "step": 2890
+    },
+    {
+      "epoch": 14.683544303797468,
+      "grad_norm": 543441.6169659087,
+      "learning_rate": 1.0122218740206831e-07,
+      "logits/chosen": -1.954636812210083,
+      "logits/rejected": -1.2880172729492188,
+      "logps/chosen": -42.44208908081055,
+      "logps/rejected": -553.3023681640625,
+      "loss": 15342.95,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.17853178083896637,
+      "rewards/margins": 0.5080317258834839,
+      "rewards/rejected": -0.32949990034103394,
+      "step": 2900
+    },
+    {
+      "epoch": 14.734177215189874,
+      "grad_norm": 485286.8133606422,
+      "learning_rate": 9.96552804763397e-08,
+      "logits/chosen": -0.10534539073705673,
+      "logits/rejected": -0.22817449271678925,
+      "logps/chosen": -58.41508102416992,
+      "logps/rejected": -589.82861328125,
+      "loss": 14829.6719,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.19155274331569672,
+      "rewards/margins": 0.5371214747428894,
+      "rewards/rejected": -0.3455687165260315,
+      "step": 2910
+    },
+    {
+      "epoch": 14.784810126582279,
+      "grad_norm": 443260.47292018944,
+      "learning_rate": 9.808837355061108e-08,
+      "logits/chosen": 0.06932596862316132,
+      "logits/rejected": -0.2167021781206131,
+      "logps/chosen": -47.265785217285156,
+      "logps/rejected": -564.3973388671875,
+      "loss": 15330.0641,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.1784828007221222,
+      "rewards/margins": 0.5103118419647217,
+      "rewards/rejected": -0.3318290710449219,
+      "step": 2920
+    },
+    {
+      "epoch": 14.835443037974684,
+      "grad_norm": 483368.1079372665,
+      "learning_rate": 9.652146662488248e-08,
+      "logits/chosen": -0.06790392100811005,
+      "logits/rejected": 0.29011401534080505,
+      "logps/chosen": -54.78580856323242,
+      "logps/rejected": -574.620361328125,
+      "loss": 15093.3531,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.18937523663043976,
+      "rewards/margins": 0.5257763862609863,
+      "rewards/rejected": -0.33640116453170776,
+      "step": 2930
+    },
+    {
+      "epoch": 14.886075949367088,
+      "grad_norm": 955906.0887958824,
+      "learning_rate": 9.495455969915387e-08,
+      "logits/chosen": 1.4835760593414307,
+      "logits/rejected": 1.6735947132110596,
+      "logps/chosen": -46.26830291748047,
+      "logps/rejected": -551.6137084960938,
+      "loss": 15061.7437,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.1782112419605255,
+      "rewards/margins": 0.5047247409820557,
+      "rewards/rejected": -0.32651349902153015,
+      "step": 2940
+    },
+    {
+      "epoch": 14.936708860759493,
+      "grad_norm": 389874.4777367002,
+      "learning_rate": 9.338765277342525e-08,
+      "logits/chosen": -0.45253458619117737,
+      "logits/rejected": 0.04955162853002548,
+      "logps/chosen": -44.50522994995117,
+      "logps/rejected": -556.4650268554688,
+      "loss": 15850.6094,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.1812363862991333,
+      "rewards/margins": 0.5129731893539429,
+      "rewards/rejected": -0.33173683285713196,
+      "step": 2950
+    },
+    {
+      "epoch": 14.987341772151899,
+      "grad_norm": 880494.2982969056,
+      "learning_rate": 9.182074584769664e-08,
+      "logits/chosen": -1.3114904165267944,
+      "logits/rejected": -0.3469497859477997,
+      "logps/chosen": -48.75851821899414,
+      "logps/rejected": -542.8458862304688,
+      "loss": 14465.8125,
+      "rewards/accuracies": 0.949999988079071,
+      "rewards/chosen": 0.1728508621454239,
+      "rewards/margins": 0.49641647934913635,
+      "rewards/rejected": -0.32356563210487366,
+      "step": 2960
     }
   ],
   "logging_steps": 10,