Training in progress, step 930, checkpoint

last-checkpoint/global_step930/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:b123a2cb22969eb9ccdfb292bc88c166ef1e867193095e50dbfbb01c4cca2951
+size 13476835648

last-checkpoint/global_step930/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75d6ced840634693a094a3edc203312b6532e27dd589c03696dd60e7bce660a5
+size 13476835648

last-checkpoint/global_step930/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74ed7591c60b03da0d1dc6cc3d9bfe28c9349fd72909ee0f7770ebf71cf5f300
+size 13476835648

last-checkpoint/global_step930/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:ec1a3899e3e0db79c3497e9367c84bcaa34ec1e76d90811c546d0eaeae30451b
+size 13476835648

last-checkpoint/global_step930/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a98a917211259358e0aa315ceda5d9c8adfaf7a77dfb884a74aedb59c4bd78f
+size 150693

last-checkpoint/global_step930/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85392d249fe1774166e899bdacd4ddbaeeb18826e8ac208a06a13d57095ef4c1
+size 150693

last-checkpoint/global_step930/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62deb2157fa7caf68b01c013a50a8a320c3d100a989932ea59653e9809b1d906
+size 150693

last-checkpoint/global_step930/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2266da4e0281fec19e433f5d7ad9197820be41db576884e86fe11aafcbc31b1
+size 150693

last-checkpoint/latest

@@ -1 +1 @@
-global_step868
+global_step930

last-checkpoint/model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b829a6042f593ec2bd98e0f87aff609c3d2c139cbbf8c9ab7ca3d75432e198a3
+oid sha256:ef0f4df2c2dd63dd83e130b116d935a681a8ee1ce46d3cb72eadaf7aca4d3879
 size 4938985352

last-checkpoint/model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a69fd15f74564bc9ca126d1831e9e854badf4ee7a361a2165300b97b3ead57ae
+oid sha256:bdb87e29425f917524e9cf4fb01bc4e694598be78980bb38df50ed6dd399a71f
 size 4947390880

last-checkpoint/model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:efc0b7362663abe6ceaab66b8d29d6b544536073d01b26a0b49932f1f7a31da0
+oid sha256:cbf925b489dde5633219973a1f7552c30e5d3d96df88e0f751b425f56891190c
 size 3590488816

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8639c02c997d5ec74743bd87a283daff10faa317419bf379edd99c706559f2ce
+oid sha256:e5b5b0934b8613429f71d38c4aaf1cf68b50eb415214b938124e65c4e50faae1
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:211a79d80fe07a9690b74e693f719eafa8303e6798af58a53dd105eb19c8ccc5
+oid sha256:19f63bfbf9ef9efae26be62fb115c8fa6adfa6333f3074d8511b5eeab88f0f1e
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:319730d0d11be8a12d1633e218e39729160d397a56a34d9ebd2e63d2c81fd68f
+oid sha256:ba2a5ec9c47d3c544c4e610213c6d14c67721b556bd12896b90b14fd54cf1901
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e7c1d2c0fa7220ac8b520afb2fc0958467f149187d655695c73de033474c910
+oid sha256:7e5fd0b1afc88ca5e581b87462a6a56c9a82154f0d69fe336f7b445bad77242b
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44e47707f4c92af1b42df0113d960785fdfeffc8e1f862cfb51ce7dbdb329bc2
+oid sha256:3037f93266047f9be66a7aa9272e870b0bbd8228f26f8217758f94a734d71a2f
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.7776,
+  "epoch": 2.976,
   "eval_steps": 62,
-  "global_step": 868,
+  "global_step": 930,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1536,6 +1536,127 @@
       "eval_samples_per_second": 5.923,
       "eval_steps_per_second": 0.385,
       "step": 868
+    },
+    {
+      "epoch": 2.784,
+      "grad_norm": 23.996876211666585,
+      "learning_rate": 3.919239904988123e-08,
+      "logits/generated": -0.9030688405036926,
+      "logits/real": -0.8869367837905884,
+      "logps/generated": -263.31884765625,
+      "logps/real": -231.1215057373047,
+      "loss": 0.3838,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": 2.1591622829437256,
+      "rewards/margins": 2.8290388584136963,
+      "rewards/real": 4.988200664520264,
+      "step": 870
+    },
+    {
+      "epoch": 2.816,
+      "grad_norm": 15.933355481968098,
+      "learning_rate": 3.32541567695962e-08,
+      "logits/generated": -0.9254922866821289,
+      "logits/real": -0.9375091791152954,
+      "logps/generated": -257.25921630859375,
+      "logps/real": -218.4817352294922,
+      "loss": 0.3943,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/generated": 2.433042049407959,
+      "rewards/margins": 2.4680287837982178,
+      "rewards/real": 4.901071071624756,
+      "step": 880
+    },
+    {
+      "epoch": 2.848,
+      "grad_norm": 15.729400330653592,
+      "learning_rate": 2.7315914489311164e-08,
+      "logits/generated": -0.8918999433517456,
+      "logits/real": -0.9543954730033875,
+      "logps/generated": -251.86434936523438,
+      "logps/real": -217.92153930664062,
+      "loss": 0.3792,
+      "rewards/accuracies": 0.9375,
+      "rewards/generated": 2.4217216968536377,
+      "rewards/margins": 2.721459150314331,
+      "rewards/real": 5.143180847167969,
+      "step": 890
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 14.591897761657874,
+      "learning_rate": 2.1377672209026125e-08,
+      "logits/generated": -0.9395130276679993,
+      "logits/real": -0.9594659805297852,
+      "logps/generated": -265.73876953125,
+      "logps/real": -222.98593139648438,
+      "loss": 0.3686,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/generated": 2.46151065826416,
+      "rewards/margins": 2.638683795928955,
+      "rewards/real": 5.100194931030273,
+      "step": 900
+    },
+    {
+      "epoch": 2.912,
+      "grad_norm": 20.099085379716534,
+      "learning_rate": 1.5439429928741092e-08,
+      "logits/generated": -0.950794517993927,
+      "logits/real": -0.8883028030395508,
+      "logps/generated": -254.50732421875,
+      "logps/real": -226.67514038085938,
+      "loss": 0.3771,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/generated": 2.2818403244018555,
+      "rewards/margins": 2.8682749271392822,
+      "rewards/real": 5.150115013122559,
+      "step": 910
+    },
+    {
+      "epoch": 2.944,
+      "grad_norm": 27.113250586107085,
+      "learning_rate": 9.501187648456057e-09,
+      "logits/generated": -0.9123473167419434,
+      "logits/real": -0.8874839544296265,
+      "logps/generated": -254.9096221923828,
+      "logps/real": -226.13595581054688,
+      "loss": 0.3878,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/generated": 2.072730302810669,
+      "rewards/margins": 2.4795820713043213,
+      "rewards/real": 4.55231237411499,
+      "step": 920
+    },
+    {
+      "epoch": 2.976,
+      "grad_norm": 18.727290911205092,
+      "learning_rate": 3.562945368171021e-09,
+      "logits/generated": -0.9185419082641602,
+      "logits/real": -0.9089628458023071,
+      "logps/generated": -247.39254760742188,
+      "logps/real": -213.3396453857422,
+      "loss": 0.4001,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/generated": 2.629145860671997,
+      "rewards/margins": 2.3609352111816406,
+      "rewards/real": 4.990080833435059,
+      "step": 930
+    },
+    {
+      "epoch": 2.976,
+      "eval_logits/generated": -0.874191164970398,
+      "eval_logits/real": -0.8781108260154724,
+      "eval_logps/generated": -253.27622985839844,
+      "eval_logps/real": -213.75250244140625,
+      "eval_loss": 0.552344799041748,
+      "eval_rewards/accuracies": 0.8461538553237915,
+      "eval_rewards/generated": 2.939284324645996,
+      "eval_rewards/margins": 2.054386854171753,
+      "eval_rewards/real": 4.993671894073486,
+      "eval_runtime": 34.0216,
+      "eval_samples_per_second": 5.879,
+      "eval_steps_per_second": 0.382,
+      "step": 930
     }
   ],
   "logging_steps": 10,