Training in progress, epoch 2, checkpoint

last-checkpoint/global_step468/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:d839f5a00a7aded43630c558876e185c8a6669cb21fbeeb228d6900d3f66a22b
+size 14483467880

last-checkpoint/global_step468/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:afdc58f84b76dd03a35f0122797a049802a0159e70183eaba791cfed5637f6a1
+size 14483467880

last-checkpoint/global_step468/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8133f6dcc7a394229fd788707b44b7b8a22071adf23a7371aa6c905e994bc1ec
+size 14483467880

last-checkpoint/global_step468/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84f2c88678636702976ef6685f02787dc6fa7d996d22ab4fe17e6ac31ceaf36c
+size 14483467880

last-checkpoint/global_step468/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1236a404fb9c0d8a288417e8417a5ddef7656873538097e5e6a16c543039da2d
+size 150629

last-checkpoint/global_step468/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37c9ee73c4e63dfe68ee988ca690da4e587de3100aa17a7d56469d6b3b2c9b75
+size 150629

last-checkpoint/global_step468/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdf2102bb233edc3010d34f68fbaf809675a0d153f96377e76c883390df28c09
+size 150629

last-checkpoint/global_step468/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ee0e952e65aca8c7967185632c1d14f282445bf4eded097de8b48527c3d1838
+size 150629

last-checkpoint/latest

@@ -1 +1 @@
-global_step313
+global_step468

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24d250e478ceec67ad697c5114ef84ef83195f94df1fc2e6d5423a259d96a851
+oid sha256:4671f390134f1e547cbdaf13277e9ee29979efb5999ac3a6f90c60fbaa23f1d2
 size 4943162336

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:746db1161ed175dafa557b81124b357a95144eb8daeac6de2459b049de70b0d9
+oid sha256:fd86be68e4ae8e991bc6ae78a06a869259bc03f7b9c97acdf02ec3b4c4136519
 size 4999819336

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:299dfa5fb3aa044f8ce0ab7dabebb0b9db464ada1bb1b84169a35ee309b44cbd
+oid sha256:e1697679f409ac5d7659a4d920471dab794a37d7bf34be608943582bc0ff51b2
 size 4540516344

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50ca27603f7c4a4d9a980223e1a102922ac31561acfa7da4f28042e4e8b3528b
+oid sha256:85a44fde7a8ca235658f035c0812d344f55779640520411739b17e214177e846
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c57f658c3d9b6de611da50021823095fc4067036fb9a4b34655cc2e34fbf5c2b
+oid sha256:2caad91201b0df46e56af9eecef567ed2037350c38fdc81837724a4727d0b050
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:109da2c363a1e09432fa2ea2b7dceb6f67d1f91975069fc669e71dd536d654d8
+oid sha256:73e988587366c84fd1b4fc5931610543c6da84cf501cd98c78745fd102dd5024
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bda65e490958bfe3fe01157ed131eb2354f8fc19536842d27c998ecf29d48991
+oid sha256:55af3bb04c01d99df727583b8b7a09241b4e47d58ff1b9cf5532d3fa0d878b52
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ae44c8e61affb3be5c9387c2b397935fdb2ed21d12002009ab043a8a5fffb07
+oid sha256:6ee2a07704fed3fe447ed9b1b7e4bc8211f255ba4385d30ed14a44579eb64834
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
+  "epoch": 2.9904153354632586,
   "eval_steps": 500,
-  "global_step": 313,
+  "global_step": 468,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -519,6 +519,247 @@
       "eval_samples_per_second": 16.282,
       "eval_steps_per_second": 0.57,
       "step": 313
+    },
+    {
+      "epoch": 2.0447284345047922,
+      "grad_norm": 12.799642134592919,
+      "learning_rate": 1.7577197149643706e-07,
+      "logits/chosen": -3.671875,
+      "logits/rejected": -3.625,
+      "logps/chosen": -173.0,
+      "logps/rejected": -195.0,
+      "loss": 0.1648,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/chosen": -0.2431640625,
+      "rewards/margins": 3.125,
+      "rewards/rejected": -3.359375,
+      "step": 320
+    },
+    {
+      "epoch": 2.108626198083067,
+      "grad_norm": 24.727219361838603,
+      "learning_rate": 1.6389548693586697e-07,
+      "logits/chosen": -3.703125,
+      "logits/rejected": -3.65625,
+      "logps/chosen": -165.0,
+      "logps/rejected": -176.0,
+      "loss": 0.1591,
+      "rewards/accuracies": 0.9312499761581421,
+      "rewards/chosen": -0.287109375,
+      "rewards/margins": 2.90625,
+      "rewards/rejected": -3.1875,
+      "step": 330
+    },
+    {
+      "epoch": 2.1725239616613417,
+      "grad_norm": 26.008991012968647,
+      "learning_rate": 1.520190023752969e-07,
+      "logits/chosen": -3.65625,
+      "logits/rejected": -3.609375,
+      "logps/chosen": -162.0,
+      "logps/rejected": -178.0,
+      "loss": 0.1532,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": -0.6640625,
+      "rewards/margins": 3.125,
+      "rewards/rejected": -3.796875,
+      "step": 340
+    },
+    {
+      "epoch": 2.236421725239617,
+      "grad_norm": 20.087149399156676,
+      "learning_rate": 1.4014251781472683e-07,
+      "logits/chosen": -3.734375,
+      "logits/rejected": -3.65625,
+      "logps/chosen": -165.0,
+      "logps/rejected": -177.0,
+      "loss": 0.1472,
+      "rewards/accuracies": 0.9437500238418579,
+      "rewards/chosen": -0.6171875,
+      "rewards/margins": 3.015625,
+      "rewards/rejected": -3.640625,
+      "step": 350
+    },
+    {
+      "epoch": 2.3003194888178915,
+      "grad_norm": 18.05087937817016,
+      "learning_rate": 1.2826603325415677e-07,
+      "logits/chosen": -3.796875,
+      "logits/rejected": -3.734375,
+      "logps/chosen": -166.0,
+      "logps/rejected": -193.0,
+      "loss": 0.1431,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/chosen": -0.54296875,
+      "rewards/margins": 3.0625,
+      "rewards/rejected": -3.609375,
+      "step": 360
+    },
+    {
+      "epoch": 2.364217252396166,
+      "grad_norm": 18.05523178462953,
+      "learning_rate": 1.163895486935867e-07,
+      "logits/chosen": -3.78125,
+      "logits/rejected": -3.75,
+      "logps/chosen": -177.0,
+      "logps/rejected": -185.0,
+      "loss": 0.1465,
+      "rewards/accuracies": 0.96875,
+      "rewards/chosen": -0.78125,
+      "rewards/margins": 3.15625,
+      "rewards/rejected": -3.9375,
+      "step": 370
+    },
+    {
+      "epoch": 2.428115015974441,
+      "grad_norm": 12.57343346412284,
+      "learning_rate": 1.0451306413301662e-07,
+      "logits/chosen": -3.78125,
+      "logits/rejected": -3.75,
+      "logps/chosen": -175.0,
+      "logps/rejected": -187.0,
+      "loss": 0.1316,
+      "rewards/accuracies": 0.96875,
+      "rewards/chosen": -0.828125,
+      "rewards/margins": 3.296875,
+      "rewards/rejected": -4.125,
+      "step": 380
+    },
+    {
+      "epoch": 2.4920127795527156,
+      "grad_norm": 21.734917480559936,
+      "learning_rate": 9.263657957244655e-08,
+      "logits/chosen": -3.796875,
+      "logits/rejected": -3.765625,
+      "logps/chosen": -173.0,
+      "logps/rejected": -200.0,
+      "loss": 0.1625,
+      "rewards/accuracies": 0.9437500238418579,
+      "rewards/chosen": -1.1796875,
+      "rewards/margins": 3.15625,
+      "rewards/rejected": -4.34375,
+      "step": 390
+    },
+    {
+      "epoch": 2.5559105431309903,
+      "grad_norm": 19.74378557143576,
+      "learning_rate": 8.076009501187649e-08,
+      "logits/chosen": -3.84375,
+      "logits/rejected": -3.765625,
+      "logps/chosen": -172.0,
+      "logps/rejected": -202.0,
+      "loss": 0.1368,
+      "rewards/accuracies": 0.96875,
+      "rewards/chosen": -1.09375,
+      "rewards/margins": 3.390625,
+      "rewards/rejected": -4.5,
+      "step": 400
+    },
+    {
+      "epoch": 2.619808306709265,
+      "grad_norm": 18.45734841726139,
+      "learning_rate": 6.88836104513064e-08,
+      "logits/chosen": -3.703125,
+      "logits/rejected": -3.65625,
+      "logps/chosen": -166.0,
+      "logps/rejected": -199.0,
+      "loss": 0.1322,
+      "rewards/accuracies": 0.9437500238418579,
+      "rewards/chosen": -1.03125,
+      "rewards/margins": 3.515625,
+      "rewards/rejected": -4.53125,
+      "step": 410
+    },
+    {
+      "epoch": 2.68370607028754,
+      "grad_norm": 22.52618935655582,
+      "learning_rate": 5.700712589073634e-08,
+      "logits/chosen": -3.703125,
+      "logits/rejected": -3.65625,
+      "logps/chosen": -178.0,
+      "logps/rejected": -203.0,
+      "loss": 0.1197,
+      "rewards/accuracies": 0.949999988079071,
+      "rewards/chosen": -0.6796875,
+      "rewards/margins": 3.765625,
+      "rewards/rejected": -4.4375,
+      "step": 420
+    },
+    {
+      "epoch": 2.747603833865815,
+      "grad_norm": 18.32031096894331,
+      "learning_rate": 4.5130641330166267e-08,
+      "logits/chosen": -3.765625,
+      "logits/rejected": -3.71875,
+      "logps/chosen": -177.0,
+      "logps/rejected": -195.0,
+      "loss": 0.1689,
+      "rewards/accuracies": 0.9437500238418579,
+      "rewards/chosen": -0.76953125,
+      "rewards/margins": 3.359375,
+      "rewards/rejected": -4.125,
+      "step": 430
+    },
+    {
+      "epoch": 2.8115015974440896,
+      "grad_norm": 15.907202030933394,
+      "learning_rate": 3.32541567695962e-08,
+      "logits/chosen": -3.71875,
+      "logits/rejected": -3.671875,
+      "logps/chosen": -170.0,
+      "logps/rejected": -203.0,
+      "loss": 0.1358,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/chosen": -0.85546875,
+      "rewards/margins": 3.578125,
+      "rewards/rejected": -4.4375,
+      "step": 440
+    },
+    {
+      "epoch": 2.8753993610223643,
+      "grad_norm": 15.683451965611985,
+      "learning_rate": 2.1377672209026125e-08,
+      "logits/chosen": -3.6875,
+      "logits/rejected": -3.59375,
+      "logps/chosen": -172.0,
+      "logps/rejected": -181.0,
+      "loss": 0.1541,
+      "rewards/accuracies": 0.949999988079071,
+      "rewards/chosen": -0.9375,
+      "rewards/margins": 3.53125,
+      "rewards/rejected": -4.46875,
+      "step": 450
+    },
+    {
+      "epoch": 2.939297124600639,
+      "grad_norm": 31.718439650295267,
+      "learning_rate": 9.501187648456057e-09,
+      "logits/chosen": -3.75,
+      "logits/rejected": -3.734375,
+      "logps/chosen": -171.0,
+      "logps/rejected": -190.0,
+      "loss": 0.1648,
+      "rewards/accuracies": 0.9375,
+      "rewards/chosen": -0.9140625,
+      "rewards/margins": 3.453125,
+      "rewards/rejected": -4.375,
+      "step": 460
+    },
+    {
+      "epoch": 2.9904153354632586,
+      "eval_logits/chosen": -3.828125,
+      "eval_logits/rejected": -3.859375,
+      "eval_logps/chosen": -176.0,
+      "eval_logps/rejected": -163.0,
+      "eval_loss": 0.64453125,
+      "eval_rewards/accuracies": 0.6607142686843872,
+      "eval_rewards/chosen": -1.7109375,
+      "eval_rewards/margins": 0.63671875,
+      "eval_rewards/rejected": -2.34375,
+      "eval_runtime": 15.0233,
+      "eval_samples_per_second": 13.313,
+      "eval_steps_per_second": 0.466,
+      "step": 468
     }
   ],
   "logging_steps": 10,
@@ -533,7 +774,7 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }