Training in progress, epoch 3, checkpoint

last-checkpoint/global_step312/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:e34b832689516476f4c3160a9e825d0f5c6a4bf46110e93c8a436224eeeb946e
+size 13476836524

last-checkpoint/global_step312/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21fd02119285354b31da865f5643f9827eeff5510e7c647dde5b0acfe8c59c08
+size 13476836524

last-checkpoint/global_step312/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ace484f6121a615627d11299f7483642a7d29946c694e68428d1f605bfceeeb
+size 13476836524

last-checkpoint/global_step312/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:e55adfba1f6814f3ecdc0a74819be46f9057787fe23a45055b2a2db93ae57e05
+size 13476836524

last-checkpoint/global_step312/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f6a5c6a31e37762de0e724b157b94ec5ed90da3d5fc827583a9d44c5f3b6c04
+size 150693

last-checkpoint/global_step312/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f07e6d53d37a2e0aeaa28c2a94b59c47e9d7b35318b341c58430572d6eacccf5
+size 150693

last-checkpoint/global_step312/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a83a1b7fd1f5ba1dab26d0ed7dd3bef9611bcd9c61786d7cda22850a5d497062
+size 150693

last-checkpoint/global_step312/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cb668f9a54c80963878f8fb3d009b523e1156add3cd25acbd2f4bc2d257fd8c
+size 150693

last-checkpoint/latest

@@ -1 +1 @@
-global_step104
+global_step312

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:525899e5835467badaec7d139a082288c47132537bb657e911a0edaba30d5349
+oid sha256:6f681c947638d1903de7d27854020edc8fcc94f219f97e86f1689e9708b57e3a
 size 4938985352

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb920d2b88f65d36ef9270b964bd82a1ed002d525e4693fbcd6c2d55a663118f
+oid sha256:9b1e9fa62a72922112d3175aca16f48d96b2df6dd417cae510bc666f71188d35
 size 4947390880

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41dea857cab485c54883015fce275092e3a48f464b56aad6bcdadbf6e94551c1
+oid sha256:75776f25bbc2f36d9efb6263f444323ef216c4f3fd810cc915b91da4fcec825c
 size 3590488816

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1ce770862b76ceecb3453a5d0b9ef65311eab0f6a60f52997726a89d718de08
+oid sha256:ab7cf0d34d60dfda516cf9661904550e2e294e723edd07c25c738f05e8ba92d1
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bcae6573442578b752be3e988d4ccb38056b45c31b5c02eb579ec6cebcfa62e
+oid sha256:f7b183324e8227a51a9556d86b2ad893a8c4c52205ed4a737356c6611dac7353
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:398ff45072fee2975e88b9e078f915103d75d94bc08753303cf855b915973623
+oid sha256:ac69e994090f4818cb1fa6f6cefa363178552c3c731c6507ff195bcb07fd5bef
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f4b210bc832d4c648e3bee8c21dca26e5a8b365d6ec90c638062005a052e57b
+oid sha256:f68e61b63402f8afb1f69c960f7944965655dac11e3ccf29919c282f23931f86
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e40a9f6998ec664098f440c22ab8bb6b3a25086c1612b60fa34f9d09f32efb92
+oid sha256:e3ae5e0ab275048976037621894081df2b3a1233660bc1684104adad082ec745
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 104,
+  "global_step": 312,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -188,6 +188,353 @@
       "eval_samples_per_second": 13.054,
       "eval_steps_per_second": 0.576,
       "step": 104
+    },
+    {
+      "epoch": 1.0576923076923077,
+      "grad_norm": 0.020791139211903984,
+      "learning_rate": 3.607142857142857e-07,
+      "logits/chosen": -1.6875,
+      "logits/rejected": -1.65625,
+      "logps/chosen": -148.0,
+      "logps/rejected": -226.0,
+      "loss": 0.0017,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -0.83203125,
+      "rewards/margins": 9.25,
+      "rewards/rejected": -10.0625,
+      "step": 110
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 0.04771399447611739,
+      "learning_rate": 3.4285714285714286e-07,
+      "logits/chosen": -1.6484375,
+      "logits/rejected": -1.5859375,
+      "logps/chosen": -167.0,
+      "logps/rejected": -234.0,
+      "loss": 0.0006,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.3125,
+      "rewards/margins": 9.5625,
+      "rewards/rejected": -10.875,
+      "step": 120
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.2534459160228267,
+      "learning_rate": 3.25e-07,
+      "logits/chosen": -1.671875,
+      "logits/rejected": -1.671875,
+      "logps/chosen": -165.0,
+      "logps/rejected": -234.0,
+      "loss": 0.001,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.3359375,
+      "rewards/margins": 9.3125,
+      "rewards/rejected": -10.6875,
+      "step": 130
+    },
+    {
+      "epoch": 1.3461538461538463,
+      "grad_norm": 0.07091308610265967,
+      "learning_rate": 3.0714285714285716e-07,
+      "logits/chosen": -1.6640625,
+      "logits/rejected": -1.640625,
+      "logps/chosen": -179.0,
+      "logps/rejected": -245.0,
+      "loss": 0.0006,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.25,
+      "rewards/margins": 10.125,
+      "rewards/rejected": -11.375,
+      "step": 140
+    },
+    {
+      "epoch": 1.4423076923076923,
+      "grad_norm": 0.01427238285225871,
+      "learning_rate": 2.892857142857143e-07,
+      "logits/chosen": -1.734375,
+      "logits/rejected": -1.703125,
+      "logps/chosen": -176.0,
+      "logps/rejected": -240.0,
+      "loss": 0.0007,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.25,
+      "rewards/margins": 9.75,
+      "rewards/rejected": -11.0625,
+      "step": 150
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.06865294866224332,
+      "learning_rate": 2.714285714285714e-07,
+      "logits/chosen": -1.7578125,
+      "logits/rejected": -1.6796875,
+      "logps/chosen": -183.0,
+      "logps/rejected": -241.0,
+      "loss": 0.0007,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.140625,
+      "rewards/margins": 10.25,
+      "rewards/rejected": -11.375,
+      "step": 160
+    },
+    {
+      "epoch": 1.6346153846153846,
+      "grad_norm": 0.07896118186411183,
+      "learning_rate": 2.5357142857142855e-07,
+      "logits/chosen": -1.625,
+      "logits/rejected": -1.5234375,
+      "logps/chosen": -160.0,
+      "logps/rejected": -246.0,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.2109375,
+      "rewards/margins": 10.5625,
+      "rewards/rejected": -11.75,
+      "step": 170
+    },
+    {
+      "epoch": 1.7307692307692308,
+      "grad_norm": 0.026222949044449873,
+      "learning_rate": 2.357142857142857e-07,
+      "logits/chosen": -1.671875,
+      "logits/rejected": -1.6484375,
+      "logps/chosen": -178.0,
+      "logps/rejected": -245.0,
+      "loss": 0.0015,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.3984375,
+      "rewards/margins": 10.25,
+      "rewards/rejected": -11.625,
+      "step": 180
+    },
+    {
+      "epoch": 1.8269230769230769,
+      "grad_norm": 0.011793924192378881,
+      "learning_rate": 2.1785714285714284e-07,
+      "logits/chosen": -1.671875,
+      "logits/rejected": -1.71875,
+      "logps/chosen": -159.0,
+      "logps/rejected": -245.0,
+      "loss": 0.0006,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.25,
+      "rewards/margins": 10.4375,
+      "rewards/rejected": -11.6875,
+      "step": 190
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 0.060978420362998054,
+      "learning_rate": 2e-07,
+      "logits/chosen": -1.703125,
+      "logits/rejected": -1.703125,
+      "logps/chosen": -158.0,
+      "logps/rejected": -243.0,
+      "loss": 0.0003,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.0859375,
+      "rewards/margins": 10.5,
+      "rewards/rejected": -11.5625,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_logits/chosen": -1.5,
+      "eval_logits/rejected": -1.4453125,
+      "eval_logps/chosen": -121.0,
+      "eval_logps/rejected": -248.0,
+      "eval_loss": 0.006824689917266369,
+      "eval_rewards/accuracies": 1.0,
+      "eval_rewards/chosen": -1.0078125,
+      "eval_rewards/margins": 10.875,
+      "eval_rewards/rejected": -11.875,
+      "eval_runtime": 6.628,
+      "eval_samples_per_second": 10.26,
+      "eval_steps_per_second": 0.453,
+      "step": 208
+    },
+    {
+      "epoch": 2.019230769230769,
+      "grad_norm": 0.025694364273425255,
+      "learning_rate": 1.8214285714285714e-07,
+      "logits/chosen": -1.640625,
+      "logits/rejected": -1.640625,
+      "logps/chosen": -162.0,
+      "logps/rejected": -244.0,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.015625,
+      "rewards/margins": 10.625,
+      "rewards/rejected": -11.625,
+      "step": 210
+    },
+    {
+      "epoch": 2.1153846153846154,
+      "grad_norm": 1.318722077689268,
+      "learning_rate": 1.6428571428571429e-07,
+      "logits/chosen": -1.71875,
+      "logits/rejected": -1.6796875,
+      "logps/chosen": -165.0,
+      "logps/rejected": -250.0,
+      "loss": 0.0005,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.03125,
+      "rewards/margins": 10.8125,
+      "rewards/rejected": -11.8125,
+      "step": 220
+    },
+    {
+      "epoch": 2.2115384615384617,
+      "grad_norm": 0.15842660516208706,
+      "learning_rate": 1.4642857142857143e-07,
+      "logits/chosen": -1.734375,
+      "logits/rejected": -1.7109375,
+      "logps/chosen": -179.0,
+      "logps/rejected": -255.0,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.5234375,
+      "rewards/margins": 10.75,
+      "rewards/rejected": -12.25,
+      "step": 230
+    },
+    {
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.1644318912700695,
+      "learning_rate": 1.2857142857142855e-07,
+      "logits/chosen": -1.6953125,
+      "logits/rejected": -1.734375,
+      "logps/chosen": -161.0,
+      "logps/rejected": -251.0,
+      "loss": 0.0008,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.1484375,
+      "rewards/margins": 10.5625,
+      "rewards/rejected": -11.6875,
+      "step": 240
+    },
+    {
+      "epoch": 2.4038461538461537,
+      "grad_norm": 0.1117902717935036,
+      "learning_rate": 1.107142857142857e-07,
+      "logits/chosen": -1.703125,
+      "logits/rejected": -1.6875,
+      "logps/chosen": -174.0,
+      "logps/rejected": -249.0,
+      "loss": 0.0003,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.25,
+      "rewards/margins": 10.9375,
+      "rewards/rejected": -12.1875,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.04593607447714544,
+      "learning_rate": 9.285714285714286e-08,
+      "logits/chosen": -1.6484375,
+      "logits/rejected": -1.6328125,
+      "logps/chosen": -166.0,
+      "logps/rejected": -251.0,
+      "loss": 0.0003,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.1328125,
+      "rewards/margins": 11.1875,
+      "rewards/rejected": -12.3125,
+      "step": 260
+    },
+    {
+      "epoch": 2.5961538461538463,
+      "grad_norm": 0.08723297585467851,
+      "learning_rate": 7.5e-08,
+      "logits/chosen": -1.7109375,
+      "logits/rejected": -1.65625,
+      "logps/chosen": -175.0,
+      "logps/rejected": -250.0,
+      "loss": 0.0003,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.328125,
+      "rewards/margins": 10.4375,
+      "rewards/rejected": -11.8125,
+      "step": 270
+    },
+    {
+      "epoch": 2.6923076923076925,
+      "grad_norm": 0.16735539683605713,
+      "learning_rate": 5.714285714285714e-08,
+      "logits/chosen": -1.71875,
+      "logits/rejected": -1.6875,
+      "logps/chosen": -181.0,
+      "logps/rejected": -252.0,
+      "loss": 0.0005,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.2734375,
+      "rewards/margins": 10.9375,
+      "rewards/rejected": -12.25,
+      "step": 280
+    },
+    {
+      "epoch": 2.7884615384615383,
+      "grad_norm": 0.10696117602485962,
+      "learning_rate": 3.9285714285714285e-08,
+      "logits/chosen": -1.625,
+      "logits/rejected": -1.6640625,
+      "logps/chosen": -150.0,
+      "logps/rejected": -248.0,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.109375,
+      "rewards/margins": 11.0,
+      "rewards/rejected": -12.0625,
+      "step": 290
+    },
+    {
+      "epoch": 2.8846153846153846,
+      "grad_norm": 0.00967600454300797,
+      "learning_rate": 2.142857142857143e-08,
+      "logits/chosen": -1.671875,
+      "logits/rejected": -1.671875,
+      "logps/chosen": -171.0,
+      "logps/rejected": -258.0,
+      "loss": 0.0004,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.171875,
+      "rewards/margins": 11.25,
+      "rewards/rejected": -12.4375,
+      "step": 300
+    },
+    {
+      "epoch": 2.980769230769231,
+      "grad_norm": 0.12453557861807049,
+      "learning_rate": 3.571428571428571e-09,
+      "logits/chosen": -1.6640625,
+      "logits/rejected": -1.671875,
+      "logps/chosen": -164.0,
+      "logps/rejected": -245.0,
+      "loss": 0.0006,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": -1.1796875,
+      "rewards/margins": 10.8125,
+      "rewards/rejected": -12.0,
+      "step": 310
+    },
+    {
+      "epoch": 3.0,
+      "eval_logits/chosen": -1.5,
+      "eval_logits/rejected": -1.453125,
+      "eval_logps/chosen": -121.5,
+      "eval_logps/rejected": -253.0,
+      "eval_loss": 0.004145369865000248,
+      "eval_rewards/accuracies": 1.0,
+      "eval_rewards/chosen": -1.0859375,
+      "eval_rewards/margins": 11.25,
+      "eval_rewards/rejected": -12.3125,
+      "eval_runtime": 6.3498,
+      "eval_samples_per_second": 10.709,
+      "eval_steps_per_second": 0.472,
+      "step": 312
     }
   ],
   "logging_steps": 10,
@@ -202,7 +549,7 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }