Training in progress, step 30000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65debe850847edf2f62bba0deb29483af297ffa34c6cdcd66a55832044fcc8ec
+oid sha256:5028e680b90d0d5deb58c782fcd2654d9815241c8789819c0ef4b4a5d0227dd1
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6f55f7fb7116bd42203f7b9313d6b9f08f6a023278c3f67be7ce3fba873b5dc
+oid sha256:48ec0fa394e10e50451293d1035a84eba2d3d01f6d72f6c0c1bd4739fe3ee776
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:006a29e4d6e5dacaa8f0d3ba56263cd7f1775a1129713c79ccc62a2c04246894
+oid sha256:d43991d7af84e9540d8f27d2f0d1e420a0cf71dfe1da000898814055b8f62473
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c84272d9e8a15cd78c706d16b77833d53dbfbd182e8ad79e3cd658ef6c3eaaf6
+oid sha256:ff9b79eba2b3c3487d57b5f9509b60a9305753a31bdaa8757ad442412114850e
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.46599408187516017,
-  "global_step": 20000,
+  "epoch": 0.6989911228127402,
+  "global_step": 30000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -406,11 +406,211 @@
       "eval_samples_per_second": 1670.425,
       "eval_steps_per_second": 26.599,
       "step": 20000
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00024599999999999996,
+      "loss": 0.4278,
+      "step": 20500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00025199999999999995,
+      "loss": 0.4233,
+      "step": 21000
+    },
+    {
+      "epoch": 0.49,
+      "eval_loss": 0.4032529294490814,
+      "eval_runtime": 1.2842,
+      "eval_samples_per_second": 1711.577,
+      "eval_steps_per_second": 27.254,
+      "step": 21000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.000258,
+      "loss": 0.4189,
+      "step": 21500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00026399999999999997,
+      "loss": 0.4146,
+      "step": 22000
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.39436739683151245,
+      "eval_runtime": 1.2633,
+      "eval_samples_per_second": 1739.901,
+      "eval_steps_per_second": 27.705,
+      "step": 22000
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00027,
+      "loss": 0.4109,
+      "step": 22500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.000276,
+      "loss": 0.4077,
+      "step": 23000
+    },
+    {
+      "epoch": 0.54,
+      "eval_loss": 0.38647642731666565,
+      "eval_runtime": 1.2503,
+      "eval_samples_per_second": 1758.021,
+      "eval_steps_per_second": 27.994,
+      "step": 23000
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00028199999999999997,
+      "loss": 0.4047,
+      "step": 23500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00028799999999999995,
+      "loss": 0.4019,
+      "step": 24000
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 0.381785124540329,
+      "eval_runtime": 1.2485,
+      "eval_samples_per_second": 1760.518,
+      "eval_steps_per_second": 28.034,
+      "step": 24000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.000294,
+      "loss": 0.3994,
+      "step": 24500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0003,
+      "loss": 0.3968,
+      "step": 25000
+    },
+    {
+      "epoch": 0.58,
+      "eval_loss": 0.3785216510295868,
+      "eval_runtime": 1.2913,
+      "eval_samples_per_second": 1702.153,
+      "eval_steps_per_second": 27.104,
+      "step": 25000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00029999920715161553,
+      "loss": 0.3945,
+      "step": 25500
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0002999968286151326,
+      "loss": 0.392,
+      "step": 26000
+    },
+    {
+      "epoch": 0.61,
+      "eval_loss": 0.37318962812423706,
+      "eval_runtime": 1.2758,
+      "eval_samples_per_second": 1722.782,
+      "eval_steps_per_second": 27.433,
+      "step": 26000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002999928644165624,
+      "loss": 0.3897,
+      "step": 26500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0002999873145992569,
+      "loss": 0.3869,
+      "step": 27000
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 0.3669659197330475,
+      "eval_runtime": 1.2869,
+      "eval_samples_per_second": 1707.967,
+      "eval_steps_per_second": 27.197,
+      "step": 27000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000299980179223908,
+      "loss": 0.385,
+      "step": 27500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0002999714583685469,
+      "loss": 0.3828,
+      "step": 28000
+    },
+    {
+      "epoch": 0.65,
+      "eval_loss": 0.3615981340408325,
+      "eval_runtime": 1.2961,
+      "eval_samples_per_second": 1695.914,
+      "eval_steps_per_second": 27.005,
+      "step": 28000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00029996115212854366,
+      "loss": 0.3808,
+      "step": 28500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00029994926061660554,
+      "loss": 0.3786,
+      "step": 29000
+    },
+    {
+      "epoch": 0.68,
+      "eval_loss": 0.3613170087337494,
+      "eval_runtime": 1.3013,
+      "eval_samples_per_second": 1689.114,
+      "eval_steps_per_second": 26.897,
+      "step": 29000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002999357839627762,
+      "loss": 0.377,
+      "step": 29500
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00029992072231443425,
+      "loss": 0.3751,
+      "step": 30000
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 0.35551005601882935,
+      "eval_runtime": 1.2661,
+      "eval_samples_per_second": 1736.053,
+      "eval_steps_per_second": 27.644,
+      "step": 30000
     }
   ],
   "max_steps": 500000,
   "num_train_epochs": 12,
-  "total_flos": 6.38974277549e+20,
+  "total_flos": 9.584614163235e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6f55f7fb7116bd42203f7b9313d6b9f08f6a023278c3f67be7ce3fba873b5dc
+oid sha256:48ec0fa394e10e50451293d1035a84eba2d3d01f6d72f6c0c1bd4739fe3ee776
 size 102501541