Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af09a84494208d1b177dbd17d23b3776db8fb8e0a4fb1db689b551a27a3b4c6e
+oid sha256:21ee51c8aedd1331406dff270a8984611b6b49ff01591d1f971301667cfec2bb
 size 202193937

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:039075d7b168d48cbc8a711dd247f621edb09291a9b978a19d5dd4b0e57d29a2
+oid sha256:8dbf204d94f01686d975a9600d69bdbd768ad031083dfb100f388cc4d3cc68ac
 size 102501541

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cd9583032b8deacd04dd2a81ca5c45c50f87e727dcae4885c295e3da08b639
+oid sha256:50f921c9f94cfbf967d9cad8d656604b2c1e1fb876e74b18ab76eb0d839fa8e8
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4931c97be633a6e3808391ac03b792837805ded64ff2a578babfa79e60dc22a7
+oid sha256:c84272d9e8a15cd78c706d16b77833d53dbfbd182e8ad79e3cd658ef6c3eaaf6
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.2548549875121056,
-  "global_step": 10000,
+  "epoch": 0.5097099750242112,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -206,11 +206,211 @@
       "eval_samples_per_second": 835.455,
       "eval_steps_per_second": 13.367,
       "step": 10000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00012599999999999997,
+      "loss": 0.5444,
+      "step": 10500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00013199999999999998,
+      "loss": 0.5385,
+      "step": 11000
+    },
+    {
+      "epoch": 0.28,
+      "eval_loss": 0.8903993964195251,
+      "eval_runtime": 1.1996,
+      "eval_samples_per_second": 833.63,
+      "eval_steps_per_second": 13.338,
+      "step": 11000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.000138,
+      "loss": 0.532,
+      "step": 11500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00014399999999999998,
+      "loss": 0.5276,
+      "step": 12000
+    },
+    {
+      "epoch": 0.31,
+      "eval_loss": 0.8864779472351074,
+      "eval_runtime": 1.1846,
+      "eval_samples_per_second": 844.173,
+      "eval_steps_per_second": 13.507,
+      "step": 12000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00015,
+      "loss": 0.5218,
+      "step": 12500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.000156,
+      "loss": 0.517,
+      "step": 13000
+    },
+    {
+      "epoch": 0.33,
+      "eval_loss": 0.88374263048172,
+      "eval_runtime": 1.1943,
+      "eval_samples_per_second": 837.343,
+      "eval_steps_per_second": 13.397,
+      "step": 13000
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.000162,
+      "loss": 0.5124,
+      "step": 13500
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.000168,
+      "loss": 0.5077,
+      "step": 14000
+    },
+    {
+      "epoch": 0.36,
+      "eval_loss": 0.8863275647163391,
+      "eval_runtime": 1.1844,
+      "eval_samples_per_second": 844.316,
+      "eval_steps_per_second": 13.509,
+      "step": 14000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00017399999999999997,
+      "loss": 0.5027,
+      "step": 14500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.498,
+      "step": 15000
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 0.8769687414169312,
+      "eval_runtime": 1.1484,
+      "eval_samples_per_second": 870.806,
+      "eval_steps_per_second": 13.933,
+      "step": 15000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.000186,
+      "loss": 0.4938,
+      "step": 15500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00019199999999999998,
+      "loss": 0.4897,
+      "step": 16000
+    },
+    {
+      "epoch": 0.41,
+      "eval_loss": 0.8793530464172363,
+      "eval_runtime": 1.2247,
+      "eval_samples_per_second": 816.516,
+      "eval_steps_per_second": 13.064,
+      "step": 16000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.000198,
+      "loss": 0.4849,
+      "step": 16500
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.000204,
+      "loss": 0.4791,
+      "step": 17000
+    },
+    {
+      "epoch": 0.43,
+      "eval_loss": 0.879580557346344,
+      "eval_runtime": 1.1672,
+      "eval_samples_per_second": 856.75,
+      "eval_steps_per_second": 13.708,
+      "step": 17000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00020999999999999998,
+      "loss": 0.4744,
+      "step": 17500
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00021599999999999996,
+      "loss": 0.4698,
+      "step": 18000
+    },
+    {
+      "epoch": 0.46,
+      "eval_loss": 0.8753651976585388,
+      "eval_runtime": 1.1516,
+      "eval_samples_per_second": 868.392,
+      "eval_steps_per_second": 13.894,
+      "step": 18000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00022199999999999998,
+      "loss": 0.4644,
+      "step": 18500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00022799999999999999,
+      "loss": 0.4592,
+      "step": 19000
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 0.8825145363807678,
+      "eval_runtime": 1.2668,
+      "eval_samples_per_second": 789.404,
+      "eval_steps_per_second": 12.63,
+      "step": 19000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.000234,
+      "loss": 0.4542,
+      "step": 19500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00023999999999999998,
+      "loss": 0.4489,
+      "step": 20000
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.878689169883728,
+      "eval_runtime": 1.1589,
+      "eval_samples_per_second": 862.885,
+      "eval_steps_per_second": 13.806,
+      "step": 20000
     }
   ],
   "max_steps": 500000,
   "num_train_epochs": 13,
-  "total_flos": 3.194871387745e+20,
+  "total_flos": 6.38974277549e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:039075d7b168d48cbc8a711dd247f621edb09291a9b978a19d5dd4b0e57d29a2
+oid sha256:8dbf204d94f01686d975a9600d69bdbd768ad031083dfb100f388cc4d3cc68ac
 size 102501541