Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ddf8c73a99cc42b29915ad54282c497fea591218d3df26b14997a77e4eb6d55
+oid sha256:d9c29fbdf49be880cb6493988a94306303a92a77dfa3a5685c194a3ed9741dd7
 size 50044241

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdc61f9c8be03d76c40deff98eea159770951d05b23400b4c372c8b407086f24
+oid sha256:cab920afe4d207be60e60c397c80474cec3b0781a866712c960668a023bb8b59
 size 25761253

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8e6070527688748b61c783bb22aaf8ce741c15f561e7b1d5c9a7481b92b181a
+oid sha256:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
 size 14503

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:595e6d0c765b4677e7c91d65c2aacefa0d09faec0213d2321b49d411358f597f
+oid sha256:2fa6f3c4ec253d8129a7481e01148ec46428b7a6eb1631c7fb589fd92c25c12f
 size 623

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.44610992148465384,
-  "global_step": 10000,
+  "epoch": 0.8922198429693077,
+  "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": 1027.899,
       "eval_steps_per_second": 16.11,
       "step": 10000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0005039999999999999,
+      "loss": 0.6775,
+      "step": 10500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0005279999999999999,
+      "loss": 0.6775,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "eval_loss": 0.6769479513168335,
+      "eval_runtime": 2.177,
+      "eval_samples_per_second": 1055.099,
+      "eval_steps_per_second": 16.536,
+      "step": 11000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.000552,
+      "loss": 0.6773,
+      "step": 11500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0005759999999999999,
+      "loss": 0.6773,
+      "step": 12000
+    },
+    {
+      "epoch": 0.54,
+      "eval_loss": 0.6773238182067871,
+      "eval_runtime": 2.1281,
+      "eval_samples_per_second": 1079.366,
+      "eval_steps_per_second": 16.916,
+      "step": 12000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0006,
+      "loss": 0.6773,
+      "step": 12500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0005999935478721662,
+      "loss": 0.6774,
+      "step": 13000
+    },
+    {
+      "epoch": 0.58,
+      "eval_loss": 0.677127480506897,
+      "eval_runtime": 2.1773,
+      "eval_samples_per_second": 1054.986,
+      "eval_steps_per_second": 16.534,
+      "step": 13000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.000599974191770902,
+      "loss": 0.6773,
+      "step": 13500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0005999419325429058,
+      "loss": 0.6773,
+      "step": 14000
+    },
+    {
+      "epoch": 0.62,
+      "eval_loss": 0.6771531105041504,
+      "eval_runtime": 2.1173,
+      "eval_samples_per_second": 1084.86,
+      "eval_steps_per_second": 17.003,
+      "step": 14000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0005998967715993009,
+      "loss": 0.6773,
+      "step": 14500
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0005998387109155732,
+      "loss": 0.6773,
+      "step": 15000
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 0.6771678924560547,
+      "eval_runtime": 2.1919,
+      "eval_samples_per_second": 1047.963,
+      "eval_steps_per_second": 16.424,
+      "step": 15000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.000599767753031485,
+      "loss": 0.6773,
+      "step": 15500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0005996839010509641,
+      "loss": 0.6772,
+      "step": 16000
+    },
+    {
+      "epoch": 0.71,
+      "eval_loss": 0.6776318550109863,
+      "eval_runtime": 2.199,
+      "eval_samples_per_second": 1044.559,
+      "eval_steps_per_second": 16.371,
+      "step": 16000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0005995871586419678,
+      "loss": 0.6773,
+      "step": 16500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0005994775300363225,
+      "loss": 0.6773,
+      "step": 17000
+    },
+    {
+      "epoch": 0.76,
+      "eval_loss": 0.676984429359436,
+      "eval_runtime": 2.1946,
+      "eval_samples_per_second": 1046.652,
+      "eval_steps_per_second": 16.404,
+      "step": 17000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0005993550200295384,
+      "loss": 0.6772,
+      "step": 17500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0005992196339806002,
+      "loss": 0.6772,
+      "step": 18000
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.6774880290031433,
+      "eval_runtime": 2.1027,
+      "eval_samples_per_second": 1092.415,
+      "eval_steps_per_second": 17.121,
+      "step": 18000
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0005990713778117324,
+      "loss": 0.6773,
+      "step": 18500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0005989102580081398,
+      "loss": 0.6772,
+      "step": 19000
+    },
+    {
+      "epoch": 0.85,
+      "eval_loss": 0.676984965801239,
+      "eval_runtime": 2.1686,
+      "eval_samples_per_second": 1059.223,
+      "eval_steps_per_second": 16.601,
+      "step": 19000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0005987362816177249,
+      "loss": 0.6773,
+      "step": 19500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0005985494562507783,
+      "loss": 0.6774,
+      "step": 20000
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.6769698262214661,
+      "eval_runtime": 2.1456,
+      "eval_samples_per_second": 1070.586,
+      "eval_steps_per_second": 16.779,
+      "step": 20000
     }
   ],
   "max_steps": 250000,
   "num_train_epochs": 12,
-  "total_flos": 1.6016800286580408e+20,
+  "total_flos": 3.2033600573160817e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdc61f9c8be03d76c40deff98eea159770951d05b23400b4c372c8b407086f24
+oid sha256:cab920afe4d207be60e60c397c80474cec3b0781a866712c960668a023bb8b59
 size 25761253