Training in progress, step 20000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9db25d01d664aab4cb98e082b28d4b9aabe9c3484451098075c4a39951609908
+oid sha256:7ff30f34c03f4aee5e5fd09a1aacd3d7bdd3d177d4232672590b18570ebd00ab
 size 50044241

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4777aef65282d7381884640132b86532faacb12845758f068a2aac0ed0646ae
+oid sha256:ee26862b665697be023c901e312db258d269e332dc1eb7fb87e516e00bf064a6
 size 25761253

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 size 14503

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54025b1e5c20b5516d968c152992bafd1f062d301f6ee50e676cd70067564190
+oid sha256:81b2c87ce00f88a26045daeef95098b32592be610ef6295c77558bd47e7dfa25
 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": 0.6127450980392157,
-  "global_step": 10000,
+  "epoch": 1.2254901960784315,
+  "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": 807.983,
       "eval_steps_per_second": 12.928,
       "step": 10000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0005039999999999999,
+      "loss": 0.7369,
+      "step": 10500
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0005279999999999999,
+      "loss": 0.7369,
+      "step": 11000
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 0.9027426838874817,
+      "eval_runtime": 1.2459,
+      "eval_samples_per_second": 802.657,
+      "eval_steps_per_second": 12.843,
+      "step": 11000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.000552,
+      "loss": 0.7369,
+      "step": 11500
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0005759999999999999,
+      "loss": 0.7368,
+      "step": 12000
+    },
+    {
+      "epoch": 0.74,
+      "eval_loss": 0.9022310972213745,
+      "eval_runtime": 1.2295,
+      "eval_samples_per_second": 813.368,
+      "eval_steps_per_second": 13.014,
+      "step": 12000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0006,
+      "loss": 0.7367,
+      "step": 12500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0005999935478721662,
+      "loss": 0.7368,
+      "step": 13000
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 0.8987133502960205,
+      "eval_runtime": 1.2643,
+      "eval_samples_per_second": 790.939,
+      "eval_steps_per_second": 12.655,
+      "step": 13000
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.000599974191770902,
+      "loss": 0.7375,
+      "step": 13500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0005999419325429058,
+      "loss": 0.7374,
+      "step": 14000
+    },
+    {
+      "epoch": 0.86,
+      "eval_loss": 0.9013972282409668,
+      "eval_runtime": 1.236,
+      "eval_samples_per_second": 809.042,
+      "eval_steps_per_second": 12.945,
+      "step": 14000
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0005998967715993009,
+      "loss": 0.7369,
+      "step": 14500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0005998387109155732,
+      "loss": 0.7369,
+      "step": 15000
+    },
+    {
+      "epoch": 0.92,
+      "eval_loss": 0.9001522660255432,
+      "eval_runtime": 1.235,
+      "eval_samples_per_second": 809.697,
+      "eval_steps_per_second": 12.955,
+      "step": 15000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.000599767753031485,
+      "loss": 0.737,
+      "step": 15500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0005996839010509641,
+      "loss": 0.7369,
+      "step": 16000
+    },
+    {
+      "epoch": 0.98,
+      "eval_loss": 0.9001864790916443,
+      "eval_runtime": 1.1685,
+      "eval_samples_per_second": 855.784,
+      "eval_steps_per_second": 13.693,
+      "step": 16000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.0005995871586419678,
+      "loss": 0.7369,
+      "step": 16500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0005994775300363225,
+      "loss": 0.7372,
+      "step": 17000
+    },
+    {
+      "epoch": 1.04,
+      "eval_loss": 0.9018534421920776,
+      "eval_runtime": 1.1865,
+      "eval_samples_per_second": 842.819,
+      "eval_steps_per_second": 13.485,
+      "step": 17000
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0005993550200295384,
+      "loss": 0.7371,
+      "step": 17500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0005992196339806002,
+      "loss": 0.737,
+      "step": 18000
+    },
+    {
+      "epoch": 1.1,
+      "eval_loss": 0.9001176953315735,
+      "eval_runtime": 1.2219,
+      "eval_samples_per_second": 818.375,
+      "eval_steps_per_second": 13.094,
+      "step": 18000
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0005990713778117324,
+      "loss": 0.7369,
+      "step": 18500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0005989102580081398,
+      "loss": 0.737,
+      "step": 19000
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.9006143808364868,
+      "eval_runtime": 1.2544,
+      "eval_samples_per_second": 797.163,
+      "eval_steps_per_second": 12.755,
+      "step": 19000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.0005987362816177249,
+      "loss": 0.7369,
+      "step": 19500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.0005985494562507783,
+      "loss": 0.7369,
+      "step": 20000
+    },
+    {
+      "epoch": 1.23,
+      "eval_loss": 0.9006676077842712,
+      "eval_runtime": 1.2815,
+      "eval_samples_per_second": 780.312,
+      "eval_steps_per_second": 12.485,
+      "step": 20000
     }
   ],
   "max_steps": 250000,
   "num_train_epochs": 16,
-  "total_flos": 1.6016800286580408e+20,
+  "total_flos": 3.203259952279931e+20,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4777aef65282d7381884640132b86532faacb12845758f068a2aac0ed0646ae
+oid sha256:ee26862b665697be023c901e312db258d269e332dc1eb7fb87e516e00bf064a6
 size 25761253