Model save

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dca950ade2c9b68106a456900c953cfde7fde39144bab7d02b46f0ce47c36b50
+oid sha256:8409188c6a0dfcb31868f51aacf3e6357c3bb0cba81c0b082b9a213ac4847595
 size 267978033

trainer_state.json

@@ -1,95 +1,305 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
-  "global_step": 1200,
+  "epoch": 20.0,
+  "global_step": 7660,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 1.605e-05,
-      "loss": 2.5445,
-      "step": 240
+      "learning_rate": 1.9005221932114882e-05,
+      "loss": 2.5049,
+      "step": 383
     },
     {
       "epoch": 1.0,
-      "eval_loss": 2.3416366577148438,
-      "eval_runtime": 15.709,
-      "eval_samples_per_second": 975.173,
-      "eval_steps_per_second": 15.278,
-      "step": 240
+      "eval_loss": 2.3055639266967773,
+      "eval_runtime": 6.5178,
+      "eval_samples_per_second": 940.197,
+      "eval_steps_per_second": 14.729,
+      "step": 383
     },
     {
       "epoch": 2.0,
-      "learning_rate": 1.2050000000000002e-05,
-      "loss": 2.4313,
-      "step": 480
+      "learning_rate": 1.8005221932114885e-05,
+      "loss": 2.3896,
+      "step": 766
     },
     {
       "epoch": 2.0,
-      "eval_loss": 2.306915521621704,
-      "eval_runtime": 15.7874,
-      "eval_samples_per_second": 970.331,
-      "eval_steps_per_second": 15.202,
-      "step": 480
+      "eval_loss": 2.246039390563965,
+      "eval_runtime": 6.5445,
+      "eval_samples_per_second": 936.352,
+      "eval_steps_per_second": 14.669,
+      "step": 766
     },
     {
       "epoch": 3.0,
-      "learning_rate": 8.050000000000001e-06,
-      "loss": 2.3947,
-      "step": 720
+      "learning_rate": 1.7005221932114885e-05,
+      "loss": 2.3458,
+      "step": 1149
     },
     {
       "epoch": 3.0,
-      "eval_loss": 2.283115863800049,
-      "eval_runtime": 15.8101,
-      "eval_samples_per_second": 968.938,
-      "eval_steps_per_second": 15.18,
-      "step": 720
+      "eval_loss": 2.2351295948028564,
+      "eval_runtime": 6.5331,
+      "eval_samples_per_second": 937.992,
+      "eval_steps_per_second": 14.694,
+      "step": 1149
     },
     {
       "epoch": 4.0,
-      "learning_rate": 4.05e-06,
-      "loss": 2.365,
-      "step": 960
+      "learning_rate": 1.6005221932114884e-05,
+      "loss": 2.3097,
+      "step": 1532
     },
     {
       "epoch": 4.0,
-      "eval_loss": 2.270317792892456,
-      "eval_runtime": 15.8204,
-      "eval_samples_per_second": 968.307,
-      "eval_steps_per_second": 15.17,
-      "step": 960
+      "eval_loss": 2.1917026042938232,
+      "eval_runtime": 6.5235,
+      "eval_samples_per_second": 939.379,
+      "eval_steps_per_second": 14.716,
+      "step": 1532
     },
     {
       "epoch": 5.0,
-      "learning_rate": 5.0000000000000004e-08,
-      "loss": 2.3579,
-      "step": 1200
+      "learning_rate": 1.5005221932114883e-05,
+      "loss": 2.2839,
+      "step": 1915
     },
     {
       "epoch": 5.0,
-      "eval_loss": 2.2615389823913574,
-      "eval_runtime": 15.8209,
-      "eval_samples_per_second": 968.274,
-      "eval_steps_per_second": 15.17,
-      "step": 1200
+      "eval_loss": 2.193546772003174,
+      "eval_runtime": 6.5261,
+      "eval_samples_per_second": 938.992,
+      "eval_steps_per_second": 14.71,
+      "step": 1915
     },
     {
-      "epoch": 5.0,
-      "step": 1200,
-      "total_flos": 5076760875586560.0,
-      "train_loss": 2.4186907958984376,
-      "train_runtime": 374.7542,
-      "train_samples_per_second": 204.387,
-      "train_steps_per_second": 3.202
+      "epoch": 6.0,
+      "learning_rate": 1.4005221932114883e-05,
+      "loss": 2.2611,
+      "step": 2298
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.174062490463257,
+      "eval_runtime": 6.5308,
+      "eval_samples_per_second": 938.316,
+      "eval_steps_per_second": 14.699,
+      "step": 2298
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 1.3005221932114884e-05,
+      "loss": 2.2397,
+      "step": 2681
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.151566743850708,
+      "eval_runtime": 6.525,
+      "eval_samples_per_second": 939.154,
+      "eval_steps_per_second": 14.713,
+      "step": 2681
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.2005221932114883e-05,
+      "loss": 2.2234,
+      "step": 3064
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.14640474319458,
+      "eval_runtime": 6.531,
+      "eval_samples_per_second": 938.292,
+      "eval_steps_per_second": 14.699,
+      "step": 3064
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.1005221932114883e-05,
+      "loss": 2.2121,
+      "step": 3447
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.124241590499878,
+      "eval_runtime": 6.5412,
+      "eval_samples_per_second": 936.826,
+      "eval_steps_per_second": 14.676,
+      "step": 3447
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.0005221932114884e-05,
+      "loss": 2.2041,
+      "step": 3830
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.1360511779785156,
+      "eval_runtime": 6.5352,
+      "eval_samples_per_second": 937.687,
+      "eval_steps_per_second": 14.69,
+      "step": 3830
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 9.005221932114883e-06,
+      "loss": 2.1883,
+      "step": 4213
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.1251063346862793,
+      "eval_runtime": 6.5334,
+      "eval_samples_per_second": 937.956,
+      "eval_steps_per_second": 14.694,
+      "step": 4213
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 8.005221932114883e-06,
+      "loss": 2.185,
+      "step": 4596
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.1296956539154053,
+      "eval_runtime": 6.5234,
+      "eval_samples_per_second": 939.386,
+      "eval_steps_per_second": 14.716,
+      "step": 4596
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 7.005221932114883e-06,
+      "loss": 2.1712,
+      "step": 4979
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.1061811447143555,
+      "eval_runtime": 6.5182,
+      "eval_samples_per_second": 940.138,
+      "eval_steps_per_second": 14.728,
+      "step": 4979
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 6.005221932114883e-06,
+      "loss": 2.1648,
+      "step": 5362
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.1048877239227295,
+      "eval_runtime": 6.5157,
+      "eval_samples_per_second": 940.496,
+      "eval_steps_per_second": 14.734,
+      "step": 5362
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 5.005221932114883e-06,
+      "loss": 2.1587,
+      "step": 5745
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.106553792953491,
+      "eval_runtime": 6.52,
+      "eval_samples_per_second": 939.88,
+      "eval_steps_per_second": 14.724,
+      "step": 5745
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 4.005221932114883e-06,
+      "loss": 2.1532,
+      "step": 6128
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 2.0981085300445557,
+      "eval_runtime": 6.5377,
+      "eval_samples_per_second": 937.338,
+      "eval_steps_per_second": 14.684,
+      "step": 6128
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.005221932114883e-06,
+      "loss": 2.1472,
+      "step": 6511
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 2.0925848484039307,
+      "eval_runtime": 6.5057,
+      "eval_samples_per_second": 941.95,
+      "eval_steps_per_second": 14.756,
+      "step": 6511
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 2.005221932114883e-06,
+      "loss": 2.1462,
+      "step": 6894
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 2.083235025405884,
+      "eval_runtime": 6.5322,
+      "eval_samples_per_second": 938.118,
+      "eval_steps_per_second": 14.696,
+      "step": 6894
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 1.0052219321148825e-06,
+      "loss": 2.1437,
+      "step": 7277
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 2.093729257583618,
+      "eval_runtime": 6.5339,
+      "eval_samples_per_second": 937.883,
+      "eval_steps_per_second": 14.693,
+      "step": 7277
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 5.2219321148825064e-09,
+      "loss": 2.1386,
+      "step": 7660
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 2.0927391052246094,
+      "eval_runtime": 6.5139,
+      "eval_samples_per_second": 940.762,
+      "eval_steps_per_second": 14.738,
+      "step": 7660
+    },
+    {
+      "epoch": 20.0,
+      "step": 7660,
+      "total_flos": 3.24907393591296e+16,
+      "train_loss": 2.228582644151334,
+      "train_runtime": 1800.6594,
+      "train_samples_per_second": 272.234,
+      "train_steps_per_second": 4.254
     }
   ],
-  "max_steps": 1200,
-  "num_train_epochs": 5,
-  "total_flos": 5076760875586560.0,
+  "max_steps": 7660,
+  "num_train_epochs": 20,
+  "total_flos": 3.24907393591296e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:209ad11520d3ca630da20a305bc405621a2e543e197e02fb3076d702b5d285c7
+oid sha256:b75bdebedc9876d7e86cf6128582b4594c481973d82cb1a9b2304d2f02c28ada
 size 3963