Training in progress, step 32000, checkpoint

checkpoint-32000/trainer_state.json

@@ -11,9 +11,9 @@
     {
       "epoch": 2.5,
       "eval_loss": 7.232339382171631,
-      "eval_runtime": 1.6226,
-      "eval_samples_per_second": 1272.646,
-      "eval_steps_per_second": 80.118,
+      "eval_runtime": 3.1848,
+      "eval_samples_per_second": 648.395,
+      "eval_steps_per_second": 40.819,
       "step": 2000
     },
     {
@@ -26,17 +26,17 @@
     {
       "epoch": 5.0,
       "eval_loss": 5.976784706115723,
-      "eval_runtime": 1.6382,
-      "eval_samples_per_second": 1260.5,
-      "eval_steps_per_second": 79.353,
+      "eval_runtime": 3.0375,
+      "eval_samples_per_second": 679.836,
+      "eval_steps_per_second": 42.798,
       "step": 4000
     },
     {
       "epoch": 7.5,
       "eval_loss": 5.826813220977783,
-      "eval_runtime": 3.0373,
-      "eval_samples_per_second": 679.889,
-      "eval_steps_per_second": 42.802,
+      "eval_runtime": 2.9574,
+      "eval_samples_per_second": 698.238,
+      "eval_steps_per_second": 43.957,
       "step": 6000
     },
     {
@@ -49,17 +49,17 @@
     {
       "epoch": 10.0,
       "eval_loss": 5.744414806365967,
-      "eval_runtime": 3.0316,
-      "eval_samples_per_second": 681.153,
-      "eval_steps_per_second": 42.881,
+      "eval_runtime": 2.5747,
+      "eval_samples_per_second": 802.048,
+      "eval_steps_per_second": 50.492,
       "step": 8000
     },
     {
       "epoch": 12.5,
       "eval_loss": 5.670751571655273,
-      "eval_runtime": 3.1272,
-      "eval_samples_per_second": 660.333,
-      "eval_steps_per_second": 41.571,
+      "eval_runtime": 1.618,
+      "eval_samples_per_second": 1276.247,
+      "eval_steps_per_second": 80.345,
       "step": 10000
     },
     {
@@ -72,17 +72,17 @@
     {
       "epoch": 15.0,
       "eval_loss": 5.57132625579834,
-      "eval_runtime": 3.0078,
-      "eval_samples_per_second": 686.558,
-      "eval_steps_per_second": 43.222,
+      "eval_runtime": 2.872,
+      "eval_samples_per_second": 719.021,
+      "eval_steps_per_second": 45.265,
       "step": 12000
     },
     {
       "epoch": 17.5,
       "eval_loss": 5.526998043060303,
-      "eval_runtime": 3.0343,
-      "eval_samples_per_second": 680.548,
-      "eval_steps_per_second": 42.843,
+      "eval_runtime": 3.0079,
+      "eval_samples_per_second": 686.536,
+      "eval_steps_per_second": 43.22,
       "step": 14000
     },
     {
@@ -95,17 +95,17 @@
     {
       "epoch": 20.0,
       "eval_loss": 5.448328971862793,
-      "eval_runtime": 3.0113,
-      "eval_samples_per_second": 685.757,
-      "eval_steps_per_second": 43.171,
+      "eval_runtime": 2.9966,
+      "eval_samples_per_second": 689.104,
+      "eval_steps_per_second": 43.382,
       "step": 16000
     },
     {
       "epoch": 22.5,
       "eval_loss": 5.3926005363464355,
-      "eval_runtime": 3.0569,
-      "eval_samples_per_second": 675.527,
-      "eval_steps_per_second": 42.527,
+      "eval_runtime": 3.0264,
+      "eval_samples_per_second": 682.332,
+      "eval_steps_per_second": 42.955,
       "step": 18000
     },
     {
@@ -118,17 +118,17 @@
     {
       "epoch": 25.0,
       "eval_loss": 5.231536388397217,
-      "eval_runtime": 2.8975,
-      "eval_samples_per_second": 712.671,
-      "eval_steps_per_second": 44.865,
+      "eval_runtime": 3.0403,
+      "eval_samples_per_second": 679.22,
+      "eval_steps_per_second": 42.76,
       "step": 20000
     },
     {
       "epoch": 27.5,
       "eval_loss": 4.905914306640625,
-      "eval_runtime": 3.0834,
-      "eval_samples_per_second": 669.706,
-      "eval_steps_per_second": 42.161,
+      "eval_runtime": 2.9951,
+      "eval_samples_per_second": 689.451,
+      "eval_steps_per_second": 43.404,
       "step": 22000
     },
     {
@@ -141,17 +141,17 @@
     {
       "epoch": 30.0,
       "eval_loss": 4.167966842651367,
-      "eval_runtime": 2.6664,
-      "eval_samples_per_second": 774.461,
-      "eval_steps_per_second": 48.755,
+      "eval_runtime": 2.4346,
+      "eval_samples_per_second": 848.205,
+      "eval_steps_per_second": 53.398,
       "step": 24000
     },
     {
       "epoch": 32.5,
       "eval_loss": 3.6409149169921875,
-      "eval_runtime": 2.5685,
-      "eval_samples_per_second": 803.957,
-      "eval_steps_per_second": 50.612,
+      "eval_runtime": 2.8198,
+      "eval_samples_per_second": 732.313,
+      "eval_steps_per_second": 46.102,
       "step": 26000
     },
     {
@@ -164,17 +164,17 @@
     {
       "epoch": 35.0,
       "eval_loss": 3.2063941955566406,
-      "eval_runtime": 2.8723,
-      "eval_samples_per_second": 718.927,
-      "eval_steps_per_second": 45.259,
+      "eval_runtime": 2.8452,
+      "eval_samples_per_second": 725.772,
+      "eval_steps_per_second": 45.69,
       "step": 28000
     },
     {
       "epoch": 37.5,
       "eval_loss": 3.0010480880737305,
-      "eval_runtime": 2.9312,
-      "eval_samples_per_second": 704.488,
-      "eval_steps_per_second": 44.35,
+      "eval_runtime": 3.0576,
+      "eval_samples_per_second": 675.356,
+      "eval_steps_per_second": 42.516,
       "step": 30000
     },
     {
@@ -187,9 +187,9 @@
     {
       "epoch": 40.0,
       "eval_loss": 2.850881338119507,
-      "eval_runtime": 3.0495,
-      "eval_samples_per_second": 677.15,
-      "eval_steps_per_second": 42.629,
+      "eval_runtime": 3.0048,
+      "eval_samples_per_second": 687.232,
+      "eval_steps_per_second": 43.264,
       "step": 32000
     }
   ],