Training in progress, step 500, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01fa6858421fdb015bb654a425c3f690795d9e1cf96f456f6b27a6326b7d244c
+oid sha256:8dacb143c7ab743e9aaddb52ceaf8b2d3084857f91906714dacd8f4f10e36a82
 size 386379

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4223bec74f7da5212aa0de8c7257e806561a45c5fccd2838265f2d310699b11
+oid sha256:a2c3217f16b577cba185ae98ad2a5674a8db2b9276059cf21350bdea694fad62
 size 1540661735

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b815622b7dfb0359193f2c3e6cdc190a4c07e6c28cfbaf526d6da260173bd4a4
+oid sha256:5fb8116194227284009af0f32eb6c9bf39b120912678b54e07fcb6539f29b5b2
 size 14455

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94b7b1a014f5dad743694fda5858015bfe3bd4c82eac608d649238b91858a560
+oid sha256:b3161ebc22e167d90d7ee5cc1bfde367c6af5f0b04db44c912617eb26a593223
 size 1465

last-checkpoint/trainer_state.json

@@ -2,172 +2,92 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 0.5,
   "eval_steps": 100,
-  "global_step": 1000,
+  "global_step": 500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.1,
-      "grad_norm": 2.191904306411743,
+      "grad_norm": 1.297243595123291,
       "learning_rate": 0.0003973800426880847,
-      "loss": 9.861549072265625,
+      "loss": 9.179060668945313,
       "step": 100
     },
     {
       "epoch": 0.1,
-      "eval_accuracy": 0.09354794520547945,
-      "eval_loss": 9.128084182739258,
-      "eval_runtime": 32.2982,
-      "eval_samples_per_second": 30.962,
-      "eval_steps_per_second": 30.962,
+      "eval_accuracy": 0.09053816046966733,
+      "eval_loss": 8.474776268005371,
+      "eval_runtime": 32.6103,
+      "eval_samples_per_second": 30.665,
+      "eval_steps_per_second": 30.665,
       "step": 100
     },
     {
       "epoch": 0.2,
-      "grad_norm": 1.9173266887664795,
+      "grad_norm": 1.142166256904602,
       "learning_rate": 0.0003762085737488283,
-      "loss": 8.851971435546876,
+      "loss": 8.30135498046875,
       "step": 200
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.10289236790606654,
-      "eval_loss": 8.73751163482666,
-      "eval_runtime": 26.6022,
-      "eval_samples_per_second": 37.591,
-      "eval_steps_per_second": 37.591,
+      "eval_accuracy": 0.10549706457925637,
+      "eval_loss": 8.138654708862305,
+      "eval_runtime": 26.7248,
+      "eval_samples_per_second": 37.418,
+      "eval_steps_per_second": 37.418,
       "step": 200
     },
     {
       "epoch": 0.3,
-      "grad_norm": 2.392211437225342,
+      "grad_norm": 1.286571741104126,
       "learning_rate": 0.00033594217168615465,
-      "loss": 8.558801879882813,
+      "loss": 7.987710571289062,
       "step": 300
     },
     {
       "epoch": 0.3,
-      "eval_accuracy": 0.10008610567514677,
-      "eval_loss": 8.624308586120605,
-      "eval_runtime": 26.8044,
-      "eval_samples_per_second": 37.307,
-      "eval_steps_per_second": 37.307,
+      "eval_accuracy": 0.11159295499021527,
+      "eval_loss": 7.991443157196045,
+      "eval_runtime": 26.5785,
+      "eval_samples_per_second": 37.624,
+      "eval_steps_per_second": 37.624,
       "step": 300
     },
     {
       "epoch": 0.4,
-      "grad_norm": 3.655690908432007,
+      "grad_norm": 1.262372612953186,
       "learning_rate": 0.00028094432596115747,
-      "loss": 8.405433349609375,
+      "loss": 7.810901489257812,
       "step": 400
     },
     {
       "epoch": 0.4,
-      "eval_accuracy": 0.10092172211350293,
-      "eval_loss": 8.508211135864258,
-      "eval_runtime": 26.8389,
-      "eval_samples_per_second": 37.259,
-      "eval_steps_per_second": 37.259,
+      "eval_accuracy": 0.10483365949119373,
+      "eval_loss": 7.9381818771362305,
+      "eval_runtime": 26.6986,
+      "eval_samples_per_second": 37.455,
+      "eval_steps_per_second": 37.455,
       "step": 400
     },
     {
       "epoch": 0.5,
-      "grad_norm": 8.362494468688965,
+      "grad_norm": 1.2876827716827393,
       "learning_rate": 0.00021717490653764342,
-      "loss": 8.498648681640624,
+      "loss": 7.900021362304687,
       "step": 500
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.07594911937377691,
-      "eval_loss": 9.306892395019531,
-      "eval_runtime": 26.9571,
-      "eval_samples_per_second": 37.096,
-      "eval_steps_per_second": 37.096,
+      "eval_accuracy": 0.11435420743639922,
+      "eval_loss": 7.851724624633789,
+      "eval_runtime": 26.5522,
+      "eval_samples_per_second": 37.662,
+      "eval_steps_per_second": 37.662,
       "step": 500
-    },
-    {
-      "epoch": 0.6,
-      "grad_norm": 31.606178283691406,
-      "learning_rate": 0.00015154431949464275,
-      "loss": 11.733912353515626,
-      "step": 600
-    },
-    {
-      "epoch": 0.6,
-      "eval_accuracy": 0.05017808219178082,
-      "eval_loss": 12.344472885131836,
-      "eval_runtime": 27.8267,
-      "eval_samples_per_second": 35.937,
-      "eval_steps_per_second": 35.937,
-      "step": 600
-    },
-    {
-      "epoch": 0.7,
-      "grad_norm": 60.602169036865234,
-      "learning_rate": 9.11646573017482e-05,
-      "loss": 12.59990234375,
-      "step": 700
-    },
-    {
-      "epoch": 0.7,
-      "eval_accuracy": 0.07920547945205479,
-      "eval_loss": 11.4829683303833,
-      "eval_runtime": 26.8514,
-      "eval_samples_per_second": 37.242,
-      "eval_steps_per_second": 37.242,
-      "step": 700
-    },
-    {
-      "epoch": 0.8,
-      "grad_norm": 28.155149459838867,
-      "learning_rate": 4.2578993244549506e-05,
-      "loss": 10.478193359375,
-      "step": 800
-    },
-    {
-      "epoch": 0.8,
-      "eval_accuracy": 0.08408610567514677,
-      "eval_loss": 9.93655776977539,
-      "eval_runtime": 26.6366,
-      "eval_samples_per_second": 37.542,
-      "eval_steps_per_second": 37.542,
-      "step": 800
-    },
-    {
-      "epoch": 0.9,
-      "grad_norm": 26.28119468688965,
-      "learning_rate": 1.1052337907897503e-05,
-      "loss": 9.994468383789062,
-      "step": 900
-    },
-    {
-      "epoch": 0.9,
-      "eval_accuracy": 0.0641252446183953,
-      "eval_loss": 9.959942817687988,
-      "eval_runtime": 26.6315,
-      "eval_samples_per_second": 37.549,
-      "eval_steps_per_second": 37.549,
-      "step": 900
-    },
-    {
-      "epoch": 1.0,
-      "grad_norm": 41.307456970214844,
-      "learning_rate": 1.093583978573065e-09,
-      "loss": 9.751447143554687,
-      "step": 1000
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.06885127201565558,
-      "eval_loss": 9.537232398986816,
-      "eval_runtime": 26.8522,
-      "eval_samples_per_second": 37.241,
-      "eval_steps_per_second": 37.241,
-      "step": 1000
     }
   ],
   "logging_steps": 100,
@@ -182,7 +102,7 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }