End of training

README.md

@@ -4,6 +4,8 @@ license: apache-2.0
 base_model: albert/albert-base-v2
 tags:
 - generated_from_trainer
+datasets:
+- squad
 model-index:
 - name: squad_albert_finetuned
   results: []
@@ -14,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # squad_albert_finetuned
 
-This model is a fine-tuned version of [albert/albert-base-v2](https://huggingface.co/albert/albert-base-v2) on an unknown dataset.
+This model is a fine-tuned version of [albert/albert-base-v2](https://huggingface.co/albert/albert-base-v2) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 6.0,
-    "eval_exact_match": 79.30936613055819,
-    "eval_f1": 88.03839918996134,
-    "eval_runtime": 37.1649,
+    "epoch": 10.0,
+    "eval_exact_match": 78.23084200567644,
+    "eval_f1": 87.49820647381922,
+    "eval_runtime": 36.1971,
     "eval_samples": 10808,
-    "eval_samples_per_second": 290.812,
-    "eval_steps_per_second": 36.351,
-    "total_flos": 8808588640659456.0,
-    "train_loss": 0.4993116462560624,
-    "train_runtime": 4288.5384,
+    "eval_samples_per_second": 298.588,
+    "eval_steps_per_second": 37.323,
+    "total_flos": 1.468098106776576e+16,
+    "train_loss": 0.3405133172148832,
+    "train_runtime": 6971.8302,
     "train_samples": 88638,
-    "train_samples_per_second": 124.011,
-    "train_steps_per_second": 3.875
+    "train_samples_per_second": 127.137,
+    "train_steps_per_second": 3.973
 }

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:503e0e42fd89214862dba2efd2fd5bed48232796cb1ef8fb1f702b3933cf1300
-size 48528698
+oid sha256:9418b1b4eadc20d7bbd888a9ce103ab9967bcf7a2029a72ec4196614fd782925
+size 49124229

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_exact_match": 79.30936613055819,
-    "eval_f1": 88.03839918996134,
-    "eval_runtime": 37.1649,
+    "epoch": 10.0,
+    "eval_exact_match": 78.23084200567644,
+    "eval_f1": 87.49820647381922,
+    "eval_runtime": 36.1971,
     "eval_samples": 10808,
-    "eval_samples_per_second": 290.812,
-    "eval_steps_per_second": 36.351
+    "eval_samples_per_second": 298.588,
+    "eval_steps_per_second": 37.323
 }

runs/Feb07_00-35-09_xgpi5/events.out.tfevents.1738866795.xgpi5.32501.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12f3fa65d1eadffdc85e71c2f9b05d990cfc42a63727827b12e74f07ce8e6ba7
+size 418

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "total_flos": 8808588640659456.0,
-    "train_loss": 0.4993116462560624,
-    "train_runtime": 4288.5384,
+    "epoch": 10.0,
+    "total_flos": 1.468098106776576e+16,
+    "train_loss": 0.3405133172148832,
+    "train_runtime": 6971.8302,
     "train_samples": 88638,
-    "train_samples_per_second": 124.011,
-    "train_steps_per_second": 3.875
+    "train_samples_per_second": 127.137,
+    "train_steps_per_second": 3.973
 }

trainer_state.json

@@ -1,258 +1,412 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 6.0,
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 16620,
+  "global_step": 27700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.18050541516245489,
-      "grad_norm": 31.118467330932617,
-      "learning_rate": 2.909747292418773e-05,
-      "loss": 1.3055,
+      "grad_norm": 31.159393310546875,
+      "learning_rate": 2.9458483754512636e-05,
+      "loss": 1.3045,
       "step": 500
     },
     {
       "epoch": 0.36101083032490977,
-      "grad_norm": 16.015565872192383,
-      "learning_rate": 2.819494584837545e-05,
-      "loss": 1.0185,
+      "grad_norm": 20.769277572631836,
+      "learning_rate": 2.8916967509025274e-05,
+      "loss": 1.0186,
       "step": 1000
     },
     {
       "epoch": 0.5415162454873647,
-      "grad_norm": 16.465599060058594,
-      "learning_rate": 2.729241877256318e-05,
-      "loss": 0.9546,
+      "grad_norm": 16.63530731201172,
+      "learning_rate": 2.837545126353791e-05,
+      "loss": 0.9621,
       "step": 1500
     },
     {
       "epoch": 0.7220216606498195,
-      "grad_norm": 11.709087371826172,
-      "learning_rate": 2.6389891696750903e-05,
-      "loss": 0.9259,
+      "grad_norm": 13.377410888671875,
+      "learning_rate": 2.783393501805054e-05,
+      "loss": 0.9288,
       "step": 2000
     },
     {
       "epoch": 0.9025270758122743,
-      "grad_norm": 22.127365112304688,
-      "learning_rate": 2.548736462093863e-05,
-      "loss": 0.8966,
+      "grad_norm": 20.618640899658203,
+      "learning_rate": 2.729241877256318e-05,
+      "loss": 0.9028,
       "step": 2500
     },
     {
       "epoch": 1.0830324909747293,
-      "grad_norm": 14.972929954528809,
-      "learning_rate": 2.4584837545126353e-05,
-      "loss": 0.7928,
+      "grad_norm": 14.885836601257324,
+      "learning_rate": 2.6750902527075813e-05,
+      "loss": 0.796,
       "step": 3000
     },
     {
       "epoch": 1.263537906137184,
-      "grad_norm": 16.17904281616211,
-      "learning_rate": 2.368231046931408e-05,
-      "loss": 0.6899,
+      "grad_norm": 14.349594116210938,
+      "learning_rate": 2.6209386281588448e-05,
+      "loss": 0.6996,
       "step": 3500
     },
     {
       "epoch": 1.444043321299639,
-      "grad_norm": 48.85356521606445,
-      "learning_rate": 2.2779783393501805e-05,
-      "loss": 0.7077,
+      "grad_norm": 19.31614875793457,
+      "learning_rate": 2.5667870036101083e-05,
+      "loss": 0.7184,
       "step": 4000
     },
     {
       "epoch": 1.6245487364620939,
-      "grad_norm": 26.525949478149414,
-      "learning_rate": 2.1877256317689534e-05,
-      "loss": 0.6929,
+      "grad_norm": 40.80540466308594,
+      "learning_rate": 2.512635379061372e-05,
+      "loss": 0.7076,
       "step": 4500
     },
     {
       "epoch": 1.8050541516245486,
-      "grad_norm": 17.478593826293945,
-      "learning_rate": 2.0974729241877255e-05,
-      "loss": 0.6838,
+      "grad_norm": 13.077240943908691,
+      "learning_rate": 2.4584837545126353e-05,
+      "loss": 0.6985,
       "step": 5000
     },
     {
       "epoch": 1.9855595667870036,
-      "grad_norm": 28.06412124633789,
-      "learning_rate": 2.0072202166064983e-05,
-      "loss": 0.7085,
+      "grad_norm": 15.288841247558594,
+      "learning_rate": 2.4043321299638987e-05,
+      "loss": 0.7239,
       "step": 5500
     },
     {
       "epoch": 2.1660649819494586,
-      "grad_norm": 13.594704627990723,
-      "learning_rate": 1.9169675090252708e-05,
-      "loss": 0.5315,
+      "grad_norm": 13.094775199890137,
+      "learning_rate": 2.3501805054151626e-05,
+      "loss": 0.5435,
       "step": 6000
     },
     {
       "epoch": 2.3465703971119134,
-      "grad_norm": 18.916522979736328,
-      "learning_rate": 1.8267148014440436e-05,
-      "loss": 0.5234,
+      "grad_norm": 15.337054252624512,
+      "learning_rate": 2.296028880866426e-05,
+      "loss": 0.5383,
       "step": 6500
     },
     {
       "epoch": 2.527075812274368,
-      "grad_norm": 27.207935333251953,
-      "learning_rate": 1.7364620938628157e-05,
-      "loss": 0.5258,
+      "grad_norm": 18.61672019958496,
+      "learning_rate": 2.2418772563176895e-05,
+      "loss": 0.5502,
       "step": 7000
     },
     {
       "epoch": 2.707581227436823,
-      "grad_norm": 14.564919471740723,
-      "learning_rate": 1.6462093862815885e-05,
-      "loss": 0.5304,
+      "grad_norm": 30.570707321166992,
+      "learning_rate": 2.1877256317689534e-05,
+      "loss": 0.5557,
       "step": 7500
     },
     {
       "epoch": 2.888086642599278,
-      "grad_norm": 26.674354553222656,
-      "learning_rate": 1.555956678700361e-05,
-      "loss": 0.5306,
+      "grad_norm": 23.822866439819336,
+      "learning_rate": 2.133574007220217e-05,
+      "loss": 0.5791,
       "step": 8000
     },
     {
       "epoch": 3.068592057761733,
-      "grad_norm": 17.092002868652344,
-      "learning_rate": 1.4657039711191336e-05,
-      "loss": 0.4566,
+      "grad_norm": 22.35167121887207,
+      "learning_rate": 2.07942238267148e-05,
+      "loss": 0.5003,
       "step": 8500
     },
     {
       "epoch": 3.2490974729241877,
-      "grad_norm": 26.201404571533203,
-      "learning_rate": 1.3754512635379063e-05,
-      "loss": 0.3703,
+      "grad_norm": 27.621000289916992,
+      "learning_rate": 2.0252707581227438e-05,
+      "loss": 0.4026,
       "step": 9000
     },
     {
       "epoch": 3.4296028880866425,
-      "grad_norm": 7.88193416595459,
-      "learning_rate": 1.2851985559566788e-05,
-      "loss": 0.3834,
+      "grad_norm": 9.574864387512207,
+      "learning_rate": 1.9711191335740073e-05,
+      "loss": 0.4173,
       "step": 9500
     },
     {
       "epoch": 3.6101083032490973,
-      "grad_norm": 17.27417755126953,
-      "learning_rate": 1.1949458483754514e-05,
-      "loss": 0.3878,
+      "grad_norm": 15.56059741973877,
+      "learning_rate": 1.9169675090252708e-05,
+      "loss": 0.4248,
       "step": 10000
     },
     {
       "epoch": 3.7906137184115525,
-      "grad_norm": 19.228748321533203,
-      "learning_rate": 1.1046931407942239e-05,
-      "loss": 0.3887,
+      "grad_norm": 24.032983779907227,
+      "learning_rate": 1.8628158844765346e-05,
+      "loss": 0.4293,
       "step": 10500
     },
     {
       "epoch": 3.9711191335740073,
-      "grad_norm": 25.48954963684082,
-      "learning_rate": 1.0144404332129965e-05,
-      "loss": 0.3818,
+      "grad_norm": 22.849510192871094,
+      "learning_rate": 1.808664259927798e-05,
+      "loss": 0.4242,
       "step": 11000
     },
     {
       "epoch": 4.1516245487364625,
-      "grad_norm": 27.043731689453125,
-      "learning_rate": 9.24187725631769e-06,
-      "loss": 0.2671,
+      "grad_norm": 20.892309188842773,
+      "learning_rate": 1.7545126353790612e-05,
+      "loss": 0.3113,
       "step": 11500
     },
     {
       "epoch": 4.332129963898917,
-      "grad_norm": 19.578210830688477,
-      "learning_rate": 8.339350180505416e-06,
-      "loss": 0.2483,
+      "grad_norm": 30.581384658813477,
+      "learning_rate": 1.7003610108303247e-05,
+      "loss": 0.2946,
       "step": 12000
     },
     {
       "epoch": 4.512635379061372,
-      "grad_norm": 21.307836532592773,
-      "learning_rate": 7.436823104693141e-06,
-      "loss": 0.2513,
+      "grad_norm": 21.06529426574707,
+      "learning_rate": 1.6462093862815885e-05,
+      "loss": 0.3022,
       "step": 12500
     },
     {
       "epoch": 4.693140794223827,
-      "grad_norm": 21.601924896240234,
-      "learning_rate": 6.534296028880867e-06,
-      "loss": 0.2517,
+      "grad_norm": 28.848838806152344,
+      "learning_rate": 1.592057761732852e-05,
+      "loss": 0.3097,
       "step": 13000
     },
     {
       "epoch": 4.873646209386282,
-      "grad_norm": 20.635147094726562,
-      "learning_rate": 5.631768953068592e-06,
-      "loss": 0.2527,
+      "grad_norm": 22.211469650268555,
+      "learning_rate": 1.5379061371841155e-05,
+      "loss": 0.3149,
       "step": 13500
     },
     {
       "epoch": 5.054151624548736,
-      "grad_norm": 21.017452239990234,
-      "learning_rate": 4.729241877256318e-06,
-      "loss": 0.2183,
+      "grad_norm": 15.2411470413208,
+      "learning_rate": 1.483754512635379e-05,
+      "loss": 0.2761,
       "step": 14000
     },
     {
       "epoch": 5.234657039711191,
-      "grad_norm": 18.656475067138672,
-      "learning_rate": 3.826714801444043e-06,
-      "loss": 0.1441,
+      "grad_norm": 22.11151885986328,
+      "learning_rate": 1.4296028880866426e-05,
+      "loss": 0.1936,
       "step": 14500
     },
     {
       "epoch": 5.415162454873646,
-      "grad_norm": 16.917116165161133,
-      "learning_rate": 2.924187725631769e-06,
-      "loss": 0.1417,
+      "grad_norm": 19.266048431396484,
+      "learning_rate": 1.3754512635379063e-05,
+      "loss": 0.1974,
       "step": 15000
     },
     {
       "epoch": 5.595667870036101,
-      "grad_norm": 25.11797332763672,
-      "learning_rate": 2.0216606498194946e-06,
-      "loss": 0.1363,
+      "grad_norm": 31.92010498046875,
+      "learning_rate": 1.3212996389891696e-05,
+      "loss": 0.2037,
       "step": 15500
     },
     {
       "epoch": 5.776173285198556,
-      "grad_norm": 25.358491897583008,
-      "learning_rate": 1.1191335740072204e-06,
-      "loss": 0.1341,
+      "grad_norm": 23.033552169799805,
+      "learning_rate": 1.2671480144404333e-05,
+      "loss": 0.1987,
       "step": 16000
     },
     {
       "epoch": 5.956678700361011,
-      "grad_norm": 24.253881454467773,
-      "learning_rate": 2.1660649819494586e-07,
-      "loss": 0.1334,
+      "grad_norm": 26.96350860595703,
+      "learning_rate": 1.2129963898916967e-05,
+      "loss": 0.2043,
       "step": 16500
     },
     {
-      "epoch": 6.0,
-      "step": 16620,
-      "total_flos": 8808588640659456.0,
-      "train_loss": 0.4993116462560624,
-      "train_runtime": 4288.5384,
-      "train_samples_per_second": 124.011,
-      "train_steps_per_second": 3.875
+      "epoch": 6.137184115523466,
+      "grad_norm": 7.384120941162109,
+      "learning_rate": 1.1588447653429602e-05,
+      "loss": 0.141,
+      "step": 17000
+    },
+    {
+      "epoch": 6.317689530685921,
+      "grad_norm": 9.360136985778809,
+      "learning_rate": 1.1046931407942239e-05,
+      "loss": 0.1172,
+      "step": 17500
+    },
+    {
+      "epoch": 6.498194945848375,
+      "grad_norm": 12.749518394470215,
+      "learning_rate": 1.0505415162454874e-05,
+      "loss": 0.1122,
+      "step": 18000
+    },
+    {
+      "epoch": 6.67870036101083,
+      "grad_norm": 22.48105812072754,
+      "learning_rate": 9.96389891696751e-06,
+      "loss": 0.115,
+      "step": 18500
+    },
+    {
+      "epoch": 6.859205776173285,
+      "grad_norm": 17.317697525024414,
+      "learning_rate": 9.422382671480145e-06,
+      "loss": 0.1235,
+      "step": 19000
+    },
+    {
+      "epoch": 7.03971119133574,
+      "grad_norm": 14.928522109985352,
+      "learning_rate": 8.88086642599278e-06,
+      "loss": 0.1057,
+      "step": 19500
+    },
+    {
+      "epoch": 7.2202166064981945,
+      "grad_norm": 32.07418441772461,
+      "learning_rate": 8.339350180505416e-06,
+      "loss": 0.0537,
+      "step": 20000
+    },
+    {
+      "epoch": 7.40072202166065,
+      "grad_norm": 27.14509391784668,
+      "learning_rate": 7.79783393501805e-06,
+      "loss": 0.0562,
+      "step": 20500
+    },
+    {
+      "epoch": 7.581227436823105,
+      "grad_norm": 8.308027267456055,
+      "learning_rate": 7.256317689530686e-06,
+      "loss": 0.0635,
+      "step": 21000
+    },
+    {
+      "epoch": 7.76173285198556,
+      "grad_norm": 33.829307556152344,
+      "learning_rate": 6.714801444043321e-06,
+      "loss": 0.0589,
+      "step": 21500
+    },
+    {
+      "epoch": 7.9422382671480145,
+      "grad_norm": 38.503910064697266,
+      "learning_rate": 6.173285198555957e-06,
+      "loss": 0.0612,
+      "step": 22000
+    },
+    {
+      "epoch": 8.12274368231047,
+      "grad_norm": 55.769901275634766,
+      "learning_rate": 5.631768953068592e-06,
+      "loss": 0.0355,
+      "step": 22500
+    },
+    {
+      "epoch": 8.303249097472925,
+      "grad_norm": 35.5570182800293,
+      "learning_rate": 5.090252707581227e-06,
+      "loss": 0.0256,
+      "step": 23000
+    },
+    {
+      "epoch": 8.483754512635379,
+      "grad_norm": 2.204684019088745,
+      "learning_rate": 4.548736462093863e-06,
+      "loss": 0.028,
+      "step": 23500
+    },
+    {
+      "epoch": 8.664259927797834,
+      "grad_norm": 0.09754960983991623,
+      "learning_rate": 4.0072202166064985e-06,
+      "loss": 0.0287,
+      "step": 24000
+    },
+    {
+      "epoch": 8.844765342960288,
+      "grad_norm": 5.562902927398682,
+      "learning_rate": 3.4657039711191337e-06,
+      "loss": 0.0284,
+      "step": 24500
+    },
+    {
+      "epoch": 9.025270758122744,
+      "grad_norm": 0.08126406371593475,
+      "learning_rate": 2.924187725631769e-06,
+      "loss": 0.0247,
+      "step": 25000
+    },
+    {
+      "epoch": 9.205776173285198,
+      "grad_norm": 1.6277568340301514,
+      "learning_rate": 2.3826714801444047e-06,
+      "loss": 0.0104,
+      "step": 25500
+    },
+    {
+      "epoch": 9.386281588447654,
+      "grad_norm": 0.16050289571285248,
+      "learning_rate": 1.8411552346570397e-06,
+      "loss": 0.0105,
+      "step": 26000
+    },
+    {
+      "epoch": 9.566787003610107,
+      "grad_norm": 1.6261656284332275,
+      "learning_rate": 1.299638989169675e-06,
+      "loss": 0.0108,
+      "step": 26500
+    },
+    {
+      "epoch": 9.747292418772563,
+      "grad_norm": 0.2381311058998108,
+      "learning_rate": 7.581227436823105e-07,
+      "loss": 0.0093,
+      "step": 27000
+    },
+    {
+      "epoch": 9.927797833935019,
+      "grad_norm": 1.3903249502182007,
+      "learning_rate": 2.1660649819494586e-07,
+      "loss": 0.0086,
+      "step": 27500
+    },
+    {
+      "epoch": 10.0,
+      "step": 27700,
+      "total_flos": 1.468098106776576e+16,
+      "train_loss": 0.3405133172148832,
+      "train_runtime": 6971.8302,
+      "train_samples_per_second": 127.137,
+      "train_steps_per_second": 3.973
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16620,
+  "max_steps": 27700,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -266,7 +420,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8808588640659456.0,
+  "total_flos": 1.468098106776576e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null