End of training

README.md

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

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 1.0,
-    "eval_exact_match": 89.3282876064333,
-    "eval_f1": 95.19020584030557,
-    "eval_runtime": 549.793,
+    "epoch": 6.0,
+    "eval_exact_match": 86.20624408703878,
+    "eval_f1": 93.68689780009221,
+    "eval_runtime": 564.5392,
     "eval_samples": 10808,
-    "eval_samples_per_second": 19.658,
-    "eval_steps_per_second": 2.457,
-    "total_flos": 4.123581730562765e+16,
-    "train_loss": 0.7573485501836784,
-    "train_runtime": 13379.8798,
+    "eval_samples_per_second": 19.145,
+    "eval_steps_per_second": 2.393,
+    "total_flos": 2.474149038337659e+17,
+    "train_loss": 0.28699065291924597,
+    "train_runtime": 82040.4271,
     "train_samples": 88638,
-    "train_samples_per_second": 6.625,
-    "train_steps_per_second": 0.207
+    "train_samples_per_second": 6.483,
+    "train_steps_per_second": 0.203
 }

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:959509544b9a21fd469d787e8cd834ab0fc90c4c3ed3ac6f634ceb426e377c9a
-size 48902870
+oid sha256:de3eedb04e5d94b4069007fd3f856822afad024ecde7080a727d00967f931aa8
+size 49408811

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.0,
-    "eval_exact_match": 89.3282876064333,
-    "eval_f1": 95.19020584030557,
-    "eval_runtime": 549.793,
+    "epoch": 6.0,
+    "eval_exact_match": 86.20624408703878,
+    "eval_f1": 93.68689780009221,
+    "eval_runtime": 564.5392,
     "eval_samples": 10808,
-    "eval_samples_per_second": 19.658,
-    "eval_steps_per_second": 2.457
+    "eval_samples_per_second": 19.145,
+    "eval_steps_per_second": 2.393
 }

runs/Feb02_18-12-39_xgpi9/events.out.tfevents.1738573893.xgpi9.431917.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b72ed57a6b0eb4ef19b064ff609daeae59ba456fd2619e6e24c82bc0ca64e2e1
+size 418

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.0,
-    "total_flos": 4.123581730562765e+16,
-    "train_loss": 0.7573485501836784,
-    "train_runtime": 13379.8798,
+    "epoch": 6.0,
+    "total_flos": 2.474149038337659e+17,
+    "train_loss": 0.28699065291924597,
+    "train_runtime": 82040.4271,
     "train_samples": 88638,
-    "train_samples_per_second": 6.625,
-    "train_steps_per_second": 0.207
+    "train_samples_per_second": 6.483,
+    "train_steps_per_second": 0.203
 }

trainer_state.json

@@ -1,62 +1,258 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 6.0,
   "eval_steps": 500,
-  "global_step": 2770,
+  "global_step": 16620,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.18050541516245489,
-      "grad_norm": 9.565498352050781,
-      "learning_rate": 2.4584837545126353e-05,
-      "loss": 1.0408,
+      "grad_norm": 9.628807067871094,
+      "learning_rate": 2.909747292418773e-05,
+      "loss": 1.0651,
       "step": 500
     },
     {
       "epoch": 0.36101083032490977,
-      "grad_norm": 3.8937153816223145,
-      "learning_rate": 1.9169675090252708e-05,
-      "loss": 0.7637,
+      "grad_norm": 4.762765884399414,
+      "learning_rate": 2.819494584837545e-05,
+      "loss": 0.7814,
       "step": 1000
     },
     {
       "epoch": 0.5415162454873647,
-      "grad_norm": 5.798653602600098,
-      "learning_rate": 1.3754512635379063e-05,
-      "loss": 0.716,
+      "grad_norm": 6.628077030181885,
+      "learning_rate": 2.729241877256318e-05,
+      "loss": 0.7341,
       "step": 1500
     },
     {
       "epoch": 0.7220216606498195,
-      "grad_norm": 3.4039201736450195,
-      "learning_rate": 8.339350180505416e-06,
-      "loss": 0.681,
+      "grad_norm": 5.3279032707214355,
+      "learning_rate": 2.6389891696750903e-05,
+      "loss": 0.7158,
       "step": 2000
     },
     {
       "epoch": 0.9025270758122743,
-      "grad_norm": 9.457993507385254,
-      "learning_rate": 2.924187725631769e-06,
-      "loss": 0.6554,
+      "grad_norm": 7.846349239349365,
+      "learning_rate": 2.548736462093863e-05,
+      "loss": 0.6904,
       "step": 2500
     },
     {
-      "epoch": 1.0,
-      "step": 2770,
-      "total_flos": 4.123581730562765e+16,
-      "train_loss": 0.7573485501836784,
-      "train_runtime": 13379.8798,
-      "train_samples_per_second": 6.625,
-      "train_steps_per_second": 0.207
+      "epoch": 1.0830324909747293,
+      "grad_norm": 4.142075538635254,
+      "learning_rate": 2.4584837545126353e-05,
+      "loss": 0.5878,
+      "step": 3000
+    },
+    {
+      "epoch": 1.263537906137184,
+      "grad_norm": 4.67156457901001,
+      "learning_rate": 2.368231046931408e-05,
+      "loss": 0.4853,
+      "step": 3500
+    },
+    {
+      "epoch": 1.444043321299639,
+      "grad_norm": 7.2608442306518555,
+      "learning_rate": 2.2779783393501805e-05,
+      "loss": 0.4992,
+      "step": 4000
+    },
+    {
+      "epoch": 1.6245487364620939,
+      "grad_norm": 5.837532043457031,
+      "learning_rate": 2.1877256317689534e-05,
+      "loss": 0.4827,
+      "step": 4500
+    },
+    {
+      "epoch": 1.8050541516245486,
+      "grad_norm": 5.671542167663574,
+      "learning_rate": 2.0974729241877255e-05,
+      "loss": 0.479,
+      "step": 5000
+    },
+    {
+      "epoch": 1.9855595667870036,
+      "grad_norm": 3.212416172027588,
+      "learning_rate": 2.0072202166064983e-05,
+      "loss": 0.4951,
+      "step": 5500
+    },
+    {
+      "epoch": 2.1660649819494586,
+      "grad_norm": 12.710512161254883,
+      "learning_rate": 1.9169675090252708e-05,
+      "loss": 0.286,
+      "step": 6000
+    },
+    {
+      "epoch": 2.3465703971119134,
+      "grad_norm": 7.159419059753418,
+      "learning_rate": 1.8267148014440436e-05,
+      "loss": 0.2685,
+      "step": 6500
+    },
+    {
+      "epoch": 2.527075812274368,
+      "grad_norm": 4.714431285858154,
+      "learning_rate": 1.7364620938628157e-05,
+      "loss": 0.2771,
+      "step": 7000
+    },
+    {
+      "epoch": 2.707581227436823,
+      "grad_norm": 5.595752716064453,
+      "learning_rate": 1.6462093862815885e-05,
+      "loss": 0.2697,
+      "step": 7500
+    },
+    {
+      "epoch": 2.888086642599278,
+      "grad_norm": 6.116200923919678,
+      "learning_rate": 1.555956678700361e-05,
+      "loss": 0.2755,
+      "step": 8000
+    },
+    {
+      "epoch": 3.068592057761733,
+      "grad_norm": 5.073014736175537,
+      "learning_rate": 1.4657039711191336e-05,
+      "loss": 0.2092,
+      "step": 8500
+    },
+    {
+      "epoch": 3.2490974729241877,
+      "grad_norm": 4.608720779418945,
+      "learning_rate": 1.3754512635379063e-05,
+      "loss": 0.1039,
+      "step": 9000
+    },
+    {
+      "epoch": 3.4296028880866425,
+      "grad_norm": 5.174805641174316,
+      "learning_rate": 1.2851985559566788e-05,
+      "loss": 0.1124,
+      "step": 9500
+    },
+    {
+      "epoch": 3.6101083032490973,
+      "grad_norm": 6.413619518280029,
+      "learning_rate": 1.1949458483754514e-05,
+      "loss": 0.1159,
+      "step": 10000
+    },
+    {
+      "epoch": 3.7906137184115525,
+      "grad_norm": 7.532567024230957,
+      "learning_rate": 1.1046931407942239e-05,
+      "loss": 0.1098,
+      "step": 10500
+    },
+    {
+      "epoch": 3.9711191335740073,
+      "grad_norm": 7.561917781829834,
+      "learning_rate": 1.0144404332129965e-05,
+      "loss": 0.1121,
+      "step": 11000
+    },
+    {
+      "epoch": 4.1516245487364625,
+      "grad_norm": 2.8685948848724365,
+      "learning_rate": 9.24187725631769e-06,
+      "loss": 0.0497,
+      "step": 11500
+    },
+    {
+      "epoch": 4.332129963898917,
+      "grad_norm": 3.023860454559326,
+      "learning_rate": 8.339350180505416e-06,
+      "loss": 0.0472,
+      "step": 12000
+    },
+    {
+      "epoch": 4.512635379061372,
+      "grad_norm": 8.299504280090332,
+      "learning_rate": 7.436823104693141e-06,
+      "loss": 0.043,
+      "step": 12500
+    },
+    {
+      "epoch": 4.693140794223827,
+      "grad_norm": 5.012452602386475,
+      "learning_rate": 6.534296028880867e-06,
+      "loss": 0.046,
+      "step": 13000
+    },
+    {
+      "epoch": 4.873646209386282,
+      "grad_norm": 0.5658218264579773,
+      "learning_rate": 5.631768953068592e-06,
+      "loss": 0.0477,
+      "step": 13500
+    },
+    {
+      "epoch": 5.054151624548736,
+      "grad_norm": 0.48821431398391724,
+      "learning_rate": 4.729241877256318e-06,
+      "loss": 0.0338,
+      "step": 14000
+    },
+    {
+      "epoch": 5.234657039711191,
+      "grad_norm": 4.726902484893799,
+      "learning_rate": 3.826714801444043e-06,
+      "loss": 0.0211,
+      "step": 14500
+    },
+    {
+      "epoch": 5.415162454873646,
+      "grad_norm": 0.05472424998879433,
+      "learning_rate": 2.924187725631769e-06,
+      "loss": 0.0247,
+      "step": 15000
+    },
+    {
+      "epoch": 5.595667870036101,
+      "grad_norm": 1.5054607391357422,
+      "learning_rate": 2.0216606498194946e-06,
+      "loss": 0.0224,
+      "step": 15500
+    },
+    {
+      "epoch": 5.776173285198556,
+      "grad_norm": 0.04949120432138443,
+      "learning_rate": 1.1191335740072204e-06,
+      "loss": 0.0234,
+      "step": 16000
+    },
+    {
+      "epoch": 5.956678700361011,
+      "grad_norm": 0.00990414246916771,
+      "learning_rate": 2.1660649819494586e-07,
+      "loss": 0.0202,
+      "step": 16500
+    },
+    {
+      "epoch": 6.0,
+      "step": 16620,
+      "total_flos": 2.474149038337659e+17,
+      "train_loss": 0.28699065291924597,
+      "train_runtime": 82040.4271,
+      "train_samples_per_second": 6.483,
+      "train_steps_per_second": 0.203
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2770,
+  "max_steps": 16620,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 6,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -70,7 +266,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.123581730562765e+16,
+  "total_flos": 2.474149038337659e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null