Training in progress, step 11000, checkpoint

checkpoint-11000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1282e431c0766394b0b9982866ca8928da790c153c887bcabe43b42905ec398f
+oid sha256:ab2fd93d4ab5f1b4d6a7f3f793b12150496b54f9a29ab927f3e2e80265281ba9
 size 932401285

checkpoint-11000/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1c69939721df6bf123d1ddcfba9212ea779ddac9a1eba35150d8640d97f3c32
+oid sha256:7540852996d6b0cd785ef4b838cd09e4e4c75abacfbe02e6eccb22df1424c962
 size 466205145

checkpoint-11000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48cef4cfc12a208886eda95f82de9fbb17ba1372c1aa374d56d6cf9b4874b72a
+oid sha256:693029a75d1e2e45e457961ad897378394cc27d0de3b7136a2137a033d74aa1e
 size 14575

checkpoint-11000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e73e2de94eb5dd5ff0f761943ef6dbf1b30e2c58b99c2d43fe0702122f979ac
+oid sha256:4fd73144a9d8c21ba2b547686a21730d0edd0e96dacf9d27da87a56e292daa8c
 size 627

checkpoint-11000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.1149991750717163,
-  "best_model_checkpoint": "/kaggle/output/checkpoint-9000",
-  "epoch": 176.0,
+  "best_metric": 1.0987293720245361,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-11000",
+  "epoch": 0.44817470664928294,
   "eval_steps": 1000,
   "global_step": 11000,
   "is_hyper_param_search": false,
@@ -9,180 +9,180 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.02,
+      "epoch": 0.0,
       "learning_rate": 2.7777777777777777e-11,
-      "loss": 1.2395,
+      "loss": 1.029,
       "step": 1
     },
     {
-      "epoch": 16.0,
+      "epoch": 0.04,
       "learning_rate": 2.7638888888888893e-08,
-      "loss": 1.1422,
+      "loss": 1.181,
       "step": 1000
     },
     {
-      "epoch": 16.0,
-      "eval_accuracy": 0.33073852295409184,
-      "eval_loss": 1.1557353734970093,
-      "eval_runtime": 54.1805,
-      "eval_samples_per_second": 92.469,
-      "eval_steps_per_second": 11.572,
+      "epoch": 0.04,
+      "eval_accuracy": 0.3273453093812375,
+      "eval_loss": 1.1529844999313354,
+      "eval_runtime": 54.2837,
+      "eval_samples_per_second": 92.293,
+      "eval_steps_per_second": 11.55,
       "step": 1000
     },
     {
-      "epoch": 32.0,
+      "epoch": 0.08,
       "learning_rate": 5.541666666666667e-08,
-      "loss": 1.1401,
+      "loss": 1.1527,
       "step": 2000
     },
     {
-      "epoch": 32.0,
-      "eval_accuracy": 0.3311377245508982,
-      "eval_loss": 1.1507551670074463,
-      "eval_runtime": 53.98,
-      "eval_samples_per_second": 92.812,
-      "eval_steps_per_second": 11.615,
+      "epoch": 0.08,
+      "eval_accuracy": 0.33013972055888224,
+      "eval_loss": 1.1351025104522705,
+      "eval_runtime": 54.0518,
+      "eval_samples_per_second": 92.689,
+      "eval_steps_per_second": 11.6,
       "step": 2000
     },
     {
-      "epoch": 48.0,
+      "epoch": 0.12,
       "learning_rate": 8.316666666666666e-08,
-      "loss": 1.1287,
+      "loss": 1.142,
       "step": 3000
     },
     {
-      "epoch": 48.0,
-      "eval_accuracy": 0.3321357285429142,
-      "eval_loss": 1.1441665887832642,
-      "eval_runtime": 53.9873,
-      "eval_samples_per_second": 92.8,
-      "eval_steps_per_second": 11.614,
+      "epoch": 0.12,
+      "eval_accuracy": 0.3317365269461078,
+      "eval_loss": 1.127414345741272,
+      "eval_runtime": 54.0871,
+      "eval_samples_per_second": 92.628,
+      "eval_steps_per_second": 11.592,
       "step": 3000
     },
     {
-      "epoch": 64.0,
-      "learning_rate": 1.1094444444444445e-07,
-      "loss": 1.117,
+      "epoch": 0.16,
+      "learning_rate": 1.1091666666666668e-07,
+      "loss": 1.1371,
       "step": 4000
     },
     {
-      "epoch": 64.0,
-      "eval_accuracy": 0.3287425149700599,
-      "eval_loss": 1.134506344795227,
-      "eval_runtime": 54.0067,
-      "eval_samples_per_second": 92.766,
-      "eval_steps_per_second": 11.61,
+      "epoch": 0.16,
+      "eval_accuracy": 0.331936127744511,
+      "eval_loss": 1.1194497346878052,
+      "eval_runtime": 54.3907,
+      "eval_samples_per_second": 92.111,
+      "eval_steps_per_second": 11.528,
       "step": 4000
     },
     {
-      "epoch": 80.0,
-      "learning_rate": 1.3872222222222222e-07,
-      "loss": 1.1111,
+      "epoch": 0.2,
+      "learning_rate": 1.3869444444444447e-07,
+      "loss": 1.1246,
       "step": 5000
     },
     {
-      "epoch": 80.0,
-      "eval_accuracy": 0.32954091816367265,
-      "eval_loss": 1.1283142566680908,
-      "eval_runtime": 54.0921,
-      "eval_samples_per_second": 92.62,
-      "eval_steps_per_second": 11.591,
+      "epoch": 0.2,
+      "eval_accuracy": 0.3345309381237525,
+      "eval_loss": 1.1131775379180908,
+      "eval_runtime": 54.3492,
+      "eval_samples_per_second": 92.182,
+      "eval_steps_per_second": 11.537,
       "step": 5000
     },
     {
-      "epoch": 96.0,
-      "learning_rate": 1.6650000000000002e-07,
-      "loss": 1.1053,
+      "epoch": 0.24,
+      "learning_rate": 1.664722222222222e-07,
+      "loss": 1.1201,
       "step": 6000
     },
     {
-      "epoch": 96.0,
-      "eval_accuracy": 0.32894211576846305,
-      "eval_loss": 1.1218452453613281,
-      "eval_runtime": 54.0607,
-      "eval_samples_per_second": 92.674,
-      "eval_steps_per_second": 11.598,
+      "epoch": 0.24,
+      "eval_accuracy": 0.34151696606786425,
+      "eval_loss": 1.1093512773513794,
+      "eval_runtime": 54.3536,
+      "eval_samples_per_second": 92.174,
+      "eval_steps_per_second": 11.536,
       "step": 6000
     },
     {
-      "epoch": 112.0,
-      "learning_rate": 1.9422222222222223e-07,
-      "loss": 1.0952,
+      "epoch": 0.29,
+      "learning_rate": 1.9425e-07,
+      "loss": 1.115,
       "step": 7000
     },
     {
-      "epoch": 112.0,
-      "eval_accuracy": 0.33073852295409184,
-      "eval_loss": 1.118302583694458,
-      "eval_runtime": 53.9252,
-      "eval_samples_per_second": 92.906,
-      "eval_steps_per_second": 11.627,
+      "epoch": 0.29,
+      "eval_accuracy": 0.34510978043912177,
+      "eval_loss": 1.1066845655441284,
+      "eval_runtime": 53.9977,
+      "eval_samples_per_second": 92.782,
+      "eval_steps_per_second": 11.612,
       "step": 7000
     },
     {
-      "epoch": 128.0,
-      "learning_rate": 2.2197222222222225e-07,
-      "loss": 1.0909,
+      "epoch": 0.33,
+      "learning_rate": 2.2202777777777778e-07,
+      "loss": 1.1125,
       "step": 8000
     },
     {
-      "epoch": 128.0,
-      "eval_accuracy": 0.3281437125748503,
-      "eval_loss": 1.1167163848876953,
-      "eval_runtime": 54.095,
-      "eval_samples_per_second": 92.615,
-      "eval_steps_per_second": 11.591,
+      "epoch": 0.33,
+      "eval_accuracy": 0.3449101796407186,
+      "eval_loss": 1.1031057834625244,
+      "eval_runtime": 54.2562,
+      "eval_samples_per_second": 92.34,
+      "eval_steps_per_second": 11.556,
       "step": 8000
     },
     {
-      "epoch": 144.0,
-      "learning_rate": 2.4975e-07,
-      "loss": 1.0838,
+      "epoch": 0.37,
+      "learning_rate": 2.498055555555556e-07,
+      "loss": 1.1072,
       "step": 9000
     },
     {
-      "epoch": 144.0,
-      "eval_accuracy": 0.3317365269461078,
-      "eval_loss": 1.1149991750717163,
-      "eval_runtime": 54.0536,
-      "eval_samples_per_second": 92.686,
-      "eval_steps_per_second": 11.6,
+      "epoch": 0.37,
+      "eval_accuracy": 0.34870259481037924,
+      "eval_loss": 1.1044224500656128,
+      "eval_runtime": 54.0795,
+      "eval_samples_per_second": 92.641,
+      "eval_steps_per_second": 11.594,
       "step": 9000
     },
     {
-      "epoch": 160.0,
-      "learning_rate": 2.775277777777778e-07,
-      "loss": 1.0746,
+      "epoch": 0.41,
+      "learning_rate": 2.7755555555555554e-07,
+      "loss": 1.1064,
       "step": 10000
     },
     {
-      "epoch": 160.0,
-      "eval_accuracy": 0.3347305389221557,
-      "eval_loss": 1.1178398132324219,
-      "eval_runtime": 54.1273,
-      "eval_samples_per_second": 92.56,
-      "eval_steps_per_second": 11.584,
+      "epoch": 0.41,
+      "eval_accuracy": 0.3499001996007984,
+      "eval_loss": 1.1001031398773193,
+      "eval_runtime": 54.0869,
+      "eval_samples_per_second": 92.629,
+      "eval_steps_per_second": 11.592,
       "step": 10000
     },
     {
-      "epoch": 176.0,
-      "learning_rate": 3.0530555555555556e-07,
-      "loss": 1.0722,
+      "epoch": 0.45,
+      "learning_rate": 3.053333333333333e-07,
+      "loss": 1.1053,
       "step": 11000
     },
     {
-      "epoch": 176.0,
-      "eval_accuracy": 0.3373253493013972,
-      "eval_loss": 1.1214321851730347,
-      "eval_runtime": 54.046,
-      "eval_samples_per_second": 92.699,
-      "eval_steps_per_second": 11.601,
+      "epoch": 0.45,
+      "eval_accuracy": 0.36087824351297404,
+      "eval_loss": 1.0987293720245361,
+      "eval_runtime": 53.9688,
+      "eval_samples_per_second": 92.831,
+      "eval_steps_per_second": 11.618,
       "step": 11000
     }
   ],
   "logging_steps": 1000,
   "max_steps": 10000000,
-  "num_train_epochs": 161291,
+  "num_train_epochs": 408,
   "save_steps": 1000,
   "total_flos": 2.2993906434048e+16,
   "trial_name": null,

checkpoint-11000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb2cd1b7e648a386ccb79e47a451842e11001981c4363e993b678e0818db5cab
+oid sha256:5520446b18da09f4a43173b75839457981a344c0e709fd4bbcb249e3f60c859e
 size 4091