Training in progress, step 300, checkpoint

checkpoint-300/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d27840db45fa9f67419833b94359281e32f6bdaf5c4aefe20c0976ecee23be1
+oid sha256:690f09f9159427695d39ed774d84663044c4bf791a59678cb51fefc40a846afa
 size 14244

checkpoint-300/trainer_state.json

@@ -23,10 +23,14 @@
     {
       "epoch": 0.06,
       "eval_accuracy": 0.44479243019924036,
+      "eval_combined_score": 0.3403229798934559,
+      "eval_f1": 0.2738667532127967,
       "eval_loss": 1.065091609954834,
-      "eval_runtime": 50.348,
-      "eval_samples_per_second": 596.131,
-      "eval_steps_per_second": 1.172,
+      "eval_precision": 0.19784030596254612,
+      "eval_recall": 0.44479243019924036,
+      "eval_runtime": 49.6621,
+      "eval_samples_per_second": 604.364,
+      "eval_steps_per_second": 1.188,
       "step": 20
     },
     {
@@ -44,10 +48,14 @@
     {
       "epoch": 0.12,
       "eval_accuracy": 0.5033650962883988,
+      "eval_combined_score": 0.4633375988688383,
+      "eval_f1": 0.4137746858205889,
       "eval_loss": 1.0188277959823608,
-      "eval_runtime": 49.614,
-      "eval_samples_per_second": 604.95,
-      "eval_steps_per_second": 1.189,
+      "eval_precision": 0.43284551707796665,
+      "eval_recall": 0.5033650962883988,
+      "eval_runtime": 49.1985,
+      "eval_samples_per_second": 610.06,
+      "eval_steps_per_second": 1.199,
       "step": 40
     },
     {
@@ -65,10 +73,14 @@
     {
       "epoch": 0.18,
       "eval_accuracy": 0.5279203038581995,
+      "eval_combined_score": 0.48433586826626607,
+      "eval_f1": 0.4596383010419703,
       "eval_loss": 0.9871189594268799,
-      "eval_runtime": 49.5201,
-      "eval_samples_per_second": 606.098,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.421864564306695,
+      "eval_recall": 0.5279203038581995,
+      "eval_runtime": 49.0237,
+      "eval_samples_per_second": 612.235,
+      "eval_steps_per_second": 1.204,
       "step": 60
     },
     {
@@ -86,10 +98,14 @@
     {
       "epoch": 0.24,
       "eval_accuracy": 0.5308189511561271,
+      "eval_combined_score": 0.4866197693698143,
+      "eval_f1": 0.46528335920671143,
       "eval_loss": 0.9888613224029541,
-      "eval_runtime": 49.4469,
-      "eval_samples_per_second": 606.994,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.4195578159602916,
+      "eval_recall": 0.5308189511561271,
+      "eval_runtime": 49.518,
+      "eval_samples_per_second": 606.124,
+      "eval_steps_per_second": 1.191,
       "step": 80
     },
     {
@@ -107,10 +123,14 @@
     {
       "epoch": 0.3,
       "eval_accuracy": 0.5307856333710935,
+      "eval_combined_score": 0.48644275806230897,
+      "eval_f1": 0.4666072639999521,
       "eval_loss": 0.9762536883354187,
-      "eval_runtime": 49.4804,
-      "eval_samples_per_second": 606.583,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4175925015070968,
+      "eval_recall": 0.5307856333710935,
+      "eval_runtime": 49.4629,
+      "eval_samples_per_second": 606.798,
+      "eval_steps_per_second": 1.193,
       "step": 100
     },
     {
@@ -128,10 +148,14 @@
     {
       "epoch": 0.36,
       "eval_accuracy": 0.5387819017791697,
+      "eval_combined_score": 0.4944622623129249,
+      "eval_f1": 0.47105988260831677,
       "eval_loss": 0.9713281989097595,
-      "eval_runtime": 49.4983,
-      "eval_samples_per_second": 606.364,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4292253630850433,
+      "eval_recall": 0.5387819017791697,
+      "eval_runtime": 49.0682,
+      "eval_samples_per_second": 611.68,
+      "eval_steps_per_second": 1.202,
       "step": 120
     },
     {
@@ -149,10 +173,14 @@
     {
       "epoch": 0.42,
       "eval_accuracy": 0.5312520823615646,
+      "eval_combined_score": 0.48712844254868687,
+      "eval_f1": 0.4673598197970026,
       "eval_loss": 0.9766249656677246,
-      "eval_runtime": 49.5213,
-      "eval_samples_per_second": 606.083,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.41864978567461564,
+      "eval_recall": 0.5312520823615646,
+      "eval_runtime": 49.1859,
+      "eval_samples_per_second": 610.215,
+      "eval_steps_per_second": 1.2,
       "step": 140
     },
     {
@@ -170,10 +198,14 @@
     {
       "epoch": 0.48,
       "eval_accuracy": 0.5398147531152129,
+      "eval_combined_score": 0.49477075635813,
+      "eval_f1": 0.4751263005883661,
       "eval_loss": 0.9589501619338989,
-      "eval_runtime": 49.4977,
-      "eval_samples_per_second": 606.372,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4243272186137281,
+      "eval_recall": 0.5398147531152129,
+      "eval_runtime": 49.0847,
+      "eval_samples_per_second": 611.473,
+      "eval_steps_per_second": 1.202,
       "step": 160
     },
     {
@@ -191,10 +223,14 @@
     {
       "epoch": 0.54,
       "eval_accuracy": 0.5423469047777704,
+      "eval_combined_score": 0.49718028497336725,
+      "eval_f1": 0.4771728160733735,
       "eval_loss": 0.953514814376831,
-      "eval_runtime": 49.4511,
-      "eval_samples_per_second": 606.943,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.42685451426455484,
+      "eval_recall": 0.5423469047777704,
+      "eval_runtime": 49.0032,
+      "eval_samples_per_second": 612.491,
+      "eval_steps_per_second": 1.204,
       "step": 180
     },
     {
@@ -212,10 +248,14 @@
     {
       "epoch": 0.6,
       "eval_accuracy": 0.567201972412874,
+      "eval_combined_score": 0.5200447629299639,
+      "eval_f1": 0.49911443945569844,
       "eval_loss": 0.9273685812950134,
-      "eval_runtime": 49.4122,
-      "eval_samples_per_second": 607.421,
-      "eval_steps_per_second": 1.194,
+      "eval_precision": 0.44666066743840943,
+      "eval_recall": 0.567201972412874,
+      "eval_runtime": 49.1912,
+      "eval_samples_per_second": 610.15,
+      "eval_steps_per_second": 1.199,
       "step": 200
     },
     {
@@ -233,10 +273,14 @@
     {
       "epoch": 0.66,
       "eval_accuracy": 0.573598987139335,
+      "eval_combined_score": 0.5278568904198743,
+      "eval_f1": 0.5026102591641352,
       "eval_loss": 0.912590503692627,
-      "eval_runtime": 49.2827,
-      "eval_samples_per_second": 609.017,
-      "eval_steps_per_second": 1.197,
+      "eval_precision": 0.4616193282366919,
+      "eval_recall": 0.573598987139335,
+      "eval_runtime": 49.2154,
+      "eval_samples_per_second": 609.849,
+      "eval_steps_per_second": 1.199,
       "step": 220
     },
     {
@@ -254,10 +298,14 @@
     {
       "epoch": 0.72,
       "eval_accuracy": 0.5759645498767242,
+      "eval_combined_score": 0.5280177627115059,
+      "eval_f1": 0.5069205876947673,
       "eval_loss": 0.9052607417106628,
-      "eval_runtime": 49.4457,
-      "eval_samples_per_second": 607.01,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.45322136339780783,
+      "eval_recall": 0.5759645498767242,
+      "eval_runtime": 49.1399,
+      "eval_samples_per_second": 610.787,
+      "eval_steps_per_second": 1.201,
       "step": 240
     },
     {
@@ -275,10 +323,14 @@
     {
       "epoch": 0.78,
       "eval_accuracy": 0.5766975411474645,
+      "eval_combined_score": 0.534067292009198,
+      "eval_f1": 0.5018221478654629,
       "eval_loss": 0.9178985953330994,
-      "eval_runtime": 49.3132,
-      "eval_samples_per_second": 608.641,
-      "eval_steps_per_second": 1.196,
+      "eval_precision": 0.4810519378764,
+      "eval_recall": 0.5766975411474645,
+      "eval_runtime": 49.2092,
+      "eval_samples_per_second": 609.927,
+      "eval_steps_per_second": 1.199,
       "step": 260
     },
     {
@@ -296,10 +348,14 @@
     {
       "epoch": 0.84,
       "eval_accuracy": 0.58915839275005,
+      "eval_combined_score": 0.5407092882328752,
+      "eval_f1": 0.5183161389995846,
       "eval_loss": 0.8937407732009888,
-      "eval_runtime": 49.4342,
-      "eval_samples_per_second": 607.15,
-      "eval_steps_per_second": 1.194,
+      "eval_precision": 0.46620422843181647,
+      "eval_recall": 0.58915839275005,
+      "eval_runtime": 49.2523,
+      "eval_samples_per_second": 609.393,
+      "eval_steps_per_second": 1.198,
       "step": 280
     },
     {
@@ -317,18 +373,22 @@
     {
       "epoch": 0.9,
       "eval_accuracy": 0.613013926834144,
+      "eval_combined_score": 0.5923759012239074,
+      "eval_f1": 0.5579282363395321,
       "eval_loss": 0.8468813300132751,
-      "eval_runtime": 49.5046,
-      "eval_samples_per_second": 606.287,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.5855475148878095,
+      "eval_recall": 0.613013926834144,
+      "eval_runtime": 49.1896,
+      "eval_samples_per_second": 610.17,
+      "eval_steps_per_second": 1.199,
       "step": 300
     }
   ],
   "logging_steps": 10,
-  "max_steps": 996,
-  "num_train_epochs": 3,
+  "max_steps": 1660,
+  "num_train_epochs": 5,
   "save_steps": 100,
-  "total_flos": 4.04142195474432e+16,
+  "total_flos": 4.04142209630208e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-300/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f1be80ebe52f6e43af0b8aa087e72fad77310d5998b6e0b8f66a6a1d53be7b7
+oid sha256:ae98e7de1b05a570517ae68653e3c31b639a52c739d05197601f467f38b01c66
 size 4536