Training in progress, epoch 0

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 16.842105263157894,
     "total_flos": 3.991922735555543e+17,
-    "train_loss": 0.10449823420494794,
-    "train_runtime": 538.8903,
-    "train_samples_per_second": 11.134,
-    "train_steps_per_second": 0.148
+    "train_loss": 0.12353107887029183,
+    "train_runtime": 494.7238,
+    "train_samples_per_second": 12.128,
+    "train_steps_per_second": 0.162
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18c8155d03fc5528407ef2294a6aa58559af47aa42136038b881476e8d3d78ac
+oid sha256:f2e5a8ee9ad333b5c0c0005abff277e39347028c249f1d95c9cd00f3cfa93c8d
 size 349854120

runs/Jun26_08-18-44_a4870aece808/events.out.tfevents.1719389935.a4870aece808.34.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b835b2d21c865040d98b50e5a55de806c41f65d71e94657021c94300cee191c2
+size 5057

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 16.842105263157894,
     "total_flos": 3.991922735555543e+17,
-    "train_loss": 0.10449823420494794,
-    "train_runtime": 538.8903,
-    "train_samples_per_second": 11.134,
-    "train_steps_per_second": 0.148
+    "train_loss": 0.12353107887029183,
+    "train_runtime": 494.7238,
+    "train_samples_per_second": 12.128,
+    "train_steps_per_second": 0.162
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.9857142857142858,
-  "best_model_checkpoint": "document-spoof-clip/checkpoint-9",
+  "best_model_checkpoint": "document-spoof-clip/checkpoint-52",
   "epoch": 16.842105263157894,
   "eval_steps": 500,
   "global_step": 80,
@@ -10,221 +10,221 @@
   "log_history": [
     {
       "epoch": 0.8421052631578947,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.2112182229757309,
-      "eval_runtime": 5.2092,
-      "eval_samples_per_second": 13.438,
-      "eval_steps_per_second": 0.96,
+      "eval_accuracy": 0.8285714285714286,
+      "eval_loss": 0.5162299275398254,
+      "eval_runtime": 4.4211,
+      "eval_samples_per_second": 15.833,
+      "eval_steps_per_second": 1.131,
       "step": 4
     },
     {
       "epoch": 1.8947368421052633,
-      "eval_accuracy": 0.9857142857142858,
-      "eval_loss": 0.12266874313354492,
-      "eval_runtime": 5.2038,
-      "eval_samples_per_second": 13.452,
-      "eval_steps_per_second": 0.961,
+      "eval_accuracy": 0.7714285714285715,
+      "eval_loss": 0.5837788581848145,
+      "eval_runtime": 4.4097,
+      "eval_samples_per_second": 15.874,
+      "eval_steps_per_second": 1.134,
       "step": 9
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 4.123354434967041,
+      "grad_norm": 50.87126541137695,
       "learning_rate": 4.8611111111111115e-05,
-      "loss": 0.295,
+      "loss": 0.4911,
       "step": 10
     },
     {
       "epoch": 2.9473684210526314,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.12026520818471909,
-      "eval_runtime": 5.3187,
-      "eval_samples_per_second": 13.161,
-      "eval_steps_per_second": 0.94,
+      "eval_accuracy": 0.9428571428571428,
+      "eval_loss": 0.15884919464588165,
+      "eval_runtime": 4.3778,
+      "eval_samples_per_second": 15.99,
+      "eval_steps_per_second": 1.142,
       "step": 14
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9714285714285714,
-      "eval_loss": 0.06349408626556396,
-      "eval_runtime": 5.1916,
-      "eval_samples_per_second": 13.483,
-      "eval_steps_per_second": 0.963,
+      "eval_accuracy": 0.9428571428571428,
+      "eval_loss": 0.20427295565605164,
+      "eval_runtime": 4.3631,
+      "eval_samples_per_second": 16.044,
+      "eval_steps_per_second": 1.146,
       "step": 19
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 14.247283935546875,
+      "grad_norm": 0.6421628594398499,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 0.0962,
+      "loss": 0.0819,
       "step": 20
     },
     {
       "epoch": 4.842105263157895,
-      "eval_accuracy": 0.9428571428571428,
-      "eval_loss": 0.2938525080680847,
-      "eval_runtime": 5.3233,
-      "eval_samples_per_second": 13.15,
-      "eval_steps_per_second": 0.939,
+      "eval_accuracy": 0.9714285714285714,
+      "eval_loss": 0.12607304751873016,
+      "eval_runtime": 4.3716,
+      "eval_samples_per_second": 16.013,
+      "eval_steps_per_second": 1.144,
       "step": 23
     },
     {
       "epoch": 5.894736842105263,
-      "eval_accuracy": 0.9285714285714286,
-      "eval_loss": 0.24831673502922058,
-      "eval_runtime": 5.2672,
-      "eval_samples_per_second": 13.29,
-      "eval_steps_per_second": 0.949,
+      "eval_accuracy": 0.9142857142857143,
+      "eval_loss": 0.6375396251678467,
+      "eval_runtime": 4.4518,
+      "eval_samples_per_second": 15.724,
+      "eval_steps_per_second": 1.123,
       "step": 28
     },
     {
       "epoch": 6.315789473684211,
-      "grad_norm": 36.67098617553711,
+      "grad_norm": 27.80014991760254,
       "learning_rate": 3.472222222222222e-05,
-      "loss": 0.163,
+      "loss": 0.1739,
       "step": 30
     },
     {
       "epoch": 6.947368421052632,
-      "eval_accuracy": 0.9857142857142858,
-      "eval_loss": 0.07119516283273697,
-      "eval_runtime": 5.2492,
-      "eval_samples_per_second": 13.335,
-      "eval_steps_per_second": 0.953,
+      "eval_accuracy": 0.8428571428571429,
+      "eval_loss": 1.199052095413208,
+      "eval_runtime": 4.4441,
+      "eval_samples_per_second": 15.751,
+      "eval_steps_per_second": 1.125,
       "step": 33
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9714285714285714,
-      "eval_loss": 0.04743731766939163,
-      "eval_runtime": 5.2254,
-      "eval_samples_per_second": 13.396,
-      "eval_steps_per_second": 0.957,
+      "eval_accuracy": 0.8571428571428571,
+      "eval_loss": 0.8591070175170898,
+      "eval_runtime": 4.4267,
+      "eval_samples_per_second": 15.813,
+      "eval_steps_per_second": 1.13,
       "step": 38
     },
     {
       "epoch": 8.421052631578947,
-      "grad_norm": 14.154749870300293,
+      "grad_norm": 0.22448770701885223,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.0646,
+      "loss": 0.2177,
       "step": 40
     },
     {
       "epoch": 8.842105263157894,
-      "eval_accuracy": 0.9428571428571428,
-      "eval_loss": 0.20123328268527985,
-      "eval_runtime": 5.2671,
-      "eval_samples_per_second": 13.29,
-      "eval_steps_per_second": 0.949,
+      "eval_accuracy": 0.9571428571428572,
+      "eval_loss": 0.25587812066078186,
+      "eval_runtime": 4.4536,
+      "eval_samples_per_second": 15.718,
+      "eval_steps_per_second": 1.123,
       "step": 42
     },
     {
       "epoch": 9.894736842105264,
-      "eval_accuracy": 0.9,
-      "eval_loss": 0.3587024211883545,
-      "eval_runtime": 5.2483,
-      "eval_samples_per_second": 13.338,
-      "eval_steps_per_second": 0.953,
+      "eval_accuracy": 0.9285714285714286,
+      "eval_loss": 0.14307884871959686,
+      "eval_runtime": 4.4003,
+      "eval_samples_per_second": 15.908,
+      "eval_steps_per_second": 1.136,
       "step": 47
     },
     {
       "epoch": 10.526315789473685,
-      "grad_norm": 3.9370455741882324,
+      "grad_norm": 20.779386520385742,
       "learning_rate": 2.0833333333333336e-05,
-      "loss": 0.1048,
+      "loss": 0.022,
       "step": 50
     },
     {
       "epoch": 10.947368421052632,
       "eval_accuracy": 0.9857142857142858,
-      "eval_loss": 0.04267547279596329,
-      "eval_runtime": 5.2718,
-      "eval_samples_per_second": 13.278,
-      "eval_steps_per_second": 0.948,
+      "eval_loss": 0.10236798971891403,
+      "eval_runtime": 4.4339,
+      "eval_samples_per_second": 15.788,
+      "eval_steps_per_second": 1.128,
       "step": 52
     },
     {
       "epoch": 12.0,
       "eval_accuracy": 0.9857142857142858,
-      "eval_loss": 0.01490915659815073,
-      "eval_runtime": 5.2034,
-      "eval_samples_per_second": 13.453,
-      "eval_steps_per_second": 0.961,
+      "eval_loss": 0.11362212896347046,
+      "eval_runtime": 4.4292,
+      "eval_samples_per_second": 15.804,
+      "eval_steps_per_second": 1.129,
       "step": 57
     },
     {
       "epoch": 12.631578947368421,
-      "grad_norm": 5.009355545043945,
+      "grad_norm": 1.5777080059051514,
       "learning_rate": 1.388888888888889e-05,
-      "loss": 0.0519,
+      "loss": 0.0007,
       "step": 60
     },
     {
       "epoch": 12.842105263157894,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.16158795356750488,
-      "eval_runtime": 5.284,
-      "eval_samples_per_second": 13.248,
-      "eval_steps_per_second": 0.946,
+      "eval_accuracy": 0.9714285714285714,
+      "eval_loss": 0.16094441711902618,
+      "eval_runtime": 4.4586,
+      "eval_samples_per_second": 15.7,
+      "eval_steps_per_second": 1.121,
       "step": 61
     },
     {
       "epoch": 13.894736842105264,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.2285740226507187,
-      "eval_runtime": 5.3185,
-      "eval_samples_per_second": 13.161,
-      "eval_steps_per_second": 0.94,
+      "eval_accuracy": 0.9857142857142858,
+      "eval_loss": 0.1507013887166977,
+      "eval_runtime": 4.409,
+      "eval_samples_per_second": 15.877,
+      "eval_steps_per_second": 1.134,
       "step": 66
     },
     {
       "epoch": 14.736842105263158,
-      "grad_norm": 1.6222240924835205,
+      "grad_norm": 0.21966008841991425,
       "learning_rate": 6.944444444444445e-06,
-      "loss": 0.0151,
+      "loss": 0.0006,
       "step": 70
     },
     {
       "epoch": 14.947368421052632,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.13689564168453217,
-      "eval_runtime": 5.3199,
-      "eval_samples_per_second": 13.158,
-      "eval_steps_per_second": 0.94,
+      "eval_accuracy": 0.9714285714285714,
+      "eval_loss": 0.22763238847255707,
+      "eval_runtime": 4.4029,
+      "eval_samples_per_second": 15.899,
+      "eval_steps_per_second": 1.136,
       "step": 71
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.21541821956634521,
-      "eval_runtime": 5.3461,
-      "eval_samples_per_second": 13.094,
-      "eval_steps_per_second": 0.935,
+      "eval_accuracy": 0.9714285714285714,
+      "eval_loss": 0.17074808478355408,
+      "eval_runtime": 4.3943,
+      "eval_samples_per_second": 15.93,
+      "eval_steps_per_second": 1.138,
       "step": 76
     },
     {
       "epoch": 16.842105263157894,
-      "grad_norm": 17.900920867919922,
+      "grad_norm": 0.0005645102355629206,
       "learning_rate": 0.0,
-      "loss": 0.0455,
+      "loss": 0.0002,
       "step": 80
     },
     {
       "epoch": 16.842105263157894,
-      "eval_accuracy": 0.9428571428571428,
-      "eval_loss": 0.2586957514286041,
-      "eval_runtime": 5.2529,
-      "eval_samples_per_second": 13.326,
-      "eval_steps_per_second": 0.952,
+      "eval_accuracy": 0.9857142857142858,
+      "eval_loss": 0.1557733714580536,
+      "eval_runtime": 4.4555,
+      "eval_samples_per_second": 15.711,
+      "eval_steps_per_second": 1.122,
       "step": 80
     },
     {
       "epoch": 16.842105263157894,
       "step": 80,
       "total_flos": 3.991922735555543e+17,
-      "train_loss": 0.10449823420494794,
-      "train_runtime": 538.8903,
-      "train_samples_per_second": 11.134,
-      "train_steps_per_second": 0.148
+      "train_loss": 0.12353107887029183,
+      "train_runtime": 494.7238,
+      "train_samples_per_second": 12.128,
+      "train_steps_per_second": 0.162
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:341ac03c36391a89b9a6524e771b5f51aee8eddc2e6bef5b9480cfd87fe64767
+oid sha256:0adfb5a838a23c5a99c53c5a96f08885303059143263c6bf84380b33c3e00a67
 size 5112