Training in progress, step 4000, checkpoint

checkpoint-4000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fd6b8f2b0e94dfe260a1dd8a49b317198033c4155d6d15e067bbad16ecbc290
+oid sha256:a53a44a87be1556699b5ebb8f459829fab751e1f0506e219021531fa5e57beca
 size 1711848436

checkpoint-4000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6194ce851679c69cccd1917b3eba2c3b2c9df0696f497da3cbb1251ad8fb2686
+oid sha256:e3f834d3f879fb57e4bbec33dc78495480335c1135c0e946a96de87d59c7623b
 size 3424043887

checkpoint-4000/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.0176074504852295,
+  "best_metric": 1.0195527076721191,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l17-l/checkpoint-4000",
   "epoch": 0.42662116040955633,
   "eval_steps": 500,
@@ -10,122 +10,122 @@
   "log_history": [
     {
       "epoch": 0.05332764505119454,
-      "grad_norm": 247.07174682617188,
+      "grad_norm": 246.1711883544922,
       "learning_rate": 9.822241183162683e-08,
       "loss": 0.4515,
       "step": 500
     },
     {
       "epoch": 0.05332764505119454,
-      "eval_loss": 1.38526451587677,
-      "eval_runtime": 63.6956,
-      "eval_samples_per_second": 15.496,
-      "eval_steps_per_second": 1.947,
+      "eval_loss": 1.3850886821746826,
+      "eval_runtime": 63.8439,
+      "eval_samples_per_second": 15.46,
+      "eval_steps_per_second": 1.942,
       "step": 500
     },
     {
       "epoch": 0.10665529010238908,
-      "grad_norm": 72.11767578125,
+      "grad_norm": 81.48503875732422,
       "learning_rate": 9.64448236632537e-08,
-      "loss": 0.414,
+      "loss": 0.4148,
       "step": 1000
     },
     {
       "epoch": 0.10665529010238908,
-      "eval_loss": 1.284197211265564,
-      "eval_runtime": 62.7418,
-      "eval_samples_per_second": 15.731,
-      "eval_steps_per_second": 1.976,
+      "eval_loss": 1.284144401550293,
+      "eval_runtime": 63.853,
+      "eval_samples_per_second": 15.457,
+      "eval_steps_per_second": 1.942,
       "step": 1000
     },
     {
       "epoch": 0.1599829351535836,
-      "grad_norm": 812.2269287109375,
+      "grad_norm": 718.367919921875,
       "learning_rate": 9.466723549488054e-08,
-      "loss": 0.3279,
+      "loss": 0.3281,
       "step": 1500
     },
     {
       "epoch": 0.1599829351535836,
-      "eval_loss": 1.2100127935409546,
-      "eval_runtime": 62.6215,
-      "eval_samples_per_second": 15.761,
-      "eval_steps_per_second": 1.98,
+      "eval_loss": 1.2112621068954468,
+      "eval_runtime": 64.127,
+      "eval_samples_per_second": 15.391,
+      "eval_steps_per_second": 1.934,
       "step": 1500
     },
     {
       "epoch": 0.21331058020477817,
-      "grad_norm": 4.459943771362305,
+      "grad_norm": 4.5923566818237305,
       "learning_rate": 9.288964732650739e-08,
       "loss": 0.2912,
       "step": 2000
     },
     {
       "epoch": 0.21331058020477817,
-      "eval_loss": 1.1584910154342651,
-      "eval_runtime": 64.0453,
-      "eval_samples_per_second": 15.411,
-      "eval_steps_per_second": 1.936,
+      "eval_loss": 1.159011960029602,
+      "eval_runtime": 63.7062,
+      "eval_samples_per_second": 15.493,
+      "eval_steps_per_second": 1.946,
       "step": 2000
     },
     {
       "epoch": 0.2666382252559727,
-      "grad_norm": 328.06475830078125,
+      "grad_norm": 319.4483642578125,
       "learning_rate": 9.111205915813424e-08,
-      "loss": 0.3056,
+      "loss": 0.3073,
       "step": 2500
     },
     {
       "epoch": 0.2666382252559727,
-      "eval_loss": 1.1090214252471924,
-      "eval_runtime": 63.2171,
-      "eval_samples_per_second": 15.613,
-      "eval_steps_per_second": 1.961,
+      "eval_loss": 1.1095446348190308,
+      "eval_runtime": 63.9925,
+      "eval_samples_per_second": 15.424,
+      "eval_steps_per_second": 1.938,
       "step": 2500
     },
     {
       "epoch": 0.3199658703071672,
-      "grad_norm": 0.00026008329587057233,
+      "grad_norm": 0.00027541385497897863,
       "learning_rate": 8.933447098976109e-08,
-      "loss": 0.2453,
+      "loss": 0.2446,
       "step": 3000
     },
     {
       "epoch": 0.3199658703071672,
-      "eval_loss": 1.0675184726715088,
-      "eval_runtime": 63.9757,
-      "eval_samples_per_second": 15.428,
-      "eval_steps_per_second": 1.938,
+      "eval_loss": 1.069318175315857,
+      "eval_runtime": 64.1107,
+      "eval_samples_per_second": 15.395,
+      "eval_steps_per_second": 1.934,
       "step": 3000
     },
     {
       "epoch": 0.37329351535836175,
-      "grad_norm": 7.824947834014893,
+      "grad_norm": 5.2035698890686035,
       "learning_rate": 8.755688282138793e-08,
-      "loss": 0.2721,
+      "loss": 0.273,
       "step": 3500
     },
     {
       "epoch": 0.37329351535836175,
-      "eval_loss": 1.0416266918182373,
-      "eval_runtime": 63.8858,
-      "eval_samples_per_second": 15.449,
-      "eval_steps_per_second": 1.941,
+      "eval_loss": 1.043074369430542,
+      "eval_runtime": 63.9447,
+      "eval_samples_per_second": 15.435,
+      "eval_steps_per_second": 1.939,
       "step": 3500
     },
     {
       "epoch": 0.42662116040955633,
-      "grad_norm": 4.335076332092285,
+      "grad_norm": 3.6792104244232178,
       "learning_rate": 8.57792946530148e-08,
-      "loss": 0.1925,
+      "loss": 0.193,
       "step": 4000
     },
     {
       "epoch": 0.42662116040955633,
-      "eval_loss": 1.0176074504852295,
-      "eval_runtime": 64.023,
-      "eval_samples_per_second": 15.416,
-      "eval_steps_per_second": 1.937,
+      "eval_loss": 1.0195527076721191,
+      "eval_runtime": 63.8076,
+      "eval_samples_per_second": 15.468,
+      "eval_steps_per_second": 1.943,
       "step": 4000
     }
   ],

checkpoint-4000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b52c451dc1fefed1a64b0a23a69d323685c58d3023f0abb7cb18bc82e1be68c7
+oid sha256:52c8db40c27633302368150be011282e1650a78d29cdfc5e366df6966e8e6c70
 size 4847