Training in progress, step 1500, checkpoint

checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f71b7d16d35f6949a3cb164eff0be104b449928ac0073f1ff7106afb3c0ca41
+oid sha256:deef5e72497c2c412d5c6b94e5f107663f588746e0fcfad74d6a55bc82d6be8f
 size 1711848436

checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9a08584f648ea86293b3bc37047bfe09ecba72061f45edba1dd341099d7d25e
+oid sha256:02973dfea932a3705a53959d9055939400d5f57032c7c9f0f521b0211c6b8c1e
 size 3424043887

checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1afba236d9bd01288d52a656be0f18e43a3e5d3b8d7dcb257421d5609a4e90ff
+oid sha256:0188aa9042551e68a55846eb798ea49922f6f93f8aeefe144a7e9bb547463316
 size 623

checkpoint-1500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.8726764917373657,
+  "best_metric": 0.9162159562110901,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-1500",
-  "epoch": 0.2763448784082535,
+  "epoch": 0.16886187098953057,
   "eval_steps": 500,
   "global_step": 1500,
   "is_hyper_param_search": false,
@@ -9,53 +9,53 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.09211495946941783,
-      "grad_norm": 3.0293657779693604,
-      "learning_rate": 4.884856300663227e-07,
-      "loss": 0.271,
+      "epoch": 0.056287290329843524,
+      "grad_norm": 150.1642303466797,
+      "learning_rate": 4.929640887087696e-07,
+      "loss": 0.3486,
       "step": 500
     },
     {
-      "epoch": 0.09211495946941783,
-      "eval_loss": 1.069305419921875,
-      "eval_runtime": 75.1409,
-      "eval_samples_per_second": 16.063,
-      "eval_steps_per_second": 2.01,
+      "epoch": 0.056287290329843524,
+      "eval_loss": 1.1265727281570435,
+      "eval_runtime": 123.7806,
+      "eval_samples_per_second": 15.948,
+      "eval_steps_per_second": 1.995,
       "step": 500
     },
     {
-      "epoch": 0.18422991893883567,
-      "grad_norm": 7.3040852546691895,
-      "learning_rate": 4.769712601326456e-07,
-      "loss": 0.2493,
+      "epoch": 0.11257458065968705,
+      "grad_norm": 366.6913146972656,
+      "learning_rate": 4.859281774175391e-07,
+      "loss": 0.2733,
       "step": 1000
     },
     {
-      "epoch": 0.18422991893883567,
-      "eval_loss": 0.9427077770233154,
-      "eval_runtime": 74.8926,
-      "eval_samples_per_second": 16.116,
-      "eval_steps_per_second": 2.016,
+      "epoch": 0.11257458065968705,
+      "eval_loss": 0.9742079377174377,
+      "eval_runtime": 125.8047,
+      "eval_samples_per_second": 15.691,
+      "eval_steps_per_second": 1.963,
       "step": 1000
     },
     {
-      "epoch": 0.2763448784082535,
-      "grad_norm": 0.020736657083034515,
-      "learning_rate": 4.654568901989683e-07,
-      "loss": 0.2348,
+      "epoch": 0.16886187098953057,
+      "grad_norm": 555.5016479492188,
+      "learning_rate": 4.788922661263087e-07,
+      "loss": 0.1851,
       "step": 1500
     },
     {
-      "epoch": 0.2763448784082535,
-      "eval_loss": 0.8726764917373657,
-      "eval_runtime": 76.6621,
-      "eval_samples_per_second": 15.744,
-      "eval_steps_per_second": 1.97,
+      "epoch": 0.16886187098953057,
+      "eval_loss": 0.9162159562110901,
+      "eval_runtime": 126.7606,
+      "eval_samples_per_second": 15.573,
+      "eval_steps_per_second": 1.949,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 21712,
+  "max_steps": 35532,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,

checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf76416e4a857628175385db7f617e54451cd7ffb62c1f9590ee9f30268f8c5b
+oid sha256:319be2a769ef7117b6826908976f080f9082f159f4f0ad75da1a151ad29d9bff
 size 4847