Training in progress, step 500, checkpoint

checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f21b49c0e93ba4ffbfa11cc5a64cd7511696f04646f816b4359b18da32decd9
+oid sha256:bcf5fc3e19aca2619838b3f048fcb5d5e14018da5a618daf89fbd24e8f2a7b85
 size 1711848436

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb5da3de3001f4e94199ad96d1f2d7da990214f2d36c42fd5023d535debbc620
+oid sha256:8697e90ef84d2bb76c6534faa0572a6f0cf7b40e49247ed31346287833be5b50
 size 3424043887

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0d0fe33a70a763e3f9ca1bb2fbbdb7e04c2f9b1497054a82af8aff31228e46b
+oid sha256:b06b4d941cc8457a577e94374b8b4010790505f9026ab40db1cc3e46e71acfa9
 size 623

checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.0680681467056274,
+  "best_metric": 1.1163015365600586,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-500",
-  "epoch": 0.09211495946941783,
+  "epoch": 0.05332764505119454,
   "eval_steps": 500,
   "global_step": 500,
   "is_hyper_param_search": false,
@@ -9,23 +9,23 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.09211495946941783,
-      "grad_norm": 2.6780805587768555,
-      "learning_rate": 4.84647506755097e-07,
-      "loss": 0.2705,
+      "epoch": 0.05332764505119454,
+      "grad_norm": 414.62841796875,
+      "learning_rate": 4.911120591581342e-07,
+      "loss": 0.3812,
       "step": 500
     },
     {
-      "epoch": 0.09211495946941783,
-      "eval_loss": 1.0680681467056274,
-      "eval_runtime": 74.5723,
-      "eval_samples_per_second": 16.186,
-      "eval_steps_per_second": 2.025,
+      "epoch": 0.05332764505119454,
+      "eval_loss": 1.1163015365600586,
+      "eval_runtime": 61.8962,
+      "eval_samples_per_second": 15.946,
+      "eval_steps_per_second": 2.003,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16284,
+  "max_steps": 28128,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,

checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea34fd4333686a44a60d290faa4ae8e00df783131842d9ae3ef1b9e874a39349
+oid sha256:bd16a1abd112ae60a2700f590eb3bd73e96acd9a0a37b30bbfa3f8ee2dc23bc1
 size 4847