Training in progress, step 500, checkpoint

checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6332b2596ff662a3647930ab43c232fb5a367238ae67a0b3d69c1dd5c33df23b
+oid sha256:72a2a45df86236d32ccb6580cd1cde7ffea386c67f70eee3e8c4c59938073880
 size 1711848436

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa7f59fa2510b1ba99ebe53ca9dcfe10ab8ec7c1dac5d148ca0fb4c095a2bb2a
+oid sha256:f7eadcacb2a21d50fd9e4e30249ae76931b0ff6959ece88ae728a9505c82638d
 size 3424043887

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8cab932f983b1770fe5d2da2ce2853a2a3ccb29212b3da036d5eefe0f8fb07b
+oid sha256:33ec7254cf8c7ec07fb654b081c0a6c1ac995a94986780a571410952577c255e
 size 623

checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.164457082748413,
+  "best_metric": 1.2187492847442627,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-500",
-  "epoch": 0.09211495946941783,
+  "epoch": 0.056287290329843524,
   "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": 8.437943458557129,
-      "learning_rate": 2.907885040530582e-07,
-      "loss": 0.3017,
+      "epoch": 0.056287290329843524,
+      "grad_norm": 115.51148986816406,
+      "learning_rate": 2.9437127096701565e-07,
+      "loss": 0.374,
       "step": 500
     },
     {
-      "epoch": 0.09211495946941783,
-      "eval_loss": 1.164457082748413,
-      "eval_runtime": 75.2697,
-      "eval_samples_per_second": 16.036,
-      "eval_steps_per_second": 2.006,
+      "epoch": 0.056287290329843524,
+      "eval_loss": 1.2187492847442627,
+      "eval_runtime": 129.1828,
+      "eval_samples_per_second": 15.281,
+      "eval_steps_per_second": 1.912,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16284,
+  "max_steps": 26649,
   "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 sha256:76f7dfc201ccc0039551250d06ef1282a727ca0a51016f874c2eff3c580dbf8e
+oid sha256:9a1efd6ef7fc730f5d11779313050a3d39c509af14aa1c8247c09693bef546ff
 size 4847