Training in progress, step 1000, checkpoint

checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fcbdba88f588ecaac08804f53ba000e372b6c931156133c7c5664b6885637f3
+oid sha256:27c95177c30aefec1c6f3866b4065933c983d8847771bde42fe3689646ff5c90
 size 1711848436

checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a14a7233604b8b6b2a07bfd5db29925c9a3c9a555b2d52be38c74d999f1f5335
+oid sha256:2b818d7563af51bb5de6d2a4e31515b8aae0e5271d9652b09b663f05a3ea694b
 size 3424043887

checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7e93333f68b08eb85bc39aa0d9b9b77fdb38db7050275ac795f2eff8ead9f37
+oid sha256:c0917f04bcb42ae07b84a7199d7afa0a42ac1d0cb18676f484cd07cf70cd510f
 size 623

checkpoint-1000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.01627779006958,
+  "best_metric": 1.0702259540557861,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-1000",
-  "epoch": 0.18422991893883567,
+  "epoch": 0.11257458065968705,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": false,
@@ -9,38 +9,38 @@
   "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": 96.92656707763672,
+      "learning_rate": 2.9437127096701565e-07,
+      "loss": 0.3748,
       "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": 1.2226382493972778,
+      "eval_runtime": 131.2779,
+      "eval_samples_per_second": 15.037,
+      "eval_steps_per_second": 1.882,
       "step": 500
     },
     {
-      "epoch": 0.18422991893883567,
-      "grad_norm": 14.740598678588867,
-      "learning_rate": 2.815770081061164e-07,
-      "loss": 0.2899,
+      "epoch": 0.11257458065968705,
+      "grad_norm": 402.31158447265625,
+      "learning_rate": 2.887425419340313e-07,
+      "loss": 0.3057,
       "step": 1000
     },
     {
-      "epoch": 0.18422991893883567,
-      "eval_loss": 1.01627779006958,
-      "eval_runtime": 76.4934,
-      "eval_samples_per_second": 15.779,
-      "eval_steps_per_second": 1.974,
+      "epoch": 0.11257458065968705,
+      "eval_loss": 1.0702259540557861,
+      "eval_runtime": 131.0109,
+      "eval_samples_per_second": 15.067,
+      "eval_steps_per_second": 1.885,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16284,
+  "max_steps": 26649,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,

checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76f7dfc201ccc0039551250d06ef1282a727ca0a51016f874c2eff3c580dbf8e
+oid sha256:074eb2b3aae1f3310fb0b27d83d259baa7ad74a32c8433c81af819e28558f6b0
 size 4847