Training in progress, step 1000, checkpoint

checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:788f0bdb5cb7299c1388496e9591f0220f6de3798b52aaebc6cf9101e3abe26b
+oid sha256:89859ea84fc4ee063d0c713a29c3d092db63ac7ea1ba5b9969a07567cde5520a
 size 1711848436

checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90402570aef546ca2b4a1975da2da745b8dba89f098a269167a8fd2216d7c17a
+oid sha256:0e6e4c47e1823e839a085f496475a9328891f7e72112502ab9862fcc40d235ff
 size 3424043887

checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8d84e87520efeee2e1bd15198308880b513e7ba7a5a6afe6524607a3f46b58d
+oid sha256:63ec54e5fbbc6940016f34020cb76815cec13cfe34b24240388c3fe6cddf3f00
 size 623

checkpoint-1000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.3513683080673218,
+  "best_metric": 1.3823094367980957,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-1000",
-  "epoch": 0.18422991893883567,
+  "epoch": 0.10665529010238908,
   "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": 19.808820724487305,
-      "learning_rate": 4.84647506755097e-08,
-      "loss": 0.3758,
+      "epoch": 0.05332764505119454,
+      "grad_norm": 217.40109252929688,
+      "learning_rate": 4.911120591581342e-08,
+      "loss": 0.4667,
       "step": 500
     },
     {
-      "epoch": 0.09211495946941783,
-      "eval_loss": 1.418845295906067,
-      "eval_runtime": 75.2377,
-      "eval_samples_per_second": 16.042,
-      "eval_steps_per_second": 2.007,
+      "epoch": 0.05332764505119454,
+      "eval_loss": 1.4427576065063477,
+      "eval_runtime": 60.328,
+      "eval_samples_per_second": 16.361,
+      "eval_steps_per_second": 2.055,
       "step": 500
     },
     {
-      "epoch": 0.18422991893883567,
-      "grad_norm": 76.37991333007812,
-      "learning_rate": 4.69295013510194e-08,
-      "loss": 0.4108,
+      "epoch": 0.10665529010238908,
+      "grad_norm": 70.38941955566406,
+      "learning_rate": 4.822241183162685e-08,
+      "loss": 0.4534,
       "step": 1000
     },
     {
-      "epoch": 0.18422991893883567,
-      "eval_loss": 1.3513683080673218,
-      "eval_runtime": 74.9686,
-      "eval_samples_per_second": 16.1,
-      "eval_steps_per_second": 2.014,
+      "epoch": 0.10665529010238908,
+      "eval_loss": 1.3823094367980957,
+      "eval_runtime": 61.4884,
+      "eval_samples_per_second": 16.052,
+      "eval_steps_per_second": 2.017,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16284,
+  "max_steps": 28128,
   "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:cbcece3d46fd39b0ba80eb3e495fdf46c9f891af5b4bddce9de77ad5a467b625
+oid sha256:590a3eed3f4b2ba9fcb0b831d5e4c3f2410a46be8f57155ebf9fdd97c6c96524
 size 4847