Training in progress, step 1000, checkpoint

checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de4f0f16864b1067d54276933e69fec46ec093fab2fcf8e71c66c06ae6298920
+oid sha256:51f483e00995d1286d4da9b53956ffd9ea573042fc6a53446b8be32a056a6c7e
 size 1711848436

checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f563923907ca6b8b7c8a7450fbc31c166d2a965dfe85800c5c7b49b007c9619
+oid sha256:f9ae42e9ef6b012886a0fb38fb825ea5f5c25b39ae0cf48871830669adf961f2
 size 3424043887

checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee5399c4f6e4a2e2c6513d2d5f800f2319cd6309dd8cc6ec5578f329b64b44c1
+oid sha256:5bbb70b1f37c82106b875b3d2ba55d0e1aaa83cd68bbd9ced3b25fda608e6570
 size 623

checkpoint-1000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.9742079377174377,
+  "best_metric": 0.9588962197303772,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-1000",
-  "epoch": 0.11257458065968705,
+  "epoch": 0.10665529010238908,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": false,
@@ -9,40 +9,40 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.056287290329843524,
-      "grad_norm": 150.1642303466797,
-      "learning_rate": 4.929640887087696e-07,
-      "loss": 0.3486,
+      "epoch": 0.05332764505119454,
+      "grad_norm": 414.2552795410156,
+      "learning_rate": 4.911120591581342e-07,
+      "loss": 0.3809,
       "step": 500
     },
     {
-      "epoch": 0.056287290329843524,
-      "eval_loss": 1.1265727281570435,
-      "eval_runtime": 123.7806,
-      "eval_samples_per_second": 15.948,
-      "eval_steps_per_second": 1.995,
+      "epoch": 0.05332764505119454,
+      "eval_loss": 1.1109352111816406,
+      "eval_runtime": 63.6734,
+      "eval_samples_per_second": 15.501,
+      "eval_steps_per_second": 1.947,
       "step": 500
     },
     {
-      "epoch": 0.11257458065968705,
-      "grad_norm": 366.6913146972656,
-      "learning_rate": 4.859281774175391e-07,
-      "loss": 0.2733,
+      "epoch": 0.10665529010238908,
+      "grad_norm": 45.880027770996094,
+      "learning_rate": 4.822241183162685e-07,
+      "loss": 0.2649,
       "step": 1000
     },
     {
-      "epoch": 0.11257458065968705,
-      "eval_loss": 0.9742079377174377,
-      "eval_runtime": 125.8047,
-      "eval_samples_per_second": 15.691,
-      "eval_steps_per_second": 1.963,
+      "epoch": 0.10665529010238908,
+      "eval_loss": 0.9588962197303772,
+      "eval_runtime": 62.6825,
+      "eval_samples_per_second": 15.746,
+      "eval_steps_per_second": 1.978,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 35532,
+  "max_steps": 28128,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {

checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:319be2a769ef7117b6826908976f080f9082f159f4f0ad75da1a151ad29d9bff
+oid sha256:057f6bbf1ff3e34abda0863071bbf5fdcc7edb1102f04dba05c7a15087bc3396
 size 4847