Training in progress, step 1500, checkpoint

checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bc6ef0fb2d0de8583064965462a8322160870fdecc136ecc27fee9449ffdb91
+oid sha256:5c8b4cb4d1f4cfd101f0d2631eef3ac3f977fdf8e85ec2d0696ff5f4097dbd03
 size 1711848436

checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea6b54653101e8b5bc17ea55096415d30ab7bccdc3f7ca222a39829d957efd75
+oid sha256:35b163731354c7912cfc04f9a3c3560dbfc0eae3b335e58c3c35d74c2de6fb45
 size 3424043887

checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff842a8e23d621812b6dd05250b27f72fa947b902effd523ae43a66bdf6c7c9b
+oid sha256:1aecc206de7b6adcc6d85263026356af89ba2890d2afe7d28c4d28048ac70f1c
 size 623

checkpoint-1500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.2895005941390991,
+  "best_metric": 1.3328827619552612,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-1500",
-  "epoch": 0.2763448784082535,
+  "epoch": 0.1599829351535836,
   "eval_steps": 500,
   "global_step": 1500,
   "is_hyper_param_search": false,
@@ -9,53 +9,53 @@
   "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
     },
     {
-      "epoch": 0.2763448784082535,
-      "grad_norm": 0.5795987844467163,
-      "learning_rate": 4.539425202652911e-08,
-      "loss": 0.4335,
+      "epoch": 0.1599829351535836,
+      "grad_norm": 389.9967956542969,
+      "learning_rate": 4.733361774744027e-08,
+      "loss": 0.3753,
       "step": 1500
     },
     {
-      "epoch": 0.2763448784082535,
-      "eval_loss": 1.2895005941390991,
-      "eval_runtime": 76.713,
-      "eval_samples_per_second": 15.734,
-      "eval_steps_per_second": 1.968,
+      "epoch": 0.1599829351535836,
+      "eval_loss": 1.3328827619552612,
+      "eval_runtime": 63.0378,
+      "eval_samples_per_second": 15.657,
+      "eval_steps_per_second": 1.967,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 16284,
+  "max_steps": 28128,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,

checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbcece3d46fd39b0ba80eb3e495fdf46c9f891af5b4bddce9de77ad5a467b625
+oid sha256:590a3eed3f4b2ba9fcb0b831d5e4c3f2410a46be8f57155ebf9fdd97c6c96524
 size 4847