Training in progress, step 2000, checkpoint

checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce7427f17e610840ae15a294c32662423fa04d17300a1f606d319c26e430c61b
+oid sha256:62d108d1039c932b8e27f5c3af19e82fa70d2f6fecc0509ab6ef1d3467d6c53e
 size 1711848436

checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35608c706f9c1ca63b26c1cc0eb7993277202ae49b919ba8b8b5e8f121534af1
+oid sha256:7021356be905f6baa7bb35b42dda949627fabf841e336680a60caa358eddc411
 size 3424043887

checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d22c46c1f58428bfd4d5d34a0b5151ad1bfaf1d465059a0b6460303733e4abb6
+oid sha256:9fedc9d05d8c53e20b8390de6f1f8081f1a35d4ebd835fabc1f6569e6c78c6e0
 size 623

checkpoint-2000/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 1.234910249710083,
+  "best_metric": 1.2384389638900757,
   "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-2000",
-  "epoch": 0.3683919690550746,
+  "epoch": 0.36845983787767134,
   "eval_steps": 500,
   "global_step": 2000,
   "is_hyper_param_search": false,
@@ -9,68 +9,68 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.09209799226376865,
-      "grad_norm": 0.0014697719598188996,
-      "learning_rate": 4.884877509670289e-08,
-      "loss": 0.4588,
+      "epoch": 0.09211495946941783,
+      "grad_norm": 20.28152847290039,
+      "learning_rate": 4.8848563006632274e-08,
+      "loss": 0.3758,
       "step": 500
     },
     {
-      "epoch": 0.09209799226376865,
-      "eval_loss": 1.4194883108139038,
-      "eval_runtime": 78.7703,
-      "eval_samples_per_second": 15.31,
-      "eval_steps_per_second": 1.917,
+      "epoch": 0.09211495946941783,
+      "eval_loss": 1.418367862701416,
+      "eval_runtime": 73.4439,
+      "eval_samples_per_second": 16.434,
+      "eval_steps_per_second": 2.056,
       "step": 500
     },
     {
-      "epoch": 0.1841959845275373,
-      "grad_norm": 301.76043701171875,
-      "learning_rate": 4.769755019340578e-08,
-      "loss": 0.4255,
+      "epoch": 0.18422991893883567,
+      "grad_norm": 47.29538345336914,
+      "learning_rate": 4.7697126013264556e-08,
+      "loss": 0.4104,
       "step": 1000
     },
     {
-      "epoch": 0.1841959845275373,
-      "eval_loss": 1.3416553735733032,
-      "eval_runtime": 77.0051,
-      "eval_samples_per_second": 15.661,
-      "eval_steps_per_second": 1.961,
+      "epoch": 0.18422991893883567,
+      "eval_loss": 1.3491536378860474,
+      "eval_runtime": 74.3404,
+      "eval_samples_per_second": 16.236,
+      "eval_steps_per_second": 2.031,
       "step": 1000
     },
     {
-      "epoch": 0.27629397679130596,
-      "grad_norm": 35.28853225708008,
-      "learning_rate": 4.654632529010867e-08,
-      "loss": 0.3724,
+      "epoch": 0.2763448784082535,
+      "grad_norm": 0.5450155138969421,
+      "learning_rate": 4.6545689019896826e-08,
+      "loss": 0.4325,
       "step": 1500
     },
     {
-      "epoch": 0.27629397679130596,
-      "eval_loss": 1.2872973680496216,
-      "eval_runtime": 77.5439,
-      "eval_samples_per_second": 15.552,
-      "eval_steps_per_second": 1.947,
+      "epoch": 0.2763448784082535,
+      "eval_loss": 1.2877168655395508,
+      "eval_runtime": 75.3303,
+      "eval_samples_per_second": 16.023,
+      "eval_steps_per_second": 2.005,
       "step": 1500
     },
     {
-      "epoch": 0.3683919690550746,
-      "grad_norm": 338.9349060058594,
-      "learning_rate": 4.539510038681156e-08,
-      "loss": 0.3251,
+      "epoch": 0.36845983787767134,
+      "grad_norm": 0.11449883133172989,
+      "learning_rate": 4.539425202652911e-08,
+      "loss": 0.3423,
       "step": 2000
     },
     {
-      "epoch": 0.3683919690550746,
-      "eval_loss": 1.234910249710083,
-      "eval_runtime": 78.7715,
-      "eval_samples_per_second": 15.31,
-      "eval_steps_per_second": 1.917,
+      "epoch": 0.36845983787767134,
+      "eval_loss": 1.2384389638900757,
+      "eval_runtime": 76.113,
+      "eval_samples_per_second": 15.858,
+      "eval_steps_per_second": 1.984,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 21716,
+  "max_steps": 21712,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
   "save_steps": 500,

checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80d6f131f320de965fa9a783a60b2b8044e1bed4ee6dc17249153cae96f66b5c
+oid sha256:595bb6f39a89ec7030e395471c015665757cb5cacaafa6bba1613fd32b629686
 size 4847