Training in progress, step 72

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:905a43307f614d4f5d5bc9343535d19dfe7851432eb2dd6fa364e60c0180fd0c
+oid sha256:ea1ae08fd2d617b5d9fc0cd0b277111933d62bcd9827957104f08a7ca24187a6
 size 605156676

run-wkp17m0i/checkpoint-72/config.json

@@ -0,0 +1,22 @@
+{
+  "_name_or_path": "openai/clip-vit-base-patch32",
+  "architectures": [
+    "CLIPModel"
+  ],
+  "initializer_factor": 1.0,
+  "logit_scale_init_value": 2.6592,
+  "model_type": "clip",
+  "projection_dim": 512,
+  "text_config": {
+    "bos_token_id": 0,
+    "dropout": 0.0,
+    "eos_token_id": 2,
+    "model_type": "clip_text_model"
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.0.dev0",
+  "vision_config": {
+    "dropout": 0.0,
+    "model_type": "clip_vision_model"
+  }
+}

run-wkp17m0i/checkpoint-72/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea1ae08fd2d617b5d9fc0cd0b277111933d62bcd9827957104f08a7ca24187a6
+size 605156676

run-wkp17m0i/checkpoint-72/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01d3f5e1d30020a00db5fced074ad106848602c80c51b834c5085a9c96a0f9c3
+size 1210551612

run-wkp17m0i/checkpoint-72/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6179d8d41a2087eda18ebeb05db182a38648e16abba6571e45df9e7473ad3af0
+size 14244

run-wkp17m0i/checkpoint-72/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2b5128c0ff0a0ed9004850a0d9afe73129d4c3d81234d8b229345604368be76
+size 1064

run-wkp17m0i/checkpoint-72/trainer_state.json

@@ -0,0 +1,400 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 3,
+  "global_step": 72,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.125,
+      "grad_norm": 85.22277069091797,
+      "learning_rate": 1.0340384721335406e-06,
+      "loss": 2.3801,
+      "step": 3
+    },
+    {
+      "epoch": 0.125,
+      "eval_loss": 1.1969482898712158,
+      "eval_runtime": 129.4293,
+      "eval_samples_per_second": 2.318,
+      "eval_steps_per_second": 0.077,
+      "step": 3
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 83.9678726196289,
+      "learning_rate": 9.890802776929518e-07,
+      "loss": 2.075,
+      "step": 6
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 1.1689711809158325,
+      "eval_runtime": 45.8928,
+      "eval_samples_per_second": 6.537,
+      "eval_steps_per_second": 0.218,
+      "step": 6
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 90.35476684570312,
+      "learning_rate": 9.44122083252363e-07,
+      "loss": 2.0115,
+      "step": 9
+    },
+    {
+      "epoch": 0.375,
+      "eval_loss": 1.143768072128296,
+      "eval_runtime": 45.4543,
+      "eval_samples_per_second": 6.6,
+      "eval_steps_per_second": 0.22,
+      "step": 9
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 85.89755249023438,
+      "learning_rate": 8.991638888117744e-07,
+      "loss": 1.9893,
+      "step": 12
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.12371027469635,
+      "eval_runtime": 42.5719,
+      "eval_samples_per_second": 7.047,
+      "eval_steps_per_second": 0.235,
+      "step": 12
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 78.18171691894531,
+      "learning_rate": 8.542056943711856e-07,
+      "loss": 2.1389,
+      "step": 15
+    },
+    {
+      "epoch": 0.625,
+      "eval_loss": 1.105021357536316,
+      "eval_runtime": 42.4153,
+      "eval_samples_per_second": 7.073,
+      "eval_steps_per_second": 0.236,
+      "step": 15
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 86.41735076904297,
+      "learning_rate": 8.092474999305969e-07,
+      "loss": 2.2852,
+      "step": 18
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.0870113372802734,
+      "eval_runtime": 42.3738,
+      "eval_samples_per_second": 7.08,
+      "eval_steps_per_second": 0.236,
+      "step": 18
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 77.25016784667969,
+      "learning_rate": 7.642893054900082e-07,
+      "loss": 1.9602,
+      "step": 21
+    },
+    {
+      "epoch": 0.875,
+      "eval_loss": 1.0731050968170166,
+      "eval_runtime": 42.5284,
+      "eval_samples_per_second": 7.054,
+      "eval_steps_per_second": 0.235,
+      "step": 21
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 99.76058959960938,
+      "learning_rate": 7.193311110494194e-07,
+      "loss": 1.744,
+      "step": 24
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.0609346628189087,
+      "eval_runtime": 42.3344,
+      "eval_samples_per_second": 7.086,
+      "eval_steps_per_second": 0.236,
+      "step": 24
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 74.53927612304688,
+      "learning_rate": 6.743729166088308e-07,
+      "loss": 1.6082,
+      "step": 27
+    },
+    {
+      "epoch": 1.125,
+      "eval_loss": 1.0511267185211182,
+      "eval_runtime": 42.0569,
+      "eval_samples_per_second": 7.133,
+      "eval_steps_per_second": 0.238,
+      "step": 27
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 75.2535171508789,
+      "learning_rate": 6.294147221682421e-07,
+      "loss": 1.6979,
+      "step": 30
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 1.0436177253723145,
+      "eval_runtime": 42.4265,
+      "eval_samples_per_second": 7.071,
+      "eval_steps_per_second": 0.236,
+      "step": 30
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 63.06973648071289,
+      "learning_rate": 5.844565277276533e-07,
+      "loss": 1.3643,
+      "step": 33
+    },
+    {
+      "epoch": 1.375,
+      "eval_loss": 1.0362929105758667,
+      "eval_runtime": 42.3973,
+      "eval_samples_per_second": 7.076,
+      "eval_steps_per_second": 0.236,
+      "step": 33
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 71.73262023925781,
+      "learning_rate": 5.394983332870646e-07,
+      "loss": 1.7137,
+      "step": 36
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 1.0296127796173096,
+      "eval_runtime": 42.4202,
+      "eval_samples_per_second": 7.072,
+      "eval_steps_per_second": 0.236,
+      "step": 36
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 71.17877960205078,
+      "learning_rate": 4.945401388464759e-07,
+      "loss": 1.856,
+      "step": 39
+    },
+    {
+      "epoch": 1.625,
+      "eval_loss": 1.0239351987838745,
+      "eval_runtime": 42.7112,
+      "eval_samples_per_second": 7.024,
+      "eval_steps_per_second": 0.234,
+      "step": 39
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 63.988853454589844,
+      "learning_rate": 4.495819444058872e-07,
+      "loss": 1.5669,
+      "step": 42
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.0187673568725586,
+      "eval_runtime": 43.2279,
+      "eval_samples_per_second": 6.94,
+      "eval_steps_per_second": 0.231,
+      "step": 42
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 66.5676498413086,
+      "learning_rate": 4.0462374996529847e-07,
+      "loss": 1.6256,
+      "step": 45
+    },
+    {
+      "epoch": 1.875,
+      "eval_loss": 1.0137732028961182,
+      "eval_runtime": 45.6562,
+      "eval_samples_per_second": 6.571,
+      "eval_steps_per_second": 0.219,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 117.69229125976562,
+      "learning_rate": 3.596655555247097e-07,
+      "loss": 1.4703,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.0091525316238403,
+      "eval_runtime": 42.4335,
+      "eval_samples_per_second": 7.07,
+      "eval_steps_per_second": 0.236,
+      "step": 48
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 79.91377258300781,
+      "learning_rate": 3.1470736108412106e-07,
+      "loss": 1.6785,
+      "step": 51
+    },
+    {
+      "epoch": 2.125,
+      "eval_loss": 1.005091905593872,
+      "eval_runtime": 42.3521,
+      "eval_samples_per_second": 7.083,
+      "eval_steps_per_second": 0.236,
+      "step": 51
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 67.18437957763672,
+      "learning_rate": 2.697491666435323e-07,
+      "loss": 1.5906,
+      "step": 54
+    },
+    {
+      "epoch": 2.25,
+      "eval_loss": 1.001765251159668,
+      "eval_runtime": 51.8178,
+      "eval_samples_per_second": 5.79,
+      "eval_steps_per_second": 0.193,
+      "step": 54
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 81.36692810058594,
+      "learning_rate": 2.247909722029436e-07,
+      "loss": 1.4478,
+      "step": 57
+    },
+    {
+      "epoch": 2.375,
+      "eval_loss": 0.9993329644203186,
+      "eval_runtime": 47.5657,
+      "eval_samples_per_second": 6.307,
+      "eval_steps_per_second": 0.21,
+      "step": 57
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 65.09678649902344,
+      "learning_rate": 1.7983277776235486e-07,
+      "loss": 1.4731,
+      "step": 60
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 0.9975587725639343,
+      "eval_runtime": 42.7263,
+      "eval_samples_per_second": 7.021,
+      "eval_steps_per_second": 0.234,
+      "step": 60
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 66.04352569580078,
+      "learning_rate": 1.3487458332176616e-07,
+      "loss": 1.4989,
+      "step": 63
+    },
+    {
+      "epoch": 2.625,
+      "eval_loss": 0.9965038299560547,
+      "eval_runtime": 46.2804,
+      "eval_samples_per_second": 6.482,
+      "eval_steps_per_second": 0.216,
+      "step": 63
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 57.59038162231445,
+      "learning_rate": 8.991638888117743e-08,
+      "loss": 1.2743,
+      "step": 66
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 0.9956594109535217,
+      "eval_runtime": 46.4651,
+      "eval_samples_per_second": 6.456,
+      "eval_steps_per_second": 0.215,
+      "step": 66
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 63.55073165893555,
+      "learning_rate": 4.4958194440588714e-08,
+      "loss": 1.362,
+      "step": 69
+    },
+    {
+      "epoch": 2.875,
+      "eval_loss": 0.9950646758079529,
+      "eval_runtime": 50.7079,
+      "eval_samples_per_second": 5.916,
+      "eval_steps_per_second": 0.197,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 90.75367736816406,
+      "learning_rate": 0.0,
+      "loss": 1.3301,
+      "step": 72
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.9947907328605652,
+      "eval_runtime": 42.1412,
+      "eval_samples_per_second": 7.119,
+      "eval_steps_per_second": 0.237,
+      "step": 72
+    }
+  ],
+  "logging_steps": 3,
+  "max_steps": 72,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 523328480700102.0,
+  "train_batch_size": 128,
+  "trial_name": null,
+  "trial_params": {
+    "_wandb": {},
+    "assignments": {},
+    "decay": 0.01,
+    "learning_rate": 1.0789966665741292e-06,
+    "metric": "eval/loss",
+    "per_device_train_batch_size": 128
+  }
+}

run-wkp17m0i/checkpoint-72/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d86bb05d90cbbd00a1f1cf5f3c393b725a27d5d5db68a93d4689e5f26efeed3
+size 5112

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8dac4926eb2a580edb47f1290a3e9a1126fe3c128c5136263e3409a2c3f1104
+oid sha256:2d86bb05d90cbbd00a1f1cf5f3c393b725a27d5d5db68a93d4689e5f26efeed3
 size 5112