Training in progress, step 72

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55c33c7b3596e8ba7511ab8ff8c828a3f17a882d1e28479f917f1ea2162eae5a
+oid sha256:64905993dda8d89fa0069bfc1c4e77cbb782c39b0b6fb9cd7937ffac94d61a3b
 size 605156676

run-w7wglxd2/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-w7wglxd2/checkpoint-72/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64905993dda8d89fa0069bfc1c4e77cbb782c39b0b6fb9cd7937ffac94d61a3b
+size 605156676

run-w7wglxd2/checkpoint-72/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d28ae51be69b0b202c1bdda0ce970ac376bfd7639eb1e8c4332a304ccdee7fd9
+size 1210551612

run-w7wglxd2/checkpoint-72/rng_state.pth

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

run-w7wglxd2/checkpoint-72/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:283ca5e628b8eff22a22adc04176859b08fdaa57a3657c77bc06e99e0c8e270e
+size 1064

run-w7wglxd2/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": 54.31294631958008,
+      "learning_rate": 6.447096451332965e-05,
+      "loss": 3.0024,
+      "step": 3
+    },
+    {
+      "epoch": 0.125,
+      "eval_loss": 2.225553035736084,
+      "eval_runtime": 106.6156,
+      "eval_samples_per_second": 2.814,
+      "eval_steps_per_second": 0.047,
+      "step": 3
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 45.66840362548828,
+      "learning_rate": 6.166787909970662e-05,
+      "loss": 2.7338,
+      "step": 6
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 2.1332547664642334,
+      "eval_runtime": 55.7845,
+      "eval_samples_per_second": 5.378,
+      "eval_steps_per_second": 0.09,
+      "step": 6
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 34.8353157043457,
+      "learning_rate": 5.886479368608359e-05,
+      "loss": 2.6629,
+      "step": 9
+    },
+    {
+      "epoch": 0.375,
+      "eval_loss": 2.0127696990966797,
+      "eval_runtime": 55.1254,
+      "eval_samples_per_second": 5.442,
+      "eval_steps_per_second": 0.091,
+      "step": 9
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 37.54914474487305,
+      "learning_rate": 5.606170827246056e-05,
+      "loss": 2.4975,
+      "step": 12
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.880717396736145,
+      "eval_runtime": 57.0215,
+      "eval_samples_per_second": 5.261,
+      "eval_steps_per_second": 0.088,
+      "step": 12
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 41.305789947509766,
+      "learning_rate": 5.325862285883753e-05,
+      "loss": 2.6358,
+      "step": 15
+    },
+    {
+      "epoch": 0.625,
+      "eval_loss": 1.8344547748565674,
+      "eval_runtime": 55.994,
+      "eval_samples_per_second": 5.358,
+      "eval_steps_per_second": 0.089,
+      "step": 15
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 34.800052642822266,
+      "learning_rate": 5.04555374452145e-05,
+      "loss": 2.5937,
+      "step": 18
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.7536548376083374,
+      "eval_runtime": 55.2931,
+      "eval_samples_per_second": 5.426,
+      "eval_steps_per_second": 0.09,
+      "step": 18
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 35.47196578979492,
+      "learning_rate": 4.765245203159148e-05,
+      "loss": 2.3824,
+      "step": 21
+    },
+    {
+      "epoch": 0.875,
+      "eval_loss": 1.6810842752456665,
+      "eval_runtime": 54.3115,
+      "eval_samples_per_second": 5.524,
+      "eval_steps_per_second": 0.092,
+      "step": 21
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 43.1494026184082,
+      "learning_rate": 4.484936661796845e-05,
+      "loss": 1.9198,
+      "step": 24
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.645703673362732,
+      "eval_runtime": 54.8632,
+      "eval_samples_per_second": 5.468,
+      "eval_steps_per_second": 0.091,
+      "step": 24
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 22.77340316772461,
+      "learning_rate": 4.2046281204345424e-05,
+      "loss": 0.8103,
+      "step": 27
+    },
+    {
+      "epoch": 1.125,
+      "eval_loss": 1.616202473640442,
+      "eval_runtime": 77.5743,
+      "eval_samples_per_second": 3.867,
+      "eval_steps_per_second": 0.064,
+      "step": 27
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 26.93900489807129,
+      "learning_rate": 3.92431957907224e-05,
+      "loss": 0.739,
+      "step": 30
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 1.716407299041748,
+      "eval_runtime": 54.0253,
+      "eval_samples_per_second": 5.553,
+      "eval_steps_per_second": 0.093,
+      "step": 30
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 29.243497848510742,
+      "learning_rate": 3.6440110377099364e-05,
+      "loss": 0.5812,
+      "step": 33
+    },
+    {
+      "epoch": 1.375,
+      "eval_loss": 1.7478954792022705,
+      "eval_runtime": 54.2895,
+      "eval_samples_per_second": 5.526,
+      "eval_steps_per_second": 0.092,
+      "step": 33
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 27.55021858215332,
+      "learning_rate": 3.363702496347634e-05,
+      "loss": 0.6582,
+      "step": 36
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 1.8055189847946167,
+      "eval_runtime": 60.1066,
+      "eval_samples_per_second": 4.991,
+      "eval_steps_per_second": 0.083,
+      "step": 36
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 31.617809295654297,
+      "learning_rate": 3.083393954985331e-05,
+      "loss": 0.6917,
+      "step": 39
+    },
+    {
+      "epoch": 1.625,
+      "eval_loss": 1.754695177078247,
+      "eval_runtime": 61.8934,
+      "eval_samples_per_second": 4.847,
+      "eval_steps_per_second": 0.081,
+      "step": 39
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 27.27933120727539,
+      "learning_rate": 2.803085413623028e-05,
+      "loss": 0.465,
+      "step": 42
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.8375868797302246,
+      "eval_runtime": 54.5547,
+      "eval_samples_per_second": 5.499,
+      "eval_steps_per_second": 0.092,
+      "step": 42
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 27.845285415649414,
+      "learning_rate": 2.522776872260725e-05,
+      "loss": 0.542,
+      "step": 45
+    },
+    {
+      "epoch": 1.875,
+      "eval_loss": 1.7997920513153076,
+      "eval_runtime": 57.6703,
+      "eval_samples_per_second": 5.202,
+      "eval_steps_per_second": 0.087,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 27.056135177612305,
+      "learning_rate": 2.2424683308984225e-05,
+      "loss": 0.4544,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8440531492233276,
+      "eval_runtime": 64.2548,
+      "eval_samples_per_second": 4.669,
+      "eval_steps_per_second": 0.078,
+      "step": 48
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 13.831292152404785,
+      "learning_rate": 1.96215978953612e-05,
+      "loss": 0.1484,
+      "step": 51
+    },
+    {
+      "epoch": 2.125,
+      "eval_loss": 1.8647875785827637,
+      "eval_runtime": 70.5693,
+      "eval_samples_per_second": 4.251,
+      "eval_steps_per_second": 0.071,
+      "step": 51
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 12.969108581542969,
+      "learning_rate": 1.681851248173817e-05,
+      "loss": 0.2212,
+      "step": 54
+    },
+    {
+      "epoch": 2.25,
+      "eval_loss": 1.8213293552398682,
+      "eval_runtime": 73.4614,
+      "eval_samples_per_second": 4.084,
+      "eval_steps_per_second": 0.068,
+      "step": 54
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 16.65746307373047,
+      "learning_rate": 1.401542706811514e-05,
+      "loss": 0.1074,
+      "step": 57
+    },
+    {
+      "epoch": 2.375,
+      "eval_loss": 1.8436734676361084,
+      "eval_runtime": 54.1609,
+      "eval_samples_per_second": 5.539,
+      "eval_steps_per_second": 0.092,
+      "step": 57
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 6.605945110321045,
+      "learning_rate": 1.1212341654492112e-05,
+      "loss": 0.1042,
+      "step": 60
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 1.9006266593933105,
+      "eval_runtime": 54.4605,
+      "eval_samples_per_second": 5.509,
+      "eval_steps_per_second": 0.092,
+      "step": 60
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 14.282174110412598,
+      "learning_rate": 8.409256240869084e-06,
+      "loss": 0.1364,
+      "step": 63
+    },
+    {
+      "epoch": 2.625,
+      "eval_loss": 1.9413175582885742,
+      "eval_runtime": 54.1998,
+      "eval_samples_per_second": 5.535,
+      "eval_steps_per_second": 0.092,
+      "step": 63
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 16.284683227539062,
+      "learning_rate": 5.606170827246056e-06,
+      "loss": 0.1525,
+      "step": 66
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 1.9419138431549072,
+      "eval_runtime": 57.8254,
+      "eval_samples_per_second": 5.188,
+      "eval_steps_per_second": 0.086,
+      "step": 66
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 9.005942344665527,
+      "learning_rate": 2.803085413623028e-06,
+      "loss": 0.0863,
+      "step": 69
+    },
+    {
+      "epoch": 2.875,
+      "eval_loss": 1.9320255517959595,
+      "eval_runtime": 54.3706,
+      "eval_samples_per_second": 5.518,
+      "eval_steps_per_second": 0.092,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 4.704849720001221,
+      "learning_rate": 0.0,
+      "loss": 0.1073,
+      "step": 72
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.9295581579208374,
+      "eval_runtime": 56.1985,
+      "eval_samples_per_second": 5.338,
+      "eval_steps_per_second": 0.089,
+      "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": 6.727404992695267e-05,
+    "metric": "eval/loss",
+    "per_device_train_batch_size": 128
+  }
+}

run-w7wglxd2/checkpoint-72/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b4ae93199136b276d497431026e0a0d864b65a61d59e1d211b97e8ec7f2ce20
+size 5112

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad85aef7981b402b6659fe135471870edc8d4c5a17d7751a9f23b1dee3e99366
+oid sha256:4b4ae93199136b276d497431026e0a0d864b65a61d59e1d211b97e8ec7f2ce20
 size 5112