Training in progress, step 72

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7479e4640aa8b3ee33e7cfabc86c5808c4ce220f08794cbde2199f6097706825
+oid sha256:953d2a47e129f038cfc3fa14cfe8b8a15e6c896e6eb73cac2575d92591de414c
 size 605156676

run-rhd3gatg/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-rhd3gatg/checkpoint-72/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:953d2a47e129f038cfc3fa14cfe8b8a15e6c896e6eb73cac2575d92591de414c
+size 605156676

run-rhd3gatg/checkpoint-72/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a7354dbf5499e12e7695fd493928a483d4b5b60181073c13e81fbad03f63cfd
+size 1210551612

run-rhd3gatg/checkpoint-72/rng_state.pth

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

run-rhd3gatg/checkpoint-72/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66a6c96124ce3934b673fee34285e1447799ac7f681fe76742e32596eb4271de
+size 1064

run-rhd3gatg/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": 86.39575958251953,
+      "learning_rate": 2.5582874144977848e-06,
+      "loss": 2.3537,
+      "step": 3
+    },
+    {
+      "epoch": 0.125,
+      "eval_loss": 1.1495916843414307,
+      "eval_runtime": 41.913,
+      "eval_samples_per_second": 7.158,
+      "eval_steps_per_second": 0.239,
+      "step": 3
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 98.42198944091797,
+      "learning_rate": 2.4470575269109243e-06,
+      "loss": 2.0005,
+      "step": 6
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 1.1086667776107788,
+      "eval_runtime": 42.2,
+      "eval_samples_per_second": 7.109,
+      "eval_steps_per_second": 0.237,
+      "step": 6
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 91.29363250732422,
+      "learning_rate": 2.3358276393240644e-06,
+      "loss": 1.9471,
+      "step": 9
+    },
+    {
+      "epoch": 0.375,
+      "eval_loss": 1.0748331546783447,
+      "eval_runtime": 42.7563,
+      "eval_samples_per_second": 7.017,
+      "eval_steps_per_second": 0.234,
+      "step": 9
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 86.57069396972656,
+      "learning_rate": 2.224597751737204e-06,
+      "loss": 1.8691,
+      "step": 12
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.0461446046829224,
+      "eval_runtime": 42.0948,
+      "eval_samples_per_second": 7.127,
+      "eval_steps_per_second": 0.238,
+      "step": 12
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 70.24944305419922,
+      "learning_rate": 2.1133678641503435e-06,
+      "loss": 2.0024,
+      "step": 15
+    },
+    {
+      "epoch": 0.625,
+      "eval_loss": 1.0137351751327515,
+      "eval_runtime": 42.3079,
+      "eval_samples_per_second": 7.091,
+      "eval_steps_per_second": 0.236,
+      "step": 15
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 80.9498291015625,
+      "learning_rate": 2.0021379765634836e-06,
+      "loss": 2.1394,
+      "step": 18
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 0.9847450852394104,
+      "eval_runtime": 41.9761,
+      "eval_samples_per_second": 7.147,
+      "eval_steps_per_second": 0.238,
+      "step": 18
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 74.85993194580078,
+      "learning_rate": 1.8909080889766234e-06,
+      "loss": 1.8312,
+      "step": 21
+    },
+    {
+      "epoch": 0.875,
+      "eval_loss": 0.9646272659301758,
+      "eval_runtime": 43.1304,
+      "eval_samples_per_second": 6.956,
+      "eval_steps_per_second": 0.232,
+      "step": 21
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 88.5673599243164,
+      "learning_rate": 1.7796782013897632e-06,
+      "loss": 1.6205,
+      "step": 24
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.9487736225128174,
+      "eval_runtime": 43.314,
+      "eval_samples_per_second": 6.926,
+      "eval_steps_per_second": 0.231,
+      "step": 24
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 54.09762954711914,
+      "learning_rate": 1.668448313802903e-06,
+      "loss": 1.2486,
+      "step": 27
+    },
+    {
+      "epoch": 1.125,
+      "eval_loss": 0.9389265775680542,
+      "eval_runtime": 56.1124,
+      "eval_samples_per_second": 5.346,
+      "eval_steps_per_second": 0.178,
+      "step": 27
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 60.164310455322266,
+      "learning_rate": 1.5572184262160428e-06,
+      "loss": 1.3264,
+      "step": 30
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 0.9309024810791016,
+      "eval_runtime": 42.4471,
+      "eval_samples_per_second": 7.068,
+      "eval_steps_per_second": 0.236,
+      "step": 30
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 47.892051696777344,
+      "learning_rate": 1.4459885386291826e-06,
+      "loss": 1.0128,
+      "step": 33
+    },
+    {
+      "epoch": 1.375,
+      "eval_loss": 0.9257251024246216,
+      "eval_runtime": 42.7266,
+      "eval_samples_per_second": 7.021,
+      "eval_steps_per_second": 0.234,
+      "step": 33
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 54.91218185424805,
+      "learning_rate": 1.3347586510423224e-06,
+      "loss": 1.3253,
+      "step": 36
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 0.9215033054351807,
+      "eval_runtime": 42.303,
+      "eval_samples_per_second": 7.092,
+      "eval_steps_per_second": 0.236,
+      "step": 36
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 58.73004150390625,
+      "learning_rate": 1.2235287634554622e-06,
+      "loss": 1.3876,
+      "step": 39
+    },
+    {
+      "epoch": 1.625,
+      "eval_loss": 0.9176895022392273,
+      "eval_runtime": 42.0984,
+      "eval_samples_per_second": 7.126,
+      "eval_steps_per_second": 0.238,
+      "step": 39
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 62.26673126220703,
+      "learning_rate": 1.112298875868602e-06,
+      "loss": 1.1772,
+      "step": 42
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 0.9114839434623718,
+      "eval_runtime": 42.1241,
+      "eval_samples_per_second": 7.122,
+      "eval_steps_per_second": 0.237,
+      "step": 42
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 56.1391716003418,
+      "learning_rate": 1.0010689882817418e-06,
+      "loss": 1.2588,
+      "step": 45
+    },
+    {
+      "epoch": 1.875,
+      "eval_loss": 0.9046874046325684,
+      "eval_runtime": 42.2604,
+      "eval_samples_per_second": 7.099,
+      "eval_steps_per_second": 0.237,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 95.35293579101562,
+      "learning_rate": 8.898391006948816e-07,
+      "loss": 1.0958,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.8994578123092651,
+      "eval_runtime": 42.1798,
+      "eval_samples_per_second": 7.112,
+      "eval_steps_per_second": 0.237,
+      "step": 48
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 44.073333740234375,
+      "learning_rate": 7.786092131080214e-07,
+      "loss": 1.1684,
+      "step": 51
+    },
+    {
+      "epoch": 2.125,
+      "eval_loss": 0.8956994414329529,
+      "eval_runtime": 42.2987,
+      "eval_samples_per_second": 7.092,
+      "eval_steps_per_second": 0.236,
+      "step": 51
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 49.152503967285156,
+      "learning_rate": 6.673793255211612e-07,
+      "loss": 1.0725,
+      "step": 54
+    },
+    {
+      "epoch": 2.25,
+      "eval_loss": 0.8932783007621765,
+      "eval_runtime": 61.307,
+      "eval_samples_per_second": 4.893,
+      "eval_steps_per_second": 0.163,
+      "step": 54
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 53.29318618774414,
+      "learning_rate": 5.56149437934301e-07,
+      "loss": 0.952,
+      "step": 57
+    },
+    {
+      "epoch": 2.375,
+      "eval_loss": 0.891077995300293,
+      "eval_runtime": 42.0381,
+      "eval_samples_per_second": 7.136,
+      "eval_steps_per_second": 0.238,
+      "step": 57
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 48.242835998535156,
+      "learning_rate": 4.449195503474408e-07,
+      "loss": 0.9573,
+      "step": 60
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 0.8885931372642517,
+      "eval_runtime": 42.1072,
+      "eval_samples_per_second": 7.125,
+      "eval_steps_per_second": 0.237,
+      "step": 60
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 47.384883880615234,
+      "learning_rate": 3.336896627605806e-07,
+      "loss": 1.0175,
+      "step": 63
+    },
+    {
+      "epoch": 2.625,
+      "eval_loss": 0.8877026438713074,
+      "eval_runtime": 42.4373,
+      "eval_samples_per_second": 7.069,
+      "eval_steps_per_second": 0.236,
+      "step": 63
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 47.30453872680664,
+      "learning_rate": 2.224597751737204e-07,
+      "loss": 0.8272,
+      "step": 66
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 0.8868218064308167,
+      "eval_runtime": 42.3346,
+      "eval_samples_per_second": 7.086,
+      "eval_steps_per_second": 0.236,
+      "step": 66
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 48.2218017578125,
+      "learning_rate": 1.112298875868602e-07,
+      "loss": 0.8745,
+      "step": 69
+    },
+    {
+      "epoch": 2.875,
+      "eval_loss": 0.8861154913902283,
+      "eval_runtime": 42.454,
+      "eval_samples_per_second": 7.066,
+      "eval_steps_per_second": 0.236,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 71.5477523803711,
+      "learning_rate": 0.0,
+      "loss": 0.8584,
+      "step": 72
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.8858494162559509,
+      "eval_runtime": 42.4248,
+      "eval_samples_per_second": 7.071,
+      "eval_steps_per_second": 0.236,
+      "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": 0.001,
+    "learning_rate": 2.6695173020846447e-06,
+    "metric": "eval/loss",
+    "per_device_train_batch_size": 128
+  }
+}

run-rhd3gatg/checkpoint-72/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29aee2a71a1aa3886315b9d3f5f8c00151e191143b415b0176733d90bd738ed0
+size 5112

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2de738072ec265745ad8ca14cff5e9c8d4d119457b017a3a3ec0e72524183c6b
+oid sha256:29aee2a71a1aa3886315b9d3f5f8c00151e191143b415b0176733d90bd738ed0
 size 5112