Training in progress, step 72

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74d09b7e21715c345df914e334fa0c9663d9cebfebfebd86d03ea69315b85e9f
+oid sha256:99f4991491121c3e19a754c6cd6ff6924a2e6c62b1a96ce3f51dab3abb7d90cb
 size 605156676

run-4np4i6fn/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-4np4i6fn/checkpoint-72/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99f4991491121c3e19a754c6cd6ff6924a2e6c62b1a96ce3f51dab3abb7d90cb
+size 605156676

run-4np4i6fn/checkpoint-72/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f65b2a21efc0b6a6d04866aed93bde42e9086f1ebb47a3f15054dbf025d44f41
+size 1210551612

run-4np4i6fn/checkpoint-72/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a5b912e901318a0a151cc175f3548a076ce9b3d3e9c276004710a62bd845a2a
+size 14244

run-4np4i6fn/checkpoint-72/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a8b8b3b161beff8eb071e56ebcbff7b8eaf12632190f32fbef9ca85448c9efb
+size 1064

run-4np4i6fn/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": 108.35137176513672,
+      "learning_rate": 6.506514193452409e-06,
+      "loss": 2.3054,
+      "step": 3
+    },
+    {
+      "epoch": 0.125,
+      "eval_loss": 1.4701541662216187,
+      "eval_runtime": 59.4443,
+      "eval_samples_per_second": 5.047,
+      "eval_steps_per_second": 0.084,
+      "step": 3
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 84.46858215332031,
+      "learning_rate": 6.223622271997956e-06,
+      "loss": 1.8783,
+      "step": 6
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 1.3813910484313965,
+      "eval_runtime": 60.894,
+      "eval_samples_per_second": 4.927,
+      "eval_steps_per_second": 0.082,
+      "step": 6
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 86.31482696533203,
+      "learning_rate": 5.940730350543504e-06,
+      "loss": 1.8718,
+      "step": 9
+    },
+    {
+      "epoch": 0.375,
+      "eval_loss": 1.354865550994873,
+      "eval_runtime": 68.7391,
+      "eval_samples_per_second": 4.364,
+      "eval_steps_per_second": 0.073,
+      "step": 9
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 87.9017333984375,
+      "learning_rate": 5.657838429089051e-06,
+      "loss": 1.7861,
+      "step": 12
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.2925242185592651,
+      "eval_runtime": 56.8444,
+      "eval_samples_per_second": 5.278,
+      "eval_steps_per_second": 0.088,
+      "step": 12
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 58.45953369140625,
+      "learning_rate": 5.374946507634598e-06,
+      "loss": 1.8629,
+      "step": 15
+    },
+    {
+      "epoch": 0.625,
+      "eval_loss": 1.239138126373291,
+      "eval_runtime": 52.9516,
+      "eval_samples_per_second": 5.666,
+      "eval_steps_per_second": 0.094,
+      "step": 15
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 61.15017318725586,
+      "learning_rate": 5.092054586180146e-06,
+      "loss": 1.9993,
+      "step": 18
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.1964143514633179,
+      "eval_runtime": 57.4229,
+      "eval_samples_per_second": 5.224,
+      "eval_steps_per_second": 0.087,
+      "step": 18
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 54.41535949707031,
+      "learning_rate": 4.8091626647256935e-06,
+      "loss": 1.7431,
+      "step": 21
+    },
+    {
+      "epoch": 0.875,
+      "eval_loss": 1.168664574623108,
+      "eval_runtime": 56.4862,
+      "eval_samples_per_second": 5.311,
+      "eval_steps_per_second": 0.089,
+      "step": 21
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 68.75323486328125,
+      "learning_rate": 4.526270743271241e-06,
+      "loss": 1.5175,
+      "step": 24
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.1468874216079712,
+      "eval_runtime": 93.0841,
+      "eval_samples_per_second": 3.223,
+      "eval_steps_per_second": 0.054,
+      "step": 24
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 31.90779685974121,
+      "learning_rate": 4.243378821816788e-06,
+      "loss": 0.9257,
+      "step": 27
+    },
+    {
+      "epoch": 1.125,
+      "eval_loss": 1.1273722648620605,
+      "eval_runtime": 69.2243,
+      "eval_samples_per_second": 4.334,
+      "eval_steps_per_second": 0.072,
+      "step": 27
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 33.2850341796875,
+      "learning_rate": 3.960486900362336e-06,
+      "loss": 0.9284,
+      "step": 30
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 1.1108582019805908,
+      "eval_runtime": 57.9469,
+      "eval_samples_per_second": 5.177,
+      "eval_steps_per_second": 0.086,
+      "step": 30
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 31.100393295288086,
+      "learning_rate": 3.677594978907883e-06,
+      "loss": 0.6838,
+      "step": 33
+    },
+    {
+      "epoch": 1.375,
+      "eval_loss": 1.1016932725906372,
+      "eval_runtime": 54.0914,
+      "eval_samples_per_second": 5.546,
+      "eval_steps_per_second": 0.092,
+      "step": 33
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 34.479129791259766,
+      "learning_rate": 3.3947030574534306e-06,
+      "loss": 0.8734,
+      "step": 36
+    },
+    {
+      "epoch": 1.5,
+      "eval_loss": 1.0961642265319824,
+      "eval_runtime": 67.7251,
+      "eval_samples_per_second": 4.43,
+      "eval_steps_per_second": 0.074,
+      "step": 36
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 42.74652099609375,
+      "learning_rate": 3.111811135998978e-06,
+      "loss": 0.8897,
+      "step": 39
+    },
+    {
+      "epoch": 1.625,
+      "eval_loss": 1.091117024421692,
+      "eval_runtime": 69.4499,
+      "eval_samples_per_second": 4.32,
+      "eval_steps_per_second": 0.072,
+      "step": 39
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 38.22075653076172,
+      "learning_rate": 2.8289192145445255e-06,
+      "loss": 0.7344,
+      "step": 42
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 1.089510440826416,
+      "eval_runtime": 53.1755,
+      "eval_samples_per_second": 5.642,
+      "eval_steps_per_second": 0.094,
+      "step": 42
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 46.27302551269531,
+      "learning_rate": 2.546027293090073e-06,
+      "loss": 0.8112,
+      "step": 45
+    },
+    {
+      "epoch": 1.875,
+      "eval_loss": 1.0918387174606323,
+      "eval_runtime": 59.2852,
+      "eval_samples_per_second": 5.06,
+      "eval_steps_per_second": 0.084,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 57.83619689941406,
+      "learning_rate": 2.2631353716356204e-06,
+      "loss": 0.6833,
+      "step": 48
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.0950809717178345,
+      "eval_runtime": 59.0606,
+      "eval_samples_per_second": 5.08,
+      "eval_steps_per_second": 0.085,
+      "step": 48
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 29.00582504272461,
+      "learning_rate": 1.980243450181168e-06,
+      "loss": 0.5381,
+      "step": 51
+    },
+    {
+      "epoch": 2.125,
+      "eval_loss": 1.0971535444259644,
+      "eval_runtime": 53.5954,
+      "eval_samples_per_second": 5.597,
+      "eval_steps_per_second": 0.093,
+      "step": 51
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 29.384523391723633,
+      "learning_rate": 1.6973515287267153e-06,
+      "loss": 0.4857,
+      "step": 54
+    },
+    {
+      "epoch": 2.25,
+      "eval_loss": 1.0965510606765747,
+      "eval_runtime": 53.671,
+      "eval_samples_per_second": 5.59,
+      "eval_steps_per_second": 0.093,
+      "step": 54
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 27.365966796875,
+      "learning_rate": 1.4144596072722627e-06,
+      "loss": 0.3852,
+      "step": 57
+    },
+    {
+      "epoch": 2.375,
+      "eval_loss": 1.09633207321167,
+      "eval_runtime": 82.6569,
+      "eval_samples_per_second": 3.629,
+      "eval_steps_per_second": 0.06,
+      "step": 57
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 29.962955474853516,
+      "learning_rate": 1.1315676858178102e-06,
+      "loss": 0.3716,
+      "step": 60
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 1.0960321426391602,
+      "eval_runtime": 61.9006,
+      "eval_samples_per_second": 4.846,
+      "eval_steps_per_second": 0.081,
+      "step": 60
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 31.133134841918945,
+      "learning_rate": 8.486757643633576e-07,
+      "loss": 0.4541,
+      "step": 63
+    },
+    {
+      "epoch": 2.625,
+      "eval_loss": 1.0945457220077515,
+      "eval_runtime": 54.0827,
+      "eval_samples_per_second": 5.547,
+      "eval_steps_per_second": 0.092,
+      "step": 63
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 29.13813018798828,
+      "learning_rate": 5.657838429089051e-07,
+      "loss": 0.3268,
+      "step": 66
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 1.094507098197937,
+      "eval_runtime": 54.443,
+      "eval_samples_per_second": 5.51,
+      "eval_steps_per_second": 0.092,
+      "step": 66
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 29.24846076965332,
+      "learning_rate": 2.8289192145445255e-07,
+      "loss": 0.3394,
+      "step": 69
+    },
+    {
+      "epoch": 2.875,
+      "eval_loss": 1.0951770544052124,
+      "eval_runtime": 58.0985,
+      "eval_samples_per_second": 5.164,
+      "eval_steps_per_second": 0.086,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 32.50340270996094,
+      "learning_rate": 0.0,
+      "loss": 0.3173,
+      "step": 72
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.0956674814224243,
+      "eval_runtime": 56.9634,
+      "eval_samples_per_second": 5.267,
+      "eval_steps_per_second": 0.088,
+      "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": 6.789406114906861e-06,
+    "metric": "eval/loss",
+    "per_device_train_batch_size": 128
+  }
+}

run-4np4i6fn/checkpoint-72/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6152f4acfabcfd032efc0c5944ecf32b2674146585dc5f327e029bb2b73a2b9
+size 5112

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f32e56a92fc888c2246e5b72e7d93d10f0ea6484a0289837bffd6e64aaa0db1c
+oid sha256:b6152f4acfabcfd032efc0c5944ecf32b2674146585dc5f327e029bb2b73a2b9
 size 5112