Upload batch 7/7: vsfc/mBERT/max/seed_42 ... vsfc/mBERT/multi_branch_average/seed_44

mbert_rtx6000_metrics/result_ablation_mbert/vsfc/mBERT/multi_branch_average/seed_44/all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9374605180037903,
+    "eval_loss": 0.2184186577796936,
+    "eval_runtime": 5.0938,
+    "eval_samples": 1583,
+    "eval_samples_per_second": 310.769,
+    "eval_steps_per_second": 3.337,
+    "total_flos": 1057513611971040.0,
+    "train_loss": 0.20509551093501857,
+    "train_runtime": 66.8774,
+    "train_samples": 11426,
+    "train_samples_per_second": 512.55,
+    "train_steps_per_second": 10.721
+}

mbert_rtx6000_metrics/result_ablation_mbert/vsfc/mBERT/multi_branch_average/seed_44/eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9374605180037903,
+    "eval_loss": 0.2184186577796936,
+    "eval_runtime": 5.0938,
+    "eval_samples": 1583,
+    "eval_samples_per_second": 310.769,
+    "eval_steps_per_second": 3.337
+}

mbert_rtx6000_metrics/result_ablation_mbert/vsfc/mBERT/multi_branch_average/seed_44/train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "total_flos": 1057513611971040.0,
+    "train_loss": 0.20509551093501857,
+    "train_runtime": 66.8774,
+    "train_samples": 11426,
+    "train_samples_per_second": 512.55,
+    "train_steps_per_second": 10.721
+}

mbert_rtx6000_metrics/result_ablation_mbert/vsfc/mBERT/multi_branch_average/seed_44/trainer_state.json

@@ -0,0 +1,265 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 717,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.10460251046025104,
+      "grad_norm": 9.980685234069824,
+      "learning_rate": 2.8953974895397488e-05,
+      "loss": 0.5075,
+      "step": 25
+    },
+    {
+      "epoch": 0.20920502092050208,
+      "grad_norm": 2.346827507019043,
+      "learning_rate": 2.7907949790794982e-05,
+      "loss": 0.3532,
+      "step": 50
+    },
+    {
+      "epoch": 0.3138075313807531,
+      "grad_norm": 5.71251916885376,
+      "learning_rate": 2.686192468619247e-05,
+      "loss": 0.3529,
+      "step": 75
+    },
+    {
+      "epoch": 0.41841004184100417,
+      "grad_norm": 2.6974854469299316,
+      "learning_rate": 2.5815899581589957e-05,
+      "loss": 0.2815,
+      "step": 100
+    },
+    {
+      "epoch": 0.5230125523012552,
+      "grad_norm": 4.73166561126709,
+      "learning_rate": 2.4769874476987448e-05,
+      "loss": 0.2831,
+      "step": 125
+    },
+    {
+      "epoch": 0.6276150627615062,
+      "grad_norm": 5.596301555633545,
+      "learning_rate": 2.372384937238494e-05,
+      "loss": 0.2614,
+      "step": 150
+    },
+    {
+      "epoch": 0.7322175732217573,
+      "grad_norm": 5.316343307495117,
+      "learning_rate": 2.2677824267782426e-05,
+      "loss": 0.3032,
+      "step": 175
+    },
+    {
+      "epoch": 0.8368200836820083,
+      "grad_norm": 3.482184410095215,
+      "learning_rate": 2.1631799163179917e-05,
+      "loss": 0.235,
+      "step": 200
+    },
+    {
+      "epoch": 0.9414225941422594,
+      "grad_norm": 9.44145679473877,
+      "learning_rate": 2.0585774058577408e-05,
+      "loss": 0.2311,
+      "step": 225
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9216677195198989,
+      "eval_loss": 0.2288883477449417,
+      "eval_runtime": 5.3667,
+      "eval_samples_per_second": 294.966,
+      "eval_steps_per_second": 3.168,
+      "step": 239
+    },
+    {
+      "epoch": 1.0460251046025104,
+      "grad_norm": 3.8848302364349365,
+      "learning_rate": 1.9539748953974895e-05,
+      "loss": 0.2114,
+      "step": 250
+    },
+    {
+      "epoch": 1.1506276150627615,
+      "grad_norm": 7.083329677581787,
+      "learning_rate": 1.8493723849372386e-05,
+      "loss": 0.2069,
+      "step": 275
+    },
+    {
+      "epoch": 1.2552301255230125,
+      "grad_norm": 3.5117275714874268,
+      "learning_rate": 1.7447698744769877e-05,
+      "loss": 0.196,
+      "step": 300
+    },
+    {
+      "epoch": 1.3598326359832635,
+      "grad_norm": 1.9568407535552979,
+      "learning_rate": 1.6401673640167365e-05,
+      "loss": 0.1913,
+      "step": 325
+    },
+    {
+      "epoch": 1.4644351464435146,
+      "grad_norm": 2.5776121616363525,
+      "learning_rate": 1.5355648535564852e-05,
+      "loss": 0.1961,
+      "step": 350
+    },
+    {
+      "epoch": 1.5690376569037658,
+      "grad_norm": 5.712918758392334,
+      "learning_rate": 1.4309623430962344e-05,
+      "loss": 0.1827,
+      "step": 375
+    },
+    {
+      "epoch": 1.6736401673640167,
+      "grad_norm": 4.35663366317749,
+      "learning_rate": 1.3263598326359832e-05,
+      "loss": 0.1893,
+      "step": 400
+    },
+    {
+      "epoch": 1.778242677824268,
+      "grad_norm": 5.765503883361816,
+      "learning_rate": 1.2217573221757323e-05,
+      "loss": 0.1839,
+      "step": 425
+    },
+    {
+      "epoch": 1.8828451882845187,
+      "grad_norm": 11.028792381286621,
+      "learning_rate": 1.1171548117154812e-05,
+      "loss": 0.1852,
+      "step": 450
+    },
+    {
+      "epoch": 1.98744769874477,
+      "grad_norm": 4.499820232391357,
+      "learning_rate": 1.0125523012552301e-05,
+      "loss": 0.1645,
+      "step": 475
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9317751105495894,
+      "eval_loss": 0.2380005568265915,
+      "eval_runtime": 4.9568,
+      "eval_samples_per_second": 319.359,
+      "eval_steps_per_second": 3.43,
+      "step": 478
+    },
+    {
+      "epoch": 2.092050209205021,
+      "grad_norm": 6.939774990081787,
+      "learning_rate": 9.079497907949792e-06,
+      "loss": 0.1375,
+      "step": 500
+    },
+    {
+      "epoch": 2.196652719665272,
+      "grad_norm": 3.196690320968628,
+      "learning_rate": 8.03347280334728e-06,
+      "loss": 0.1073,
+      "step": 525
+    },
+    {
+      "epoch": 2.301255230125523,
+      "grad_norm": 0.9964064359664917,
+      "learning_rate": 6.98744769874477e-06,
+      "loss": 0.1336,
+      "step": 550
+    },
+    {
+      "epoch": 2.405857740585774,
+      "grad_norm": 7.170077800750732,
+      "learning_rate": 5.941422594142259e-06,
+      "loss": 0.1212,
+      "step": 575
+    },
+    {
+      "epoch": 2.510460251046025,
+      "grad_norm": 5.591251850128174,
+      "learning_rate": 4.895397489539749e-06,
+      "loss": 0.1272,
+      "step": 600
+    },
+    {
+      "epoch": 2.6150627615062763,
+      "grad_norm": 8.385252952575684,
+      "learning_rate": 3.849372384937238e-06,
+      "loss": 0.1093,
+      "step": 625
+    },
+    {
+      "epoch": 2.719665271966527,
+      "grad_norm": 8.233146667480469,
+      "learning_rate": 2.803347280334728e-06,
+      "loss": 0.1332,
+      "step": 650
+    },
+    {
+      "epoch": 2.8242677824267783,
+      "grad_norm": 3.9085960388183594,
+      "learning_rate": 1.7573221757322176e-06,
+      "loss": 0.1009,
+      "step": 675
+    },
+    {
+      "epoch": 2.928870292887029,
+      "grad_norm": 3.9862301349639893,
+      "learning_rate": 7.112970711297071e-07,
+      "loss": 0.1136,
+      "step": 700
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9374605180037903,
+      "eval_loss": 0.2184186577796936,
+      "eval_runtime": 4.926,
+      "eval_samples_per_second": 321.354,
+      "eval_steps_per_second": 3.451,
+      "step": 717
+    },
+    {
+      "epoch": 3.0,
+      "step": 717,
+      "total_flos": 1057513611971040.0,
+      "train_loss": 0.20509551093501857,
+      "train_runtime": 66.8774,
+      "train_samples_per_second": 512.55,
+      "train_steps_per_second": 10.721
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 717,
+  "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": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1057513611971040.0,
+  "train_batch_size": 48,
+  "trial_name": null,
+  "trial_params": null
+}