| { | |
| "best_metric": 0.5434782608695652, | |
| "best_model_checkpoint": "swiftformer-xs-DMAE\\checkpoint-3", | |
| "epoch": 34.285714285714285, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.3591431379318237, | |
| "eval_runtime": 0.6507, | |
| "eval_samples_per_second": 70.698, | |
| "eval_steps_per_second": 4.611, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5434782608695652, | |
| "eval_loss": 1.231246829032898, | |
| "eval_runtime": 0.6722, | |
| "eval_samples_per_second": 68.437, | |
| "eval_steps_per_second": 4.463, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.006666666666666667, | |
| "loss": 1.2973, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_accuracy": 0.391304347826087, | |
| "eval_loss": 2.0533361434936523, | |
| "eval_runtime": 0.5966, | |
| "eval_samples_per_second": 77.098, | |
| "eval_steps_per_second": 5.028, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.13043478260869565, | |
| "eval_loss": 3.3852343559265137, | |
| "eval_runtime": 0.6126, | |
| "eval_samples_per_second": 75.084, | |
| "eval_steps_per_second": 4.897, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_accuracy": 0.2826086956521739, | |
| "eval_loss": 5.402902126312256, | |
| "eval_runtime": 0.6162, | |
| "eval_samples_per_second": 74.656, | |
| "eval_steps_per_second": 4.869, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.007407407407407408, | |
| "loss": 1.3932, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.15217391304347827, | |
| "eval_loss": 6.73091983795166, | |
| "eval_runtime": 0.6251, | |
| "eval_samples_per_second": 73.584, | |
| "eval_steps_per_second": 4.799, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.1893627643585205, | |
| "eval_runtime": 0.5871, | |
| "eval_samples_per_second": 78.345, | |
| "eval_steps_per_second": 5.109, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.3668298721313477, | |
| "eval_runtime": 0.8242, | |
| "eval_samples_per_second": 55.811, | |
| "eval_steps_per_second": 3.64, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 0.006666666666666667, | |
| "loss": 1.224, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "eval_accuracy": 0.30434782608695654, | |
| "eval_loss": 1.4738078117370605, | |
| "eval_runtime": 0.5976, | |
| "eval_samples_per_second": 76.969, | |
| "eval_steps_per_second": 5.02, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4782608695652174, | |
| "eval_loss": 3.30267333984375, | |
| "eval_runtime": 0.6467, | |
| "eval_samples_per_second": 71.136, | |
| "eval_steps_per_second": 4.639, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.5451834201812744, | |
| "eval_runtime": 0.5736, | |
| "eval_samples_per_second": 80.19, | |
| "eval_steps_per_second": 5.23, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 0.005925925925925926, | |
| "loss": 1.1773, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.32608695652173914, | |
| "eval_loss": 2.5105574131011963, | |
| "eval_runtime": 0.6352, | |
| "eval_samples_per_second": 72.423, | |
| "eval_steps_per_second": 4.723, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 2.3560242652893066, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 73.82, | |
| "eval_steps_per_second": 4.814, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.3445404767990112, | |
| "eval_runtime": 0.6242, | |
| "eval_samples_per_second": 73.699, | |
| "eval_steps_per_second": 4.806, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 0.005185185185185185, | |
| "loss": 1.2552, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "eval_accuracy": 0.43478260869565216, | |
| "eval_loss": 1.264084815979004, | |
| "eval_runtime": 0.6312, | |
| "eval_samples_per_second": 72.882, | |
| "eval_steps_per_second": 4.753, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.2809749841690063, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 73.116, | |
| "eval_steps_per_second": 4.768, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "eval_accuracy": 0.43478260869565216, | |
| "eval_loss": 1.2486095428466797, | |
| "eval_runtime": 0.6246, | |
| "eval_samples_per_second": 73.641, | |
| "eval_steps_per_second": 4.803, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 0.0044444444444444444, | |
| "loss": 1.1857, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.3768922090530396, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 74.415, | |
| "eval_steps_per_second": 4.853, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.86, | |
| "eval_accuracy": 0.391304347826087, | |
| "eval_loss": 1.3311198949813843, | |
| "eval_runtime": 0.5861, | |
| "eval_samples_per_second": 78.481, | |
| "eval_steps_per_second": 5.118, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.003703703703703704, | |
| "loss": 1.2054, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.30434782608695654, | |
| "eval_loss": 1.3728384971618652, | |
| "eval_runtime": 0.6246, | |
| "eval_samples_per_second": 73.641, | |
| "eval_steps_per_second": 4.803, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2437033653259277, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 74.417, | |
| "eval_steps_per_second": 4.853, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.2751320600509644, | |
| "eval_runtime": 0.6296, | |
| "eval_samples_per_second": 73.057, | |
| "eval_steps_per_second": 4.765, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "learning_rate": 0.002962962962962963, | |
| "loss": 1.2037, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2529469728469849, | |
| "eval_runtime": 0.6047, | |
| "eval_samples_per_second": 76.076, | |
| "eval_steps_per_second": 4.961, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2354108095169067, | |
| "eval_runtime": 0.6081, | |
| "eval_samples_per_second": 75.641, | |
| "eval_steps_per_second": 4.933, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.86, | |
| "eval_accuracy": 0.43478260869565216, | |
| "eval_loss": 1.23885977268219, | |
| "eval_runtime": 0.6132, | |
| "eval_samples_per_second": 75.022, | |
| "eval_steps_per_second": 4.893, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "learning_rate": 0.0022222222222222222, | |
| "loss": 1.1888, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.43478260869565216, | |
| "eval_loss": 1.2455952167510986, | |
| "eval_runtime": 0.6196, | |
| "eval_samples_per_second": 74.236, | |
| "eval_steps_per_second": 4.841, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2312220335006714, | |
| "eval_runtime": 0.5921, | |
| "eval_samples_per_second": 77.685, | |
| "eval_steps_per_second": 5.066, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2452948093414307, | |
| "eval_runtime": 0.6407, | |
| "eval_samples_per_second": 71.802, | |
| "eval_steps_per_second": 4.683, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 0.0014814814814814814, | |
| "loss": 1.1872, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2506513595581055, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 75.766, | |
| "eval_steps_per_second": 4.941, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.41304347826086957, | |
| "eval_loss": 1.2598766088485718, | |
| "eval_runtime": 0.6437, | |
| "eval_samples_per_second": 71.467, | |
| "eval_steps_per_second": 4.661, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2493034601211548, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 74.415, | |
| "eval_steps_per_second": 4.853, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.43, | |
| "learning_rate": 0.0007407407407407407, | |
| "loss": 1.1796, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2410696744918823, | |
| "eval_runtime": 0.6317, | |
| "eval_samples_per_second": 72.825, | |
| "eval_steps_per_second": 4.749, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2361904382705688, | |
| "eval_runtime": 0.5841, | |
| "eval_samples_per_second": 78.749, | |
| "eval_steps_per_second": 5.136, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2300723791122437, | |
| "eval_runtime": 0.8292, | |
| "eval_samples_per_second": 55.475, | |
| "eval_steps_per_second": 3.618, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "learning_rate": 0.0, | |
| "loss": 1.199, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.2346495389938354, | |
| "eval_runtime": 0.6237, | |
| "eval_samples_per_second": 73.759, | |
| "eval_steps_per_second": 4.81, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "step": 120, | |
| "total_flos": 2.0027429927092224e+16, | |
| "train_loss": 1.2247031291325887, | |
| "train_runtime": 132.0853, | |
| "train_samples_per_second": 64.504, | |
| "train_steps_per_second": 0.909 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 2.0027429927092224e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |