| { | |
| "best_metric": 0.49422508478164673, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-4/checkpoint-229", | |
| "epoch": 34.0, | |
| "eval_steps": 500, | |
| "global_step": 289, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 2.2058663368225098, | |
| "learning_rate": 8.333333333333334e-08, | |
| "loss": 2.3266, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 3.2251737117767334, | |
| "learning_rate": 1.6666666666666668e-07, | |
| "loss": 2.8101, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 2.427489757537842, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 2.4918, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 2.110276460647583, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 2.416, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 2.285691499710083, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 2.5495, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 2.5076982975006104, | |
| "eval_runtime": 31.8608, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 2.757939577102661, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 2.4129, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 3.1854565143585205, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 2.6068, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 2.928661346435547, | |
| "learning_rate": 1.1666666666666668e-06, | |
| "loss": 2.66, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 2.948481321334839, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.3733, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.4775068759918213, | |
| "eval_runtime": 31.8538, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 2.4443531036376953, | |
| "learning_rate": 1.5e-06, | |
| "loss": 2.422, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 2.2713582515716553, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 2.3315, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 2.6579792499542236, | |
| "learning_rate": 1.8333333333333333e-06, | |
| "loss": 2.5587, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 2.5377650260925293, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.46, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 2.4217796325683594, | |
| "eval_runtime": 31.8528, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 2.9668917655944824, | |
| "learning_rate": 2.166666666666667e-06, | |
| "loss": 2.5217, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 2.460824728012085, | |
| "learning_rate": 2.3333333333333336e-06, | |
| "loss": 2.2339, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 3.505033493041992, | |
| "learning_rate": 2.5e-06, | |
| "loss": 2.4788, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 2.1192469596862793, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 2.2748, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.7211689949035645, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 2.4585, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.3277575969696045, | |
| "eval_runtime": 31.8566, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 2.7326908111572266, | |
| "learning_rate": 3e-06, | |
| "loss": 2.3561, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 2.5006515979766846, | |
| "learning_rate": 3.1666666666666667e-06, | |
| "loss": 2.1582, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 2.728483200073242, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 2.2463, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 3.0043787956237793, | |
| "learning_rate": 3.5e-06, | |
| "loss": 2.2624, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "eval_loss": 2.191901683807373, | |
| "eval_runtime": 31.8501, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 2.0914738178253174, | |
| "learning_rate": 3.6666666666666666e-06, | |
| "loss": 2.0377, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 3.9215309619903564, | |
| "learning_rate": 3.833333333333334e-06, | |
| "loss": 2.3093, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 2.7781357765197754, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.0491, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 2.7123777866363525, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 2.0553, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.9704445600509644, | |
| "eval_runtime": 31.8507, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 3.2437472343444824, | |
| "learning_rate": 4.333333333333334e-06, | |
| "loss": 1.9741, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 2.4424147605895996, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.8914, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 2.605942964553833, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 1.8738, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 2.119314432144165, | |
| "learning_rate": 4.833333333333333e-06, | |
| "loss": 1.7403, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "eval_loss": 1.7065722942352295, | |
| "eval_runtime": 31.853, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.8015623092651367, | |
| "learning_rate": 5e-06, | |
| "loss": 1.6225, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 2.412097215652466, | |
| "learning_rate": 5.1666666666666675e-06, | |
| "loss": 1.6174, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 2.3760602474212646, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.5403, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 2.5603954792022705, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.4378, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.450331211090088, | |
| "learning_rate": 5.666666666666667e-06, | |
| "loss": 1.3756, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.3616687059402466, | |
| "eval_runtime": 31.8467, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.7808512449264526, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 1.2903, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 2.280879497528076, | |
| "learning_rate": 6e-06, | |
| "loss": 1.2624, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 2.584132671356201, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 1.0856, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.6824058294296265, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 1.11, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "eval_loss": 1.0612744092941284, | |
| "eval_runtime": 31.847, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 1.8082739114761353, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.9939, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 1.7640292644500732, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.9378, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 1.609857201576233, | |
| "learning_rate": 6.833333333333334e-06, | |
| "loss": 0.7986, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 1.5625256299972534, | |
| "learning_rate": 7e-06, | |
| "loss": 0.7161, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7772443890571594, | |
| "eval_runtime": 31.8458, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 1.2514125108718872, | |
| "learning_rate": 7.166666666666667e-06, | |
| "loss": 0.7059, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 0.9835730195045471, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 0.5005, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.8222648501396179, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.7141, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 1.2220810651779175, | |
| "learning_rate": 7.666666666666667e-06, | |
| "loss": 0.7609, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "eval_loss": 0.6786696910858154, | |
| "eval_runtime": 31.8483, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.6009970307350159, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.6874, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 0.7795246243476868, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7462, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 0.6197288036346436, | |
| "learning_rate": 8.166666666666668e-06, | |
| "loss": 0.6408, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.6273565292358398, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.4579, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.45476672053337097, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.4358, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.6182084083557129, | |
| "eval_runtime": 31.8454, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.7051817774772644, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 0.6431, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 0.5281614065170288, | |
| "learning_rate": 8.833333333333334e-06, | |
| "loss": 0.518, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 0.5485586524009705, | |
| "learning_rate": 9e-06, | |
| "loss": 0.4787, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 0.9739407300949097, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.4774, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "eval_loss": 0.5912042260169983, | |
| "eval_runtime": 31.8561, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 13.176470588235293, | |
| "grad_norm": 0.3852427899837494, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.476, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 13.411764705882353, | |
| "grad_norm": 0.3773607909679413, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.4432, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 13.647058823529411, | |
| "grad_norm": 0.5833092331886292, | |
| "learning_rate": 9.666666666666667e-06, | |
| "loss": 0.4677, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 13.882352941176471, | |
| "grad_norm": 0.525675356388092, | |
| "learning_rate": 9.833333333333333e-06, | |
| "loss": 0.5569, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.5746464133262634, | |
| "eval_runtime": 31.8497, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 0.46034735441207886, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6595, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 14.352941176470589, | |
| "grad_norm": 0.9925610423088074, | |
| "learning_rate": 9.999915384288723e-06, | |
| "loss": 0.5066, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 14.588235294117647, | |
| "grad_norm": 0.3283170163631439, | |
| "learning_rate": 9.999661540018812e-06, | |
| "loss": 0.5017, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 14.823529411764707, | |
| "grad_norm": 0.4828791320323944, | |
| "learning_rate": 9.999238475781957e-06, | |
| "loss": 0.427, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 14.941176470588236, | |
| "eval_loss": 0.5487004518508911, | |
| "eval_runtime": 31.8495, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 15.058823529411764, | |
| "grad_norm": 0.44912636280059814, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.4705, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 15.294117647058824, | |
| "grad_norm": 0.43985074758529663, | |
| "learning_rate": 9.997884750411004e-06, | |
| "loss": 0.3865, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 15.529411764705882, | |
| "grad_norm": 1.4928995370864868, | |
| "learning_rate": 9.99695413509548e-06, | |
| "loss": 0.483, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 15.764705882352942, | |
| "grad_norm": 0.3390346169471741, | |
| "learning_rate": 9.995854391448607e-06, | |
| "loss": 0.5068, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.33858582377433777, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 0.4672, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.533892035484314, | |
| "eval_runtime": 31.8489, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.235294117647058, | |
| "grad_norm": 0.3130759298801422, | |
| "learning_rate": 9.993147673772869e-06, | |
| "loss": 0.3805, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 0.32997503876686096, | |
| "learning_rate": 9.991540791356342e-06, | |
| "loss": 0.6362, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 16.705882352941178, | |
| "grad_norm": 1.398141622543335, | |
| "learning_rate": 9.989764963830038e-06, | |
| "loss": 0.4943, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "grad_norm": 0.19720911979675293, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.3495, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "eval_loss": 0.552502453327179, | |
| "eval_runtime": 31.8645, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 17.176470588235293, | |
| "grad_norm": 0.21927881240844727, | |
| "learning_rate": 9.985706719584888e-06, | |
| "loss": 0.4342, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 17.41176470588235, | |
| "grad_norm": 0.2400996834039688, | |
| "learning_rate": 9.98342444022253e-06, | |
| "loss": 0.4801, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 0.2719495892524719, | |
| "learning_rate": 9.980973490458728e-06, | |
| "loss": 0.4199, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 17.88235294117647, | |
| "grad_norm": 0.18624863028526306, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 0.4731, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.5323101878166199, | |
| "eval_runtime": 31.8439, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 18.11764705882353, | |
| "grad_norm": 0.18268544971942902, | |
| "learning_rate": 9.975565917255017e-06, | |
| "loss": 0.3459, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 18.352941176470587, | |
| "grad_norm": 0.30436864495277405, | |
| "learning_rate": 9.972609476841368e-06, | |
| "loss": 0.58, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 18.58823529411765, | |
| "grad_norm": 0.2947259843349457, | |
| "learning_rate": 9.9694847320726e-06, | |
| "loss": 0.4085, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 18.823529411764707, | |
| "grad_norm": 0.24882009625434875, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.3913, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 18.941176470588236, | |
| "eval_loss": 0.5243125557899475, | |
| "eval_runtime": 31.8447, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 19.058823529411764, | |
| "grad_norm": 0.31586578488349915, | |
| "learning_rate": 9.962730758206612e-06, | |
| "loss": 0.5022, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 19.294117647058822, | |
| "grad_norm": 0.21429723501205444, | |
| "learning_rate": 9.959101757706308e-06, | |
| "loss": 0.3682, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 19.529411764705884, | |
| "grad_norm": 0.29345616698265076, | |
| "learning_rate": 9.955304910036993e-06, | |
| "loss": 0.4185, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 19.764705882352942, | |
| "grad_norm": 0.24088309705257416, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.4027, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.3225376307964325, | |
| "learning_rate": 9.947208192904722e-06, | |
| "loss": 0.5624, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.5252914428710938, | |
| "eval_runtime": 31.8447, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.235294117647058, | |
| "grad_norm": 0.24431054294109344, | |
| "learning_rate": 9.942908597485558e-06, | |
| "loss": 0.3565, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 20.470588235294116, | |
| "grad_norm": 0.30302417278289795, | |
| "learning_rate": 9.938441702975689e-06, | |
| "loss": 0.4944, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 20.705882352941178, | |
| "grad_norm": 0.25407275557518005, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.3397, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "grad_norm": 0.3202497363090515, | |
| "learning_rate": 9.929006627092298e-06, | |
| "loss": 0.4684, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "eval_loss": 0.5222189426422119, | |
| "eval_runtime": 31.8492, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 21.176470588235293, | |
| "grad_norm": 0.2713940441608429, | |
| "learning_rate": 9.924038765061042e-06, | |
| "loss": 0.5016, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 21.41176470588235, | |
| "grad_norm": 0.2259199619293213, | |
| "learning_rate": 9.918904242612794e-06, | |
| "loss": 0.5315, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 21.647058823529413, | |
| "grad_norm": 0.2454088032245636, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 0.474, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 21.88235294117647, | |
| "grad_norm": 0.18309438228607178, | |
| "learning_rate": 9.908135917238321e-06, | |
| "loss": 0.3029, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.5099583864212036, | |
| "eval_runtime": 31.8371, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 22.11764705882353, | |
| "grad_norm": 0.3067823648452759, | |
| "learning_rate": 9.902502478779897e-06, | |
| "loss": 0.3838, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 22.352941176470587, | |
| "grad_norm": 0.2594364881515503, | |
| "learning_rate": 9.896703108827758e-06, | |
| "loss": 0.48, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 22.58823529411765, | |
| "grad_norm": 0.27693891525268555, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.4363, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 22.823529411764707, | |
| "grad_norm": 0.304267019033432, | |
| "learning_rate": 9.884607365200355e-06, | |
| "loss": 0.3522, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 22.941176470588236, | |
| "eval_loss": 0.5085048079490662, | |
| "eval_runtime": 31.8502, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 23.058823529411764, | |
| "grad_norm": 0.17085126042366028, | |
| "learning_rate": 9.878311400921072e-06, | |
| "loss": 0.2722, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 23.294117647058822, | |
| "grad_norm": 0.4169885218143463, | |
| "learning_rate": 9.871850323926178e-06, | |
| "loss": 0.4118, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 0.25028589367866516, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.3777, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.764705882352942, | |
| "grad_norm": 0.3203675150871277, | |
| "learning_rate": 9.858433712104403e-06, | |
| "loss": 0.5195, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.3470858931541443, | |
| "learning_rate": 9.851478631379982e-06, | |
| "loss": 0.3855, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.49713531136512756, | |
| "eval_runtime": 31.8498, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.235294117647058, | |
| "grad_norm": 0.3214435875415802, | |
| "learning_rate": 9.844359346129504e-06, | |
| "loss": 0.476, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 24.470588235294116, | |
| "grad_norm": 0.3332065939903259, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.4383, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 24.705882352941178, | |
| "grad_norm": 0.22289708256721497, | |
| "learning_rate": 9.829629131445342e-06, | |
| "loss": 0.3256, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "grad_norm": 0.25234198570251465, | |
| "learning_rate": 9.822018700574696e-06, | |
| "loss": 0.317, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "eval_loss": 0.504925549030304, | |
| "eval_runtime": 31.8487, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 25.176470588235293, | |
| "grad_norm": 0.3678690493106842, | |
| "learning_rate": 9.81424506228719e-06, | |
| "loss": 0.4398, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 25.41176470588235, | |
| "grad_norm": 0.31491589546203613, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.3335, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 25.647058823529413, | |
| "grad_norm": 0.3429551124572754, | |
| "learning_rate": 9.798209221411748e-06, | |
| "loss": 0.4467, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 25.88235294117647, | |
| "grad_norm": 0.30439844727516174, | |
| "learning_rate": 9.789947561577445e-06, | |
| "loss": 0.338, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.5015532970428467, | |
| "eval_runtime": 31.8415, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 26.11764705882353, | |
| "grad_norm": 0.2530220150947571, | |
| "learning_rate": 9.781523779815178e-06, | |
| "loss": 0.3514, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 26.352941176470587, | |
| "grad_norm": 0.2835986018180847, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.3022, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 26.58823529411765, | |
| "grad_norm": 0.3710886538028717, | |
| "learning_rate": 9.764190996439181e-06, | |
| "loss": 0.4567, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 26.823529411764707, | |
| "grad_norm": 0.3602848947048187, | |
| "learning_rate": 9.755282581475769e-06, | |
| "loss": 0.391, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 26.941176470588236, | |
| "eval_loss": 0.49422508478164673, | |
| "eval_runtime": 31.8515, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 27.058823529411764, | |
| "grad_norm": 0.3394179940223694, | |
| "learning_rate": 9.74621321786517e-06, | |
| "loss": 0.3986, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 27.294117647058822, | |
| "grad_norm": 0.4131983518600464, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 0.359, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 27.529411764705884, | |
| "grad_norm": 0.35208216309547424, | |
| "learning_rate": 9.727592877996585e-06, | |
| "loss": 0.2981, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 27.764705882352942, | |
| "grad_norm": 0.42573440074920654, | |
| "learning_rate": 9.718042531967918e-06, | |
| "loss": 0.4258, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.3431633412837982, | |
| "learning_rate": 9.708332497729378e-06, | |
| "loss": 0.3964, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.5010454058647156, | |
| "eval_runtime": 31.8416, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.235294117647058, | |
| "grad_norm": 0.31860944628715515, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.2962, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 28.470588235294116, | |
| "grad_norm": 0.34731313586235046, | |
| "learning_rate": 9.688434684610725e-06, | |
| "loss": 0.3673, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 28.705882352941178, | |
| "grad_norm": 0.36615273356437683, | |
| "learning_rate": 9.678247579197658e-06, | |
| "loss": 0.5272, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "grad_norm": 0.2973138689994812, | |
| "learning_rate": 9.667902132486009e-06, | |
| "loss": 0.2951, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "eval_loss": 0.509841799736023, | |
| "eval_runtime": 31.843, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 29.176470588235293, | |
| "grad_norm": 0.3237411379814148, | |
| "learning_rate": 9.657398694630713e-06, | |
| "loss": 0.2387, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 29.41176470588235, | |
| "grad_norm": 0.31316205859184265, | |
| "learning_rate": 9.646737621134112e-06, | |
| "loss": 0.3701, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 29.647058823529413, | |
| "grad_norm": 0.41230136156082153, | |
| "learning_rate": 9.635919272833938e-06, | |
| "loss": 0.4157, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 29.88235294117647, | |
| "grad_norm": 0.4581688940525055, | |
| "learning_rate": 9.62494401589108e-06, | |
| "loss": 0.4021, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.5068490505218506, | |
| "eval_runtime": 31.8451, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 30.11764705882353, | |
| "grad_norm": 0.3983169198036194, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.3278, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 30.352941176470587, | |
| "grad_norm": 0.26158151030540466, | |
| "learning_rate": 9.602524267262202e-06, | |
| "loss": 0.1625, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 30.58823529411765, | |
| "grad_norm": 0.4597890377044678, | |
| "learning_rate": 9.591080534401371e-06, | |
| "loss": 0.4323, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 30.823529411764707, | |
| "grad_norm": 0.4650469422340393, | |
| "learning_rate": 9.579481410522556e-06, | |
| "loss": 0.4021, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 30.941176470588236, | |
| "eval_loss": 0.507030189037323, | |
| "eval_runtime": 31.8372, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 31.058823529411764, | |
| "grad_norm": 0.41834914684295654, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.3066, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 31.294117647058822, | |
| "grad_norm": 0.40843620896339417, | |
| "learning_rate": 9.555818565306086e-06, | |
| "loss": 0.288, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 31.529411764705884, | |
| "grad_norm": 0.3444281816482544, | |
| "learning_rate": 9.543755644867823e-06, | |
| "loss": 0.3391, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 31.764705882352942, | |
| "grad_norm": 0.42617908120155334, | |
| "learning_rate": 9.531538935183252e-06, | |
| "loss": 0.396, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.47771233320236206, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.3456, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.5024969577789307, | |
| "eval_runtime": 31.841, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 32.23529411764706, | |
| "grad_norm": 0.5327393412590027, | |
| "learning_rate": 9.506645807227311e-06, | |
| "loss": 0.3042, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 32.470588235294116, | |
| "grad_norm": 0.35559046268463135, | |
| "learning_rate": 9.493970231495836e-06, | |
| "loss": 0.386, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 32.705882352941174, | |
| "grad_norm": 0.3480581045150757, | |
| "learning_rate": 9.481142551569318e-06, | |
| "loss": 0.2265, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 32.94117647058823, | |
| "grad_norm": 0.5482520461082458, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 0.4431, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 32.94117647058823, | |
| "eval_loss": 0.5049856901168823, | |
| "eval_runtime": 31.8427, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 33.1764705882353, | |
| "grad_norm": 0.3856636583805084, | |
| "learning_rate": 9.45503262094184e-06, | |
| "loss": 0.1491, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 33.411764705882355, | |
| "grad_norm": 0.5228236317634583, | |
| "learning_rate": 9.441751253965022e-06, | |
| "loss": 0.3196, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 33.64705882352941, | |
| "grad_norm": 0.38300809264183044, | |
| "learning_rate": 9.428319550211531e-06, | |
| "loss": 0.2612, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 33.88235294117647, | |
| "grad_norm": 0.5598320364952087, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.4131, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.5093716382980347, | |
| "eval_runtime": 31.8401, | |
| "eval_samples_per_second": 1.131, | |
| "eval_steps_per_second": 1.131, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "step": 289, | |
| "total_flos": 3.6265283397720474e+17, | |
| "train_loss": 0.8757064017663777, | |
| "train_runtime": 13296.8495, | |
| "train_samples_per_second": 1.534, | |
| "train_steps_per_second": 0.09 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.6265283397720474e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |