| { | |
| "best_metric": 0.7155756950378418, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-4/checkpoint-246", | |
| "epoch": 36.0, | |
| "eval_steps": 500, | |
| "global_step": 306, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 0.6938014030456543, | |
| "learning_rate": 8.333333333333334e-08, | |
| "loss": 1.1573, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 0.7594343423843384, | |
| "learning_rate": 1.6666666666666668e-07, | |
| "loss": 1.1671, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.7093557715415955, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 1.1466, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.6927754878997803, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.1113, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.6476206183433533, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 1.0917, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 1.1325923204421997, | |
| "eval_runtime": 37.009, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.6894296407699585, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 1.1424, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.6622559428215027, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.1225, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.6495747566223145, | |
| "learning_rate": 1.1666666666666668e-06, | |
| "loss": 1.0675, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.7559149861335754, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.1392, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1271247863769531, | |
| "eval_runtime": 36.9985, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.6588976979255676, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.1406, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.6868743300437927, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 1.0912, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.699292778968811, | |
| "learning_rate": 1.8333333333333333e-06, | |
| "loss": 1.1073, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.7510226964950562, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.1572, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 1.1161284446716309, | |
| "eval_runtime": 37.0046, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.7101291418075562, | |
| "learning_rate": 2.166666666666667e-06, | |
| "loss": 1.0988, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.5981112718582153, | |
| "learning_rate": 2.3333333333333336e-06, | |
| "loss": 1.0899, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.7123465538024902, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.1263, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.6823212504386902, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.1119, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.7324620485305786, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 1.0856, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.0970431566238403, | |
| "eval_runtime": 37.0081, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 0.7198031544685364, | |
| "learning_rate": 3e-06, | |
| "loss": 1.1012, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.7194771766662598, | |
| "learning_rate": 3.1666666666666667e-06, | |
| "loss": 1.0752, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.702494740486145, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.0602, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 0.6754999160766602, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.0789, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "eval_loss": 1.0719497203826904, | |
| "eval_runtime": 37.0094, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 0.7691217660903931, | |
| "learning_rate": 3.6666666666666666e-06, | |
| "loss": 1.0709, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 0.6451730132102966, | |
| "learning_rate": 3.833333333333334e-06, | |
| "loss": 1.0372, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 0.6900641322135925, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9947, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.6281608939170837, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.0546, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.0336723327636719, | |
| "eval_runtime": 37.0003, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.6548106074333191, | |
| "learning_rate": 4.333333333333334e-06, | |
| "loss": 1.0604, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.6569029688835144, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.9777, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 0.6138703227043152, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 1.0248, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 0.5528715252876282, | |
| "learning_rate": 4.833333333333333e-06, | |
| "loss": 0.9806, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "eval_loss": 0.9922827482223511, | |
| "eval_runtime": 37.0043, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.5543631315231323, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0264, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 0.5159863233566284, | |
| "learning_rate": 5.1666666666666675e-06, | |
| "loss": 0.9702, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 0.5009753704071045, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.9452, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 0.49283164739608765, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.9367, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.47310400009155273, | |
| "learning_rate": 5.666666666666667e-06, | |
| "loss": 0.9617, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9472079873085022, | |
| "eval_runtime": 37.0171, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.3864011764526367, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.9326, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.37244734168052673, | |
| "learning_rate": 6e-06, | |
| "loss": 0.948, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.3857339322566986, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 0.8741, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 0.39583349227905273, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 0.933, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "eval_loss": 0.912886381149292, | |
| "eval_runtime": 37.0055, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 0.30479201674461365, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.8891, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.30881431698799133, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.848, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.3065626919269562, | |
| "learning_rate": 6.833333333333334e-06, | |
| "loss": 0.8886, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.3278353214263916, | |
| "learning_rate": 7e-06, | |
| "loss": 0.9005, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8825066089630127, | |
| "eval_runtime": 36.9987, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 0.3112061023712158, | |
| "learning_rate": 7.166666666666667e-06, | |
| "loss": 0.8631, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 0.2803957164287567, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 0.8166, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.2763586938381195, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.8208, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 0.29917630553245544, | |
| "learning_rate": 7.666666666666667e-06, | |
| "loss": 0.9183, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "eval_loss": 0.8577808141708374, | |
| "eval_runtime": 37.0008, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.2617398202419281, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.8786, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 0.251274436712265, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8257, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 0.27396726608276367, | |
| "learning_rate": 8.166666666666668e-06, | |
| "loss": 0.8913, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.27446839213371277, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.7921, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.24932074546813965, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.7739, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8334917426109314, | |
| "eval_runtime": 37.0127, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.297094464302063, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 0.8754, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 0.23659107089042664, | |
| "learning_rate": 8.833333333333334e-06, | |
| "loss": 0.764, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 0.23405446112155914, | |
| "learning_rate": 9e-06, | |
| "loss": 0.8218, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 0.23257797956466675, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.7649, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "eval_loss": 0.8164898157119751, | |
| "eval_runtime": 37.01, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 13.176470588235293, | |
| "grad_norm": 0.23545639216899872, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.7674, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 13.411764705882353, | |
| "grad_norm": 0.26143553853034973, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.7662, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 13.647058823529411, | |
| "grad_norm": 0.21934638917446136, | |
| "learning_rate": 9.666666666666667e-06, | |
| "loss": 0.7928, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 13.882352941176471, | |
| "grad_norm": 0.2396870255470276, | |
| "learning_rate": 9.833333333333333e-06, | |
| "loss": 0.8197, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.8008810877799988, | |
| "eval_runtime": 36.9979, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 0.24245895445346832, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8115, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 14.352941176470589, | |
| "grad_norm": 0.23691204190254211, | |
| "learning_rate": 9.999915384288723e-06, | |
| "loss": 0.7565, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 14.588235294117647, | |
| "grad_norm": 0.21797290444374084, | |
| "learning_rate": 9.999661540018812e-06, | |
| "loss": 0.7858, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 14.823529411764707, | |
| "grad_norm": 0.23253759741783142, | |
| "learning_rate": 9.999238475781957e-06, | |
| "loss": 0.7488, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 14.941176470588236, | |
| "eval_loss": 0.7888637185096741, | |
| "eval_runtime": 36.999, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 15.058823529411764, | |
| "grad_norm": 0.2564374804496765, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.7646, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 15.294117647058824, | |
| "grad_norm": 0.23032395541667938, | |
| "learning_rate": 9.997884750411004e-06, | |
| "loss": 0.7598, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 15.529411764705882, | |
| "grad_norm": 0.22293731570243835, | |
| "learning_rate": 9.99695413509548e-06, | |
| "loss": 0.733, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 15.764705882352942, | |
| "grad_norm": 0.2295634150505066, | |
| "learning_rate": 9.995854391448607e-06, | |
| "loss": 0.7611, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.24928726255893707, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 0.7651, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.7769894003868103, | |
| "eval_runtime": 36.9969, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.235294117647058, | |
| "grad_norm": 0.21673335134983063, | |
| "learning_rate": 9.993147673772869e-06, | |
| "loss": 0.7225, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 0.22104425728321075, | |
| "learning_rate": 9.991540791356342e-06, | |
| "loss": 0.7727, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 16.705882352941178, | |
| "grad_norm": 0.23531343042850494, | |
| "learning_rate": 9.989764963830038e-06, | |
| "loss": 0.7589, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "grad_norm": 0.21621249616146088, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.6992, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "eval_loss": 0.7678544521331787, | |
| "eval_runtime": 36.9951, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 17.176470588235293, | |
| "grad_norm": 0.242046520113945, | |
| "learning_rate": 9.985706719584888e-06, | |
| "loss": 0.7617, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 17.41176470588235, | |
| "grad_norm": 0.21592232584953308, | |
| "learning_rate": 9.98342444022253e-06, | |
| "loss": 0.6826, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 0.2592144310474396, | |
| "learning_rate": 9.980973490458728e-06, | |
| "loss": 0.7452, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 17.88235294117647, | |
| "grad_norm": 0.2321804314851761, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 0.7609, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.7580164670944214, | |
| "eval_runtime": 37.0023, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 18.11764705882353, | |
| "grad_norm": 0.2417711317539215, | |
| "learning_rate": 9.975565917255017e-06, | |
| "loss": 0.6975, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 18.352941176470587, | |
| "grad_norm": 0.2503942549228668, | |
| "learning_rate": 9.972609476841368e-06, | |
| "loss": 0.7259, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 18.58823529411765, | |
| "grad_norm": 0.23440955579280853, | |
| "learning_rate": 9.9694847320726e-06, | |
| "loss": 0.7242, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 18.823529411764707, | |
| "grad_norm": 0.2478140890598297, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.6868, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 18.941176470588236, | |
| "eval_loss": 0.7510344386100769, | |
| "eval_runtime": 37.0057, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 19.058823529411764, | |
| "grad_norm": 0.2523338496685028, | |
| "learning_rate": 9.962730758206612e-06, | |
| "loss": 0.7132, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 19.294117647058822, | |
| "grad_norm": 0.2554718255996704, | |
| "learning_rate": 9.959101757706308e-06, | |
| "loss": 0.7306, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 19.529411764705884, | |
| "grad_norm": 0.2540837526321411, | |
| "learning_rate": 9.955304910036993e-06, | |
| "loss": 0.6959, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 19.764705882352942, | |
| "grad_norm": 0.23473863303661346, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.677, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.26500314474105835, | |
| "learning_rate": 9.947208192904722e-06, | |
| "loss": 0.7077, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7436654567718506, | |
| "eval_runtime": 37.0063, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.235294117647058, | |
| "grad_norm": 0.2688054144382477, | |
| "learning_rate": 9.942908597485558e-06, | |
| "loss": 0.7092, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 20.470588235294116, | |
| "grad_norm": 0.24925930798053741, | |
| "learning_rate": 9.938441702975689e-06, | |
| "loss": 0.7176, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 20.705882352941178, | |
| "grad_norm": 0.2611039876937866, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.6457, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "grad_norm": 0.2844160497188568, | |
| "learning_rate": 9.929006627092298e-06, | |
| "loss": 0.6862, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "eval_loss": 0.7379278540611267, | |
| "eval_runtime": 36.9999, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 21.176470588235293, | |
| "grad_norm": 0.27488094568252563, | |
| "learning_rate": 9.924038765061042e-06, | |
| "loss": 0.7067, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 21.41176470588235, | |
| "grad_norm": 0.24594415724277496, | |
| "learning_rate": 9.918904242612794e-06, | |
| "loss": 0.6842, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 21.647058823529413, | |
| "grad_norm": 0.2662781774997711, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 0.6661, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 21.88235294117647, | |
| "grad_norm": 0.2941681146621704, | |
| "learning_rate": 9.908135917238321e-06, | |
| "loss": 0.6939, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.7319392561912537, | |
| "eval_runtime": 37.0166, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 22.11764705882353, | |
| "grad_norm": 0.3132288157939911, | |
| "learning_rate": 9.902502478779897e-06, | |
| "loss": 0.6804, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 22.352941176470587, | |
| "grad_norm": 0.2927018105983734, | |
| "learning_rate": 9.896703108827758e-06, | |
| "loss": 0.649, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 22.58823529411765, | |
| "grad_norm": 0.28771597146987915, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.6525, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 22.823529411764707, | |
| "grad_norm": 0.31475573778152466, | |
| "learning_rate": 9.884607365200355e-06, | |
| "loss": 0.6564, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 22.941176470588236, | |
| "eval_loss": 0.7275494933128357, | |
| "eval_runtime": 37.002, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 23.058823529411764, | |
| "grad_norm": 0.29836949706077576, | |
| "learning_rate": 9.878311400921072e-06, | |
| "loss": 0.6561, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 23.294117647058822, | |
| "grad_norm": 0.30295827984809875, | |
| "learning_rate": 9.871850323926178e-06, | |
| "loss": 0.6815, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 0.2973706126213074, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.6261, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.764705882352942, | |
| "grad_norm": 0.32414522767066956, | |
| "learning_rate": 9.858433712104403e-06, | |
| "loss": 0.6752, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.30349239706993103, | |
| "learning_rate": 9.851478631379982e-06, | |
| "loss": 0.6446, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.7236446738243103, | |
| "eval_runtime": 37.004, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.235294117647058, | |
| "grad_norm": 0.30559128522872925, | |
| "learning_rate": 9.844359346129504e-06, | |
| "loss": 0.6672, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 24.470588235294116, | |
| "grad_norm": 0.2979726195335388, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.6283, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 24.705882352941178, | |
| "grad_norm": 0.3105686604976654, | |
| "learning_rate": 9.829629131445342e-06, | |
| "loss": 0.6412, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "grad_norm": 0.3595413863658905, | |
| "learning_rate": 9.822018700574696e-06, | |
| "loss": 0.6304, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "eval_loss": 0.7200143337249756, | |
| "eval_runtime": 37.0035, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 25.176470588235293, | |
| "grad_norm": 0.3050694763660431, | |
| "learning_rate": 9.81424506228719e-06, | |
| "loss": 0.6179, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 25.41176470588235, | |
| "grad_norm": 0.3428696095943451, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.6223, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 25.647058823529413, | |
| "grad_norm": 0.3474690616130829, | |
| "learning_rate": 9.798209221411748e-06, | |
| "loss": 0.652, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 25.88235294117647, | |
| "grad_norm": 0.38103392720222473, | |
| "learning_rate": 9.789947561577445e-06, | |
| "loss": 0.6583, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.7177853584289551, | |
| "eval_runtime": 37.0015, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 26.11764705882353, | |
| "grad_norm": 0.3361453413963318, | |
| "learning_rate": 9.781523779815178e-06, | |
| "loss": 0.6239, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 26.352941176470587, | |
| "grad_norm": 0.38879382610321045, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.6229, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 26.58823529411765, | |
| "grad_norm": 0.3591279685497284, | |
| "learning_rate": 9.764190996439181e-06, | |
| "loss": 0.6374, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 26.823529411764707, | |
| "grad_norm": 0.3179728388786316, | |
| "learning_rate": 9.755282581475769e-06, | |
| "loss": 0.5974, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 26.941176470588236, | |
| "eval_loss": 0.7159412503242493, | |
| "eval_runtime": 36.9995, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 27.058823529411764, | |
| "grad_norm": 0.3596520721912384, | |
| "learning_rate": 9.74621321786517e-06, | |
| "loss": 0.6042, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 27.294117647058822, | |
| "grad_norm": 0.36043640971183777, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 0.6058, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 27.529411764705884, | |
| "grad_norm": 0.35985061526298523, | |
| "learning_rate": 9.727592877996585e-06, | |
| "loss": 0.5959, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 27.764705882352942, | |
| "grad_norm": 0.3502049446105957, | |
| "learning_rate": 9.718042531967918e-06, | |
| "loss": 0.5894, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.3679797053337097, | |
| "learning_rate": 9.708332497729378e-06, | |
| "loss": 0.6266, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.7158757448196411, | |
| "eval_runtime": 37.0093, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.235294117647058, | |
| "grad_norm": 0.35959428548812866, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.5731, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 28.470588235294116, | |
| "grad_norm": 0.3938836455345154, | |
| "learning_rate": 9.688434684610725e-06, | |
| "loss": 0.5868, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 28.705882352941178, | |
| "grad_norm": 0.6573047637939453, | |
| "learning_rate": 9.678247579197658e-06, | |
| "loss": 0.6126, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "grad_norm": 0.3609618842601776, | |
| "learning_rate": 9.667902132486009e-06, | |
| "loss": 0.6081, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "eval_loss": 0.7155756950378418, | |
| "eval_runtime": 37.01, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 29.176470588235293, | |
| "grad_norm": 0.38799017667770386, | |
| "learning_rate": 9.657398694630713e-06, | |
| "loss": 0.5564, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 29.41176470588235, | |
| "grad_norm": 0.4135487675666809, | |
| "learning_rate": 9.646737621134112e-06, | |
| "loss": 0.6149, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 29.647058823529413, | |
| "grad_norm": 0.3742540180683136, | |
| "learning_rate": 9.635919272833938e-06, | |
| "loss": 0.5747, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 29.88235294117647, | |
| "grad_norm": 0.40949445962905884, | |
| "learning_rate": 9.62494401589108e-06, | |
| "loss": 0.5853, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.7164749503135681, | |
| "eval_runtime": 37.0011, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 30.11764705882353, | |
| "grad_norm": 0.4069327414035797, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.5611, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 30.352941176470587, | |
| "grad_norm": 0.4262481927871704, | |
| "learning_rate": 9.602524267262202e-06, | |
| "loss": 0.5845, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 30.58823529411765, | |
| "grad_norm": 0.42745286226272583, | |
| "learning_rate": 9.591080534401371e-06, | |
| "loss": 0.5742, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 30.823529411764707, | |
| "grad_norm": 0.4548822343349457, | |
| "learning_rate": 9.579481410522556e-06, | |
| "loss": 0.54, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 30.941176470588236, | |
| "eval_loss": 0.7200043201446533, | |
| "eval_runtime": 37.0013, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 31.058823529411764, | |
| "grad_norm": 0.41799986362457275, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.5325, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 31.294117647058822, | |
| "grad_norm": 0.4568859338760376, | |
| "learning_rate": 9.555818565306086e-06, | |
| "loss": 0.5688, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 31.529411764705884, | |
| "grad_norm": 0.441699355840683, | |
| "learning_rate": 9.543755644867823e-06, | |
| "loss": 0.5628, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 31.764705882352942, | |
| "grad_norm": 0.45061326026916504, | |
| "learning_rate": 9.531538935183252e-06, | |
| "loss": 0.5419, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.4943307638168335, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.5438, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.7231061458587646, | |
| "eval_runtime": 37.0086, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 32.23529411764706, | |
| "grad_norm": 0.46770891547203064, | |
| "learning_rate": 9.506645807227311e-06, | |
| "loss": 0.5162, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 32.470588235294116, | |
| "grad_norm": 0.48194897174835205, | |
| "learning_rate": 9.493970231495836e-06, | |
| "loss": 0.5715, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 32.705882352941174, | |
| "grad_norm": 0.47527700662612915, | |
| "learning_rate": 9.481142551569318e-06, | |
| "loss": 0.5233, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 32.94117647058823, | |
| "grad_norm": 0.5583000779151917, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 0.5163, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 32.94117647058823, | |
| "eval_loss": 0.7267724275588989, | |
| "eval_runtime": 37.0063, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 33.1764705882353, | |
| "grad_norm": 0.5567420125007629, | |
| "learning_rate": 9.45503262094184e-06, | |
| "loss": 0.5146, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 33.411764705882355, | |
| "grad_norm": 0.5160362124443054, | |
| "learning_rate": 9.441751253965022e-06, | |
| "loss": 0.5227, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 33.64705882352941, | |
| "grad_norm": 0.5851587057113647, | |
| "learning_rate": 9.428319550211531e-06, | |
| "loss": 0.4936, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 33.88235294117647, | |
| "grad_norm": 0.564764142036438, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.5302, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.7345450520515442, | |
| "eval_runtime": 37.0031, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 34.11764705882353, | |
| "grad_norm": 0.5420270562171936, | |
| "learning_rate": 9.401006955900555e-06, | |
| "loss": 0.5254, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 34.35294117647059, | |
| "grad_norm": 0.5466279983520508, | |
| "learning_rate": 9.38712698977291e-06, | |
| "loss": 0.5236, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 34.588235294117645, | |
| "grad_norm": 0.5974857807159424, | |
| "learning_rate": 9.37309853569698e-06, | |
| "loss": 0.4812, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 34.8235294117647, | |
| "grad_norm": 0.6072189807891846, | |
| "learning_rate": 9.358922068483813e-06, | |
| "loss": 0.4865, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 34.94117647058823, | |
| "eval_loss": 0.7542312145233154, | |
| "eval_runtime": 37.0073, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 35.05882352941177, | |
| "grad_norm": 0.7493066191673279, | |
| "learning_rate": 9.344598067954151e-06, | |
| "loss": 0.4651, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 35.294117647058826, | |
| "grad_norm": 0.6359327435493469, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.5043, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 35.529411764705884, | |
| "grad_norm": 0.5997797846794128, | |
| "learning_rate": 9.315509411179182e-06, | |
| "loss": 0.4598, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 35.76470588235294, | |
| "grad_norm": 0.586499810218811, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.4543, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.6805587410926819, | |
| "learning_rate": 9.285836503510562e-06, | |
| "loss": 0.4774, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.7611451148986816, | |
| "eval_runtime": 37.0054, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.973, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "step": 306, | |
| "total_flos": 4.686344037995643e+17, | |
| "train_loss": 0.7647335636459924, | |
| "train_runtime": 16723.6429, | |
| "train_samples_per_second": 1.22, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "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": 4.686344037995643e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |