| { | |
| "best_metric": 1.141607403755188, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_180-samples_config-1/checkpoint-51", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.058823529411764705, | |
| "grad_norm": 5.94310998916626, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 2.2854, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 4.818267822265625, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 2.0252, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 5.333231449127197, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 2.192, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 5.257590293884277, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 2.3389, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 5.808491230010986, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 2.1288, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 5.0901618003845215, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 2.153, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 4.486069202423096, | |
| "learning_rate": 1.411764705882353e-05, | |
| "loss": 1.9168, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 4.509524345397949, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 1.8902, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 3.2386724948883057, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 1.6974, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.5618224143981934, | |
| "eval_runtime": 14.5608, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 2.472, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 2.549015760421753, | |
| "learning_rate": 2.1176470588235296e-05, | |
| "loss": 1.5601, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 2.3952205181121826, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 1.5338, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 1.7811028957366943, | |
| "learning_rate": 2.5882352941176475e-05, | |
| "loss": 1.3579, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.6944749355316162, | |
| "learning_rate": 2.823529411764706e-05, | |
| "loss": 1.4151, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 1.640507459640503, | |
| "learning_rate": 3.058823529411765e-05, | |
| "loss": 1.3185, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.781482219696045, | |
| "learning_rate": 3.294117647058824e-05, | |
| "loss": 1.2674, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 1.622686743736267, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 1.2291, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.5880047082901, | |
| "learning_rate": 3.7647058823529415e-05, | |
| "loss": 1.2269, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4253026247024536, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1802, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.2333735227584839, | |
| "eval_runtime": 14.5754, | |
| "eval_samples_per_second": 2.47, | |
| "eval_steps_per_second": 2.47, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 1.3464667797088623, | |
| "learning_rate": 4.235294117647059e-05, | |
| "loss": 1.0786, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 1.4025360345840454, | |
| "learning_rate": 4.470588235294118e-05, | |
| "loss": 1.1678, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.4293924570083618, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.9548, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 1.6406131982803345, | |
| "learning_rate": 4.9411764705882355e-05, | |
| "loss": 1.1246, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 1.757340908050537, | |
| "learning_rate": 5.176470588235295e-05, | |
| "loss": 1.0679, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 1.5292569398880005, | |
| "learning_rate": 5.411764705882353e-05, | |
| "loss": 1.055, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 1.6216624975204468, | |
| "learning_rate": 5.647058823529412e-05, | |
| "loss": 1.04, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 1.6907811164855957, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.9812, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.141607403755188, | |
| "eval_runtime": 14.5508, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 1.5789411067962646, | |
| "learning_rate": 6.11764705882353e-05, | |
| "loss": 0.9357, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 1.6847856044769287, | |
| "learning_rate": 6.352941176470588e-05, | |
| "loss": 0.844, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 1.7474318742752075, | |
| "learning_rate": 6.588235294117648e-05, | |
| "loss": 0.8198, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 1.8489917516708374, | |
| "learning_rate": 6.823529411764707e-05, | |
| "loss": 0.8127, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 2.2157130241394043, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.8208, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 2.560288667678833, | |
| "learning_rate": 7.294117647058823e-05, | |
| "loss": 0.7702, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 2.116132974624634, | |
| "learning_rate": 7.529411764705883e-05, | |
| "loss": 0.7093, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 2.6108651161193848, | |
| "learning_rate": 7.764705882352942e-05, | |
| "loss": 0.7901, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.8218464851379395, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6749, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1867635250091553, | |
| "eval_runtime": 14.5837, | |
| "eval_samples_per_second": 2.469, | |
| "eval_steps_per_second": 2.469, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 1.9257030487060547, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.5332, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 2.3828821182250977, | |
| "learning_rate": 8.470588235294118e-05, | |
| "loss": 0.5186, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 2.9151997566223145, | |
| "learning_rate": 8.705882352941177e-05, | |
| "loss": 0.5018, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 3.63409161567688, | |
| "learning_rate": 8.941176470588236e-05, | |
| "loss": 0.4376, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 3.7016890048980713, | |
| "learning_rate": 9.176470588235295e-05, | |
| "loss": 0.4247, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 4.303072452545166, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.4546, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 3.1941800117492676, | |
| "learning_rate": 9.647058823529412e-05, | |
| "loss": 0.4463, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 3.1212875843048096, | |
| "learning_rate": 9.882352941176471e-05, | |
| "loss": 0.4963, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.3243592977523804, | |
| "eval_runtime": 14.5574, | |
| "eval_samples_per_second": 2.473, | |
| "eval_steps_per_second": 2.473, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0588235294117645, | |
| "grad_norm": 2.331570863723755, | |
| "learning_rate": 9.99995783847866e-05, | |
| "loss": 0.3668, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 1.8265957832336426, | |
| "learning_rate": 9.999620550574153e-05, | |
| "loss": 0.241, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 2.5654234886169434, | |
| "learning_rate": 9.998945997517956e-05, | |
| "loss": 0.2197, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 4.596362113952637, | |
| "learning_rate": 9.997934224814173e-05, | |
| "loss": 0.2602, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.529411764705882, | |
| "grad_norm": 5.800356864929199, | |
| "learning_rate": 9.996585300715116e-05, | |
| "loss": 0.2459, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 4.117295265197754, | |
| "learning_rate": 9.994899316216708e-05, | |
| "loss": 0.1648, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.764705882352941, | |
| "grad_norm": 4.657989025115967, | |
| "learning_rate": 9.992876385052345e-05, | |
| "loss": 0.2517, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 3.0068447589874268, | |
| "learning_rate": 9.990516643685222e-05, | |
| "loss": 0.2779, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.5185630321502686, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.2296, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.5520074367523193, | |
| "eval_runtime": 14.5572, | |
| "eval_samples_per_second": 2.473, | |
| "eval_steps_per_second": 2.473, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 1.8949757814407349, | |
| "learning_rate": 9.984787389787688e-05, | |
| "loss": 0.1204, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 6.235294117647059, | |
| "grad_norm": 2.084357976913452, | |
| "learning_rate": 9.981418263742148e-05, | |
| "loss": 0.1057, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 2.355532169342041, | |
| "learning_rate": 9.977713100437509e-05, | |
| "loss": 0.1279, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 2.634601354598999, | |
| "learning_rate": 9.973672149817232e-05, | |
| "loss": 0.1167, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 3.470742702484131, | |
| "learning_rate": 9.96929568447637e-05, | |
| "loss": 0.1036, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 3.562553882598877, | |
| "learning_rate": 9.964583999643174e-05, | |
| "loss": 0.1264, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 3.067589282989502, | |
| "learning_rate": 9.95953741315919e-05, | |
| "loss": 0.1229, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "grad_norm": 4.300107479095459, | |
| "learning_rate": 9.954156265457801e-05, | |
| "loss": 0.126, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.8394368886947632, | |
| "eval_runtime": 14.5639, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 2.472, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.2025381326675415, | |
| "learning_rate": 9.948440919541278e-05, | |
| "loss": 0.1183, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.176470588235294, | |
| "grad_norm": 2.12835955619812, | |
| "learning_rate": 9.942391760956277e-05, | |
| "loss": 0.0738, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 1.7308270931243896, | |
| "learning_rate": 9.936009197767845e-05, | |
| "loss": 0.0601, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 7.411764705882353, | |
| "grad_norm": 1.6031392812728882, | |
| "learning_rate": 9.929293660531888e-05, | |
| "loss": 0.0587, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 2.102191686630249, | |
| "learning_rate": 9.922245602266118e-05, | |
| "loss": 0.0554, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 2.9458324909210205, | |
| "learning_rate": 9.91486549841951e-05, | |
| "loss": 0.0719, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 2.767437696456909, | |
| "learning_rate": 9.90715384684021e-05, | |
| "loss": 0.0813, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 7.882352941176471, | |
| "grad_norm": 1.5180050134658813, | |
| "learning_rate": 9.899111167741966e-05, | |
| "loss": 0.0644, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.601783275604248, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.1044, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.9316412210464478, | |
| "eval_runtime": 14.5864, | |
| "eval_samples_per_second": 2.468, | |
| "eval_steps_per_second": 2.468, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 8.117647058823529, | |
| "grad_norm": 1.442734956741333, | |
| "learning_rate": 9.882034919459555e-05, | |
| "loss": 0.0412, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 1.7208150625228882, | |
| "learning_rate": 9.873002502207503e-05, | |
| "loss": 0.0245, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 2.408069372177124, | |
| "learning_rate": 9.863641361223024e-05, | |
| "loss": 0.0513, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 1.7206476926803589, | |
| "learning_rate": 9.853952127991372e-05, | |
| "loss": 0.0602, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.588235294117647, | |
| "grad_norm": 2.3409576416015625, | |
| "learning_rate": 9.843935456130295e-05, | |
| "loss": 0.0507, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 2.0807151794433594, | |
| "learning_rate": 9.833592021345937e-05, | |
| "loss": 0.0286, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 3.4072768688201904, | |
| "learning_rate": 9.822922521387276e-05, | |
| "loss": 0.0464, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 3.362941265106201, | |
| "learning_rate": 9.811927675999036e-05, | |
| "loss": 0.0735, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.0808990001678467, | |
| "eval_runtime": 14.5855, | |
| "eval_samples_per_second": 2.468, | |
| "eval_steps_per_second": 2.468, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 9.058823529411764, | |
| "grad_norm": 1.9431918859481812, | |
| "learning_rate": 9.800608226873142e-05, | |
| "loss": 0.0302, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 2.4730467796325684, | |
| "learning_rate": 9.788964937598689e-05, | |
| "loss": 0.0443, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 9.294117647058824, | |
| "grad_norm": 1.5794306993484497, | |
| "learning_rate": 9.776998593610428e-05, | |
| "loss": 0.0627, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 3.127790689468384, | |
| "learning_rate": 9.764710002135784e-05, | |
| "loss": 0.0535, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.529411764705882, | |
| "grad_norm": 1.959427833557129, | |
| "learning_rate": 9.752099992140399e-05, | |
| "loss": 0.0337, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 2.6444616317749023, | |
| "learning_rate": 9.739169414272217e-05, | |
| "loss": 0.0403, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 9.764705882352942, | |
| "grad_norm": 2.717651128768921, | |
| "learning_rate": 9.725919140804099e-05, | |
| "loss": 0.0456, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 1.916383147239685, | |
| "learning_rate": 9.71235006557497e-05, | |
| "loss": 0.0402, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.9298330545425415, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0393, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.041475534439087, | |
| "eval_runtime": 14.5657, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 2.472, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 170, | |
| "total_flos": 2.1328997535186944e+16, | |
| "train_loss": 0.6198049567858962, | |
| "train_runtime": 1615.3335, | |
| "train_samples_per_second": 4.21, | |
| "train_steps_per_second": 0.526 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 850, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "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": 2.1328997535186944e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |