| { | |
| "best_metric": 1.258855938911438, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-1_180-samples_config-4/checkpoint-187", | |
| "epoch": 28.941176470588236, | |
| "eval_steps": 500, | |
| "global_step": 246, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 1.5944128036499023, | |
| "learning_rate": 8.333333333333334e-08, | |
| "loss": 2.0252, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 1.674605369567871, | |
| "learning_rate": 1.6666666666666668e-07, | |
| "loss": 2.0106, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 1.6583189964294434, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 2.0484, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 1.7035460472106934, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 2.1137, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 1.8433172702789307, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 2.0972, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 2.0717952251434326, | |
| "eval_runtime": 14.5106, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.5908912420272827, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 2.0639, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.6493914127349854, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 2.1013, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.6529414653778076, | |
| "learning_rate": 1.1666666666666668e-06, | |
| "loss": 2.1046, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.7618305683135986, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.0234, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.054532766342163, | |
| "eval_runtime": 14.511, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 1.626803994178772, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.994, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.7270427942276, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 2.0662, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 1.6685792207717896, | |
| "learning_rate": 1.8333333333333333e-06, | |
| "loss": 2.1119, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 1.6774814128875732, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.0324, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 2.0288267135620117, | |
| "eval_runtime": 14.5108, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 1.728971242904663, | |
| "learning_rate": 2.166666666666667e-06, | |
| "loss": 2.0096, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 1.5353363752365112, | |
| "learning_rate": 2.3333333333333336e-06, | |
| "loss": 2.0035, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 1.7659140825271606, | |
| "learning_rate": 2.5e-06, | |
| "loss": 2.0145, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 1.7244279384613037, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.9661, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.814552664756775, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 2.0064, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9798120260238647, | |
| "eval_runtime": 14.4915, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 1.6728545427322388, | |
| "learning_rate": 3e-06, | |
| "loss": 1.9509, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 1.800223469734192, | |
| "learning_rate": 3.1666666666666667e-06, | |
| "loss": 1.9526, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 1.652232050895691, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.922, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 1.769357681274414, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.9611, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "eval_loss": 1.9139456748962402, | |
| "eval_runtime": 14.5046, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 2.482, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 1.7674305438995361, | |
| "learning_rate": 3.6666666666666666e-06, | |
| "loss": 1.8548, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 1.7830374240875244, | |
| "learning_rate": 3.833333333333334e-06, | |
| "loss": 1.8919, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 2.0368635654449463, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.8827, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 1.5133813619613647, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.8283, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.8090087175369263, | |
| "eval_runtime": 14.4945, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 1.5872511863708496, | |
| "learning_rate": 4.333333333333334e-06, | |
| "loss": 1.8319, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 1.550522804260254, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.7823, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 1.5826510190963745, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 1.7759, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.338010549545288, | |
| "learning_rate": 4.833333333333333e-06, | |
| "loss": 1.6817, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "eval_loss": 1.7011380195617676, | |
| "eval_runtime": 14.5144, | |
| "eval_samples_per_second": 2.48, | |
| "eval_steps_per_second": 2.48, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 1.1569616794586182, | |
| "learning_rate": 5e-06, | |
| "loss": 1.6941, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 1.2016438245773315, | |
| "learning_rate": 5.1666666666666675e-06, | |
| "loss": 1.7058, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 1.1135125160217285, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.6247, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 0.8475298285484314, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.5891, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.755041778087616, | |
| "learning_rate": 5.666666666666667e-06, | |
| "loss": 1.5762, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.6084668636322021, | |
| "eval_runtime": 14.5036, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 2.482, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.7044244408607483, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 1.5742, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 0.6197890043258667, | |
| "learning_rate": 6e-06, | |
| "loss": 1.5981, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 0.6438341736793518, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 1.5194, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 0.627289891242981, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 1.5529, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "eval_loss": 1.5658942461013794, | |
| "eval_runtime": 14.5082, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 0.6046221852302551, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.5489, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.5346683263778687, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.5338, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 0.5408790111541748, | |
| "learning_rate": 6.833333333333334e-06, | |
| "loss": 1.48, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 0.5899155735969543, | |
| "learning_rate": 7e-06, | |
| "loss": 1.4817, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.5206388235092163, | |
| "eval_runtime": 14.5107, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 0.5904850363731384, | |
| "learning_rate": 7.166666666666667e-06, | |
| "loss": 1.4943, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 0.5600600838661194, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 1.4419, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 0.4918341636657715, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.3978, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 0.5680522322654724, | |
| "learning_rate": 7.666666666666667e-06, | |
| "loss": 1.5125, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "eval_loss": 1.4815658330917358, | |
| "eval_runtime": 14.5146, | |
| "eval_samples_per_second": 2.48, | |
| "eval_steps_per_second": 2.48, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.4833405911922455, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 1.5109, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 0.5413841605186462, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.4351, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 0.5074775218963623, | |
| "learning_rate": 8.166666666666668e-06, | |
| "loss": 1.4643, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.5865756869316101, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.3969, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.5256586074829102, | |
| "learning_rate": 8.5e-06, | |
| "loss": 1.3226, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.4351813793182373, | |
| "eval_runtime": 14.5096, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.235294117647058, | |
| "grad_norm": 0.5593088269233704, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 1.4301, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 12.470588235294118, | |
| "grad_norm": 0.5165168642997742, | |
| "learning_rate": 8.833333333333334e-06, | |
| "loss": 1.2965, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 12.705882352941176, | |
| "grad_norm": 0.5415379405021667, | |
| "learning_rate": 9e-06, | |
| "loss": 1.3613, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "grad_norm": 0.5435191988945007, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 1.3823, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.941176470588236, | |
| "eval_loss": 1.3951374292373657, | |
| "eval_runtime": 14.5488, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 13.176470588235293, | |
| "grad_norm": 0.5483547449111938, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 1.3189, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 13.411764705882353, | |
| "grad_norm": 0.6265339851379395, | |
| "learning_rate": 9.5e-06, | |
| "loss": 1.3235, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 13.647058823529411, | |
| "grad_norm": 0.5347606539726257, | |
| "learning_rate": 9.666666666666667e-06, | |
| "loss": 1.3265, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 13.882352941176471, | |
| "grad_norm": 0.5550796389579773, | |
| "learning_rate": 9.833333333333333e-06, | |
| "loss": 1.2564, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.358044981956482, | |
| "eval_runtime": 14.5168, | |
| "eval_samples_per_second": 2.48, | |
| "eval_steps_per_second": 2.48, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 14.117647058823529, | |
| "grad_norm": 0.5553519129753113, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3011, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 14.352941176470589, | |
| "grad_norm": 0.5510237812995911, | |
| "learning_rate": 9.999915384288723e-06, | |
| "loss": 1.2783, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 14.588235294117647, | |
| "grad_norm": 0.5882507562637329, | |
| "learning_rate": 9.999661540018812e-06, | |
| "loss": 1.3425, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 14.823529411764707, | |
| "grad_norm": 0.5551594495773315, | |
| "learning_rate": 9.999238475781957e-06, | |
| "loss": 1.1936, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 14.941176470588236, | |
| "eval_loss": 1.330453872680664, | |
| "eval_runtime": 14.5004, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 15.058823529411764, | |
| "grad_norm": 0.6728988885879517, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 1.2968, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 15.294117647058824, | |
| "grad_norm": 0.6072357892990112, | |
| "learning_rate": 9.997884750411004e-06, | |
| "loss": 1.2174, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 15.529411764705882, | |
| "grad_norm": 0.6069150567054749, | |
| "learning_rate": 9.99695413509548e-06, | |
| "loss": 1.2296, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 15.764705882352942, | |
| "grad_norm": 0.567636251449585, | |
| "learning_rate": 9.995854391448607e-06, | |
| "loss": 1.2153, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.6102450489997864, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 1.2322, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.306054949760437, | |
| "eval_runtime": 14.4899, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 16.235294117647058, | |
| "grad_norm": 0.6264614462852478, | |
| "learning_rate": 9.993147673772869e-06, | |
| "loss": 1.1767, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 16.470588235294116, | |
| "grad_norm": 0.5897552967071533, | |
| "learning_rate": 9.991540791356342e-06, | |
| "loss": 1.2402, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 16.705882352941178, | |
| "grad_norm": 0.5707640647888184, | |
| "learning_rate": 9.989764963830038e-06, | |
| "loss": 1.1951, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "grad_norm": 0.5615460276603699, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 1.1389, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 16.941176470588236, | |
| "eval_loss": 1.291010856628418, | |
| "eval_runtime": 14.4995, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 17.176470588235293, | |
| "grad_norm": 0.6482234597206116, | |
| "learning_rate": 9.985706719584888e-06, | |
| "loss": 1.1827, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 17.41176470588235, | |
| "grad_norm": 0.7177907824516296, | |
| "learning_rate": 9.98342444022253e-06, | |
| "loss": 1.1384, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 0.6740428805351257, | |
| "learning_rate": 9.980973490458728e-06, | |
| "loss": 1.1597, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 17.88235294117647, | |
| "grad_norm": 0.6563602089881897, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 1.2119, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.2774587869644165, | |
| "eval_runtime": 14.492, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 18.11764705882353, | |
| "grad_norm": 0.697040319442749, | |
| "learning_rate": 9.975565917255017e-06, | |
| "loss": 1.1046, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 18.352941176470587, | |
| "grad_norm": 0.6884188652038574, | |
| "learning_rate": 9.972609476841368e-06, | |
| "loss": 1.175, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 18.58823529411765, | |
| "grad_norm": 0.6756815910339355, | |
| "learning_rate": 9.9694847320726e-06, | |
| "loss": 1.1308, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 18.823529411764707, | |
| "grad_norm": 0.7454915046691895, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.0796, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 18.941176470588236, | |
| "eval_loss": 1.2671666145324707, | |
| "eval_runtime": 14.491, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 19.058823529411764, | |
| "grad_norm": 0.7441277503967285, | |
| "learning_rate": 9.962730758206612e-06, | |
| "loss": 1.085, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 19.294117647058822, | |
| "grad_norm": 0.7544702887535095, | |
| "learning_rate": 9.959101757706308e-06, | |
| "loss": 1.1764, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 19.529411764705884, | |
| "grad_norm": 0.7890501022338867, | |
| "learning_rate": 9.955304910036993e-06, | |
| "loss": 1.0506, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 19.764705882352942, | |
| "grad_norm": 0.7971728444099426, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 1.058, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.7496370077133179, | |
| "learning_rate": 9.947208192904722e-06, | |
| "loss": 1.088, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.2627261877059937, | |
| "eval_runtime": 14.4935, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 20.235294117647058, | |
| "grad_norm": 0.8145542740821838, | |
| "learning_rate": 9.942908597485558e-06, | |
| "loss": 1.0338, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 20.470588235294116, | |
| "grad_norm": 0.7999013662338257, | |
| "learning_rate": 9.938441702975689e-06, | |
| "loss": 1.1188, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 20.705882352941178, | |
| "grad_norm": 0.8285914659500122, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 1.0152, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "grad_norm": 0.8708339333534241, | |
| "learning_rate": 9.929006627092298e-06, | |
| "loss": 1.0344, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 20.941176470588236, | |
| "eval_loss": 1.2630927562713623, | |
| "eval_runtime": 14.4971, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 21.176470588235293, | |
| "grad_norm": 0.7319707274436951, | |
| "learning_rate": 9.924038765061042e-06, | |
| "loss": 1.0431, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 21.41176470588235, | |
| "grad_norm": 0.9208843111991882, | |
| "learning_rate": 9.918904242612794e-06, | |
| "loss": 1.0477, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 21.647058823529413, | |
| "grad_norm": 0.9072057604789734, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 1.0263, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 21.88235294117647, | |
| "grad_norm": 0.873211145401001, | |
| "learning_rate": 9.908135917238321e-06, | |
| "loss": 1.0175, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.258855938911438, | |
| "eval_runtime": 14.489, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 22.11764705882353, | |
| "grad_norm": 0.9776601791381836, | |
| "learning_rate": 9.902502478779897e-06, | |
| "loss": 0.9928, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 22.352941176470587, | |
| "grad_norm": 0.9805429577827454, | |
| "learning_rate": 9.896703108827758e-06, | |
| "loss": 0.9626, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 22.58823529411765, | |
| "grad_norm": 0.8789134621620178, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.9428, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 22.823529411764707, | |
| "grad_norm": 1.0301721096038818, | |
| "learning_rate": 9.884607365200355e-06, | |
| "loss": 0.9509, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 22.941176470588236, | |
| "eval_loss": 1.2707229852676392, | |
| "eval_runtime": 14.5009, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 23.058823529411764, | |
| "grad_norm": 1.148660659790039, | |
| "learning_rate": 9.878311400921072e-06, | |
| "loss": 0.994, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 23.294117647058822, | |
| "grad_norm": 1.0924886465072632, | |
| "learning_rate": 9.871850323926178e-06, | |
| "loss": 1.04, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 1.0336880683898926, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.897, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.764705882352942, | |
| "grad_norm": 1.0547744035720825, | |
| "learning_rate": 9.858433712104403e-06, | |
| "loss": 0.9449, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 1.0507797002792358, | |
| "learning_rate": 9.851478631379982e-06, | |
| "loss": 0.8574, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.2783534526824951, | |
| "eval_runtime": 14.4915, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 24.235294117647058, | |
| "grad_norm": 1.1866151094436646, | |
| "learning_rate": 9.844359346129504e-06, | |
| "loss": 0.9402, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 24.470588235294116, | |
| "grad_norm": 1.207543134689331, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.8642, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 24.705882352941178, | |
| "grad_norm": 1.3277732133865356, | |
| "learning_rate": 9.829629131445342e-06, | |
| "loss": 0.8979, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "grad_norm": 1.2087572813034058, | |
| "learning_rate": 9.822018700574696e-06, | |
| "loss": 0.8673, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 24.941176470588236, | |
| "eval_loss": 1.2985085248947144, | |
| "eval_runtime": 14.5378, | |
| "eval_samples_per_second": 2.476, | |
| "eval_steps_per_second": 2.476, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 25.176470588235293, | |
| "grad_norm": 1.185181975364685, | |
| "learning_rate": 9.81424506228719e-06, | |
| "loss": 0.839, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 25.41176470588235, | |
| "grad_norm": 1.223609209060669, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.8175, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 25.647058823529413, | |
| "grad_norm": 1.1787949800491333, | |
| "learning_rate": 9.798209221411748e-06, | |
| "loss": 0.8512, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 25.88235294117647, | |
| "grad_norm": 1.4942697286605835, | |
| "learning_rate": 9.789947561577445e-06, | |
| "loss": 0.8657, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.32999849319458, | |
| "eval_runtime": 14.4921, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 26.11764705882353, | |
| "grad_norm": 1.2106794118881226, | |
| "learning_rate": 9.781523779815178e-06, | |
| "loss": 0.8279, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 26.352941176470587, | |
| "grad_norm": 1.2683756351470947, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.8012, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 26.58823529411765, | |
| "grad_norm": 1.3693079948425293, | |
| "learning_rate": 9.764190996439181e-06, | |
| "loss": 0.7962, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 26.823529411764707, | |
| "grad_norm": 1.341029405593872, | |
| "learning_rate": 9.755282581475769e-06, | |
| "loss": 0.7453, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 26.941176470588236, | |
| "eval_loss": 1.3724702596664429, | |
| "eval_runtime": 14.5144, | |
| "eval_samples_per_second": 2.48, | |
| "eval_steps_per_second": 2.48, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 27.058823529411764, | |
| "grad_norm": 1.5795172452926636, | |
| "learning_rate": 9.74621321786517e-06, | |
| "loss": 0.7852, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 27.294117647058822, | |
| "grad_norm": 1.4557024240493774, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 0.7336, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 27.529411764705884, | |
| "grad_norm": 1.538028597831726, | |
| "learning_rate": 9.727592877996585e-06, | |
| "loss": 0.7007, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 27.764705882352942, | |
| "grad_norm": 1.7676023244857788, | |
| "learning_rate": 9.718042531967918e-06, | |
| "loss": 0.7119, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 1.5406033992767334, | |
| "learning_rate": 9.708332497729378e-06, | |
| "loss": 0.7771, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.382300853729248, | |
| "eval_runtime": 14.493, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 28.235294117647058, | |
| "grad_norm": 1.6042836904525757, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.6619, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 28.470588235294116, | |
| "grad_norm": 1.640479326248169, | |
| "learning_rate": 9.688434684610725e-06, | |
| "loss": 0.7059, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 28.705882352941178, | |
| "grad_norm": 1.919432282447815, | |
| "learning_rate": 9.678247579197658e-06, | |
| "loss": 0.6835, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "grad_norm": 2.0142574310302734, | |
| "learning_rate": 9.667902132486009e-06, | |
| "loss": 0.6941, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "eval_loss": 1.450831651687622, | |
| "eval_runtime": 14.51, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 28.941176470588236, | |
| "step": 246, | |
| "total_flos": 6.277931324145664e+16, | |
| "train_loss": 1.3461502527560645, | |
| "train_runtime": 4652.7461, | |
| "train_samples_per_second": 4.385, | |
| "train_steps_per_second": 0.258 | |
| } | |
| ], | |
| "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": 6.277931324145664e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |