aliREA/Decoder_test
06
1{2 "best_metric": null,3 "best_model_checkpoint": null,4 "epoch": 5.4,5 "eval_steps": 500,6 "global_step": 18,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "epoch": 0.3,13 "grad_norm": 0.55078125,14 "learning_rate": 0.00018888888888888888,15 "loss": 0.4255,16 "step": 117 },18 {19 "epoch": 0.6,20 "grad_norm": 1.03125,21 "learning_rate": 0.00017777777777777779,22 "loss": 0.355,23 "step": 224 },25 {26 "epoch": 0.9,27 "grad_norm": 1.25,28 "learning_rate": 0.0001666666666666667,29 "loss": 0.3059,30 "step": 331 },32 {33 "epoch": 1.2,34 "grad_norm": 0.6953125,35 "learning_rate": 0.00015555555555555556,36 "loss": 0.2006,37 "step": 438 },39 {40 "epoch": 1.5,41 "grad_norm": 0.326171875,42 "learning_rate": 0.00014444444444444444,43 "loss": 0.216,44 "step": 545 },46 {47 "epoch": 1.8,48 "grad_norm": 0.283203125,49 "learning_rate": 0.00013333333333333334,50 "loss": 0.1792,51 "step": 652 },53 {54 "epoch": 2.1,55 "grad_norm": 0.375,56 "learning_rate": 0.00012222222222222224,57 "loss": 0.181,58 "step": 759 },60 {61 "epoch": 2.4,62 "grad_norm": 0.2119140625,63 "learning_rate": 0.00011111111111111112,64 "loss": 0.1302,65 "step": 866 },67 {68 "epoch": 2.7,69 "grad_norm": 0.21875,70 "learning_rate": 0.0001,71 "loss": 0.1395,72 "step": 973 },74 {75 "epoch": 3.0,76 "grad_norm": 0.26171875,77 "learning_rate": 8.888888888888889e-05,78 "loss": 0.1223,79 "step": 1080 },81 {82 "epoch": 3.3,83 "grad_norm": 0.1748046875,84 "learning_rate": 7.777777777777778e-05,85 "loss": 0.1013,86 "step": 1187 },88 {89 "epoch": 3.6,90 "grad_norm": 0.197265625,91 "learning_rate": 6.666666666666667e-05,92 "loss": 0.104,93 "step": 1294 },95 {96 "epoch": 3.9,97 "grad_norm": 0.201171875,98 "learning_rate": 5.555555555555556e-05,99 "loss": 0.0926,100 "step": 13101 },102 {103 "epoch": 4.2,104 "grad_norm": 0.279296875,105 "learning_rate": 4.4444444444444447e-05,106 "loss": 0.0812,107 "step": 14108 },109 {110 "epoch": 4.5,111 "grad_norm": 0.1748046875,112 "learning_rate": 3.3333333333333335e-05,113 "loss": 0.0791,114 "step": 15115 },116 {117 "epoch": 4.8,118 "grad_norm": 0.1767578125,119 "learning_rate": 2.2222222222222223e-05,120 "loss": 0.0817,121 "step": 16122 },123 {124 "epoch": 5.1,125 "grad_norm": 0.23046875,126 "learning_rate": 1.1111111111111112e-05,127 "loss": 0.0699,128 "step": 17129 },130 {131 "epoch": 5.4,132 "grad_norm": 0.169921875,133 "learning_rate": 0.0,134 "loss": 0.0721,135 "step": 18136 }137 ],138 "logging_steps": 1,139 "max_steps": 18,140 "num_input_tokens_seen": 0,141 "num_train_epochs": 6,142 "save_steps": 500,143 "stateful_callbacks": {144 "TrainerControl": {145 "args": {146 "should_epoch_stop": false,147 "should_evaluate": false,148 "should_log": false,149 "should_save": true,150 "should_training_stop": true151 },152 "attributes": {}153 }154 },155 "total_flos": 3.1696816177152e+16,156 "train_batch_size": 6,157 "trial_name": null,158 "trial_params": null159}160 