| { | |
| "best_metric": 0.6156604453386092, | |
| "best_model_checkpoint": "./runtime-ar/checkpoint-152", | |
| "epoch": 3.9950900163666123, | |
| "global_step": 152, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5e-05, | |
| "loss": 3.3108, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0001, | |
| "loss": 4.8935, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.6355, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0002, | |
| "loss": 3.4513, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0001999048221581858, | |
| "loss": 3.2178, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 3.0584, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 2.908, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 2.7828, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00019762960071199333, | |
| "loss": 2.6914, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 2.6232, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 2.5839, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 2.5405, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 2.4946, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.000190630778703665, | |
| "loss": 2.4245, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00018870108331782217, | |
| "loss": 2.4099, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 2.3972, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0001843391445812886, | |
| "loss": 2.3371, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0001819152044288992, | |
| "loss": 2.276, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 2.3067, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5594373263429553, | |
| "eval_loss": 2.3559441566467285, | |
| "eval_runtime": 87.3704, | |
| "eval_samples_per_second": 6.192, | |
| "eval_steps_per_second": 3.102, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 2.3629, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 2.0942, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 2.0975, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00016755902076156604, | |
| "loss": 2.0554, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 2.0336, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 2.0584, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 1.9959, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0001537299608346824, | |
| "loss": 1.9855, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.9941, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00014617486132350343, | |
| "loss": 1.9589, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00014226182617406996, | |
| "loss": 1.9918, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.9673, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 1.9626, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00013007057995042732, | |
| "loss": 1.9327, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 1.9334, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00012164396139381029, | |
| "loss": 1.9405, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.8394, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 1.9284, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010871557427476583, | |
| "loss": 1.9243, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5974633090939628, | |
| "eval_loss": 2.1282806396484375, | |
| "eval_runtime": 88.3059, | |
| "eval_samples_per_second": 6.126, | |
| "eval_steps_per_second": 3.069, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00010436193873653361, | |
| "loss": 1.8745, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0001, | |
| "loss": 1.6456, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 1.6496, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 9.128442572523417e-05, | |
| "loss": 1.6571, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 1.6366, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 1.6033, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 7.835603860618972e-05, | |
| "loss": 1.6042, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 1.6111, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 6.992942004957271e-05, | |
| "loss": 1.6058, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 1.6079, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 1.5789, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 5.773817382593008e-05, | |
| "loss": 1.6267, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 1.5811, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.5605, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 4.6270039165317605e-05, | |
| "loss": 1.5828, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.264235636489542e-05, | |
| "loss": 1.5579, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 1.5991, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.5641, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.244097923843398e-05, | |
| "loss": 1.5971, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6139908093264839, | |
| "eval_loss": 2.075925827026367, | |
| "eval_runtime": 88.4185, | |
| "eval_samples_per_second": 6.119, | |
| "eval_steps_per_second": 3.065, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 1.5486, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 2.6272266318987603e-05, | |
| "loss": 1.4341, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 1.4085, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 1.3819, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.808479557110081e-05, | |
| "loss": 1.3935, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 1.566085541871145e-05, | |
| "loss": 1.4073, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 1.419, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.129891668217783e-05, | |
| "loss": 1.3807, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 1.3881, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.3674, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 1.3847, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 4.628304925177318e-06, | |
| "loss": 1.3734, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 1.3969, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 1.3544, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 1.4021, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 1.4061, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 3.805301908254455e-07, | |
| "loss": 1.3621, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 9.517784181422019e-08, | |
| "loss": 1.3759, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.4139, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6156604453386092, | |
| "eval_loss": 2.101821184158325, | |
| "eval_runtime": 88.2947, | |
| "eval_samples_per_second": 6.127, | |
| "eval_steps_per_second": 3.069, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 152, | |
| "total_flos": 8.481201929060352e+16, | |
| "train_loss": 1.970301656346572, | |
| "train_runtime": 9467.2073, | |
| "train_samples_per_second": 2.062, | |
| "train_steps_per_second": 0.016 | |
| } | |
| ], | |
| "max_steps": 152, | |
| "num_train_epochs": 4, | |
| "total_flos": 8.481201929060352e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |