| { | |
| "best_metric": 1.2113300561904907, | |
| "best_model_checkpoint": "cars-countries/checkpoint-78", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 78, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": Infinity, | |
| "learning_rate": 0.0, | |
| "loss": 1.781, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 9.921228408813477, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.5569, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 7.8902506828308105, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.5381, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 9.90084457397461, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.5947, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 10.790699005126953, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.3412, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 9.438414573669434, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.5891, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 14.554863929748535, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.5911, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 11.261236190795898, | |
| "learning_rate": 4.375e-05, | |
| "loss": 1.6023, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 10.684854507446289, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5315, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 8.2311429977417, | |
| "learning_rate": 4.928571428571429e-05, | |
| "loss": 1.4993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 8.519936561584473, | |
| "learning_rate": 4.8571428571428576e-05, | |
| "loss": 1.5576, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 7.320441246032715, | |
| "learning_rate": 4.785714285714286e-05, | |
| "loss": 1.5395, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 19.024276733398438, | |
| "learning_rate": 4.714285714285714e-05, | |
| "loss": 1.5867, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 8.801590919494629, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 1.2448, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 8.235298156738281, | |
| "learning_rate": 4.5714285714285716e-05, | |
| "loss": 1.4907, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 9.275227546691895, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4567, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 9.200013160705566, | |
| "learning_rate": 4.428571428571428e-05, | |
| "loss": 1.6051, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 7.5380120277404785, | |
| "learning_rate": 4.3571428571428576e-05, | |
| "loss": 1.3018, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 9.66368579864502, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.2577, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 6.4247260093688965, | |
| "learning_rate": 4.214285714285714e-05, | |
| "loss": 1.3549, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 7.779500484466553, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 1.3177, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 9.328712463378906, | |
| "learning_rate": 4.0714285714285717e-05, | |
| "loss": 1.9277, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 9.957062721252441, | |
| "learning_rate": 4e-05, | |
| "loss": 1.543, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 11.664092063903809, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 1.4666, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 7.613643646240234, | |
| "learning_rate": 3.857142857142858e-05, | |
| "loss": 1.6037, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 16.03534698486328, | |
| "learning_rate": 3.785714285714286e-05, | |
| "loss": 1.2129, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.39215686274509803, | |
| "eval_f1_macro": 0.17172161172161174, | |
| "eval_f1_micro": 0.39215686274509803, | |
| "eval_f1_weighted": 0.26543130072541843, | |
| "eval_loss": 1.3392212390899658, | |
| "eval_precision_macro": 0.32999999999999996, | |
| "eval_precision_micro": 0.39215686274509803, | |
| "eval_precision_weighted": 0.4598039215686274, | |
| "eval_recall_macro": 0.2365079365079365, | |
| "eval_recall_micro": 0.39215686274509803, | |
| "eval_recall_weighted": 0.39215686274509803, | |
| "eval_runtime": 0.9569, | |
| "eval_samples_per_second": 53.295, | |
| "eval_steps_per_second": 4.18, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 7.75885009765625, | |
| "learning_rate": 3.7142857142857143e-05, | |
| "loss": 1.3228, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 8.9802827835083, | |
| "learning_rate": 3.642857142857143e-05, | |
| "loss": 1.2236, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 11.71562671661377, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.6211, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 8.007612228393555, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0492, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 8.764470100402832, | |
| "learning_rate": 3.428571428571429e-05, | |
| "loss": 1.2495, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 9.003642082214355, | |
| "learning_rate": 3.357142857142857e-05, | |
| "loss": 1.4517, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 8.626983642578125, | |
| "learning_rate": 3.285714285714286e-05, | |
| "loss": 1.188, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 6.939526557922363, | |
| "learning_rate": 3.2142857142857144e-05, | |
| "loss": 1.1346, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 8.789880752563477, | |
| "learning_rate": 3.142857142857143e-05, | |
| "loss": 1.4955, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 8.395410537719727, | |
| "learning_rate": 3.071428571428572e-05, | |
| "loss": 1.1301, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 9.601696968078613, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3088, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 9.192988395690918, | |
| "learning_rate": 2.9285714285714288e-05, | |
| "loss": 1.3069, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 10.932126998901367, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.5361, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 8.457172393798828, | |
| "learning_rate": 2.785714285714286e-05, | |
| "loss": 1.0766, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 8.215372085571289, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 1.1533, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 9.27467155456543, | |
| "learning_rate": 2.642857142857143e-05, | |
| "loss": 1.2253, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 8.538656234741211, | |
| "learning_rate": 2.5714285714285714e-05, | |
| "loss": 1.5666, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 9.004700660705566, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2428, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 9.92335319519043, | |
| "learning_rate": 2.4285714285714288e-05, | |
| "loss": 1.3294, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 6.969778537750244, | |
| "learning_rate": 2.357142857142857e-05, | |
| "loss": 0.892, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 9.29692554473877, | |
| "learning_rate": 2.2857142857142858e-05, | |
| "loss": 0.8958, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 8.588207244873047, | |
| "learning_rate": 2.214285714285714e-05, | |
| "loss": 0.965, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 7.432168483734131, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.1908, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 8.40340518951416, | |
| "learning_rate": 2.0714285714285718e-05, | |
| "loss": 1.3141, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 11.749902725219727, | |
| "learning_rate": 2e-05, | |
| "loss": 1.664, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": Infinity, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0425, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.47058823529411764, | |
| "eval_f1_macro": 0.30023902651021295, | |
| "eval_f1_micro": 0.47058823529411764, | |
| "eval_f1_weighted": 0.3905676133991189, | |
| "eval_loss": 1.2403396368026733, | |
| "eval_precision_macro": 0.4878048780487805, | |
| "eval_precision_micro": 0.47058823529411764, | |
| "eval_precision_weighted": 0.5961262553802008, | |
| "eval_recall_macro": 0.32301587301587303, | |
| "eval_recall_micro": 0.47058823529411764, | |
| "eval_recall_weighted": 0.47058823529411764, | |
| "eval_runtime": 0.9273, | |
| "eval_samples_per_second": 54.997, | |
| "eval_steps_per_second": 4.314, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 11.03630256652832, | |
| "learning_rate": 1.928571428571429e-05, | |
| "loss": 1.1866, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 7.543542861938477, | |
| "learning_rate": 1.8571428571428572e-05, | |
| "loss": 1.1695, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 8.078568458557129, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.9876, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 9.406330108642578, | |
| "learning_rate": 1.7142857142857145e-05, | |
| "loss": 1.4199, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 8.94027328491211, | |
| "learning_rate": 1.642857142857143e-05, | |
| "loss": 1.0628, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 7.433661937713623, | |
| "learning_rate": 1.5714285714285715e-05, | |
| "loss": 1.0264, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 7.517592430114746, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.8727, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 11.105234146118164, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.0733, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 8.228996276855469, | |
| "learning_rate": 1.357142857142857e-05, | |
| "loss": 0.7755, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 10.02564525604248, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 1.3597, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 7.33097505569458, | |
| "learning_rate": 1.2142857142857144e-05, | |
| "loss": 0.8739, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 11.741021156311035, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 1.3264, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 10.197936058044434, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 1.1468, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 9.967548370361328, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2385, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 6.901878356933594, | |
| "learning_rate": 9.285714285714286e-06, | |
| "loss": 0.7635, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 7.190940856933594, | |
| "learning_rate": 8.571428571428573e-06, | |
| "loss": 0.8255, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 9.144484519958496, | |
| "learning_rate": 7.857142857142858e-06, | |
| "loss": 1.0753, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 6.83949089050293, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.0511, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 11.242728233337402, | |
| "learning_rate": 6.428571428571429e-06, | |
| "loss": 1.6024, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 9.73320484161377, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0241, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 13.551000595092773, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0143, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 9.739935874938965, | |
| "learning_rate": 4.285714285714286e-06, | |
| "loss": 1.528, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 7.910869598388672, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 0.9389, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 9.845719337463379, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 1.2907, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 6.709384441375732, | |
| "learning_rate": 2.142857142857143e-06, | |
| "loss": 1.0141, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 16.175195693969727, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 1.4561, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5098039215686274, | |
| "eval_f1_macro": 0.35240829346092506, | |
| "eval_f1_micro": 0.5098039215686274, | |
| "eval_f1_weighted": 0.43996622572473965, | |
| "eval_loss": 1.2113300561904907, | |
| "eval_precision_macro": 0.443963963963964, | |
| "eval_precision_micro": 0.5098039215686274, | |
| "eval_precision_weighted": 0.5037272566684332, | |
| "eval_recall_macro": 0.36746031746031743, | |
| "eval_recall_micro": 0.5098039215686274, | |
| "eval_recall_weighted": 0.5098039215686274, | |
| "eval_runtime": 0.9222, | |
| "eval_samples_per_second": 55.303, | |
| "eval_steps_per_second": 4.337, | |
| "step": 78 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 78, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 4.696140837323981e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |