| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 46.51162790697674, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_bp": 0.8731878729849682, | |
| "eval_counts": [ | |
| 4461, | |
| 3931, | |
| 3406, | |
| 2881 | |
| ], | |
| "eval_loss": 0.004611688666045666, | |
| "eval_precisions": [ | |
| 99.33199732798931, | |
| 99.14249684741488, | |
| 99.04041872637394, | |
| 98.90147614143494 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 16.8, | |
| "eval_samples_per_second": 31.31, | |
| "eval_score": 86.5363874068898, | |
| "eval_steps_per_second": 1.012, | |
| "eval_sys_len": 4491, | |
| "eval_totals": [ | |
| 4491, | |
| 3965, | |
| 3439, | |
| 2913 | |
| ], | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bp": 0.869649689238195, | |
| "eval_counts": [ | |
| 4458, | |
| 3930, | |
| 3403, | |
| 2876 | |
| ], | |
| "eval_loss": 0.00696711428463459, | |
| "eval_precisions": [ | |
| 99.62011173184358, | |
| 99.51886553557863, | |
| 99.41571720712825, | |
| 99.27511218501898 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.788, | |
| "eval_samples_per_second": 38.149, | |
| "eval_score": 86.49307063670169, | |
| "eval_steps_per_second": 1.233, | |
| "eval_sys_len": 4475, | |
| "eval_totals": [ | |
| 4475, | |
| 3949, | |
| 3423, | |
| 2897 | |
| ], | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bp": 0.8727461954138965, | |
| "eval_counts": [ | |
| 4463, | |
| 3936, | |
| 3409, | |
| 2882 | |
| ], | |
| "eval_loss": 0.0054754531010985374, | |
| "eval_precisions": [ | |
| 99.42080641568278, | |
| 99.31869795609387, | |
| 99.18533604887983, | |
| 99.00377877018207 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.887, | |
| "eval_samples_per_second": 37.877, | |
| "eval_score": 86.60437841944254, | |
| "eval_steps_per_second": 1.224, | |
| "eval_sys_len": 4489, | |
| "eval_totals": [ | |
| 4489, | |
| 3963, | |
| 3437, | |
| 2911 | |
| ], | |
| "step": 387 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 1.8449612403100777e-05, | |
| "loss": 0.0338, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bp": 0.8703139283627056, | |
| "eval_counts": [ | |
| 4464, | |
| 3938, | |
| 3409, | |
| 2881 | |
| ], | |
| "eval_loss": 0.0021970090456306934, | |
| "eval_precisions": [ | |
| 99.68736042876284, | |
| 99.64574898785425, | |
| 99.50379451255108, | |
| 99.34482758620689 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1368, | |
| "eval_samples_per_second": 37.208, | |
| "eval_score": 86.63569782205596, | |
| "eval_steps_per_second": 1.203, | |
| "eval_sys_len": 4478, | |
| "eval_totals": [ | |
| 4478, | |
| 3952, | |
| 3426, | |
| 2900 | |
| ], | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bp": 0.8707565416150743, | |
| "eval_counts": [ | |
| 4461, | |
| 3933, | |
| 3406, | |
| 2880 | |
| ], | |
| "eval_loss": 0.0019001211039721966, | |
| "eval_precisions": [ | |
| 99.57589285714286, | |
| 99.46889226100151, | |
| 99.35822637106185, | |
| 99.24190213645761 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.879, | |
| "eval_samples_per_second": 37.899, | |
| "eval_score": 86.56290964643597, | |
| "eval_steps_per_second": 1.225, | |
| "eval_sys_len": 4480, | |
| "eval_totals": [ | |
| 4480, | |
| 3954, | |
| 3428, | |
| 2902 | |
| ], | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.0021749669685959816, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1959, | |
| "eval_samples_per_second": 37.053, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.198, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bp": 0.8703139283627056, | |
| "eval_counts": [ | |
| 4464, | |
| 3938, | |
| 3412, | |
| 2886 | |
| ], | |
| "eval_loss": 0.0006092642433941364, | |
| "eval_precisions": [ | |
| 99.68736042876284, | |
| 99.64574898785425, | |
| 99.59136018680677, | |
| 99.51724137931035 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.778, | |
| "eval_samples_per_second": 38.177, | |
| "eval_score": 86.69232496739014, | |
| "eval_steps_per_second": 1.234, | |
| "eval_sys_len": 4478, | |
| "eval_totals": [ | |
| 4478, | |
| 3952, | |
| 3426, | |
| 2900 | |
| ], | |
| "step": 903 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 1.689922480620155e-05, | |
| "loss": 0.0059, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bp": 0.8714201423501817, | |
| "eval_counts": [ | |
| 4460, | |
| 3930, | |
| 3404, | |
| 2878 | |
| ], | |
| "eval_loss": 0.0026030810549855232, | |
| "eval_precisions": [ | |
| 99.48695070265447, | |
| 99.31766489764973, | |
| 99.21305741766248, | |
| 99.07056798623064 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1647, | |
| "eval_samples_per_second": 37.135, | |
| "eval_score": 86.50757156803533, | |
| "eval_steps_per_second": 1.2, | |
| "eval_sys_len": 4483, | |
| "eval_totals": [ | |
| 4483, | |
| 3957, | |
| 3431, | |
| 2905 | |
| ], | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bp": 0.8711989846507546, | |
| "eval_counts": [ | |
| 4459, | |
| 3929, | |
| 3403, | |
| 2877 | |
| ], | |
| "eval_loss": 0.00256777903996408, | |
| "eval_precisions": [ | |
| 99.48683623382419, | |
| 99.31749241658241, | |
| 99.21282798833819, | |
| 99.0702479338843 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7621, | |
| "eval_samples_per_second": 38.221, | |
| "eval_score": 86.4854345384993, | |
| "eval_steps_per_second": 1.235, | |
| "eval_sys_len": 4482, | |
| "eval_totals": [ | |
| 4482, | |
| 3956, | |
| 3430, | |
| 2904 | |
| ], | |
| "step": 1161 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3410, | |
| 2883 | |
| ], | |
| "eval_loss": 0.0008762977086007595, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.56204379562044, | |
| 99.44808554674026 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.033, | |
| "eval_samples_per_second": 37.483, | |
| "eval_score": 86.64880068746125, | |
| "eval_steps_per_second": 1.211, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bp": 0.8707565416150743, | |
| "eval_counts": [ | |
| 4462, | |
| 3935, | |
| 3408, | |
| 2881 | |
| ], | |
| "eval_loss": 0.0005689110257662833, | |
| "eval_precisions": [ | |
| 99.59821428571429, | |
| 99.51947395042994, | |
| 99.41656942823803, | |
| 99.276361130255 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8477, | |
| "eval_samples_per_second": 37.985, | |
| "eval_score": 86.5989861295871, | |
| "eval_steps_per_second": 1.228, | |
| "eval_sys_len": 4480, | |
| "eval_totals": [ | |
| 4480, | |
| 3954, | |
| 3428, | |
| 2902 | |
| ], | |
| "step": 1419 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 1.5348837209302328e-05, | |
| "loss": 0.0042, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bp": 0.8711989846507546, | |
| "eval_counts": [ | |
| 4459, | |
| 3929, | |
| 3403, | |
| 2877 | |
| ], | |
| "eval_loss": 0.0013145459815859795, | |
| "eval_precisions": [ | |
| 99.48683623382419, | |
| 99.31749241658241, | |
| 99.21282798833819, | |
| 99.0702479338843 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7281, | |
| "eval_samples_per_second": 38.316, | |
| "eval_score": 86.4854345384993, | |
| "eval_steps_per_second": 1.238, | |
| "eval_sys_len": 4482, | |
| "eval_totals": [ | |
| 4482, | |
| 3956, | |
| 3430, | |
| 2904 | |
| ], | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3410, | |
| 2883 | |
| ], | |
| "eval_loss": 0.0008758930489420891, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.56204379562044, | |
| 99.44808554674026 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.731, | |
| "eval_samples_per_second": 38.308, | |
| "eval_score": 86.64880068746125, | |
| "eval_steps_per_second": 1.238, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 1677 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.0003304094134364277, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.021, | |
| "eval_samples_per_second": 37.515, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.212, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 1806 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.0008369011338800192, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8925, | |
| "eval_samples_per_second": 37.862, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.224, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 1.3798449612403102e-05, | |
| "loss": 0.0027, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.00020534679060801864, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.9128, | |
| "eval_samples_per_second": 37.807, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.222, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2064 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.001328099868260324, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.6909, | |
| "eval_samples_per_second": 38.42, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.242, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2193 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.0003365726734045893, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7109, | |
| "eval_samples_per_second": 38.364, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.24, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 8.775618334766477e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.0994, | |
| "eval_samples_per_second": 37.307, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.206, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2451 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 1.2248062015503876e-05, | |
| "loss": 0.002, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 8.96168130566366e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8102, | |
| "eval_samples_per_second": 38.088, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.231, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.0007285438477993011, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7462, | |
| "eval_samples_per_second": 38.265, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.237, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2709 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_bp": 0.8703139283627056, | |
| "eval_counts": [ | |
| 4464, | |
| 3938, | |
| 3412, | |
| 2886 | |
| ], | |
| "eval_loss": 0.0018083422910422087, | |
| "eval_precisions": [ | |
| 99.68736042876284, | |
| 99.64574898785425, | |
| 99.59136018680677, | |
| 99.51724137931035 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8084, | |
| "eval_samples_per_second": 38.093, | |
| "eval_score": 86.69232496739014, | |
| "eval_steps_per_second": 1.231, | |
| "eval_sys_len": 4478, | |
| "eval_totals": [ | |
| 4478, | |
| 3952, | |
| 3426, | |
| 2900 | |
| ], | |
| "step": 2838 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.000638504687231034, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.995, | |
| "eval_samples_per_second": 37.585, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.215, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 2967 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "learning_rate": 1.0697674418604651e-05, | |
| "loss": 0.0015, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.00022906862432137132, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7342, | |
| "eval_samples_per_second": 38.299, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.238, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.000275774480542168, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1769, | |
| "eval_samples_per_second": 37.103, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.199, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.00025378959253430367, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.944, | |
| "eval_samples_per_second": 37.722, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.219, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3354 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 2.7398107704357244e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.9791, | |
| "eval_samples_per_second": 37.628, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.216, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3483 | |
| }, | |
| { | |
| "epoch": 27.13, | |
| "learning_rate": 9.147286821705427e-06, | |
| "loss": 0.0014, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 6.660177314188331e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1404, | |
| "eval_samples_per_second": 37.198, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.202, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3612 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.00022675798390991986, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.9231, | |
| "eval_samples_per_second": 37.779, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.221, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3741 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 5.096692984807305e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.3573, | |
| "eval_samples_per_second": 36.636, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.184, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 9.942305041477084e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7812, | |
| "eval_samples_per_second": 38.168, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.234, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 3999 | |
| }, | |
| { | |
| "epoch": 31.01, | |
| "learning_rate": 7.596899224806202e-06, | |
| "loss": 0.0009, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 0.00020378571934998035, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1117, | |
| "eval_samples_per_second": 37.274, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.205, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4128 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 4.814989733858965e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.7623, | |
| "eval_samples_per_second": 38.22, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.235, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4257 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 5.921188130741939e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.904, | |
| "eval_samples_per_second": 37.831, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.223, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4386 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "learning_rate": 6.046511627906977e-06, | |
| "loss": 0.0008, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 2.948127621493768e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.005, | |
| "eval_samples_per_second": 37.558, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.214, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4515 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 1.501874066889286e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8563, | |
| "eval_samples_per_second": 37.961, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.227, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 3.573419598978944e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.0819, | |
| "eval_samples_per_second": 37.353, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.207, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4773 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 1.292789511353476e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8393, | |
| "eval_samples_per_second": 38.008, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.228, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 4902 | |
| }, | |
| { | |
| "epoch": 38.76, | |
| "learning_rate": 4.4961240310077525e-06, | |
| "loss": 0.0006, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 2.826396666932851e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.2505, | |
| "eval_samples_per_second": 36.911, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.193, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5031 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 9.735503226693254e-06, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8339, | |
| "eval_samples_per_second": 38.023, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.229, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 5.421350579126738e-06, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.9445, | |
| "eval_samples_per_second": 37.721, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.219, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5289 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 1.1437254215707071e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8327, | |
| "eval_samples_per_second": 38.026, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.229, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 42.64, | |
| "learning_rate": 2.9457364341085276e-06, | |
| "loss": 0.0004, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 1.4262999684433453e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8768, | |
| "eval_samples_per_second": 37.905, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.225, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5547 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 3.849239874398336e-06, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.9608, | |
| "eval_samples_per_second": 37.677, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.218, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5676 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 1.1063038982683793e-05, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 13.8436, | |
| "eval_samples_per_second": 37.996, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.228, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5805 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_bp": 0.8700925578924447, | |
| "eval_counts": [ | |
| 4463, | |
| 3937, | |
| 3411, | |
| 2885 | |
| ], | |
| "eval_loss": 7.035921953502111e-06, | |
| "eval_precisions": [ | |
| 99.6872905963815, | |
| 99.64565932675272, | |
| 99.5912408759124, | |
| 99.51707485339773 | |
| ], | |
| "eval_ref_len": 5100, | |
| "eval_runtime": 14.1871, | |
| "eval_samples_per_second": 37.076, | |
| "eval_score": 86.6701772747815, | |
| "eval_steps_per_second": 1.198, | |
| "eval_sys_len": 4477, | |
| "eval_totals": [ | |
| 4477, | |
| 3951, | |
| 3425, | |
| 2899 | |
| ], | |
| "step": 5934 | |
| }, | |
| { | |
| "epoch": 46.51, | |
| "learning_rate": 1.3953488372093025e-06, | |
| "loss": 0.0002, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 2.06928054010368e+16, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |