| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.218796253204346, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5604, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6626643397324026, | |
| "eval_loss": 0.4996059536933899, | |
| "eval_precision": 0.6671439480717831, | |
| "eval_recall": 0.6592107655937443, | |
| "eval_runtime": 5.1871, | |
| "eval_samples_per_second": 76.921, | |
| "eval_steps_per_second": 9.639, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.024437427520752, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4842, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7543859649122807, | |
| "eval_f1": 0.7247113489157984, | |
| "eval_loss": 0.4520498514175415, | |
| "eval_precision": 0.7169408246101261, | |
| "eval_recall": 0.7462265866521185, | |
| "eval_runtime": 5.0789, | |
| "eval_samples_per_second": 78.561, | |
| "eval_steps_per_second": 9.845, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.7485127449035645, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4079, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8320802005012531, | |
| "eval_f1": 0.800852224051491, | |
| "eval_loss": 0.37486591935157776, | |
| "eval_precision": 0.7962609970674487, | |
| "eval_recall": 0.8061920349154392, | |
| "eval_runtime": 5.055, | |
| "eval_samples_per_second": 78.932, | |
| "eval_steps_per_second": 9.891, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.6576170921325684, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3378, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8122237052238519, | |
| "eval_loss": 0.36237815022468567, | |
| "eval_precision": 0.8184491978609625, | |
| "eval_recall": 0.8068285142753229, | |
| "eval_runtime": 5.0874, | |
| "eval_samples_per_second": 78.43, | |
| "eval_steps_per_second": 9.828, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.3436532020568848, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3146, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8234833375639119, | |
| "eval_loss": 0.36204174160957336, | |
| "eval_precision": 0.8130172220979647, | |
| "eval_recall": 0.8393344244408074, | |
| "eval_runtime": 5.0836, | |
| "eval_samples_per_second": 78.488, | |
| "eval_steps_per_second": 9.836, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.493219375610352, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2935, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8252554744525548, | |
| "eval_loss": 0.35182783007621765, | |
| "eval_precision": 0.8157894736842105, | |
| "eval_recall": 0.8386070194580832, | |
| "eval_runtime": 5.0469, | |
| "eval_samples_per_second": 79.059, | |
| "eval_steps_per_second": 9.907, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.3236638307571411, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2842, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.3307338058948517, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0617, | |
| "eval_samples_per_second": 78.827, | |
| "eval_steps_per_second": 9.878, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.9696807861328125, | |
| "learning_rate": 3e-05, | |
| "loss": 0.267, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.3191296458244324, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.1096, | |
| "eval_samples_per_second": 78.089, | |
| "eval_steps_per_second": 9.786, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.35373067855835, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2598, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8401647707947546, | |
| "eval_loss": 0.31739577651023865, | |
| "eval_precision": 0.8393298751432535, | |
| "eval_recall": 0.8410165484633569, | |
| "eval_runtime": 5.1627, | |
| "eval_samples_per_second": 77.285, | |
| "eval_steps_per_second": 9.685, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.5958967208862305, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2557, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8367430441898527, | |
| "eval_loss": 0.30755287408828735, | |
| "eval_precision": 0.8367430441898527, | |
| "eval_recall": 0.8367430441898527, | |
| "eval_runtime": 5.0513, | |
| "eval_samples_per_second": 78.99, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.4262058734893799, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2341, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.31438612937927246, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 5.0545, | |
| "eval_samples_per_second": 78.939, | |
| "eval_steps_per_second": 9.892, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.927000999450684, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2352, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8409485773595975, | |
| "eval_loss": 0.31350696086883545, | |
| "eval_precision": 0.8385304659498208, | |
| "eval_recall": 0.8435170030914712, | |
| "eval_runtime": 5.0621, | |
| "eval_samples_per_second": 78.821, | |
| "eval_steps_per_second": 9.877, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.771910309791565, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2335, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8469505178365937, | |
| "eval_loss": 0.30345895886421204, | |
| "eval_precision": 0.844489247311828, | |
| "eval_recall": 0.8495635570103655, | |
| "eval_runtime": 5.0626, | |
| "eval_samples_per_second": 78.814, | |
| "eval_steps_per_second": 9.876, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 11.13768482208252, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.232, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8450716845878137, | |
| "eval_loss": 0.3012339174747467, | |
| "eval_precision": 0.8404471544715447, | |
| "eval_recall": 0.8502909619930896, | |
| "eval_runtime": 5.0472, | |
| "eval_samples_per_second": 79.054, | |
| "eval_steps_per_second": 9.906, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.7109107971191406, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.221, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8424651921601347, | |
| "eval_loss": 0.3049696981906891, | |
| "eval_precision": 0.8372140762463343, | |
| "eval_recall": 0.8485179123476996, | |
| "eval_runtime": 5.0811, | |
| "eval_samples_per_second": 78.526, | |
| "eval_steps_per_second": 9.84, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.484745979309082, | |
| "learning_rate": 1e-05, | |
| "loss": 0.216, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8457993935430168, | |
| "eval_loss": 0.3015521466732025, | |
| "eval_precision": 0.8398540145985401, | |
| "eval_recall": 0.8527914166212038, | |
| "eval_runtime": 5.0477, | |
| "eval_samples_per_second": 79.046, | |
| "eval_steps_per_second": 9.906, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 10.693771362304688, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2096, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8435671632788467, | |
| "eval_loss": 0.2880638837814331, | |
| "eval_precision": 0.8419117647058824, | |
| "eval_recall": 0.8452900527368612, | |
| "eval_runtime": 5.1498, | |
| "eval_samples_per_second": 77.478, | |
| "eval_steps_per_second": 9.709, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 5.451054096221924, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2184, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8465112150085814, | |
| "eval_loss": 0.29656729102134705, | |
| "eval_precision": 0.8393498147290412, | |
| "eval_recall": 0.855291871249318, | |
| "eval_runtime": 5.0876, | |
| "eval_samples_per_second": 78.426, | |
| "eval_steps_per_second": 9.828, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.6623899936676025, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2134, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8409485773595975, | |
| "eval_loss": 0.2884305715560913, | |
| "eval_precision": 0.8385304659498208, | |
| "eval_recall": 0.8435170030914712, | |
| "eval_runtime": 5.0715, | |
| "eval_samples_per_second": 78.675, | |
| "eval_steps_per_second": 9.859, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 10.805831909179688, | |
| "learning_rate": 0.0, | |
| "loss": 0.2077, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8383403361344538, | |
| "eval_loss": 0.289480060338974, | |
| "eval_precision": 0.8352012604792199, | |
| "eval_recall": 0.8417439534460811, | |
| "eval_runtime": 5.0647, | |
| "eval_samples_per_second": 78.78, | |
| "eval_steps_per_second": 9.872, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7635661149264000.0, | |
| "train_loss": 0.28429694019380164, | |
| "train_runtime": 1955.8264, | |
| "train_samples_per_second": 37.202, | |
| "train_steps_per_second": 1.248 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7635661149264000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |