{ "best_metric": null, "best_model_checkpoint": null, "epoch": 9.0, "global_step": 225, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.04, "learning_rate": 1.9920000000000002e-05, "loss": 2.3122, "step": 1 }, { "epoch": 0.52, "learning_rate": 1.896e-05, "loss": 2.2992, "step": 13 }, { "epoch": 0.52, "eval_accuracy": 0.1, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 24.785266876220703, "eval_f1": 0.018181818181818184, "eval_gpu_ram_allocated": 2.091515064239502, "eval_gpu_ram_cached": 24.6484375, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 50, "eval_loss": 2.303118944168091, "eval_precision": 0.01, "eval_recall": 0.1, "eval_runtime": 0.3632, "eval_samples_per_second": 550.672, "eval_steps_per_second": 19.274, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 3.9340057373046875, "step": 13 }, { "epoch": 1.04, "learning_rate": 1.792e-05, "loss": 2.3096, "step": 26 }, { "epoch": 1.04, "eval_accuracy": 0.1, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 29.620643615722656, "eval_f1": 0.018181818181818184, "eval_gpu_ram_allocated": 2.0915164947509766, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 43, "eval_loss": 2.2983736991882324, "eval_precision": 0.01, "eval_recall": 0.1, "eval_runtime": 0.3843, "eval_samples_per_second": 520.372, "eval_steps_per_second": 18.213, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.119495391845703, "step": 26 }, { "epoch": 1.56, "learning_rate": 1.688e-05, "loss": 2.2906, "step": 39 }, { "epoch": 1.56, "eval_accuracy": 0.145, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 29.620647430419922, "eval_f1": 0.06484230055658627, "eval_gpu_ram_allocated": 2.0915350914001465, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 51, "eval_loss": 2.2851924896240234, "eval_precision": 0.05246212121212121, "eval_recall": 0.145, "eval_runtime": 0.3683, "eval_samples_per_second": 542.987, "eval_steps_per_second": 19.005, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.205013275146484, "step": 39 }, { "epoch": 2.08, "learning_rate": 1.584e-05, "loss": 2.2723, "step": 52 }, { "epoch": 2.08, "eval_accuracy": 0.225, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69240951538086, "eval_f1": 0.12825672433668248, "eval_gpu_ram_allocated": 2.091507911682129, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 43, "eval_loss": 2.2198123931884766, "eval_precision": 0.16249866509230126, "eval_recall": 0.225, "eval_runtime": 0.3837, "eval_samples_per_second": 521.197, "eval_steps_per_second": 18.242, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.216480255126953, "step": 52 }, { "epoch": 2.6, "learning_rate": 1.48e-05, "loss": 2.1387, "step": 65 }, { "epoch": 2.6, "eval_accuracy": 0.335, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69245147705078, "eval_f1": 0.2579535245737298, "eval_gpu_ram_allocated": 2.091599464416504, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 56, "eval_loss": 2.0293209552764893, "eval_precision": 0.2654710140004258, "eval_recall": 0.335, "eval_runtime": 0.3658, "eval_samples_per_second": 546.742, "eval_steps_per_second": 19.136, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2218017578125, "step": 65 }, { "epoch": 3.12, "learning_rate": 1.376e-05, "loss": 1.9534, "step": 78 }, { "epoch": 3.12, "eval_accuracy": 0.4, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692481994628906, "eval_f1": 0.37299130397528574, "eval_gpu_ram_allocated": 2.091500759124756, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 41, "eval_loss": 1.8756619691848755, "eval_precision": 0.44191021478521475, "eval_recall": 0.4, "eval_runtime": 0.3822, "eval_samples_per_second": 523.32, "eval_steps_per_second": 18.316, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2091827392578125, "step": 78 }, { "epoch": 3.64, "learning_rate": 1.2720000000000002e-05, "loss": 1.7689, "step": 91 }, { "epoch": 3.64, "eval_accuracy": 0.48, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692481994628906, "eval_f1": 0.4442698143694956, "eval_gpu_ram_allocated": 2.0915350914001465, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 46, "eval_loss": 1.7208746671676636, "eval_precision": 0.5197521691286822, "eval_recall": 0.48, "eval_runtime": 0.3705, "eval_samples_per_second": 539.861, "eval_steps_per_second": 18.895, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2303466796875, "step": 91 }, { "epoch": 4.16, "learning_rate": 1.1760000000000001e-05, "loss": 1.6052, "step": 104 }, { "epoch": 4.16, "eval_accuracy": 0.525, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69255828857422, "eval_f1": 0.5044390342519582, "eval_gpu_ram_allocated": 2.0915451049804688, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 45, "eval_loss": 1.6318445205688477, "eval_precision": 0.5139384762914174, "eval_recall": 0.525, "eval_runtime": 0.3806, "eval_samples_per_second": 525.479, "eval_steps_per_second": 18.392, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.229682922363281, "step": 104 }, { "epoch": 4.68, "learning_rate": 1.072e-05, "loss": 1.4606, "step": 117 }, { "epoch": 4.68, "eval_accuracy": 0.575, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69255828857422, "eval_f1": 0.5539001078732381, "eval_gpu_ram_allocated": 2.0915350914001465, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 47, "eval_loss": 1.4969176054000854, "eval_precision": 0.5787876338221166, "eval_recall": 0.575, "eval_runtime": 0.3752, "eval_samples_per_second": 533.081, "eval_steps_per_second": 18.658, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.231475830078125, "step": 117 }, { "epoch": 5.2, "learning_rate": 9.68e-06, "loss": 1.2963, "step": 130 }, { "epoch": 5.2, "eval_accuracy": 0.61, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69259262084961, "eval_f1": 0.6036944568156745, "eval_gpu_ram_allocated": 2.091599464416504, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 43, "eval_loss": 1.3920272588729858, "eval_precision": 0.6063040077990168, "eval_recall": 0.61, "eval_runtime": 0.381, "eval_samples_per_second": 524.941, "eval_steps_per_second": 18.373, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.242012023925781, "step": 130 }, { "epoch": 5.72, "learning_rate": 8.64e-06, "loss": 1.1948, "step": 143 }, { "epoch": 5.72, "eval_accuracy": 0.63, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69259262084961, "eval_f1": 0.6251401317519267, "eval_gpu_ram_allocated": 2.0915136337280273, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 48, "eval_loss": 1.3030096292495728, "eval_precision": 0.6292207792207792, "eval_recall": 0.63, "eval_runtime": 0.3633, "eval_samples_per_second": 550.566, "eval_steps_per_second": 19.27, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2687225341796875, "step": 143 }, { "epoch": 6.24, "learning_rate": 7.600000000000001e-06, "loss": 1.0248, "step": 156 }, { "epoch": 6.24, "eval_accuracy": 0.625, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69269561767578, "eval_f1": 0.6184008464653626, "eval_gpu_ram_allocated": 2.091526508331299, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 50, "eval_loss": 1.2567895650863647, "eval_precision": 0.6353763341921237, "eval_recall": 0.625, "eval_runtime": 0.3724, "eval_samples_per_second": 537.06, "eval_steps_per_second": 18.797, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.259613037109375, "step": 156 }, { "epoch": 6.76, "learning_rate": 6.560000000000001e-06, "loss": 0.9509, "step": 169 }, { "epoch": 6.76, "eval_accuracy": 0.65, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.69269561767578, "eval_f1": 0.6448307390509175, "eval_gpu_ram_allocated": 2.091519355773926, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 44, "eval_loss": 1.191145420074463, "eval_precision": 0.6551875660571312, "eval_recall": 0.65, "eval_runtime": 0.3664, "eval_samples_per_second": 545.815, "eval_steps_per_second": 19.104, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.262535095214844, "step": 169 }, { "epoch": 7.28, "learning_rate": 5.5200000000000005e-06, "loss": 0.9081, "step": 182 }, { "epoch": 7.28, "eval_accuracy": 0.655, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692707061767578, "eval_f1": 0.6441038418902195, "eval_gpu_ram_allocated": 2.0915451049804688, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 50, "eval_loss": 1.1784366369247437, "eval_precision": 0.6450486800780918, "eval_recall": 0.655, "eval_runtime": 0.3834, "eval_samples_per_second": 521.619, "eval_steps_per_second": 18.257, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.19549560546875, "step": 182 }, { "epoch": 7.8, "learning_rate": 4.48e-06, "loss": 0.7629, "step": 195 }, { "epoch": 7.8, "eval_accuracy": 0.655, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692710876464844, "eval_f1": 0.659811831437347, "eval_gpu_ram_allocated": 2.091515064239502, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 44, "eval_loss": 1.135439157485962, "eval_precision": 0.6736920486920487, "eval_recall": 0.655, "eval_runtime": 0.37, "eval_samples_per_second": 540.554, "eval_steps_per_second": 18.919, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.186817169189453, "step": 195 }, { "epoch": 8.32, "learning_rate": 3.44e-06, "loss": 0.7348, "step": 208 }, { "epoch": 8.32, "eval_accuracy": 0.65, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692718505859375, "eval_f1": 0.6430377731742748, "eval_gpu_ram_allocated": 2.0915064811706543, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 43, "eval_loss": 1.1369377374649048, "eval_precision": 0.6483042554206238, "eval_recall": 0.65, "eval_runtime": 0.3889, "eval_samples_per_second": 514.251, "eval_steps_per_second": 17.999, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2168426513671875, "step": 208 }, { "epoch": 8.84, "learning_rate": 2.4000000000000003e-06, "loss": 0.7443, "step": 221 }, { "epoch": 8.84, "eval_accuracy": 0.66, "eval_disk_space_total": 78.1898422241211, "eval_disk_space_used": 31.692718505859375, "eval_f1": 0.6531333156719726, "eval_gpu_ram_allocated": 2.091543674468994, "eval_gpu_ram_cached": 24.66015625, "eval_gpu_ram_total": 39.56402587890625, "eval_gpu_utilization": 51, "eval_loss": 1.127373456954956, "eval_precision": 0.6576212193603499, "eval_recall": 0.66, "eval_runtime": 0.368, "eval_samples_per_second": 543.468, "eval_steps_per_second": 19.021, "eval_system_ram_total": 83.48074722290039, "eval_system_ram_used": 4.2272796630859375, "step": 221 } ], "max_steps": 250, "num_train_epochs": 10, "total_flos": 489381153835392.0, "trial_name": null, "trial_params": null }