| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.728904366493225, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7312, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.7350375652313232, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7843, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.4805001020431519, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.6772, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.0534298419952393, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6122, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.8123345375061035, | |
| "learning_rate": 0.00019990212265199738, | |
| "loss": 0.5329, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.6635419130325317, | |
| "learning_rate": 0.00019960868220749448, | |
| "loss": 0.4878, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.6072973608970642, | |
| "learning_rate": 0.00019912025308994148, | |
| "loss": 0.4836, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5697150826454163, | |
| "learning_rate": 0.00019843779142227256, | |
| "loss": 0.5162, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.5166112780570984, | |
| "learning_rate": 0.0001975626331552507, | |
| "loss": 0.4825, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5054742097854614, | |
| "learning_rate": 0.00019649649145228102, | |
| "loss": 0.4564, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.4537505805492401, | |
| "learning_rate": 0.00019524145333581317, | |
| "loss": 0.4383, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.4296068251132965, | |
| "learning_rate": 0.00019379997560189675, | |
| "loss": 0.4529, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.44367527961730957, | |
| "learning_rate": 0.00019217488001088784, | |
| "loss": 0.4545, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.4168623685836792, | |
| "learning_rate": 0.0001903693477637204, | |
| "loss": 0.418, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.4231944680213928, | |
| "learning_rate": 0.0001883869132745561, | |
| "loss": 0.435, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.4075939953327179, | |
| "learning_rate": 0.00018623145725200278, | |
| "loss": 0.4274, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.36284372210502625, | |
| "learning_rate": 0.00018390719910244487, | |
| "loss": 0.3972, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.3902932405471802, | |
| "learning_rate": 0.00018141868867035745, | |
| "loss": 0.3953, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.38837161660194397, | |
| "learning_rate": 0.00017877079733177184, | |
| "loss": 0.4294, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.38330700993537903, | |
| "learning_rate": 0.0001759687084583285, | |
| "loss": 0.4155, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.3897230625152588, | |
| "learning_rate": 0.00017301790727058345, | |
| "loss": 0.3791, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.40314897894859314, | |
| "learning_rate": 0.00016992417010043142, | |
| "loss": 0.4018, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.4068446755409241, | |
| "learning_rate": 0.0001666935530836651, | |
| "loss": 0.3801, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.40863823890686035, | |
| "learning_rate": 0.0001633323803048047, | |
| "loss": 0.3879, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.3958357274532318, | |
| "learning_rate": 0.00015984723141740576, | |
| "loss": 0.3929, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.3210630416870117, | |
| "learning_rate": 0.0001562449287640781, | |
| "loss": 0.2722, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.34723371267318726, | |
| "learning_rate": 0.00015253252402142988, | |
| "loss": 0.2701, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.3267146050930023, | |
| "learning_rate": 0.00014871728439607966, | |
| "loss": 0.2654, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.3217560350894928, | |
| "learning_rate": 0.00014480667839875786, | |
| "loss": 0.2653, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3129405975341797, | |
| "learning_rate": 0.0001408083612243465, | |
| "loss": 0.2495, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.3169604241847992, | |
| "learning_rate": 0.00013673015976647568, | |
| "loss": 0.2783, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.3302832543849945, | |
| "learning_rate": 0.00013258005729601177, | |
| "loss": 0.2589, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.3463418781757355, | |
| "learning_rate": 0.0001283661778334297, | |
| "loss": 0.2453, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.3463260531425476, | |
| "learning_rate": 0.00012409677024566144, | |
| "loss": 0.242, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3702252209186554, | |
| "learning_rate": 0.00011978019209855174, | |
| "loss": 0.264, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.3509206771850586, | |
| "learning_rate": 0.00011542489329653024, | |
| "loss": 0.2593, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.3612159490585327, | |
| "learning_rate": 0.000111039399541527, | |
| "loss": 0.2411, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.3651520609855652, | |
| "learning_rate": 0.00010663229564351041, | |
| "loss": 0.2378, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.3665476441383362, | |
| "learning_rate": 0.00010221220871531869, | |
| "loss": 0.2334, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.34961438179016113, | |
| "learning_rate": 9.778779128468132e-05, | |
| "loss": 0.2307, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.379111111164093, | |
| "learning_rate": 9.336770435648964e-05, | |
| "loss": 0.2212, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.38593631982803345, | |
| "learning_rate": 8.896060045847304e-05, | |
| "loss": 0.2335, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.37961545586586, | |
| "learning_rate": 8.457510670346976e-05, | |
| "loss": 0.2306, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.3735259771347046, | |
| "learning_rate": 8.021980790144827e-05, | |
| "loss": 0.2499, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.40170496702194214, | |
| "learning_rate": 7.590322975433857e-05, | |
| "loss": 0.2275, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.3875046372413635, | |
| "learning_rate": 7.163382216657034e-05, | |
| "loss": 0.218, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.37028905749320984, | |
| "learning_rate": 6.741994270398826e-05, | |
| "loss": 0.2422, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.3737669289112091, | |
| "learning_rate": 6.326984023352435e-05, | |
| "loss": 0.2195, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.3924426734447479, | |
| "learning_rate": 5.91916387756535e-05, | |
| "loss": 0.2235, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.36921918392181396, | |
| "learning_rate": 5.5193321601242156e-05, | |
| "loss": 0.2141, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.28800830245018005, | |
| "learning_rate": 5.1282715603920374e-05, | |
| "loss": 0.1736, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.31100502610206604, | |
| "learning_rate": 4.746747597857014e-05, | |
| "loss": 0.1578, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.3049222528934479, | |
| "learning_rate": 4.375507123592194e-05, | |
| "loss": 0.1771, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.28219300508499146, | |
| "learning_rate": 4.015276858259427e-05, | |
| "loss": 0.1476, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.3022613525390625, | |
| "learning_rate": 3.6667619695195285e-05, | |
| "loss": 0.1779, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.2823966443538666, | |
| "learning_rate": 3.330644691633492e-05, | |
| "loss": 0.1501, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.2800000000000002, | |
| "grad_norm": 0.28174689412117004, | |
| "learning_rate": 3.0075829899568597e-05, | |
| "loss": 0.1511, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.2776714861392975, | |
| "learning_rate": 2.6982092729416587e-05, | |
| "loss": 0.1568, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.2745690643787384, | |
| "learning_rate": 2.403129154167153e-05, | |
| "loss": 0.1393, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.2862659692764282, | |
| "learning_rate": 2.1229202668228197e-05, | |
| "loss": 0.1568, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 0.30168795585632324, | |
| "learning_rate": 1.858131132964259e-05, | |
| "loss": 0.164, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.30739548802375793, | |
| "learning_rate": 1.609280089755515e-05, | |
| "loss": 0.1595, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 0.2983320355415344, | |
| "learning_rate": 1.3768542747997215e-05, | |
| "loss": 0.174, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.30526575446128845, | |
| "learning_rate": 1.161308672544389e-05, | |
| "loss": 0.168, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.28905755281448364, | |
| "learning_rate": 9.630652236279625e-06, | |
| "loss": 0.1557, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.29685401916503906, | |
| "learning_rate": 7.825119989112173e-06, | |
| "loss": 0.1531, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.309733510017395, | |
| "learning_rate": 6.200024398103255e-06, | |
| "loss": 0.1538, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.31325626373291016, | |
| "learning_rate": 4.758546664186869e-06, | |
| "loss": 0.16, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 0.29155057668685913, | |
| "learning_rate": 3.5035085477190143e-06, | |
| "loss": 0.1612, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.31043997406959534, | |
| "learning_rate": 2.4373668447493224e-06, | |
| "loss": 0.1655, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 0.30450770258903503, | |
| "learning_rate": 1.562208577727442e-06, | |
| "loss": 0.1581, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.2816106677055359, | |
| "learning_rate": 8.797469100585431e-07, | |
| "loss": 0.1369, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 0.3166625201702118, | |
| "learning_rate": 3.913177925055189e-07, | |
| "loss": 0.1724, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.30547070503234863, | |
| "learning_rate": 9.78773480026396e-08, | |
| "loss": 0.1555, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.29624301195144653, | |
| "learning_rate": 0.0, | |
| "loss": 0.1598, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.106714169779814e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |