| { | |
| "best_metric": 0.260405570268631, | |
| "best_model_checkpoint": "./checkpoint-huawei-noah/checkpoint-45000", | |
| "epoch": 1.9922082521692934, | |
| "eval_steps": 1000, | |
| "global_step": 45000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 13.771051406860352, | |
| "learning_rate": 1.9873510587163855e-05, | |
| "loss": 0.7093, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_LOC_f1": 0.6130136331478614, | |
| "eval_ORG_f1": 0.5179822369964004, | |
| "eval_PER_f1": 0.6815983074481896, | |
| "eval_loss": 0.5055311322212219, | |
| "eval_overall_accuracy": 0.8334903358124973, | |
| "eval_overall_f1": 0.6058789768206305, | |
| "eval_overall_precision": 0.5856612385653673, | |
| "eval_overall_recall": 0.6275425074138874, | |
| "eval_runtime": 974.5486, | |
| "eval_samples_per_second": 67.416, | |
| "eval_steps_per_second": 0.264, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 23.08890151977539, | |
| "learning_rate": 1.974702117432771e-05, | |
| "loss": 0.501, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_LOC_f1": 0.6875965698393078, | |
| "eval_ORG_f1": 0.5727174388363369, | |
| "eval_PER_f1": 0.7225324778200253, | |
| "eval_loss": 0.4574837386608124, | |
| "eval_overall_accuracy": 0.8565286056409621, | |
| "eval_overall_f1": 0.6654497292771804, | |
| "eval_overall_precision": 0.6761779246769729, | |
| "eval_overall_recall": 0.6550566427755974, | |
| "eval_runtime": 902.1725, | |
| "eval_samples_per_second": 72.824, | |
| "eval_steps_per_second": 0.285, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 20.49671745300293, | |
| "learning_rate": 1.9620531761491565e-05, | |
| "loss": 0.4505, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_LOC_f1": 0.7204635210044149, | |
| "eval_ORG_f1": 0.5725871273468899, | |
| "eval_PER_f1": 0.7433987684658955, | |
| "eval_loss": 0.4243098795413971, | |
| "eval_overall_accuracy": 0.8621471869792049, | |
| "eval_overall_f1": 0.6874798198516072, | |
| "eval_overall_precision": 0.6776606774900581, | |
| "eval_overall_recall": 0.6975877000398559, | |
| "eval_runtime": 876.2182, | |
| "eval_samples_per_second": 74.981, | |
| "eval_steps_per_second": 0.293, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 14.499987602233887, | |
| "learning_rate": 1.949404234865542e-05, | |
| "loss": 0.4268, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_LOC_f1": 0.7364175740245708, | |
| "eval_ORG_f1": 0.5861517860999524, | |
| "eval_PER_f1": 0.7518103620724146, | |
| "eval_loss": 0.4260491728782654, | |
| "eval_overall_accuracy": 0.8618999871026152, | |
| "eval_overall_f1": 0.6951561442521513, | |
| "eval_overall_precision": 0.6862764458838554, | |
| "eval_overall_recall": 0.7042686428026184, | |
| "eval_runtime": 900.7744, | |
| "eval_samples_per_second": 72.937, | |
| "eval_steps_per_second": 0.285, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.379225969314575, | |
| "learning_rate": 1.9367552935819272e-05, | |
| "loss": 0.4023, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_LOC_f1": 0.7258738398309293, | |
| "eval_ORG_f1": 0.6257472293742304, | |
| "eval_PER_f1": 0.7652099137004392, | |
| "eval_loss": 0.3998052477836609, | |
| "eval_overall_accuracy": 0.8690542423545061, | |
| "eval_overall_f1": 0.70826256852105, | |
| "eval_overall_precision": 0.7241968388877515, | |
| "eval_overall_recall": 0.6930143954388549, | |
| "eval_runtime": 879.0998, | |
| "eval_samples_per_second": 74.736, | |
| "eval_steps_per_second": 0.292, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 17.12338638305664, | |
| "learning_rate": 1.924106352298313e-05, | |
| "loss": 0.3733, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_LOC_f1": 0.7708459775529246, | |
| "eval_ORG_f1": 0.6382364519091983, | |
| "eval_PER_f1": 0.774621616895607, | |
| "eval_loss": 0.3535228669643402, | |
| "eval_overall_accuracy": 0.8847707173727775, | |
| "eval_overall_f1": 0.7347440116838627, | |
| "eval_overall_precision": 0.7320440692286544, | |
| "eval_overall_recall": 0.7374639438503577, | |
| "eval_runtime": 911.8464, | |
| "eval_samples_per_second": 72.052, | |
| "eval_steps_per_second": 0.282, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 45.46567153930664, | |
| "learning_rate": 1.9114574110146982e-05, | |
| "loss": 0.3636, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_LOC_f1": 0.7551567306422088, | |
| "eval_ORG_f1": 0.6445153971205689, | |
| "eval_PER_f1": 0.7953595339091327, | |
| "eval_loss": 0.36239349842071533, | |
| "eval_overall_accuracy": 0.8839519863237142, | |
| "eval_overall_f1": 0.7360533119495638, | |
| "eval_overall_precision": 0.7536363185945938, | |
| "eval_overall_recall": 0.7192720542041302, | |
| "eval_runtime": 934.2532, | |
| "eval_samples_per_second": 70.324, | |
| "eval_steps_per_second": 0.275, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 23.77419090270996, | |
| "learning_rate": 1.8988084697310836e-05, | |
| "loss": 0.3438, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_LOC_f1": 0.7777527075812274, | |
| "eval_ORG_f1": 0.6523009578392771, | |
| "eval_PER_f1": 0.7983586509404405, | |
| "eval_loss": 0.35361218452453613, | |
| "eval_overall_accuracy": 0.8877984416924404, | |
| "eval_overall_f1": 0.7468599888415638, | |
| "eval_overall_precision": 0.7522804372408594, | |
| "eval_overall_recall": 0.741517094161437, | |
| "eval_runtime": 964.9468, | |
| "eval_samples_per_second": 68.087, | |
| "eval_steps_per_second": 0.266, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 12.217660903930664, | |
| "learning_rate": 1.8861595284474693e-05, | |
| "loss": 0.3388, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_LOC_f1": 0.7769920027192643, | |
| "eval_ORG_f1": 0.6685048661658352, | |
| "eval_PER_f1": 0.7955169761564663, | |
| "eval_loss": 0.36109644174575806, | |
| "eval_overall_accuracy": 0.8892544552878013, | |
| "eval_overall_f1": 0.7515097063573983, | |
| "eval_overall_precision": 0.763163393230255, | |
| "eval_overall_recall": 0.7402065755608547, | |
| "eval_runtime": 935.4215, | |
| "eval_samples_per_second": 70.236, | |
| "eval_steps_per_second": 0.275, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 6.038209438323975, | |
| "learning_rate": 1.8735105871638546e-05, | |
| "loss": 0.34, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_LOC_f1": 0.7909373544148547, | |
| "eval_ORG_f1": 0.6797622124579994, | |
| "eval_PER_f1": 0.8097494515182176, | |
| "eval_loss": 0.31574323773384094, | |
| "eval_overall_accuracy": 0.8950133905201693, | |
| "eval_overall_f1": 0.7636732236459436, | |
| "eval_overall_precision": 0.7691121860312153, | |
| "eval_overall_recall": 0.7583106469503421, | |
| "eval_runtime": 875.8957, | |
| "eval_samples_per_second": 75.009, | |
| "eval_steps_per_second": 0.293, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 4.172688961029053, | |
| "learning_rate": 1.86086164588024e-05, | |
| "loss": 0.3234, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_LOC_f1": 0.7809025636945843, | |
| "eval_ORG_f1": 0.6583455896967046, | |
| "eval_PER_f1": 0.8088906849868661, | |
| "eval_loss": 0.3503168821334839, | |
| "eval_overall_accuracy": 0.8873938178535211, | |
| "eval_overall_f1": 0.7490455281410953, | |
| "eval_overall_precision": 0.7389521011220446, | |
| "eval_overall_recall": 0.7594185080353705, | |
| "eval_runtime": 876.8783, | |
| "eval_samples_per_second": 74.925, | |
| "eval_steps_per_second": 0.293, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.31278446316719055, | |
| "learning_rate": 1.8482127045966253e-05, | |
| "loss": 0.3116, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_LOC_f1": 0.8015539728054758, | |
| "eval_ORG_f1": 0.6780179794520548, | |
| "eval_PER_f1": 0.807255456750202, | |
| "eval_loss": 0.3222993314266205, | |
| "eval_overall_accuracy": 0.8972015203740747, | |
| "eval_overall_f1": 0.7650194669062593, | |
| "eval_overall_precision": 0.7538710756376371, | |
| "eval_overall_recall": 0.7765025365965696, | |
| "eval_runtime": 876.0279, | |
| "eval_samples_per_second": 74.998, | |
| "eval_steps_per_second": 0.293, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 17.869319915771484, | |
| "learning_rate": 1.8355637633130106e-05, | |
| "loss": 0.3195, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_LOC_f1": 0.8077397954482631, | |
| "eval_ORG_f1": 0.6910689787013923, | |
| "eval_PER_f1": 0.8216717076180746, | |
| "eval_loss": 0.3207678496837616, | |
| "eval_overall_accuracy": 0.8985247667722903, | |
| "eval_overall_f1": 0.7773739255449474, | |
| "eval_overall_precision": 0.7767264634475317, | |
| "eval_overall_recall": 0.7780224679632244, | |
| "eval_runtime": 885.6262, | |
| "eval_samples_per_second": 74.185, | |
| "eval_steps_per_second": 0.29, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 31.75470733642578, | |
| "learning_rate": 1.8229148220293963e-05, | |
| "loss": 0.3038, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_LOC_f1": 0.7926177313116938, | |
| "eval_ORG_f1": 0.6917118970756915, | |
| "eval_PER_f1": 0.8275611161079727, | |
| "eval_loss": 0.3353281319141388, | |
| "eval_overall_accuracy": 0.8932463223486391, | |
| "eval_overall_f1": 0.775302594053191, | |
| "eval_overall_precision": 0.7831411656868759, | |
| "eval_overall_recall": 0.7676193821647875, | |
| "eval_runtime": 929.7993, | |
| "eval_samples_per_second": 70.66, | |
| "eval_steps_per_second": 0.276, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 4.61831521987915, | |
| "learning_rate": 1.8102658807457817e-05, | |
| "loss": 0.3108, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_LOC_f1": 0.8077224388355109, | |
| "eval_ORG_f1": 0.6909212129572154, | |
| "eval_PER_f1": 0.8192221560207918, | |
| "eval_loss": 0.3174687623977661, | |
| "eval_overall_accuracy": 0.8962873233880165, | |
| "eval_overall_f1": 0.7771138787987528, | |
| "eval_overall_precision": 0.7788884364820847, | |
| "eval_overall_recall": 0.7753473887579121, | |
| "eval_runtime": 989.2234, | |
| "eval_samples_per_second": 66.416, | |
| "eval_steps_per_second": 0.26, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 12.568930625915527, | |
| "learning_rate": 1.7976169394621674e-05, | |
| "loss": 0.3071, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_LOC_f1": 0.794363707776905, | |
| "eval_ORG_f1": 0.6988674651653153, | |
| "eval_PER_f1": 0.830271581375005, | |
| "eval_loss": 0.32856282591819763, | |
| "eval_overall_accuracy": 0.8977907538395009, | |
| "eval_overall_f1": 0.7767913544942637, | |
| "eval_overall_precision": 0.7810995225703338, | |
| "eval_overall_recall": 0.772530449291712, | |
| "eval_runtime": 943.0369, | |
| "eval_samples_per_second": 69.669, | |
| "eval_steps_per_second": 0.273, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 5.375293731689453, | |
| "learning_rate": 1.7849679981785527e-05, | |
| "loss": 0.294, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_LOC_f1": 0.8179866156103497, | |
| "eval_ORG_f1": 0.7139503216415458, | |
| "eval_PER_f1": 0.8319541693189051, | |
| "eval_loss": 0.3004244267940521, | |
| "eval_overall_accuracy": 0.9056726997766982, | |
| "eval_overall_f1": 0.7925003906754269, | |
| "eval_overall_precision": 0.7971147209370665, | |
| "eval_overall_recall": 0.7879391757243317, | |
| "eval_runtime": 922.5388, | |
| "eval_samples_per_second": 71.217, | |
| "eval_steps_per_second": 0.279, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 15.245515823364258, | |
| "learning_rate": 1.772319056894938e-05, | |
| "loss": 0.3022, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_LOC_f1": 0.8157123906017681, | |
| "eval_ORG_f1": 0.7098208686331968, | |
| "eval_PER_f1": 0.8456405034257072, | |
| "eval_loss": 0.30179163813591003, | |
| "eval_overall_accuracy": 0.9067740352882565, | |
| "eval_overall_f1": 0.7949277311259202, | |
| "eval_overall_precision": 0.7816879334439119, | |
| "eval_overall_recall": 0.808623752811873, | |
| "eval_runtime": 877.0088, | |
| "eval_samples_per_second": 74.914, | |
| "eval_steps_per_second": 0.293, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 27.571258544921875, | |
| "learning_rate": 1.7596701156113234e-05, | |
| "loss": 0.2932, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_LOC_f1": 0.812069319938867, | |
| "eval_ORG_f1": 0.711144363521586, | |
| "eval_PER_f1": 0.8502128757579667, | |
| "eval_loss": 0.2992981970310211, | |
| "eval_overall_accuracy": 0.9061335916182172, | |
| "eval_overall_f1": 0.7960065863825535, | |
| "eval_overall_precision": 0.7903741392400006, | |
| "eval_overall_recall": 0.8017198867820013, | |
| "eval_runtime": 877.5146, | |
| "eval_samples_per_second": 74.871, | |
| "eval_steps_per_second": 0.293, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 6.182469844818115, | |
| "learning_rate": 1.747021174327709e-05, | |
| "loss": 0.2943, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_LOC_f1": 0.8233705131530693, | |
| "eval_ORG_f1": 0.7053565992474702, | |
| "eval_PER_f1": 0.8430738573647348, | |
| "eval_loss": 0.28671061992645264, | |
| "eval_overall_accuracy": 0.9084127618358795, | |
| "eval_overall_f1": 0.7974687389257328, | |
| "eval_overall_precision": 0.7968528685705035, | |
| "eval_overall_recall": 0.7980855620030669, | |
| "eval_runtime": 876.9521, | |
| "eval_samples_per_second": 74.919, | |
| "eval_steps_per_second": 0.293, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 37.97357177734375, | |
| "learning_rate": 1.7343722330440944e-05, | |
| "loss": 0.2782, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_LOC_f1": 0.8330484744899815, | |
| "eval_ORG_f1": 0.7210407632263661, | |
| "eval_PER_f1": 0.8489616418275103, | |
| "eval_loss": 0.30233901739120483, | |
| "eval_overall_accuracy": 0.906348548032643, | |
| "eval_overall_f1": 0.8060817413887736, | |
| "eval_overall_precision": 0.7987835211557213, | |
| "eval_overall_recall": 0.8135145541872421, | |
| "eval_runtime": 903.2924, | |
| "eval_samples_per_second": 72.734, | |
| "eval_steps_per_second": 0.285, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.23641037940979, | |
| "learning_rate": 1.7217232917604798e-05, | |
| "loss": 0.2913, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_LOC_f1": 0.8284559448131656, | |
| "eval_ORG_f1": 0.7247506440319704, | |
| "eval_PER_f1": 0.8467821489168608, | |
| "eval_loss": 0.2910088300704956, | |
| "eval_overall_accuracy": 0.908466500939486, | |
| "eval_overall_f1": 0.8031924311719724, | |
| "eval_overall_precision": 0.7945208195637806, | |
| "eval_overall_recall": 0.8120554200752534, | |
| "eval_runtime": 930.8988, | |
| "eval_samples_per_second": 70.577, | |
| "eval_steps_per_second": 0.276, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 51.232643127441406, | |
| "learning_rate": 1.709074350476865e-05, | |
| "loss": 0.254, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_LOC_f1": 0.8326820729529502, | |
| "eval_ORG_f1": 0.7281753443144438, | |
| "eval_PER_f1": 0.8538723624698639, | |
| "eval_loss": 0.3031412661075592, | |
| "eval_overall_accuracy": 0.9094907050317503, | |
| "eval_overall_f1": 0.8106029688042823, | |
| "eval_overall_precision": 0.8094644661502189, | |
| "eval_overall_recall": 0.8117446785514041, | |
| "eval_runtime": 982.2729, | |
| "eval_samples_per_second": 66.886, | |
| "eval_steps_per_second": 0.262, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.0182816982269287, | |
| "learning_rate": 1.6964254091932504e-05, | |
| "loss": 0.2412, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_LOC_f1": 0.8337292382788848, | |
| "eval_ORG_f1": 0.7265721539463927, | |
| "eval_PER_f1": 0.8438998504510589, | |
| "eval_loss": 0.2959749400615692, | |
| "eval_overall_accuracy": 0.9087023207706061, | |
| "eval_overall_f1": 0.8054477820887165, | |
| "eval_overall_precision": 0.7949914132873621, | |
| "eval_overall_recall": 0.8161828781420359, | |
| "eval_runtime": 937.946, | |
| "eval_samples_per_second": 70.047, | |
| "eval_steps_per_second": 0.274, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 4.651257038116455, | |
| "learning_rate": 1.683776467909636e-05, | |
| "loss": 0.2248, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_LOC_f1": 0.8204892221350009, | |
| "eval_ORG_f1": 0.7164497497985908, | |
| "eval_PER_f1": 0.8366153573083787, | |
| "eval_loss": 0.2870059013366699, | |
| "eval_overall_accuracy": 0.9095747909232758, | |
| "eval_overall_f1": 0.7954083144399056, | |
| "eval_overall_precision": 0.7841744564646571, | |
| "eval_overall_recall": 0.806968716434849, | |
| "eval_runtime": 880.5612, | |
| "eval_samples_per_second": 74.612, | |
| "eval_steps_per_second": 0.292, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.5376187562942505, | |
| "learning_rate": 1.6711275266260215e-05, | |
| "loss": 0.2367, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_LOC_f1": 0.8274943290245922, | |
| "eval_ORG_f1": 0.7181801646127961, | |
| "eval_PER_f1": 0.8393770384236834, | |
| "eval_loss": 0.3035840094089508, | |
| "eval_overall_accuracy": 0.9099459068505344, | |
| "eval_overall_f1": 0.7997951699758828, | |
| "eval_overall_precision": 0.7826674511984585, | |
| "eval_overall_recall": 0.8176892990076537, | |
| "eval_runtime": 877.1859, | |
| "eval_samples_per_second": 74.899, | |
| "eval_steps_per_second": 0.293, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 29.850025177001953, | |
| "learning_rate": 1.658478585342407e-05, | |
| "loss": 0.2259, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_LOC_f1": 0.8368737846096861, | |
| "eval_ORG_f1": 0.730566338210411, | |
| "eval_PER_f1": 0.8548625950405009, | |
| "eval_loss": 0.2981078624725342, | |
| "eval_overall_accuracy": 0.912430550111398, | |
| "eval_overall_f1": 0.8107802613802052, | |
| "eval_overall_precision": 0.8028279082088811, | |
| "eval_overall_recall": 0.8188917335999406, | |
| "eval_runtime": 900.9915, | |
| "eval_samples_per_second": 72.92, | |
| "eval_steps_per_second": 0.285, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 6.522253036499023, | |
| "learning_rate": 1.6458296440587925e-05, | |
| "loss": 0.2353, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_LOC_f1": 0.8361146769362431, | |
| "eval_ORG_f1": 0.7397288818401768, | |
| "eval_PER_f1": 0.851691836373991, | |
| "eval_loss": 0.2890784442424774, | |
| "eval_overall_accuracy": 0.9118988490980682, | |
| "eval_overall_f1": 0.813503140265178, | |
| "eval_overall_precision": 0.8080863288253761, | |
| "eval_overall_recall": 0.8189930623577175, | |
| "eval_runtime": 898.5924, | |
| "eval_samples_per_second": 73.114, | |
| "eval_steps_per_second": 0.286, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 18.48634910583496, | |
| "learning_rate": 1.633180702775178e-05, | |
| "loss": 0.231, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_LOC_f1": 0.8399307496708805, | |
| "eval_ORG_f1": 0.7353511607405231, | |
| "eval_PER_f1": 0.8429051875514639, | |
| "eval_loss": 0.29812344908714294, | |
| "eval_overall_accuracy": 0.9100799384971765, | |
| "eval_overall_f1": 0.8109618028412001, | |
| "eval_overall_precision": 0.8090263071203351, | |
| "eval_overall_recall": 0.8129065816405802, | |
| "eval_runtime": 911.3101, | |
| "eval_samples_per_second": 72.094, | |
| "eval_steps_per_second": 0.282, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 6.476167678833008, | |
| "learning_rate": 1.6205317614915632e-05, | |
| "loss": 0.2298, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_LOC_f1": 0.8437642148074813, | |
| "eval_ORG_f1": 0.741920341727885, | |
| "eval_PER_f1": 0.845791168353266, | |
| "eval_loss": 0.2789755165576935, | |
| "eval_overall_accuracy": 0.9128231616800994, | |
| "eval_overall_f1": 0.8156984934617233, | |
| "eval_overall_precision": 0.8007835888891781, | |
| "eval_overall_recall": 0.8311795342930293, | |
| "eval_runtime": 950.013, | |
| "eval_samples_per_second": 69.157, | |
| "eval_steps_per_second": 0.271, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.6922666430473328, | |
| "learning_rate": 1.607882820207949e-05, | |
| "loss": 0.2236, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_LOC_f1": 0.8401312445122232, | |
| "eval_ORG_f1": 0.7412946847115139, | |
| "eval_PER_f1": 0.8599101069965396, | |
| "eval_loss": 0.2861514985561371, | |
| "eval_overall_accuracy": 0.9133093425115507, | |
| "eval_overall_f1": 0.8168471254617229, | |
| "eval_overall_precision": 0.8077931685921699, | |
| "eval_overall_recall": 0.8261063411536617, | |
| "eval_runtime": 964.6377, | |
| "eval_samples_per_second": 68.108, | |
| "eval_steps_per_second": 0.266, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 10.913984298706055, | |
| "learning_rate": 1.5952338789243342e-05, | |
| "loss": 0.2164, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_LOC_f1": 0.8450018789928598, | |
| "eval_ORG_f1": 0.7453389102160086, | |
| "eval_PER_f1": 0.8475419561015748, | |
| "eval_loss": 0.29202836751937866, | |
| "eval_overall_accuracy": 0.9133548626934291, | |
| "eval_overall_f1": 0.8167286457267982, | |
| "eval_overall_precision": 0.8108745156006552, | |
| "eval_overall_recall": 0.8226679186397627, | |
| "eval_runtime": 934.7918, | |
| "eval_samples_per_second": 70.283, | |
| "eval_steps_per_second": 0.275, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 8.604541778564453, | |
| "learning_rate": 1.5825849376407196e-05, | |
| "loss": 0.2343, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_LOC_f1": 0.8327052539148251, | |
| "eval_ORG_f1": 0.7464142820374833, | |
| "eval_PER_f1": 0.8571568569804591, | |
| "eval_loss": 0.26980945467948914, | |
| "eval_overall_accuracy": 0.915141529832157, | |
| "eval_overall_f1": 0.8162628685387808, | |
| "eval_overall_precision": 0.8141182004502234, | |
| "eval_overall_recall": 0.818418866063648, | |
| "eval_runtime": 878.6904, | |
| "eval_samples_per_second": 74.77, | |
| "eval_steps_per_second": 0.292, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 20.011140823364258, | |
| "learning_rate": 1.569935996357105e-05, | |
| "loss": 0.2305, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_LOC_f1": 0.8434370154154885, | |
| "eval_ORG_f1": 0.7450794786844748, | |
| "eval_PER_f1": 0.8598302131901996, | |
| "eval_loss": 0.2736206650733948, | |
| "eval_overall_accuracy": 0.9164160949247526, | |
| "eval_overall_f1": 0.820116525352046, | |
| "eval_overall_precision": 0.8215285544822911, | |
| "eval_overall_recall": 0.818709341835942, | |
| "eval_runtime": 878.0774, | |
| "eval_samples_per_second": 74.823, | |
| "eval_steps_per_second": 0.293, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.8893330693244934, | |
| "learning_rate": 1.5572870550734906e-05, | |
| "loss": 0.218, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_LOC_f1": 0.8372996858861737, | |
| "eval_ORG_f1": 0.7351363688234623, | |
| "eval_PER_f1": 0.8452893909397927, | |
| "eval_loss": 0.277670681476593, | |
| "eval_overall_accuracy": 0.9132897435443531, | |
| "eval_overall_f1": 0.8105032765054125, | |
| "eval_overall_precision": 0.7914432306117588, | |
| "eval_overall_recall": 0.8305040092411827, | |
| "eval_runtime": 878.1965, | |
| "eval_samples_per_second": 74.812, | |
| "eval_steps_per_second": 0.293, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 2.807310104370117, | |
| "learning_rate": 1.544638113789876e-05, | |
| "loss": 0.2209, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_LOC_f1": 0.8436019819082686, | |
| "eval_ORG_f1": 0.7541017701160051, | |
| "eval_PER_f1": 0.8559255699664113, | |
| "eval_loss": 0.2975883483886719, | |
| "eval_overall_accuracy": 0.9155777649084917, | |
| "eval_overall_f1": 0.8217433690792348, | |
| "eval_overall_precision": 0.8178831213153369, | |
| "eval_overall_recall": 0.8256402288678876, | |
| "eval_runtime": 887.7886, | |
| "eval_samples_per_second": 74.004, | |
| "eval_steps_per_second": 0.289, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.7378529906272888, | |
| "learning_rate": 1.5319891725062616e-05, | |
| "loss": 0.2068, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_LOC_f1": 0.8404596277816221, | |
| "eval_ORG_f1": 0.7519756060658962, | |
| "eval_PER_f1": 0.8650594959056045, | |
| "eval_loss": 0.2906901240348816, | |
| "eval_overall_accuracy": 0.9159109473508519, | |
| "eval_overall_f1": 0.8231073274551537, | |
| "eval_overall_precision": 0.8232630746670091, | |
| "eval_overall_recall": 0.8229516391615384, | |
| "eval_runtime": 914.9939, | |
| "eval_samples_per_second": 71.804, | |
| "eval_steps_per_second": 0.281, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 4.775814056396484, | |
| "learning_rate": 1.5193402312226468e-05, | |
| "loss": 0.2222, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_LOC_f1": 0.8486954241510423, | |
| "eval_ORG_f1": 0.7403596163509645, | |
| "eval_PER_f1": 0.8627163820626227, | |
| "eval_loss": 0.2920599579811096, | |
| "eval_overall_accuracy": 0.9144530370812459, | |
| "eval_overall_f1": 0.8204945751023299, | |
| "eval_overall_precision": 0.8079300635190885, | |
| "eval_overall_recall": 0.8334560537177521, | |
| "eval_runtime": 974.0136, | |
| "eval_samples_per_second": 67.453, | |
| "eval_steps_per_second": 0.264, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 2.2256317138671875, | |
| "learning_rate": 1.5066912899390323e-05, | |
| "loss": 0.2328, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_LOC_f1": 0.849334397801749, | |
| "eval_ORG_f1": 0.753236617390506, | |
| "eval_PER_f1": 0.8693573280340553, | |
| "eval_loss": 0.29489845037460327, | |
| "eval_overall_accuracy": 0.917294255100157, | |
| "eval_overall_f1": 0.8276056778793333, | |
| "eval_overall_precision": 0.8335114765330592, | |
| "eval_overall_recall": 0.8217829808218438, | |
| "eval_runtime": 944.9062, | |
| "eval_samples_per_second": 69.531, | |
| "eval_steps_per_second": 0.272, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 16.9512996673584, | |
| "learning_rate": 1.4940423486554176e-05, | |
| "loss": 0.2229, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_LOC_f1": 0.8445889009269291, | |
| "eval_ORG_f1": 0.7452847675981278, | |
| "eval_PER_f1": 0.8508496270046708, | |
| "eval_loss": 0.27226653695106506, | |
| "eval_overall_accuracy": 0.916396495957555, | |
| "eval_overall_f1": 0.818113712374582, | |
| "eval_overall_precision": 0.8101638106341121, | |
| "eval_overall_recall": 0.8262211804124756, | |
| "eval_runtime": 919.492, | |
| "eval_samples_per_second": 71.452, | |
| "eval_steps_per_second": 0.28, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 0.6652330160140991, | |
| "learning_rate": 1.481393407371803e-05, | |
| "loss": 0.2219, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_LOC_f1": 0.8483445744353834, | |
| "eval_ORG_f1": 0.7490041659533185, | |
| "eval_PER_f1": 0.865145374272123, | |
| "eval_loss": 0.2795349359512329, | |
| "eval_overall_accuracy": 0.9183601860258099, | |
| "eval_overall_f1": 0.8247159453593771, | |
| "eval_overall_precision": 0.8203393908609086, | |
| "eval_overall_recall": 0.8291394486364527, | |
| "eval_runtime": 878.0472, | |
| "eval_samples_per_second": 74.825, | |
| "eval_steps_per_second": 0.293, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 8.537057876586914, | |
| "learning_rate": 1.4687444660881885e-05, | |
| "loss": 0.2265, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_LOC_f1": 0.8431806420528559, | |
| "eval_ORG_f1": 0.7481651106805237, | |
| "eval_PER_f1": 0.8624314888139032, | |
| "eval_loss": 0.2695271372795105, | |
| "eval_overall_accuracy": 0.9176305986662586, | |
| "eval_overall_f1": 0.8227747180347664, | |
| "eval_overall_precision": 0.8127298369558054, | |
| "eval_overall_recall": 0.8330710044381996, | |
| "eval_runtime": 878.4759, | |
| "eval_samples_per_second": 74.789, | |
| "eval_steps_per_second": 0.293, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 2.5510284900665283, | |
| "learning_rate": 1.456095524804574e-05, | |
| "loss": 0.208, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_LOC_f1": 0.8455445895423215, | |
| "eval_ORG_f1": 0.7595170903640558, | |
| "eval_PER_f1": 0.8683249226114047, | |
| "eval_loss": 0.291418194770813, | |
| "eval_overall_accuracy": 0.9158768072144431, | |
| "eval_overall_f1": 0.828613510075651, | |
| "eval_overall_precision": 0.8212624496539643, | |
| "eval_overall_recall": 0.8360973566704721, | |
| "eval_runtime": 879.88, | |
| "eval_samples_per_second": 74.669, | |
| "eval_steps_per_second": 0.292, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 40.04784393310547, | |
| "learning_rate": 1.4434465835209595e-05, | |
| "loss": 0.2201, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_LOC_f1": 0.8562697361719467, | |
| "eval_ORG_f1": 0.7598152424942263, | |
| "eval_PER_f1": 0.8635555196248839, | |
| "eval_loss": 0.26673147082328796, | |
| "eval_overall_accuracy": 0.9210597857010994, | |
| "eval_overall_f1": 0.8302701818813599, | |
| "eval_overall_precision": 0.8374682152429387, | |
| "eval_overall_recall": 0.8231948281802031, | |
| "eval_runtime": 908.7842, | |
| "eval_samples_per_second": 72.294, | |
| "eval_steps_per_second": 0.283, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 4.13014030456543, | |
| "learning_rate": 1.4307976422373449e-05, | |
| "loss": 0.2065, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_LOC_f1": 0.850374667633551, | |
| "eval_ORG_f1": 0.7572422253856926, | |
| "eval_PER_f1": 0.8662861965717501, | |
| "eval_loss": 0.260405570268631, | |
| "eval_overall_accuracy": 0.9184101317809266, | |
| "eval_overall_f1": 0.8283524878105291, | |
| "eval_overall_precision": 0.835684527492472, | |
| "eval_overall_recall": 0.821147987273108, | |
| "eval_runtime": 930.7338, | |
| "eval_samples_per_second": 70.589, | |
| "eval_steps_per_second": 0.276, | |
| "step": 45000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 158116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1.4980271080173432e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |