| { | |
| "best_metric": 0.602910749664121, | |
| "best_model_checkpoint": "phobert-legal-qa-finetuned\\checkpoint-1800", | |
| "epoch": 0.36487270917986564, | |
| "eval_steps": 200, | |
| "global_step": 2200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008292561572269675, | |
| "grad_norm": 8.954318046569824, | |
| "learning_rate": 5.417357656163627e-07, | |
| "loss": 5.4244, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01658512314453935, | |
| "grad_norm": 6.557638168334961, | |
| "learning_rate": 1.0834715312327253e-06, | |
| "loss": 5.1885, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.024877684716809022, | |
| "grad_norm": 6.624057769775391, | |
| "learning_rate": 1.6141514648977336e-06, | |
| "loss": 4.6824, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0331702462890787, | |
| "grad_norm": 6.447659015655518, | |
| "learning_rate": 2.1669430624654506e-06, | |
| "loss": 3.9743, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0331702462890787, | |
| "eval_accuracy": 0.5348311306901615, | |
| "eval_end_accuracy": 0.29080763582966224, | |
| "eval_end_f1": 0.20537819782547337, | |
| "eval_f1": 0.10898462727065836, | |
| "eval_loss": 3.2860655784606934, | |
| "eval_runtime": 127.3249, | |
| "eval_samples_per_second": 133.713, | |
| "eval_start_accuracy": 0.7788546255506608, | |
| "eval_start_f1": 0.012591056715843358, | |
| "eval_steps_per_second": 16.721, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04146280786134837, | |
| "grad_norm": 5.946743965148926, | |
| "learning_rate": 2.7197346600331676e-06, | |
| "loss": 3.0621, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.049755369433618045, | |
| "grad_norm": 7.603392601013184, | |
| "learning_rate": 3.272526257600885e-06, | |
| "loss": 1.9122, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05804793100588772, | |
| "grad_norm": 4.718493461608887, | |
| "learning_rate": 3.825317855168602e-06, | |
| "loss": 0.8663, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.0663404925781574, | |
| "grad_norm": 1.8805421590805054, | |
| "learning_rate": 4.367053620784965e-06, | |
| "loss": 0.4477, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.0663404925781574, | |
| "eval_accuracy": 0.9048458149779737, | |
| "eval_end_accuracy": 0.9175917767988253, | |
| "eval_end_f1": 0.8303333971515077, | |
| "eval_f1": 0.424233749861785, | |
| "eval_loss": 0.23289425671100616, | |
| "eval_runtime": 129.7527, | |
| "eval_samples_per_second": 131.211, | |
| "eval_start_accuracy": 0.8920998531571219, | |
| "eval_start_f1": 0.018134102572062404, | |
| "eval_steps_per_second": 16.408, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.07463305415042706, | |
| "grad_norm": 1.6680586338043213, | |
| "learning_rate": 4.919845218352681e-06, | |
| "loss": 0.2926, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.08292561572269674, | |
| "grad_norm": 1.3913531303405762, | |
| "learning_rate": 5.472636815920398e-06, | |
| "loss": 0.2317, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09121817729496641, | |
| "grad_norm": 0.8732656836509705, | |
| "learning_rate": 6.025428413488116e-06, | |
| "loss": 0.2013, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.09951073886723609, | |
| "grad_norm": 0.4801824986934662, | |
| "learning_rate": 6.578220011055833e-06, | |
| "loss": 0.1501, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.09951073886723609, | |
| "eval_accuracy": 0.9687224669603525, | |
| "eval_end_accuracy": 0.9620558002936858, | |
| "eval_end_f1": 0.9029114827270522, | |
| "eval_f1": 0.47707631709814124, | |
| "eval_loss": 0.10375536233186722, | |
| "eval_runtime": 115.3633, | |
| "eval_samples_per_second": 147.577, | |
| "eval_start_accuracy": 0.9753891336270191, | |
| "eval_start_f1": 0.051241151469230285, | |
| "eval_steps_per_second": 18.455, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.10780330043950577, | |
| "grad_norm": 3.085040807723999, | |
| "learning_rate": 7.131011608623549e-06, | |
| "loss": 0.147, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.11609586201177544, | |
| "grad_norm": 2.0271573066711426, | |
| "learning_rate": 7.672747374239912e-06, | |
| "loss": 0.1356, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.12438842358404512, | |
| "grad_norm": 0.833656370639801, | |
| "learning_rate": 8.22553897180763e-06, | |
| "loss": 0.1604, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.1326809851563148, | |
| "grad_norm": 4.890761375427246, | |
| "learning_rate": 8.778330569375346e-06, | |
| "loss": 0.2731, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.1326809851563148, | |
| "eval_accuracy": 0.9534801762114538, | |
| "eval_end_accuracy": 0.9742143906020558, | |
| "eval_end_f1": 0.9321589248479495, | |
| "eval_f1": 0.49608082825261623, | |
| "eval_loss": 0.1122935563325882, | |
| "eval_runtime": 133.8336, | |
| "eval_samples_per_second": 127.21, | |
| "eval_start_accuracy": 0.9327459618208517, | |
| "eval_start_f1": 0.06000273165728297, | |
| "eval_steps_per_second": 15.908, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.14097354672858445, | |
| "grad_norm": 1.1853266954421997, | |
| "learning_rate": 9.331122166943063e-06, | |
| "loss": 0.1276, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.14926610830085413, | |
| "grad_norm": 7.255343437194824, | |
| "learning_rate": 9.88391376451078e-06, | |
| "loss": 0.0954, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.1575586698731238, | |
| "grad_norm": 0.5000291466712952, | |
| "learning_rate": 1.0436705362078497e-05, | |
| "loss": 0.1081, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.1658512314453935, | |
| "grad_norm": 0.3002016544342041, | |
| "learning_rate": 1.0989496959646216e-05, | |
| "loss": 0.1257, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1658512314453935, | |
| "eval_accuracy": 0.9772980910425844, | |
| "eval_end_accuracy": 0.9769750367107195, | |
| "eval_end_f1": 0.9421694663520186, | |
| "eval_f1": 0.5088194880993329, | |
| "eval_loss": 0.08298086374998093, | |
| "eval_runtime": 127.1174, | |
| "eval_samples_per_second": 133.931, | |
| "eval_start_accuracy": 0.9776211453744493, | |
| "eval_start_f1": 0.07546950984664719, | |
| "eval_steps_per_second": 16.748, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17414379301766317, | |
| "grad_norm": 1.4803558588027954, | |
| "learning_rate": 1.1542288557213931e-05, | |
| "loss": 0.0988, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.18243635458993282, | |
| "grad_norm": 5.256414413452148, | |
| "learning_rate": 1.2095080154781648e-05, | |
| "loss": 0.1159, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1907289161622025, | |
| "grad_norm": 2.2532193660736084, | |
| "learning_rate": 1.2647871752349365e-05, | |
| "loss": 0.1119, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.19902147773447218, | |
| "grad_norm": 1.1581368446350098, | |
| "learning_rate": 1.3200663349917082e-05, | |
| "loss": 0.0801, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.19902147773447218, | |
| "eval_accuracy": 0.9774743024963289, | |
| "eval_end_accuracy": 0.9773274596182085, | |
| "eval_end_f1": 0.9522660725045008, | |
| "eval_f1": 0.5138705881207118, | |
| "eval_loss": 0.06253915280103683, | |
| "eval_runtime": 129.6143, | |
| "eval_samples_per_second": 131.351, | |
| "eval_start_accuracy": 0.9776211453744493, | |
| "eval_start_f1": 0.0754751037369228, | |
| "eval_steps_per_second": 16.426, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.20731403930674186, | |
| "grad_norm": 3.3523240089416504, | |
| "learning_rate": 1.3742399115533445e-05, | |
| "loss": 0.1058, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.21560660087901154, | |
| "grad_norm": 0.09368986636400223, | |
| "learning_rate": 1.429519071310116e-05, | |
| "loss": 0.1002, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.2238991624512812, | |
| "grad_norm": 2.6880834102630615, | |
| "learning_rate": 1.4847982310668878e-05, | |
| "loss": 0.0563, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.23219172402355087, | |
| "grad_norm": 10.309085845947266, | |
| "learning_rate": 1.5400773908236596e-05, | |
| "loss": 0.0973, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.23219172402355087, | |
| "eval_accuracy": 0.9787371512481644, | |
| "eval_end_accuracy": 0.978208516886931, | |
| "eval_end_f1": 0.9484799222981063, | |
| "eval_f1": 0.5438896216202269, | |
| "eval_loss": 0.0639370009303093, | |
| "eval_runtime": 128.1258, | |
| "eval_samples_per_second": 132.877, | |
| "eval_start_accuracy": 0.9792657856093979, | |
| "eval_start_f1": 0.13929932094234762, | |
| "eval_steps_per_second": 16.616, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.24048428559582055, | |
| "grad_norm": 0.07747649401426315, | |
| "learning_rate": 1.5953565505804315e-05, | |
| "loss": 0.0724, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.24877684716809023, | |
| "grad_norm": 0.076473668217659, | |
| "learning_rate": 1.650635710337203e-05, | |
| "loss": 0.0928, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2570694087403599, | |
| "grad_norm": 0.18258516490459442, | |
| "learning_rate": 1.7059148700939746e-05, | |
| "loss": 0.085, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.2653619703126296, | |
| "grad_norm": 2.0234451293945312, | |
| "learning_rate": 1.7611940298507464e-05, | |
| "loss": 0.1107, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.2653619703126296, | |
| "eval_accuracy": 0.9774449339207049, | |
| "eval_end_accuracy": 0.9773274596182085, | |
| "eval_end_f1": 0.9548822846052203, | |
| "eval_f1": 0.5144663543275407, | |
| "eval_loss": 0.06893135607242584, | |
| "eval_runtime": 137.3173, | |
| "eval_samples_per_second": 123.983, | |
| "eval_start_accuracy": 0.9775624082232012, | |
| "eval_start_f1": 0.07405042404986113, | |
| "eval_steps_per_second": 15.504, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.27365453188489924, | |
| "grad_norm": 2.9243485927581787, | |
| "learning_rate": 1.816473189607518e-05, | |
| "loss": 0.069, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.2819470934571689, | |
| "grad_norm": 0.05162263661623001, | |
| "learning_rate": 1.87175234936429e-05, | |
| "loss": 0.0815, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.2902396550294386, | |
| "grad_norm": 0.12183202058076859, | |
| "learning_rate": 1.9270315091210617e-05, | |
| "loss": 0.0741, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.29853221660170826, | |
| "grad_norm": 3.247403621673584, | |
| "learning_rate": 1.9823106688778332e-05, | |
| "loss": 0.0819, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.29853221660170826, | |
| "eval_accuracy": 0.9795007342143907, | |
| "eval_end_accuracy": 0.9793832599118942, | |
| "eval_end_f1": 0.9547976694500852, | |
| "eval_f1": 0.602910749664121, | |
| "eval_loss": 0.07215487957000732, | |
| "eval_runtime": 135.7664, | |
| "eval_samples_per_second": 125.399, | |
| "eval_start_accuracy": 0.979618208516887, | |
| "eval_start_f1": 0.25102382987815675, | |
| "eval_steps_per_second": 15.681, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.30682477817397796, | |
| "grad_norm": 0.042116910219192505, | |
| "learning_rate": 1.9958225826268585e-05, | |
| "loss": 0.0641, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.3151173397462476, | |
| "grad_norm": 0.36534908413887024, | |
| "learning_rate": 1.989679321784003e-05, | |
| "loss": 0.1036, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.32340990131851727, | |
| "grad_norm": 0.2260214239358902, | |
| "learning_rate": 1.9835360609411478e-05, | |
| "loss": 0.0751, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.331702462890787, | |
| "grad_norm": 0.03621504455804825, | |
| "learning_rate": 1.977392800098292e-05, | |
| "loss": 0.0632, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.331702462890787, | |
| "eval_accuracy": 0.9779148311306902, | |
| "eval_end_accuracy": 0.9779735682819384, | |
| "eval_end_f1": 0.9323578270443484, | |
| "eval_f1": 0.5095598212540562, | |
| "eval_loss": 0.058555059134960175, | |
| "eval_runtime": 136.5016, | |
| "eval_samples_per_second": 124.724, | |
| "eval_start_accuracy": 0.977856093979442, | |
| "eval_start_f1": 0.08676181546376394, | |
| "eval_steps_per_second": 15.597, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3399950244630566, | |
| "grad_norm": 0.4064314365386963, | |
| "learning_rate": 1.9712495392554368e-05, | |
| "loss": 0.0833, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.34828758603532634, | |
| "grad_norm": 1.703637957572937, | |
| "learning_rate": 1.9651062784125818e-05, | |
| "loss": 0.1415, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.356580147607596, | |
| "grad_norm": 0.7372889518737793, | |
| "learning_rate": 1.958963017569726e-05, | |
| "loss": 0.2726, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.36487270917986564, | |
| "grad_norm": 0.28392699360847473, | |
| "learning_rate": 1.9528197567268707e-05, | |
| "loss": 0.137, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.36487270917986564, | |
| "eval_accuracy": 0.9787958883994126, | |
| "eval_end_accuracy": 0.9785609397944199, | |
| "eval_end_f1": 0.9504899506887509, | |
| "eval_f1": 0.5319231158110048, | |
| "eval_loss": 0.05977020785212517, | |
| "eval_runtime": 136.9823, | |
| "eval_samples_per_second": 124.286, | |
| "eval_start_accuracy": 0.9790308370044053, | |
| "eval_start_f1": 0.11335628093325875, | |
| "eval_steps_per_second": 15.542, | |
| "step": 2200 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 18087, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.001 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4598822918553600.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |