| { |
| "best_global_step": 300, |
| "best_metric": 0.33115288615226746, |
| "best_model_checkpoint": "/workspace/anchor/olmo-compassion-s1/train/checkpoint-300", |
| "epoch": 0.3211131924003211, |
| "eval_steps": 50, |
| "global_step": 300, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.005351886540005352, |
| "grad_norm": 8.027827262878418, |
| "learning_rate": 1.7021276595744682e-05, |
| "loss": 7.017, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.010703773080010704, |
| "grad_norm": 6.428860187530518, |
| "learning_rate": 3.829787234042553e-05, |
| "loss": 6.4859, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.016055659620016056, |
| "grad_norm": 5.057083606719971, |
| "learning_rate": 5.9574468085106384e-05, |
| "loss": 5.498, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.02140754616002141, |
| "grad_norm": 3.510988235473633, |
| "learning_rate": 8.085106382978723e-05, |
| "loss": 4.6372, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.02675943270002676, |
| "grad_norm": 2.448411464691162, |
| "learning_rate": 0.00010212765957446809, |
| "loss": 3.9152, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.03211131924003211, |
| "grad_norm": 1.693341612815857, |
| "learning_rate": 0.00012340425531914893, |
| "loss": 3.2389, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.037463205780037465, |
| "grad_norm": 2.012845277786255, |
| "learning_rate": 0.0001446808510638298, |
| "loss": 2.8434, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.04281509232004282, |
| "grad_norm": 1.9178993701934814, |
| "learning_rate": 0.00016595744680851065, |
| "loss": 1.9668, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.048166978860048164, |
| "grad_norm": 1.9014297723770142, |
| "learning_rate": 0.0001872340425531915, |
| "loss": 1.4078, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.05351886540005352, |
| "grad_norm": 1.721352458000183, |
| "learning_rate": 0.00019999749676283775, |
| "loss": 1.1673, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05351886540005352, |
| "eval_loss": 0.7825450897216797, |
| "eval_runtime": 223.338, |
| "eval_samples_per_second": 1.719, |
| "eval_steps_per_second": 0.86, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05887075194005887, |
| "grad_norm": 1.2226594686508179, |
| "learning_rate": 0.00019996933678400946, |
| "loss": 0.8503, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.06422263848006422, |
| "grad_norm": 1.3121552467346191, |
| "learning_rate": 0.00019990989662046818, |
| "loss": 0.8787, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.06957452502006957, |
| "grad_norm": 1.5243521928787231, |
| "learning_rate": 0.00019981919487090972, |
| "loss": 0.81, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.07492641156007493, |
| "grad_norm": 1.1321091651916504, |
| "learning_rate": 0.00019969725991571128, |
| "loss": 0.6448, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.08027829810008028, |
| "grad_norm": 0.9228650331497192, |
| "learning_rate": 0.00019954412990805107, |
| "loss": 0.874, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.08563018464008564, |
| "grad_norm": 0.7880178093910217, |
| "learning_rate": 0.0001993598527619703, |
| "loss": 0.6613, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.09098207118009098, |
| "grad_norm": 0.749523937702179, |
| "learning_rate": 0.00019914448613738106, |
| "loss": 0.782, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.09633395772009633, |
| "grad_norm": 0.7473569512367249, |
| "learning_rate": 0.00019889809742202455, |
| "loss": 0.7812, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.10168584426010169, |
| "grad_norm": 0.7861048579216003, |
| "learning_rate": 0.0001986207637103857, |
| "loss": 0.6086, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.10703773080010703, |
| "grad_norm": 0.6467419862747192, |
| "learning_rate": 0.00019831257177957044, |
| "loss": 0.7206, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.10703773080010703, |
| "eval_loss": 0.4198272228240967, |
| "eval_runtime": 107.7664, |
| "eval_samples_per_second": 3.563, |
| "eval_steps_per_second": 1.782, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1123896173401124, |
| "grad_norm": 0.6676068902015686, |
| "learning_rate": 0.00019797361806215332, |
| "loss": 0.6849, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.11774150388011774, |
| "grad_norm": 0.6388771533966064, |
| "learning_rate": 0.000197604008616004, |
| "loss": 0.6971, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.1230933904201231, |
| "grad_norm": 0.7713803052902222, |
| "learning_rate": 0.00019720385909110198, |
| "loss": 0.7971, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.12844527696012845, |
| "grad_norm": 0.5955840945243835, |
| "learning_rate": 0.0001967732946933499, |
| "loss": 0.7388, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.1337971635001338, |
| "grad_norm": 0.7248641848564148, |
| "learning_rate": 0.00019631245014539698, |
| "loss": 0.4543, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.13914905004013914, |
| "grad_norm": 0.7671256065368652, |
| "learning_rate": 0.0001958214696444846, |
| "loss": 0.5128, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.1445009365801445, |
| "grad_norm": 0.5836634635925293, |
| "learning_rate": 0.0001953005068173272, |
| "loss": 0.4982, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.14985282312014986, |
| "grad_norm": 0.7129510045051575, |
| "learning_rate": 0.00019474972467204297, |
| "loss": 0.5622, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.1552047096601552, |
| "grad_norm": 0.9118379354476929, |
| "learning_rate": 0.00019416929554714888, |
| "loss": 0.872, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.16055659620016055, |
| "grad_norm": 0.6198474764823914, |
| "learning_rate": 0.0001935594010576362, |
| "loss": 0.5969, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.16055659620016055, |
| "eval_loss": 0.37001410126686096, |
| "eval_runtime": 107.8964, |
| "eval_samples_per_second": 3.559, |
| "eval_steps_per_second": 1.779, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.1659084827401659, |
| "grad_norm": 0.9263865351676941, |
| "learning_rate": 0.000192920232038144, |
| "loss": 0.6, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.17126036928017127, |
| "grad_norm": 0.6084879636764526, |
| "learning_rate": 0.0001922519884832469, |
| "loss": 0.6595, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.17661225582017662, |
| "grad_norm": 0.6593811511993408, |
| "learning_rate": 0.00019155487948487748, |
| "loss": 0.533, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.18196414236018196, |
| "grad_norm": 0.7938013672828674, |
| "learning_rate": 0.0001908291231669019, |
| "loss": 0.7408, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.1873160289001873, |
| "grad_norm": 0.5492228865623474, |
| "learning_rate": 0.00019007494661686935, |
| "loss": 0.4705, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.19266791544019266, |
| "grad_norm": 0.5744177103042603, |
| "learning_rate": 0.00018929258581495685, |
| "loss": 0.5089, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.19801980198019803, |
| "grad_norm": 0.6590476036071777, |
| "learning_rate": 0.0001884822855601316, |
| "loss": 0.6987, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.20337168852020338, |
| "grad_norm": 0.6895216107368469, |
| "learning_rate": 0.00018764429939355392, |
| "loss": 0.5449, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.20872357506020872, |
| "grad_norm": 0.6002170443534851, |
| "learning_rate": 0.00018677888951924474, |
| "loss": 0.7599, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.21407546160021407, |
| "grad_norm": 0.6684539318084717, |
| "learning_rate": 0.00018588632672204264, |
| "loss": 0.4909, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.21407546160021407, |
| "eval_loss": 0.35583770275115967, |
| "eval_runtime": 107.6184, |
| "eval_samples_per_second": 3.568, |
| "eval_steps_per_second": 1.784, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.2194273481402194, |
| "grad_norm": 0.623796820640564, |
| "learning_rate": 0.00018496689028287572, |
| "loss": 0.6567, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.2247792346802248, |
| "grad_norm": 0.4596005082130432, |
| "learning_rate": 0.00018402086789137546, |
| "loss": 0.4085, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.23013112122023013, |
| "grad_norm": 0.5475478768348694, |
| "learning_rate": 0.00018304855555585894, |
| "loss": 0.6919, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.23548300776023548, |
| "grad_norm": 0.5737990140914917, |
| "learning_rate": 0.00018205025751070875, |
| "loss": 0.4353, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.24083489430024083, |
| "grad_norm": 0.7474645376205444, |
| "learning_rate": 0.00018102628612117865, |
| "loss": 0.6445, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.2461867808402462, |
| "grad_norm": 0.5716281533241272, |
| "learning_rate": 0.0001799769617856552, |
| "loss": 0.5219, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.25153866738025155, |
| "grad_norm": 0.48242220282554626, |
| "learning_rate": 0.00017890261283540562, |
| "loss": 0.4461, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.2568905539202569, |
| "grad_norm": 0.5659968256950378, |
| "learning_rate": 0.00017780357543184397, |
| "loss": 0.5751, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.26224244046026224, |
| "grad_norm": 0.642313539981842, |
| "learning_rate": 0.0001766801934613467, |
| "loss": 0.487, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.2675943270002676, |
| "grad_norm": 0.6316558122634888, |
| "learning_rate": 0.00017553281842765169, |
| "loss": 0.6842, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.2675943270002676, |
| "eval_loss": 0.33880969882011414, |
| "eval_runtime": 107.9649, |
| "eval_samples_per_second": 3.557, |
| "eval_steps_per_second": 1.778, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.27294621354027293, |
| "grad_norm": 0.6853533983230591, |
| "learning_rate": 0.00017436180934187308, |
| "loss": 0.6, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.2782981000802783, |
| "grad_norm": 0.4678995907306671, |
| "learning_rate": 0.00017316753261016783, |
| "loss": 0.4881, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.2836499866202836, |
| "grad_norm": 0.438473105430603, |
| "learning_rate": 0.00017195036191908797, |
| "loss": 0.4314, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.289001873160289, |
| "grad_norm": 0.6265630125999451, |
| "learning_rate": 0.00017071067811865476, |
| "loss": 0.3728, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.29435375970029437, |
| "grad_norm": 0.42538031935691833, |
| "learning_rate": 0.00016944886910319173, |
| "loss": 0.3449, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.2997056462402997, |
| "grad_norm": 0.48485225439071655, |
| "learning_rate": 0.00016816532968995328, |
| "loss": 0.6331, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.30505753278030506, |
| "grad_norm": 0.5438908338546753, |
| "learning_rate": 0.00016686046149558736, |
| "loss": 0.5774, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.3104094193203104, |
| "grad_norm": 0.4062053859233856, |
| "learning_rate": 0.0001655346728104704, |
| "loss": 0.4822, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.31576130586031576, |
| "grad_norm": 0.6844729781150818, |
| "learning_rate": 0.0001641883784709541, |
| "loss": 0.8138, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.3211131924003211, |
| "grad_norm": 0.5358812212944031, |
| "learning_rate": 0.00016282199972956425, |
| "loss": 0.382, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.3211131924003211, |
| "eval_loss": 0.33115288615226746, |
| "eval_runtime": 107.5039, |
| "eval_samples_per_second": 3.572, |
| "eval_steps_per_second": 1.786, |
| "step": 300 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 935, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "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": 8.361963439519334e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
