| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9885057471264367, | |
| "eval_steps": 11, | |
| "global_step": 130, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022988505747126436, | |
| "eval_loss": 9.30986213684082, | |
| "eval_runtime": 9.0065, | |
| "eval_samples_per_second": 8.105, | |
| "eval_steps_per_second": 1.11, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06896551724137931, | |
| "grad_norm": 37.248802185058594, | |
| "learning_rate": 3e-05, | |
| "loss": 8.9493, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 9.639245986938477, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0587, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.20689655172413793, | |
| "grad_norm": 6.129356384277344, | |
| "learning_rate": 9e-05, | |
| "loss": 3.933, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.25287356321839083, | |
| "eval_loss": 3.160940408706665, | |
| "eval_runtime": 9.1581, | |
| "eval_samples_per_second": 7.971, | |
| "eval_steps_per_second": 1.092, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 5.822756767272949, | |
| "learning_rate": 9.99314767377287e-05, | |
| "loss": 3.2319, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "grad_norm": 4.974664688110352, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 2.9212, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 3.2585740089416504, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.2899, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4827586206896552, | |
| "grad_norm": 3.2496554851531982, | |
| "learning_rate": 9.794098674340965e-05, | |
| "loss": 2.1133, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5057471264367817, | |
| "eval_loss": 2.0404434204101562, | |
| "eval_runtime": 9.1631, | |
| "eval_samples_per_second": 7.967, | |
| "eval_steps_per_second": 1.091, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 3.3397107124328613, | |
| "learning_rate": 9.667902132486009e-05, | |
| "loss": 1.8379, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6206896551724138, | |
| "grad_norm": 4.2894768714904785, | |
| "learning_rate": 9.512926421749304e-05, | |
| "loss": 1.9301, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 3.3234455585479736, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.6536, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7586206896551724, | |
| "grad_norm": 2.444336414337158, | |
| "learning_rate": 9.120630943110077e-05, | |
| "loss": 1.4659, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7586206896551724, | |
| "eval_loss": 1.616043210029602, | |
| "eval_runtime": 9.1967, | |
| "eval_samples_per_second": 7.938, | |
| "eval_steps_per_second": 1.087, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 4.422275066375732, | |
| "learning_rate": 8.885729807284856e-05, | |
| "loss": 1.4601, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.896551724137931, | |
| "grad_norm": 3.7574503421783447, | |
| "learning_rate": 8.626871855061438e-05, | |
| "loss": 1.3037, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 3.0062453746795654, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.234, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.0114942528735633, | |
| "eval_loss": 1.342581868171692, | |
| "eval_runtime": 9.1668, | |
| "eval_samples_per_second": 7.964, | |
| "eval_steps_per_second": 1.091, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.0344827586206897, | |
| "grad_norm": 2.715128183364868, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 1.4425, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.103448275862069, | |
| "grad_norm": 2.441864490509033, | |
| "learning_rate": 7.723195175075136e-05, | |
| "loss": 0.9217, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.1724137931034484, | |
| "grad_norm": 2.2927913665771484, | |
| "learning_rate": 7.385793801298042e-05, | |
| "loss": 0.8214, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.2413793103448276, | |
| "grad_norm": 1.960526466369629, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.9555, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.264367816091954, | |
| "eval_loss": 1.2750840187072754, | |
| "eval_runtime": 9.1846, | |
| "eval_samples_per_second": 7.948, | |
| "eval_steps_per_second": 1.089, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.3103448275862069, | |
| "grad_norm": 3.3633315563201904, | |
| "learning_rate": 6.669034296168855e-05, | |
| "loss": 0.8391, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 1.9623112678527832, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.7523, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4482758620689655, | |
| "grad_norm": 2.6259655952453613, | |
| "learning_rate": 5.911177627460739e-05, | |
| "loss": 0.8822, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "grad_norm": 3.360589027404785, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 1.0438, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "eval_loss": 1.2139718532562256, | |
| "eval_runtime": 9.1834, | |
| "eval_samples_per_second": 7.949, | |
| "eval_steps_per_second": 1.089, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.5862068965517242, | |
| "grad_norm": 1.6269245147705078, | |
| "learning_rate": 5.1308847415393666e-05, | |
| "loss": 0.8215, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.6551724137931034, | |
| "grad_norm": 2.1838412284851074, | |
| "learning_rate": 4.738320218785281e-05, | |
| "loss": 0.8856, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 2.0462594032287598, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.7003, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.7701149425287355, | |
| "eval_loss": 1.1193917989730835, | |
| "eval_runtime": 9.1719, | |
| "eval_samples_per_second": 7.959, | |
| "eval_steps_per_second": 1.09, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.793103448275862, | |
| "grad_norm": 2.196936845779419, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.7238, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.8620689655172413, | |
| "grad_norm": 2.042219400405884, | |
| "learning_rate": 3.579923276480387e-05, | |
| "loss": 0.7039, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.9310344827586206, | |
| "grad_norm": 2.6126327514648438, | |
| "learning_rate": 3.2081602522734986e-05, | |
| "loss": 0.7641, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.926860332489014, | |
| "learning_rate": 2.8474445159585235e-05, | |
| "loss": 0.9279, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.0229885057471266, | |
| "eval_loss": 1.031583547592163, | |
| "eval_runtime": 9.1542, | |
| "eval_samples_per_second": 7.974, | |
| "eval_steps_per_second": 1.092, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "grad_norm": 1.4757606983184814, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.6411, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1379310344827585, | |
| "grad_norm": 1.4598504304885864, | |
| "learning_rate": 2.167968815375837e-05, | |
| "loss": 0.7081, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.206896551724138, | |
| "grad_norm": 1.0860507488250732, | |
| "learning_rate": 1.8533980447508137e-05, | |
| "loss": 0.611, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.2758620689655173, | |
| "grad_norm": 1.1976888179779053, | |
| "learning_rate": 1.5582271215312294e-05, | |
| "loss": 0.5523, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.2758620689655173, | |
| "eval_loss": 1.016935110092163, | |
| "eval_runtime": 9.1948, | |
| "eval_samples_per_second": 7.939, | |
| "eval_steps_per_second": 1.088, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.344827586206897, | |
| "grad_norm": 1.211234211921692, | |
| "learning_rate": 1.2842758726130283e-05, | |
| "loss": 0.6378, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.413793103448276, | |
| "grad_norm": 1.2312695980072021, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.5659, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.4827586206896552, | |
| "grad_norm": 1.7582778930664062, | |
| "learning_rate": 8.066471602728803e-06, | |
| "loss": 0.6258, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.528735632183908, | |
| "eval_loss": 1.0043144226074219, | |
| "eval_runtime": 9.1676, | |
| "eval_samples_per_second": 7.963, | |
| "eval_steps_per_second": 1.091, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.5517241379310347, | |
| "grad_norm": 1.0846282243728638, | |
| "learning_rate": 6.059144366901736e-06, | |
| "loss": 0.6884, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.6206896551724137, | |
| "grad_norm": 1.0979564189910889, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.6791, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.689655172413793, | |
| "grad_norm": 1.3624191284179688, | |
| "learning_rate": 2.8679254453910785e-06, | |
| "loss": 0.7076, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "grad_norm": 1.714447021484375, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.6285, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.781609195402299, | |
| "eval_loss": 0.9969158172607422, | |
| "eval_runtime": 9.1743, | |
| "eval_samples_per_second": 7.957, | |
| "eval_steps_per_second": 1.09, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 2.8275862068965516, | |
| "grad_norm": 1.9297926425933838, | |
| "learning_rate": 8.372546218022747e-07, | |
| "loss": 0.7666, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 2.896551724137931, | |
| "grad_norm": 1.666353702545166, | |
| "learning_rate": 2.7390523158633554e-07, | |
| "loss": 0.6233, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.9655172413793105, | |
| "grad_norm": 1.1451926231384277, | |
| "learning_rate": 1.7133751222137007e-08, | |
| "loss": 0.5853, | |
| "step": 129 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 11, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.176515519409029e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |