| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3979822456254807, | |
| "global_step": 60000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.999999999999999e-06, | |
| "loss": 0.9072, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.1999999999999999e-05, | |
| "loss": 0.6779, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.6512336730957031, | |
| "eval_runtime": 1.3899, | |
| "eval_samples_per_second": 1581.436, | |
| "eval_steps_per_second": 25.182, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.7999999999999997e-05, | |
| "loss": 0.6516, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.3999999999999997e-05, | |
| "loss": 0.6505, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.6479398608207703, | |
| "eval_runtime": 1.2819, | |
| "eval_samples_per_second": 1714.632, | |
| "eval_steps_per_second": 27.303, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 0.6487, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.5999999999999994e-05, | |
| "loss": 0.6475, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.6461147665977478, | |
| "eval_runtime": 1.2932, | |
| "eval_samples_per_second": 1699.633, | |
| "eval_steps_per_second": 27.064, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6469, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 0.6464, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 0.6451290249824524, | |
| "eval_runtime": 1.2355, | |
| "eval_samples_per_second": 1779.085, | |
| "eval_steps_per_second": 28.329, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 5.399999999999999e-05, | |
| "loss": 0.6459, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.6455, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.6444417238235474, | |
| "eval_runtime": 1.2972, | |
| "eval_samples_per_second": 1694.435, | |
| "eval_steps_per_second": 26.981, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 6.599999999999999e-05, | |
| "loss": 0.6452, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 7.199999999999999e-05, | |
| "loss": 0.6448, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 0.6438854336738586, | |
| "eval_runtime": 1.2871, | |
| "eval_samples_per_second": 1707.767, | |
| "eval_steps_per_second": 27.194, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 7.8e-05, | |
| "loss": 0.64, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.6222, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.6123253703117371, | |
| "eval_runtime": 1.2501, | |
| "eval_samples_per_second": 1758.21, | |
| "eval_steps_per_second": 27.997, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 0.6115, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.599999999999999e-05, | |
| "loss": 0.6019, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.5872684717178345, | |
| "eval_runtime": 1.2778, | |
| "eval_samples_per_second": 1720.181, | |
| "eval_steps_per_second": 27.391, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.000102, | |
| "loss": 0.5924, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00010799999999999998, | |
| "loss": 0.5838, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.5689075589179993, | |
| "eval_runtime": 1.2743, | |
| "eval_samples_per_second": 1724.838, | |
| "eval_steps_per_second": 27.466, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 0.5737, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.5612, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.5473007559776306, | |
| "eval_runtime": 1.2648, | |
| "eval_samples_per_second": 1737.882, | |
| "eval_steps_per_second": 27.673, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00012599999999999997, | |
| "loss": 0.5516, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00013199999999999998, | |
| "loss": 0.5428, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 0.5300613045692444, | |
| "eval_runtime": 1.2658, | |
| "eval_samples_per_second": 1736.485, | |
| "eval_steps_per_second": 27.651, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.000138, | |
| "loss": 0.5346, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00014399999999999998, | |
| "loss": 0.527, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.5140743255615234, | |
| "eval_runtime": 1.2609, | |
| "eval_samples_per_second": 1743.192, | |
| "eval_steps_per_second": 27.758, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00015, | |
| "loss": 0.5203, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.000156, | |
| "loss": 0.5142, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.5022566318511963, | |
| "eval_runtime": 1.2481, | |
| "eval_samples_per_second": 1761.066, | |
| "eval_steps_per_second": 28.042, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000162, | |
| "loss": 0.5079, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.000168, | |
| "loss": 0.5002, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.4841987192630768, | |
| "eval_runtime": 1.3176, | |
| "eval_samples_per_second": 1668.178, | |
| "eval_steps_per_second": 26.563, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00017399999999999997, | |
| "loss": 0.4916, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.4852, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.47066375613212585, | |
| "eval_runtime": 1.3026, | |
| "eval_samples_per_second": 1687.422, | |
| "eval_steps_per_second": 26.87, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.000186, | |
| "loss": 0.4778, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00019199999999999998, | |
| "loss": 0.4697, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.4562186300754547, | |
| "eval_runtime": 1.2895, | |
| "eval_samples_per_second": 1704.503, | |
| "eval_steps_per_second": 27.142, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.000198, | |
| "loss": 0.4636, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000204, | |
| "loss": 0.459, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.44295796751976013, | |
| "eval_runtime": 1.2516, | |
| "eval_samples_per_second": 1756.093, | |
| "eval_steps_per_second": 27.963, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.4544, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00021599999999999996, | |
| "loss": 0.45, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.4343169629573822, | |
| "eval_runtime": 1.327, | |
| "eval_samples_per_second": 1656.351, | |
| "eval_steps_per_second": 26.375, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00022199999999999998, | |
| "loss": 0.4458, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00022799999999999999, | |
| "loss": 0.4411, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.42496559023857117, | |
| "eval_runtime": 1.3285, | |
| "eval_samples_per_second": 1654.492, | |
| "eval_steps_per_second": 26.345, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000234, | |
| "loss": 0.4368, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.4322, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.4132905900478363, | |
| "eval_runtime": 1.3158, | |
| "eval_samples_per_second": 1670.425, | |
| "eval_steps_per_second": 26.599, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00024599999999999996, | |
| "loss": 0.4278, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00025199999999999995, | |
| "loss": 0.4233, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.4032529294490814, | |
| "eval_runtime": 1.2842, | |
| "eval_samples_per_second": 1711.577, | |
| "eval_steps_per_second": 27.254, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000258, | |
| "loss": 0.4189, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00026399999999999997, | |
| "loss": 0.4146, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.39436739683151245, | |
| "eval_runtime": 1.2633, | |
| "eval_samples_per_second": 1739.901, | |
| "eval_steps_per_second": 27.705, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00027, | |
| "loss": 0.4109, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.000276, | |
| "loss": 0.4077, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.38647642731666565, | |
| "eval_runtime": 1.2503, | |
| "eval_samples_per_second": 1758.021, | |
| "eval_steps_per_second": 27.994, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00028199999999999997, | |
| "loss": 0.4047, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00028799999999999995, | |
| "loss": 0.4019, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.381785124540329, | |
| "eval_runtime": 1.2485, | |
| "eval_samples_per_second": 1760.518, | |
| "eval_steps_per_second": 28.034, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.000294, | |
| "loss": 0.3994, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3968, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 0.3785216510295868, | |
| "eval_runtime": 1.2913, | |
| "eval_samples_per_second": 1702.153, | |
| "eval_steps_per_second": 27.104, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00029999920715161553, | |
| "loss": 0.3945, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0002999968286151326, | |
| "loss": 0.392, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.37318962812423706, | |
| "eval_runtime": 1.2758, | |
| "eval_samples_per_second": 1722.782, | |
| "eval_steps_per_second": 27.433, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002999928644165624, | |
| "loss": 0.3897, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0002999873145992569, | |
| "loss": 0.3869, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.3669659197330475, | |
| "eval_runtime": 1.2869, | |
| "eval_samples_per_second": 1707.967, | |
| "eval_steps_per_second": 27.197, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.000299980179223908, | |
| "loss": 0.385, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002999714583685469, | |
| "loss": 0.3828, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.3615981340408325, | |
| "eval_runtime": 1.2961, | |
| "eval_samples_per_second": 1695.914, | |
| "eval_steps_per_second": 27.005, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00029996115212854366, | |
| "loss": 0.3808, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00029994926061660554, | |
| "loss": 0.3786, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.3613170087337494, | |
| "eval_runtime": 1.3013, | |
| "eval_samples_per_second": 1689.114, | |
| "eval_steps_per_second": 26.897, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0002999357839627762, | |
| "loss": 0.377, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00029992072231443425, | |
| "loss": 0.3751, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.35551005601882935, | |
| "eval_runtime": 1.2661, | |
| "eval_samples_per_second": 1736.053, | |
| "eval_steps_per_second": 27.644, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002999040758362914, | |
| "loss": 0.3731, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00029988584471039094, | |
| "loss": 0.3713, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.35287660360336304, | |
| "eval_runtime": 1.3423, | |
| "eval_samples_per_second": 1637.545, | |
| "eval_steps_per_second": 26.076, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002998660291361054, | |
| "loss": 0.3699, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0002998446293301349, | |
| "loss": 0.3681, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.35174888372421265, | |
| "eval_runtime": 1.3006, | |
| "eval_samples_per_second": 1690.029, | |
| "eval_steps_per_second": 26.911, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0002998216455265042, | |
| "loss": 0.3667, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00029979707797656046, | |
| "loss": 0.3656, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_loss": 0.34766125679016113, | |
| "eval_runtime": 1.3363, | |
| "eval_samples_per_second": 1644.872, | |
| "eval_steps_per_second": 26.192, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00029977092694897053, | |
| "loss": 0.3641, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0002997431927297178, | |
| "loss": 0.3626, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 0.3423311412334442, | |
| "eval_runtime": 1.3208, | |
| "eval_samples_per_second": 1664.116, | |
| "eval_steps_per_second": 26.499, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00029971387562209936, | |
| "loss": 0.3616, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00029968297594672226, | |
| "loss": 0.3601, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.34427037835121155, | |
| "eval_runtime": 1.2929, | |
| "eval_samples_per_second": 1700.099, | |
| "eval_steps_per_second": 27.072, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0002996504940415005, | |
| "loss": 0.3589, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00029961643026165096, | |
| "loss": 0.3577, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.34002143144607544, | |
| "eval_runtime": 1.2974, | |
| "eval_samples_per_second": 1694.186, | |
| "eval_steps_per_second": 26.977, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00029958078497968973, | |
| "loss": 0.3568, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002995435585854278, | |
| "loss": 0.3555, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.3368704915046692, | |
| "eval_runtime": 1.3166, | |
| "eval_samples_per_second": 1669.417, | |
| "eval_steps_per_second": 26.583, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002995047514859671, | |
| "loss": 0.3542, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002994643641056959, | |
| "loss": 0.3532, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.3357352018356323, | |
| "eval_runtime": 1.2847, | |
| "eval_samples_per_second": 1710.953, | |
| "eval_steps_per_second": 27.244, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.000299422396886284, | |
| "loss": 0.3523, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002993788502866783, | |
| "loss": 0.3514, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 0.33294928073883057, | |
| "eval_runtime": 1.272, | |
| "eval_samples_per_second": 1728.032, | |
| "eval_steps_per_second": 27.516, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00029933372478309746, | |
| "loss": 0.3503, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00029928702086902664, | |
| "loss": 0.3493, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.332314670085907, | |
| "eval_runtime": 1.318, | |
| "eval_samples_per_second": 1667.699, | |
| "eval_steps_per_second": 26.556, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00029923873905521244, | |
| "loss": 0.3483, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.000299188879869657, | |
| "loss": 0.3477, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.33138296008110046, | |
| "eval_runtime": 1.2875, | |
| "eval_samples_per_second": 1707.216, | |
| "eval_steps_per_second": 27.185, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00029913744385761244, | |
| "loss": 0.3466, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00029908443158157465, | |
| "loss": 0.3459, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 0.32563212513923645, | |
| "eval_runtime": 1.2983, | |
| "eval_samples_per_second": 1692.964, | |
| "eval_steps_per_second": 26.958, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.0002990298436212775, | |
| "loss": 0.3448, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002989736805736861, | |
| "loss": 0.3441, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.32700827717781067, | |
| "eval_runtime": 1.3967, | |
| "eval_samples_per_second": 1573.716, | |
| "eval_steps_per_second": 25.059, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00029891594305299065, | |
| "loss": 0.3432, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00029885663169059926, | |
| "loss": 0.3424, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.3216361701488495, | |
| "eval_runtime": 1.4298, | |
| "eval_samples_per_second": 1537.25, | |
| "eval_steps_per_second": 24.479, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0002987957471351316, | |
| "loss": 0.3415, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.00029873329005241137, | |
| "loss": 0.3408, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.324531614780426, | |
| "eval_runtime": 1.3974, | |
| "eval_samples_per_second": 1572.95, | |
| "eval_steps_per_second": 25.047, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00029866926112545925, | |
| "loss": 0.3399, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00029860366105448534, | |
| "loss": 0.3394, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.3231622278690338, | |
| "eval_runtime": 1.3921, | |
| "eval_samples_per_second": 1578.889, | |
| "eval_steps_per_second": 25.142, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00029853649055688143, | |
| "loss": 0.3387, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00029846775036721337, | |
| "loss": 0.338, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.32254886627197266, | |
| "eval_runtime": 1.4182, | |
| "eval_samples_per_second": 1549.86, | |
| "eval_steps_per_second": 24.679, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.0002983974412372129, | |
| "loss": 0.337, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029832556393576934, | |
| "loss": 0.3363, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.3195147216320038, | |
| "eval_runtime": 1.408, | |
| "eval_samples_per_second": 1561.07, | |
| "eval_steps_per_second": 24.858, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0002982521192489214, | |
| "loss": 0.3357, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002981771079798483, | |
| "loss": 0.3353, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.3167419731616974, | |
| "eval_runtime": 1.4463, | |
| "eval_samples_per_second": 1519.733, | |
| "eval_steps_per_second": 24.2, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00029810053094886136, | |
| "loss": 0.3345, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00029802238899339473, | |
| "loss": 0.3338, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.3171909749507904, | |
| "eval_runtime": 1.4105, | |
| "eval_samples_per_second": 1558.283, | |
| "eval_steps_per_second": 24.813, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0002979426829679962, | |
| "loss": 0.3334, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0002978614137443183, | |
| "loss": 0.3325, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 0.31682637333869934, | |
| "eval_runtime": 1.4213, | |
| "eval_samples_per_second": 1546.472, | |
| "eval_steps_per_second": 24.625, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.000297778582211108, | |
| "loss": 0.3322, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00029769418927419786, | |
| "loss": 0.3316, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.31410133838653564, | |
| "eval_runtime": 1.4261, | |
| "eval_samples_per_second": 1541.269, | |
| "eval_steps_per_second": 24.542, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0002976082358564954, | |
| "loss": 0.331, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00029752072289797353, | |
| "loss": 0.3305, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.3114263713359833, | |
| "eval_runtime": 1.4182, | |
| "eval_samples_per_second": 1549.888, | |
| "eval_steps_per_second": 24.68, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00029743165135565986, | |
| "loss": 0.3301, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00029734102220362654, | |
| "loss": 0.3303, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 0.3110273778438568, | |
| "eval_runtime": 1.3899, | |
| "eval_samples_per_second": 1581.462, | |
| "eval_steps_per_second": 25.183, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00029724883643297937, | |
| "loss": 0.3289, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0002971550950518473, | |
| "loss": 0.3284, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.3112471401691437, | |
| "eval_runtime": 1.3713, | |
| "eval_samples_per_second": 1602.906, | |
| "eval_steps_per_second": 25.524, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.000297059799085371, | |
| "loss": 0.3284, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00029696294957569196, | |
| "loss": 0.3276, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.30989283323287964, | |
| "eval_runtime": 1.4309, | |
| "eval_samples_per_second": 1536.096, | |
| "eval_steps_per_second": 24.46, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00029686454758194076, | |
| "loss": 0.3269, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00029676459418022594, | |
| "loss": 0.3264, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.3089369833469391, | |
| "eval_runtime": 1.4226, | |
| "eval_samples_per_second": 1545.068, | |
| "eval_steps_per_second": 24.603, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0002966630904636219, | |
| "loss": 0.3261, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.0002965600375421569, | |
| "loss": 0.3256, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 0.31043851375579834, | |
| "eval_runtime": 1.4043, | |
| "eval_samples_per_second": 1565.245, | |
| "eval_steps_per_second": 24.924, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0002964554365428013, | |
| "loss": 0.3251, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.00029634928860945486, | |
| "loss": 0.3247, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.3088381588459015, | |
| "eval_runtime": 1.4281, | |
| "eval_samples_per_second": 1539.081, | |
| "eval_steps_per_second": 24.508, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0002962415949029343, | |
| "loss": 0.3243, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00029613235660096084, | |
| "loss": 0.3241, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.3055568337440491, | |
| "eval_runtime": 1.4302, | |
| "eval_samples_per_second": 1536.88, | |
| "eval_steps_per_second": 24.473, | |
| "step": 60000 | |
| } | |
| ], | |
| "max_steps": 500000, | |
| "num_train_epochs": 12, | |
| "total_flos": 1.9169203356560932e+21, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |