| [ | |
| { | |
| "step": 30, | |
| "loss": 1.4062, | |
| "type": "train", | |
| "grad_norm": 8.482284545898438, | |
| "learning_rate": 1.8347107438016528e-06, | |
| "epoch": 0.2475502836513667 | |
| }, | |
| { | |
| "step": 30, | |
| "loss": 1.1563361883163452, | |
| "type": "eval", | |
| "eval_runtime": 238.6436, | |
| "eval_samples_per_second": 2.891, | |
| "eval_steps_per_second": 2.891, | |
| "epoch": 0.2475502836513667 | |
| }, | |
| { | |
| "step": 60, | |
| "loss": 1.1662, | |
| "type": "train", | |
| "grad_norm": 6.751504898071289, | |
| "learning_rate": 1.6694214876033058e-06, | |
| "epoch": 0.4951005673027334 | |
| }, | |
| { | |
| "step": 60, | |
| "loss": 1.0914983749389648, | |
| "type": "eval", | |
| "eval_runtime": 248.6513, | |
| "eval_samples_per_second": 2.775, | |
| "eval_steps_per_second": 2.775, | |
| "epoch": 0.4951005673027334 | |
| }, | |
| { | |
| "step": 90, | |
| "loss": 1.1109, | |
| "type": "train", | |
| "grad_norm": 4.135236740112305, | |
| "learning_rate": 1.5041322314049587e-06, | |
| "epoch": 0.7426508509541001 | |
| }, | |
| { | |
| "step": 90, | |
| "loss": 1.0496803522109985, | |
| "type": "eval", | |
| "eval_runtime": 239.5109, | |
| "eval_samples_per_second": 2.881, | |
| "eval_steps_per_second": 2.881, | |
| "epoch": 0.7426508509541001 | |
| }, | |
| { | |
| "step": 120, | |
| "loss": 1.0639, | |
| "type": "train", | |
| "grad_norm": 5.695450782775879, | |
| "learning_rate": 1.3388429752066116e-06, | |
| "epoch": 0.9902011346054668 | |
| }, | |
| { | |
| "step": 120, | |
| "loss": 1.0101323127746582, | |
| "type": "eval", | |
| "eval_runtime": 234.2087, | |
| "eval_samples_per_second": 2.946, | |
| "eval_steps_per_second": 2.946, | |
| "epoch": 0.9902011346054668 | |
| }, | |
| { | |
| "step": 150, | |
| "loss": 0.9416, | |
| "type": "train", | |
| "grad_norm": 7.571918487548828, | |
| "learning_rate": 1.1735537190082645e-06, | |
| "epoch": 1.2377514182568334 | |
| }, | |
| { | |
| "step": 150, | |
| "loss": 1.0040597915649414, | |
| "type": "eval", | |
| "eval_runtime": 247.1746, | |
| "eval_samples_per_second": 2.792, | |
| "eval_steps_per_second": 2.792, | |
| "epoch": 1.2377514182568334 | |
| }, | |
| { | |
| "step": 180, | |
| "loss": 0.925, | |
| "type": "train", | |
| "grad_norm": 4.53548002243042, | |
| "learning_rate": 1.0082644628099172e-06, | |
| "epoch": 1.4853017019082002 | |
| }, | |
| { | |
| "step": 180, | |
| "loss": 0.9938931465148926, | |
| "type": "eval", | |
| "eval_runtime": 234.8252, | |
| "eval_samples_per_second": 2.938, | |
| "eval_steps_per_second": 2.938, | |
| "epoch": 1.4853017019082002 | |
| }, | |
| { | |
| "step": 210, | |
| "loss": 0.9033, | |
| "type": "train", | |
| "grad_norm": 5.361794948577881, | |
| "learning_rate": 8.429752066115701e-07, | |
| "epoch": 1.7328519855595668 | |
| }, | |
| { | |
| "step": 210, | |
| "loss": 0.9868502020835876, | |
| "type": "eval", | |
| "eval_runtime": 234.8772, | |
| "eval_samples_per_second": 2.938, | |
| "eval_steps_per_second": 2.938, | |
| "epoch": 1.7328519855595668 | |
| }, | |
| { | |
| "step": 240, | |
| "loss": 0.9087, | |
| "type": "train", | |
| "grad_norm": 4.525313377380371, | |
| "learning_rate": 6.776859504132231e-07, | |
| "epoch": 1.9804022692109333 | |
| }, | |
| { | |
| "step": 240, | |
| "loss": 0.9824326634407043, | |
| "type": "eval", | |
| "eval_runtime": 234.8918, | |
| "eval_samples_per_second": 2.938, | |
| "eval_steps_per_second": 2.938, | |
| "epoch": 1.9804022692109333 | |
| }, | |
| { | |
| "step": 270, | |
| "loss": 0.8166, | |
| "type": "train", | |
| "grad_norm": 4.654973030090332, | |
| "learning_rate": 5.12396694214876e-07, | |
| "epoch": 2.2279525528623 | |
| }, | |
| { | |
| "step": 270, | |
| "loss": 0.9874295592308044, | |
| "type": "eval", | |
| "eval_runtime": 243.7953, | |
| "eval_samples_per_second": 2.83, | |
| "eval_steps_per_second": 2.83, | |
| "epoch": 2.2279525528623 | |
| }, | |
| { | |
| "step": 300, | |
| "loss": 0.8226, | |
| "type": "train", | |
| "grad_norm": 5.9346442222595215, | |
| "learning_rate": 3.471074380165289e-07, | |
| "epoch": 2.475502836513667 | |
| }, | |
| { | |
| "step": 300, | |
| "loss": 0.9854046106338501, | |
| "type": "eval", | |
| "eval_runtime": 243.3847, | |
| "eval_samples_per_second": 2.835, | |
| "eval_steps_per_second": 2.835, | |
| "epoch": 2.475502836513667 | |
| }, | |
| { | |
| "step": 330, | |
| "loss": 0.8289, | |
| "type": "train", | |
| "grad_norm": 4.637845516204834, | |
| "learning_rate": 1.818181818181818e-07, | |
| "epoch": 2.7230531201650336 | |
| }, | |
| { | |
| "step": 330, | |
| "loss": 0.9841367602348328, | |
| "type": "eval", | |
| "eval_runtime": 242.6797, | |
| "eval_samples_per_second": 2.843, | |
| "eval_steps_per_second": 2.843, | |
| "epoch": 2.7230531201650336 | |
| }, | |
| { | |
| "step": 360, | |
| "loss": 0.8137, | |
| "type": "train", | |
| "grad_norm": 4.132049560546875, | |
| "learning_rate": 1.652892561983471e-08, | |
| "epoch": 2.9706034038164004 | |
| }, | |
| { | |
| "step": 360, | |
| "loss": 0.9834251403808594, | |
| "type": "eval", | |
| "eval_runtime": 242.7445, | |
| "eval_samples_per_second": 2.842, | |
| "eval_steps_per_second": 2.842, | |
| "epoch": 2.9706034038164004 | |
| }, | |
| { | |
| "step": 363, | |
| "train_runtime": 22924.4691, | |
| "train_samples_per_second": 0.507, | |
| "train_steps_per_second": 0.016, | |
| "total_flos": 7877706776576.0, | |
| "train_loss": 0.9745982139892158, | |
| "epoch": 2.9953584321815367 | |
| } | |
| ] |