| { | |
| "best_metric": 2.9399616718292236, | |
| "best_model_checkpoint": "./results/models/checkpoint-43032", | |
| "epoch": 44.0, | |
| "eval_steps": 500, | |
| "global_step": 43032, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5112474437627812, | |
| "grad_norm": 0.056640625, | |
| "learning_rate": 0.001979550102249489, | |
| "loss": 3.8421, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.714660882949829, | |
| "eval_runtime": 2.4299, | |
| "eval_samples_per_second": 205.767, | |
| "eval_steps_per_second": 0.412, | |
| "step": 978 | |
| }, | |
| { | |
| "epoch": 1.0224948875255624, | |
| "grad_norm": 0.0849609375, | |
| "learning_rate": 0.0019591002044989777, | |
| "loss": 3.7427, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5337423312883436, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 0.0019386503067484661, | |
| "loss": 3.6832, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.5354397296905518, | |
| "eval_runtime": 2.3994, | |
| "eval_samples_per_second": 208.386, | |
| "eval_steps_per_second": 0.417, | |
| "step": 1956 | |
| }, | |
| { | |
| "epoch": 2.044989775051125, | |
| "grad_norm": 0.197265625, | |
| "learning_rate": 0.001918200408997955, | |
| "loss": 3.5848, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.556237218813906, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.0018977505112474438, | |
| "loss": 3.474, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.3599166870117188, | |
| "eval_runtime": 2.2955, | |
| "eval_samples_per_second": 217.822, | |
| "eval_steps_per_second": 0.436, | |
| "step": 2934 | |
| }, | |
| { | |
| "epoch": 3.067484662576687, | |
| "grad_norm": 0.2216796875, | |
| "learning_rate": 0.0018773006134969327, | |
| "loss": 3.3901, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.5787321063394684, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.0018568507157464215, | |
| "loss": 3.3209, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.2504658699035645, | |
| "eval_runtime": 2.7262, | |
| "eval_samples_per_second": 183.409, | |
| "eval_steps_per_second": 0.367, | |
| "step": 3912 | |
| }, | |
| { | |
| "epoch": 4.08997955010225, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.00183640081799591, | |
| "loss": 3.2742, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.601226993865031, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 0.0018159509202453987, | |
| "loss": 3.2314, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.1896719932556152, | |
| "eval_runtime": 2.6127, | |
| "eval_samples_per_second": 191.371, | |
| "eval_steps_per_second": 0.383, | |
| "step": 4890 | |
| }, | |
| { | |
| "epoch": 5.112474437627812, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0017955010224948876, | |
| "loss": 3.2012, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.623721881390593, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.0017750511247443764, | |
| "loss": 3.1715, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.143582820892334, | |
| "eval_runtime": 2.3471, | |
| "eval_samples_per_second": 213.027, | |
| "eval_steps_per_second": 0.426, | |
| "step": 5868 | |
| }, | |
| { | |
| "epoch": 6.134969325153374, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.001754601226993865, | |
| "loss": 3.1533, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.6462167689161555, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0017341513292433537, | |
| "loss": 3.1316, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 3.1050572395324707, | |
| "eval_runtime": 2.7309, | |
| "eval_samples_per_second": 183.09, | |
| "eval_steps_per_second": 0.366, | |
| "step": 6846 | |
| }, | |
| { | |
| "epoch": 7.157464212678937, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.0017137014314928425, | |
| "loss": 3.1166, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.668711656441718, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 0.0016932515337423314, | |
| "loss": 3.1031, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.0762298107147217, | |
| "eval_runtime": 3.005, | |
| "eval_samples_per_second": 166.388, | |
| "eval_steps_per_second": 0.333, | |
| "step": 7824 | |
| }, | |
| { | |
| "epoch": 8.1799591002045, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.0016728016359918202, | |
| "loss": 3.0867, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.69120654396728, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0016523517382413088, | |
| "loss": 3.0785, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.063835382461548, | |
| "eval_runtime": 0.9235, | |
| "eval_samples_per_second": 541.411, | |
| "eval_steps_per_second": 1.083, | |
| "step": 8802 | |
| }, | |
| { | |
| "epoch": 9.202453987730062, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.0016319018404907975, | |
| "loss": 3.0683, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.713701431492842, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 0.0016114519427402863, | |
| "loss": 3.0582, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.043813943862915, | |
| "eval_runtime": 0.9286, | |
| "eval_samples_per_second": 538.419, | |
| "eval_steps_per_second": 1.077, | |
| "step": 9780 | |
| }, | |
| { | |
| "epoch": 10.224948875255624, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.0015910020449897751, | |
| "loss": 3.05, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.736196319018404, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 0.0015705521472392638, | |
| "loss": 3.0436, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.0265145301818848, | |
| "eval_runtime": 0.8628, | |
| "eval_samples_per_second": 579.525, | |
| "eval_steps_per_second": 1.159, | |
| "step": 10758 | |
| }, | |
| { | |
| "epoch": 11.247443762781186, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0015501022494887526, | |
| "loss": 3.0351, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.758691206543967, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.0015296523517382412, | |
| "loss": 3.0337, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.0209317207336426, | |
| "eval_runtime": 1.0057, | |
| "eval_samples_per_second": 497.152, | |
| "eval_steps_per_second": 0.994, | |
| "step": 11736 | |
| }, | |
| { | |
| "epoch": 12.269938650306749, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00150920245398773, | |
| "loss": 3.0209, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.781186094069529, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.001488752556237219, | |
| "loss": 3.0193, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.0119516849517822, | |
| "eval_runtime": 0.9281, | |
| "eval_samples_per_second": 538.717, | |
| "eval_steps_per_second": 1.077, | |
| "step": 12714 | |
| }, | |
| { | |
| "epoch": 13.292433537832311, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.0014683026584867075, | |
| "loss": 3.0169, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.803680981595091, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.0014478527607361964, | |
| "loss": 3.0083, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 3.005497455596924, | |
| "eval_runtime": 0.9352, | |
| "eval_samples_per_second": 534.66, | |
| "eval_steps_per_second": 1.069, | |
| "step": 13692 | |
| }, | |
| { | |
| "epoch": 14.314928425357873, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.001427402862985685, | |
| "loss": 3.0039, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 14.826175869120654, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.0014069529652351738, | |
| "loss": 3.0026, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 2.997267007827759, | |
| "eval_runtime": 0.8134, | |
| "eval_samples_per_second": 614.694, | |
| "eval_steps_per_second": 1.229, | |
| "step": 14670 | |
| }, | |
| { | |
| "epoch": 15.337423312883436, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0013865030674846627, | |
| "loss": 2.9966, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 15.848670756646216, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0013660531697341513, | |
| "loss": 2.9944, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 2.986419916152954, | |
| "eval_runtime": 1.0985, | |
| "eval_samples_per_second": 455.159, | |
| "eval_steps_per_second": 0.91, | |
| "step": 15648 | |
| }, | |
| { | |
| "epoch": 16.359918200409, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.0013456032719836402, | |
| "loss": 2.9894, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 16.87116564417178, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.0013251533742331288, | |
| "loss": 2.9893, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 2.9802663326263428, | |
| "eval_runtime": 0.9902, | |
| "eval_samples_per_second": 504.964, | |
| "eval_steps_per_second": 1.01, | |
| "step": 16626 | |
| }, | |
| { | |
| "epoch": 17.38241308793456, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0013047034764826176, | |
| "loss": 2.9857, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 17.893660531697343, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0012842535787321062, | |
| "loss": 2.9819, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.978501796722412, | |
| "eval_runtime": 0.9842, | |
| "eval_samples_per_second": 508.013, | |
| "eval_steps_per_second": 1.016, | |
| "step": 17604 | |
| }, | |
| { | |
| "epoch": 18.404907975460123, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.001263803680981595, | |
| "loss": 2.9785, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 18.916155419222903, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.001243353783231084, | |
| "loss": 2.9779, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.9747886657714844, | |
| "eval_runtime": 1.1337, | |
| "eval_samples_per_second": 441.025, | |
| "eval_steps_per_second": 0.882, | |
| "step": 18582 | |
| }, | |
| { | |
| "epoch": 19.427402862985684, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.0012229038854805726, | |
| "loss": 2.9719, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 19.938650306748468, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0012024539877300614, | |
| "loss": 2.9763, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.9712185859680176, | |
| "eval_runtime": 0.8388, | |
| "eval_samples_per_second": 596.063, | |
| "eval_steps_per_second": 1.192, | |
| "step": 19560 | |
| }, | |
| { | |
| "epoch": 20.449897750511248, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.00118200408997955, | |
| "loss": 2.9675, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 20.961145194274028, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.0011615541922290389, | |
| "loss": 2.9715, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 2.9683492183685303, | |
| "eval_runtime": 1.2443, | |
| "eval_samples_per_second": 401.838, | |
| "eval_steps_per_second": 0.804, | |
| "step": 20538 | |
| }, | |
| { | |
| "epoch": 21.47239263803681, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0011411042944785277, | |
| "loss": 2.9674, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 21.983640081799592, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.0011206543967280163, | |
| "loss": 2.9658, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 2.96321177482605, | |
| "eval_runtime": 1.0239, | |
| "eval_samples_per_second": 488.314, | |
| "eval_steps_per_second": 0.977, | |
| "step": 21516 | |
| }, | |
| { | |
| "epoch": 22.494887525562373, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.0011002044989775052, | |
| "loss": 2.9617, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 2.962963342666626, | |
| "eval_runtime": 1.176, | |
| "eval_samples_per_second": 425.172, | |
| "eval_steps_per_second": 0.85, | |
| "step": 22494 | |
| }, | |
| { | |
| "epoch": 23.006134969325153, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.0010797546012269938, | |
| "loss": 2.9646, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 23.517382413087933, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0010593047034764826, | |
| "loss": 2.959, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 2.9584052562713623, | |
| "eval_runtime": 0.8936, | |
| "eval_samples_per_second": 559.54, | |
| "eval_steps_per_second": 1.119, | |
| "step": 23472 | |
| }, | |
| { | |
| "epoch": 24.028629856850717, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0010388548057259715, | |
| "loss": 2.9606, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 24.539877300613497, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.00101840490797546, | |
| "loss": 2.9591, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.9545176029205322, | |
| "eval_runtime": 0.9712, | |
| "eval_samples_per_second": 514.829, | |
| "eval_steps_per_second": 1.03, | |
| "step": 24450 | |
| }, | |
| { | |
| "epoch": 25.051124744376278, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.000997955010224949, | |
| "loss": 2.9567, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 25.562372188139058, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.0009775051124744376, | |
| "loss": 2.9566, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 2.9551963806152344, | |
| "eval_runtime": 0.8333, | |
| "eval_samples_per_second": 600.037, | |
| "eval_steps_per_second": 1.2, | |
| "step": 25428 | |
| }, | |
| { | |
| "epoch": 26.073619631901842, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.0009570552147239264, | |
| "loss": 2.9551, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 26.584867075664622, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.0009366053169734151, | |
| "loss": 2.9523, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 2.957306146621704, | |
| "eval_runtime": 0.7301, | |
| "eval_samples_per_second": 684.842, | |
| "eval_steps_per_second": 1.37, | |
| "step": 26406 | |
| }, | |
| { | |
| "epoch": 27.096114519427402, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0009161554192229039, | |
| "loss": 2.9531, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 27.607361963190183, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0008957055214723927, | |
| "loss": 2.9524, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 2.9476640224456787, | |
| "eval_runtime": 0.8281, | |
| "eval_samples_per_second": 603.779, | |
| "eval_steps_per_second": 1.208, | |
| "step": 27384 | |
| }, | |
| { | |
| "epoch": 28.118609406952967, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0008752556237218813, | |
| "loss": 2.9493, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 28.629856850715747, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.0008548057259713702, | |
| "loss": 2.9487, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 2.9507575035095215, | |
| "eval_runtime": 0.9351, | |
| "eval_samples_per_second": 534.698, | |
| "eval_steps_per_second": 1.069, | |
| "step": 28362 | |
| }, | |
| { | |
| "epoch": 29.141104294478527, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0008343558282208589, | |
| "loss": 2.95, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 29.652351738241308, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0008139059304703477, | |
| "loss": 2.9467, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 2.9493441581726074, | |
| "eval_runtime": 0.9913, | |
| "eval_samples_per_second": 504.379, | |
| "eval_steps_per_second": 1.009, | |
| "step": 29340 | |
| }, | |
| { | |
| "epoch": 30.16359918200409, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.0007934560327198365, | |
| "loss": 2.9481, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 30.67484662576687, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0007730061349693251, | |
| "loss": 2.945, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 2.9469032287597656, | |
| "eval_runtime": 0.7298, | |
| "eval_samples_per_second": 685.073, | |
| "eval_steps_per_second": 1.37, | |
| "step": 30318 | |
| }, | |
| { | |
| "epoch": 31.186094069529652, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.000752556237218814, | |
| "loss": 2.9452, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 31.697341513292432, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.0007321063394683026, | |
| "loss": 2.9431, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 2.947190046310425, | |
| "eval_runtime": 0.8045, | |
| "eval_samples_per_second": 621.476, | |
| "eval_steps_per_second": 1.243, | |
| "step": 31296 | |
| }, | |
| { | |
| "epoch": 32.20858895705521, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.0007116564417177914, | |
| "loss": 2.9454, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 32.719836400818, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0006912065439672803, | |
| "loss": 2.9428, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 2.9455933570861816, | |
| "eval_runtime": 0.7507, | |
| "eval_samples_per_second": 666.034, | |
| "eval_steps_per_second": 1.332, | |
| "step": 32274 | |
| }, | |
| { | |
| "epoch": 33.23108384458078, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.0006707566462167689, | |
| "loss": 2.9424, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 33.74233128834356, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0006503067484662577, | |
| "loss": 2.9426, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 2.945394992828369, | |
| "eval_runtime": 0.91, | |
| "eval_samples_per_second": 549.459, | |
| "eval_steps_per_second": 1.099, | |
| "step": 33252 | |
| }, | |
| { | |
| "epoch": 34.25357873210634, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.0006298568507157464, | |
| "loss": 2.94, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 34.76482617586912, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.0006094069529652352, | |
| "loss": 2.9438, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 2.9434268474578857, | |
| "eval_runtime": 0.9642, | |
| "eval_samples_per_second": 518.581, | |
| "eval_steps_per_second": 1.037, | |
| "step": 34230 | |
| }, | |
| { | |
| "epoch": 35.2760736196319, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0005889570552147239, | |
| "loss": 2.939, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 35.787321063394685, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.0005685071574642127, | |
| "loss": 2.9407, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 2.9426307678222656, | |
| "eval_runtime": 0.9434, | |
| "eval_samples_per_second": 530.024, | |
| "eval_steps_per_second": 1.06, | |
| "step": 35208 | |
| }, | |
| { | |
| "epoch": 36.29856850715746, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.0005480572597137015, | |
| "loss": 2.9372, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 36.809815950920246, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0005276073619631901, | |
| "loss": 2.941, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 2.9429283142089844, | |
| "eval_runtime": 0.9172, | |
| "eval_samples_per_second": 545.157, | |
| "eval_steps_per_second": 1.09, | |
| "step": 36186 | |
| }, | |
| { | |
| "epoch": 37.32106339468303, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.000507157464212679, | |
| "loss": 2.9368, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 37.83231083844581, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00048670756646216766, | |
| "loss": 2.9385, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 2.9418914318084717, | |
| "eval_runtime": 0.9937, | |
| "eval_samples_per_second": 503.192, | |
| "eval_steps_per_second": 1.006, | |
| "step": 37164 | |
| }, | |
| { | |
| "epoch": 38.34355828220859, | |
| "grad_norm": 0.25390625, | |
| "learning_rate": 0.00046625766871165645, | |
| "loss": 2.9349, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 38.85480572597137, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.00044580777096114523, | |
| "loss": 2.9384, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 2.943824291229248, | |
| "eval_runtime": 0.9909, | |
| "eval_samples_per_second": 504.6, | |
| "eval_steps_per_second": 1.009, | |
| "step": 38142 | |
| }, | |
| { | |
| "epoch": 39.36605316973415, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00042535787321063397, | |
| "loss": 2.9386, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 39.877300613496935, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.0004049079754601227, | |
| "loss": 2.9364, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 2.941757917404175, | |
| "eval_runtime": 0.8169, | |
| "eval_samples_per_second": 612.068, | |
| "eval_steps_per_second": 1.224, | |
| "step": 39120 | |
| }, | |
| { | |
| "epoch": 40.38854805725971, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.00038445807770961143, | |
| "loss": 2.9374, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 40.899795501022496, | |
| "grad_norm": 0.2431640625, | |
| "learning_rate": 0.0003640081799591002, | |
| "loss": 2.9341, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 2.9411299228668213, | |
| "eval_runtime": 0.9525, | |
| "eval_samples_per_second": 524.937, | |
| "eval_steps_per_second": 1.05, | |
| "step": 40098 | |
| }, | |
| { | |
| "epoch": 41.41104294478528, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00034355828220858896, | |
| "loss": 2.9323, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 41.922290388548056, | |
| "grad_norm": 0.2353515625, | |
| "learning_rate": 0.00032310838445807774, | |
| "loss": 2.9361, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 2.94166898727417, | |
| "eval_runtime": 0.7476, | |
| "eval_samples_per_second": 668.768, | |
| "eval_steps_per_second": 1.338, | |
| "step": 41076 | |
| }, | |
| { | |
| "epoch": 42.43353783231084, | |
| "grad_norm": 0.2470703125, | |
| "learning_rate": 0.0003026584867075665, | |
| "loss": 2.9344, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 42.94478527607362, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.0002822085889570552, | |
| "loss": 2.9345, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 2.9405741691589355, | |
| "eval_runtime": 0.8055, | |
| "eval_samples_per_second": 620.71, | |
| "eval_steps_per_second": 1.241, | |
| "step": 42054 | |
| }, | |
| { | |
| "epoch": 43.4560327198364, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 0.000261758691206544, | |
| "loss": 2.9342, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 43.967280163599185, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00024130879345603273, | |
| "loss": 2.9347, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 2.9399616718292236, | |
| "eval_runtime": 0.7311, | |
| "eval_samples_per_second": 683.897, | |
| "eval_steps_per_second": 1.368, | |
| "step": 43032 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 48900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "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": 3.377598882696192e+17, | |
| "train_batch_size": 1024, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |