| { | |
| "best_metric": 0.7138778567314148, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-3/checkpoint-242", | |
| "epoch": 29.0, | |
| "eval_steps": 500, | |
| "global_step": 319, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 0.7728238701820374, | |
| "learning_rate": 6.060606060606061e-08, | |
| "loss": 1.102, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.8360726237297058, | |
| "learning_rate": 1.2121212121212122e-07, | |
| "loss": 1.2412, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.6940823793411255, | |
| "learning_rate": 2.4242424242424244e-07, | |
| "loss": 1.0332, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.7341020107269287, | |
| "learning_rate": 3.6363636363636366e-07, | |
| "loss": 1.1054, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.8064824938774109, | |
| "learning_rate": 4.848484848484849e-07, | |
| "loss": 1.2037, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.7021966576576233, | |
| "learning_rate": 6.060606060606061e-07, | |
| "loss": 1.0459, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.122668743133545, | |
| "eval_runtime": 24.349, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.7579166293144226, | |
| "learning_rate": 7.272727272727273e-07, | |
| "loss": 1.0994, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.6994158029556274, | |
| "learning_rate": 8.484848484848486e-07, | |
| "loss": 1.1056, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.698021650314331, | |
| "learning_rate": 9.696969696969698e-07, | |
| "loss": 1.1576, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.7830127477645874, | |
| "learning_rate": 1.090909090909091e-06, | |
| "loss": 1.1701, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.749530553817749, | |
| "learning_rate": 1.2121212121212122e-06, | |
| "loss": 1.0871, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6776620149612427, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.0223, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1149355173110962, | |
| "eval_runtime": 24.3485, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.6116415858268738, | |
| "learning_rate": 1.4545454545454546e-06, | |
| "loss": 1.0642, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.7151461839675903, | |
| "learning_rate": 1.5757575757575759e-06, | |
| "loss": 1.1885, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.7120382785797119, | |
| "learning_rate": 1.6969696969696973e-06, | |
| "loss": 1.0927, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.6961694955825806, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 1.0392, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.8099948167800903, | |
| "learning_rate": 1.9393939393939395e-06, | |
| "loss": 1.0795, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1017646789550781, | |
| "eval_runtime": 24.3583, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 0.7341113686561584, | |
| "learning_rate": 2.0606060606060607e-06, | |
| "loss": 1.1445, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.7002806067466736, | |
| "learning_rate": 2.181818181818182e-06, | |
| "loss": 1.0858, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 0.7403613328933716, | |
| "learning_rate": 2.303030303030303e-06, | |
| "loss": 1.0768, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.7391121983528137, | |
| "learning_rate": 2.4242424242424244e-06, | |
| "loss": 1.035, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 0.790711522102356, | |
| "learning_rate": 2.5454545454545456e-06, | |
| "loss": 1.1519, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.633929967880249, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.9982, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.078678011894226, | |
| "eval_runtime": 24.3517, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 0.7980412840843201, | |
| "learning_rate": 2.7878787878787885e-06, | |
| "loss": 1.1463, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.7583560347557068, | |
| "learning_rate": 2.9090909090909093e-06, | |
| "loss": 1.0093, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.7219624519348145, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 1.0371, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.7923340797424316, | |
| "learning_rate": 3.1515151515151517e-06, | |
| "loss": 1.1243, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 0.6717983484268188, | |
| "learning_rate": 3.272727272727273e-06, | |
| "loss": 0.9702, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.0444384813308716, | |
| "eval_runtime": 24.3479, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.6388964653015137, | |
| "learning_rate": 3.3939393939393946e-06, | |
| "loss": 0.9427, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 0.8183718919754028, | |
| "learning_rate": 3.5151515151515154e-06, | |
| "loss": 1.0683, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.7228857278823853, | |
| "learning_rate": 3.6363636363636366e-06, | |
| "loss": 1.0193, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 0.6639413237571716, | |
| "learning_rate": 3.757575757575758e-06, | |
| "loss": 1.0165, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.7365549802780151, | |
| "learning_rate": 3.878787878787879e-06, | |
| "loss": 1.0233, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.642576277256012, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9509, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.9990049004554749, | |
| "eval_runtime": 24.3532, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.6555075645446777, | |
| "learning_rate": 4.1212121212121215e-06, | |
| "loss": 0.9929, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 0.5816611647605896, | |
| "learning_rate": 4.242424242424243e-06, | |
| "loss": 0.9891, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.5633305907249451, | |
| "learning_rate": 4.363636363636364e-06, | |
| "loss": 0.9197, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.5987116098403931, | |
| "learning_rate": 4.4848484848484855e-06, | |
| "loss": 0.9596, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.5823273658752441, | |
| "learning_rate": 4.606060606060606e-06, | |
| "loss": 0.9573, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.9499709606170654, | |
| "eval_runtime": 24.3506, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 0.5255045890808105, | |
| "learning_rate": 4.727272727272728e-06, | |
| "loss": 0.8918, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.5354146957397461, | |
| "learning_rate": 4.848484848484849e-06, | |
| "loss": 0.9815, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 0.4640227258205414, | |
| "learning_rate": 4.9696969696969696e-06, | |
| "loss": 0.8863, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.4414072334766388, | |
| "learning_rate": 5.090909090909091e-06, | |
| "loss": 0.9009, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 0.42416912317276, | |
| "learning_rate": 5.212121212121213e-06, | |
| "loss": 0.9165, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3701015114784241, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.8624, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9071061015129089, | |
| "eval_runtime": 24.3464, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.3807128369808197, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.8405, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.39005354046821594, | |
| "learning_rate": 5.575757575757577e-06, | |
| "loss": 0.8492, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 0.3477958142757416, | |
| "learning_rate": 5.696969696969698e-06, | |
| "loss": 0.8703, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.34921929240226746, | |
| "learning_rate": 5.8181818181818185e-06, | |
| "loss": 0.9232, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 0.38983580470085144, | |
| "learning_rate": 5.93939393939394e-06, | |
| "loss": 0.8804, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8746969103813171, | |
| "eval_runtime": 24.3504, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.3402298390865326, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.8316, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.3445686399936676, | |
| "learning_rate": 6.181818181818182e-06, | |
| "loss": 0.8235, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.375674307346344, | |
| "learning_rate": 6.303030303030303e-06, | |
| "loss": 0.8982, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 0.3227982223033905, | |
| "learning_rate": 6.424242424242425e-06, | |
| "loss": 0.7633, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.38300326466560364, | |
| "learning_rate": 6.545454545454546e-06, | |
| "loss": 0.852, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.32602500915527344, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.8515, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8457457423210144, | |
| "eval_runtime": 24.3488, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.3338332176208496, | |
| "learning_rate": 6.787878787878789e-06, | |
| "loss": 0.825, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 0.44083863496780396, | |
| "learning_rate": 6.90909090909091e-06, | |
| "loss": 0.8433, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.3659360408782959, | |
| "learning_rate": 7.030303030303031e-06, | |
| "loss": 0.8081, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 0.35738250613212585, | |
| "learning_rate": 7.151515151515152e-06, | |
| "loss": 0.8144, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.33010390400886536, | |
| "learning_rate": 7.272727272727273e-06, | |
| "loss": 0.7864, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.8207831382751465, | |
| "eval_runtime": 24.3504, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.090909090909092, | |
| "grad_norm": 0.329024076461792, | |
| "learning_rate": 7.393939393939395e-06, | |
| "loss": 0.7717, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.32361382246017456, | |
| "learning_rate": 7.515151515151516e-06, | |
| "loss": 0.7642, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 11.454545454545455, | |
| "grad_norm": 0.33380502462387085, | |
| "learning_rate": 7.636363636363638e-06, | |
| "loss": 0.7893, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.3214806318283081, | |
| "learning_rate": 7.757575757575758e-06, | |
| "loss": 0.7901, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 11.818181818181818, | |
| "grad_norm": 0.3193203806877136, | |
| "learning_rate": 7.87878787878788e-06, | |
| "loss": 0.7028, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.32331448793411255, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8648, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8018015027046204, | |
| "eval_runtime": 24.3516, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.181818181818182, | |
| "grad_norm": 0.2892431616783142, | |
| "learning_rate": 8.121212121212121e-06, | |
| "loss": 0.764, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.3037986755371094, | |
| "learning_rate": 8.242424242424243e-06, | |
| "loss": 0.7343, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 12.545454545454545, | |
| "grad_norm": 0.3432358205318451, | |
| "learning_rate": 8.363636363636365e-06, | |
| "loss": 0.7794, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.3149927258491516, | |
| "learning_rate": 8.484848484848486e-06, | |
| "loss": 0.8031, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "grad_norm": 0.29204079508781433, | |
| "learning_rate": 8.606060606060606e-06, | |
| "loss": 0.736, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.786732017993927, | |
| "eval_runtime": 24.3508, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.3326067626476288, | |
| "learning_rate": 8.727272727272728e-06, | |
| "loss": 0.7661, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 13.272727272727273, | |
| "grad_norm": 0.3473084270954132, | |
| "learning_rate": 8.84848484848485e-06, | |
| "loss": 0.7727, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.28269076347351074, | |
| "learning_rate": 8.969696969696971e-06, | |
| "loss": 0.7282, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 13.636363636363637, | |
| "grad_norm": 0.32704994082450867, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.6916, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.3217592239379883, | |
| "learning_rate": 9.212121212121213e-06, | |
| "loss": 0.7475, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.3401748836040497, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.7882, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.7727737426757812, | |
| "eval_runtime": 24.3475, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.3393374979496002, | |
| "learning_rate": 9.454545454545456e-06, | |
| "loss": 0.6774, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 14.363636363636363, | |
| "grad_norm": 0.28455665707588196, | |
| "learning_rate": 9.575757575757576e-06, | |
| "loss": 0.7579, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.30775022506713867, | |
| "learning_rate": 9.696969696969698e-06, | |
| "loss": 0.7331, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 14.727272727272727, | |
| "grad_norm": 0.3098331689834595, | |
| "learning_rate": 9.81818181818182e-06, | |
| "loss": 0.7255, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.39052486419677734, | |
| "learning_rate": 9.939393939393939e-06, | |
| "loss": 0.7452, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.7604131698608398, | |
| "eval_runtime": 24.3513, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 15.090909090909092, | |
| "grad_norm": 0.31109246611595154, | |
| "learning_rate": 9.999988811118232e-06, | |
| "loss": 0.6938, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 15.272727272727273, | |
| "grad_norm": 0.3080443739891052, | |
| "learning_rate": 9.999899300364534e-06, | |
| "loss": 0.6979, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 15.454545454545455, | |
| "grad_norm": 0.37110060453414917, | |
| "learning_rate": 9.999720280459576e-06, | |
| "loss": 0.7866, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 15.636363636363637, | |
| "grad_norm": 0.9138216972351074, | |
| "learning_rate": 9.999451754608208e-06, | |
| "loss": 0.683, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 15.818181818181818, | |
| "grad_norm": 0.37108689546585083, | |
| "learning_rate": 9.99909372761763e-06, | |
| "loss": 0.7214, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.32667478919029236, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.6818, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.7484579086303711, | |
| "eval_runtime": 24.3469, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 16.181818181818183, | |
| "grad_norm": 0.32395023107528687, | |
| "learning_rate": 9.998109197458865e-06, | |
| "loss": 0.6817, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 16.363636363636363, | |
| "grad_norm": 0.33770957589149475, | |
| "learning_rate": 9.997482711915926e-06, | |
| "loss": 0.7282, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.545454545454547, | |
| "grad_norm": 0.3693891763687134, | |
| "learning_rate": 9.996766760483955e-06, | |
| "loss": 0.6857, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 16.727272727272727, | |
| "grad_norm": 0.32818594574928284, | |
| "learning_rate": 9.995961355980052e-06, | |
| "loss": 0.653, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 16.90909090909091, | |
| "grad_norm": 0.3677254915237427, | |
| "learning_rate": 9.99506651282272e-06, | |
| "loss": 0.7119, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.7387299537658691, | |
| "eval_runtime": 24.3449, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 17.09090909090909, | |
| "grad_norm": 0.33137425780296326, | |
| "learning_rate": 9.994082247031613e-06, | |
| "loss": 0.6706, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 17.272727272727273, | |
| "grad_norm": 0.34344998002052307, | |
| "learning_rate": 9.993008576227248e-06, | |
| "loss": 0.656, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 17.454545454545453, | |
| "grad_norm": 0.4169725775718689, | |
| "learning_rate": 9.991845519630679e-06, | |
| "loss": 0.6647, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 17.636363636363637, | |
| "grad_norm": 0.4314097762107849, | |
| "learning_rate": 9.99059309806317e-06, | |
| "loss": 0.6885, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 17.818181818181817, | |
| "grad_norm": 0.3693665862083435, | |
| "learning_rate": 9.989251333945813e-06, | |
| "loss": 0.6567, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.42939141392707825, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.7107, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.7306709289550781, | |
| "eval_runtime": 24.3524, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "grad_norm": 0.3910067677497864, | |
| "learning_rate": 9.986299875742612e-06, | |
| "loss": 0.6819, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.363636363636363, | |
| "grad_norm": 0.3679266571998596, | |
| "learning_rate": 9.984690234494338e-06, | |
| "loss": 0.6852, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 18.545454545454547, | |
| "grad_norm": 0.36058109998703003, | |
| "learning_rate": 9.982991356370404e-06, | |
| "loss": 0.6231, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 18.727272727272727, | |
| "grad_norm": 0.4211237132549286, | |
| "learning_rate": 9.98120327178445e-06, | |
| "loss": 0.6758, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "grad_norm": 0.3913659155368805, | |
| "learning_rate": 9.979326012747106e-06, | |
| "loss": 0.6405, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.7237685322761536, | |
| "eval_runtime": 24.3475, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 19.09090909090909, | |
| "grad_norm": 0.3846379816532135, | |
| "learning_rate": 9.977359612865424e-06, | |
| "loss": 0.5929, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 19.272727272727273, | |
| "grad_norm": 0.44087332487106323, | |
| "learning_rate": 9.975304107342268e-06, | |
| "loss": 0.627, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 19.454545454545453, | |
| "grad_norm": 0.467111200094223, | |
| "learning_rate": 9.973159532975691e-06, | |
| "loss": 0.6858, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 19.636363636363637, | |
| "grad_norm": 0.43696919083595276, | |
| "learning_rate": 9.970925928158275e-06, | |
| "loss": 0.6414, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 19.818181818181817, | |
| "grad_norm": 0.4129337966442108, | |
| "learning_rate": 9.968603332876435e-06, | |
| "loss": 0.6621, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.4260455369949341, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.6075, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7188207507133484, | |
| "eval_runtime": 24.3505, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.181818181818183, | |
| "grad_norm": 0.4215308725833893, | |
| "learning_rate": 9.963691338830045e-06, | |
| "loss": 0.611, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 20.363636363636363, | |
| "grad_norm": 0.4364253282546997, | |
| "learning_rate": 9.961102028000948e-06, | |
| "loss": 0.5843, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 20.545454545454547, | |
| "grad_norm": 0.45465582609176636, | |
| "learning_rate": 9.958423902576764e-06, | |
| "loss": 0.6556, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 20.727272727272727, | |
| "grad_norm": 0.43474891781806946, | |
| "learning_rate": 9.955657010501807e-06, | |
| "loss": 0.6197, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 20.90909090909091, | |
| "grad_norm": 0.4751872420310974, | |
| "learning_rate": 9.952801401309504e-06, | |
| "loss": 0.6323, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.7152169346809387, | |
| "eval_runtime": 24.3508, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 21.09090909090909, | |
| "grad_norm": 0.4626959264278412, | |
| "learning_rate": 9.949857126121519e-06, | |
| "loss": 0.624, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 21.272727272727273, | |
| "grad_norm": 0.45061811804771423, | |
| "learning_rate": 9.946824237646823e-06, | |
| "loss": 0.6272, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 21.454545454545453, | |
| "grad_norm": 0.4539961814880371, | |
| "learning_rate": 9.94370279018077e-06, | |
| "loss": 0.6137, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 21.636363636363637, | |
| "grad_norm": 0.5059633255004883, | |
| "learning_rate": 9.940492839604103e-06, | |
| "loss": 0.5831, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 21.818181818181817, | |
| "grad_norm": 0.46466246247291565, | |
| "learning_rate": 9.937194443381972e-06, | |
| "loss": 0.6131, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.5081361532211304, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.557, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.7138778567314148, | |
| "eval_runtime": 24.3506, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 22.181818181818183, | |
| "grad_norm": 0.4696747064590454, | |
| "learning_rate": 9.930332551777709e-06, | |
| "loss": 0.5765, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 22.363636363636363, | |
| "grad_norm": 0.4927414655685425, | |
| "learning_rate": 9.926769179238467e-06, | |
| "loss": 0.6057, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 22.545454545454547, | |
| "grad_norm": 0.5349259376525879, | |
| "learning_rate": 9.923117606737347e-06, | |
| "loss": 0.6059, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 22.727272727272727, | |
| "grad_norm": 0.5289928317070007, | |
| "learning_rate": 9.919377899645497e-06, | |
| "loss": 0.5736, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 22.90909090909091, | |
| "grad_norm": 0.5489732027053833, | |
| "learning_rate": 9.915550124911866e-06, | |
| "loss": 0.5692, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.7158213257789612, | |
| "eval_runtime": 24.36, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 23.09090909090909, | |
| "grad_norm": 0.5354622602462769, | |
| "learning_rate": 9.91163435106201e-06, | |
| "loss": 0.5341, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 23.272727272727273, | |
| "grad_norm": 0.488086074590683, | |
| "learning_rate": 9.907630648196857e-06, | |
| "loss": 0.593, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 23.454545454545453, | |
| "grad_norm": 0.545102059841156, | |
| "learning_rate": 9.903539087991462e-06, | |
| "loss": 0.568, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 23.636363636363637, | |
| "grad_norm": 0.5522411465644836, | |
| "learning_rate": 9.899359743693715e-06, | |
| "loss": 0.5498, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 23.818181818181817, | |
| "grad_norm": 0.6323682069778442, | |
| "learning_rate": 9.895092690123036e-06, | |
| "loss": 0.5401, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.5789000988006592, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.558, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.719824492931366, | |
| "eval_runtime": 24.3548, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 24.181818181818183, | |
| "grad_norm": 0.5536386966705322, | |
| "learning_rate": 9.886295762290125e-06, | |
| "loss": 0.5058, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 24.363636363636363, | |
| "grad_norm": 0.6637153029441833, | |
| "learning_rate": 9.881766045512176e-06, | |
| "loss": 0.5807, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 24.545454545454547, | |
| "grad_norm": 0.619708240032196, | |
| "learning_rate": 9.877148934427037e-06, | |
| "loss": 0.5592, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 24.727272727272727, | |
| "grad_norm": 0.6658281087875366, | |
| "learning_rate": 9.872444511691108e-06, | |
| "loss": 0.5266, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 24.90909090909091, | |
| "grad_norm": 0.6834419369697571, | |
| "learning_rate": 9.867652861523866e-06, | |
| "loss": 0.5153, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.7296044230461121, | |
| "eval_runtime": 24.3588, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 25.09090909090909, | |
| "grad_norm": 0.667942225933075, | |
| "learning_rate": 9.862774069706346e-06, | |
| "loss": 0.4993, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 25.272727272727273, | |
| "grad_norm": 0.6705055236816406, | |
| "learning_rate": 9.85780822357961e-06, | |
| "loss": 0.5251, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 25.454545454545453, | |
| "grad_norm": 0.7730400562286377, | |
| "learning_rate": 9.85275541204318e-06, | |
| "loss": 0.5414, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 25.636363636363637, | |
| "grad_norm": 0.6594474911689758, | |
| "learning_rate": 9.847615725553457e-06, | |
| "loss": 0.5029, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 25.818181818181817, | |
| "grad_norm": 0.6896001696586609, | |
| "learning_rate": 9.842389256122086e-06, | |
| "loss": 0.4927, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.714022159576416, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.4964, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.7366821765899658, | |
| "eval_runtime": 24.3662, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 26.181818181818183, | |
| "grad_norm": 0.8182728886604309, | |
| "learning_rate": 9.831676344247343e-06, | |
| "loss": 0.4601, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 26.363636363636363, | |
| "grad_norm": 0.8259897828102112, | |
| "learning_rate": 9.826190093588564e-06, | |
| "loss": 0.5021, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 26.545454545454547, | |
| "grad_norm": 0.8739567995071411, | |
| "learning_rate": 9.820617443553889e-06, | |
| "loss": 0.5133, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 26.727272727272727, | |
| "grad_norm": 0.798852801322937, | |
| "learning_rate": 9.814958493905962e-06, | |
| "loss": 0.4688, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 26.90909090909091, | |
| "grad_norm": 0.7111189365386963, | |
| "learning_rate": 9.80921334595238e-06, | |
| "loss": 0.4713, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.7402606010437012, | |
| "eval_runtime": 24.3554, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 27.09090909090909, | |
| "grad_norm": 0.7679806351661682, | |
| "learning_rate": 9.80338210254388e-06, | |
| "loss": 0.4414, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 27.272727272727273, | |
| "grad_norm": 0.7741813659667969, | |
| "learning_rate": 9.797464868072489e-06, | |
| "loss": 0.4424, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.454545454545453, | |
| "grad_norm": 0.7848456501960754, | |
| "learning_rate": 9.791461748469669e-06, | |
| "loss": 0.461, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 27.636363636363637, | |
| "grad_norm": 0.9988805055618286, | |
| "learning_rate": 9.785372851204415e-06, | |
| "loss": 0.4594, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 27.818181818181817, | |
| "grad_norm": 1.017672061920166, | |
| "learning_rate": 9.779198285281326e-06, | |
| "loss": 0.4656, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.8862157464027405, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.4144, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.7620039582252502, | |
| "eval_runtime": 24.3481, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 28.181818181818183, | |
| "grad_norm": 1.078522801399231, | |
| "learning_rate": 9.766592591146353e-06, | |
| "loss": 0.4126, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 28.363636363636363, | |
| "grad_norm": 0.8740026354789734, | |
| "learning_rate": 9.760161688604008e-06, | |
| "loss": 0.4104, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 28.545454545454547, | |
| "grad_norm": 1.2476239204406738, | |
| "learning_rate": 9.753645568738872e-06, | |
| "loss": 0.4275, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 28.727272727272727, | |
| "grad_norm": 1.1232469081878662, | |
| "learning_rate": 9.747044348203766e-06, | |
| "loss": 0.4, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 28.90909090909091, | |
| "grad_norm": 0.9635869860649109, | |
| "learning_rate": 9.740358145174999e-06, | |
| "loss": 0.4184, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.7954289317131042, | |
| "eval_runtime": 24.3558, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 319 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "step": 319, | |
| "total_flos": 2.4257124139375e+17, | |
| "train_loss": 0.757462413613699, | |
| "train_runtime": 8705.8997, | |
| "train_samples_per_second": 1.516, | |
| "train_steps_per_second": 0.19 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 1650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4257124139375e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |