| { | |
| "best_metric": 0.4847618043422699, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/mlt-Latn/checkpoint-96000", | |
| "epoch": 43.656207366984994, | |
| "eval_steps": 500, | |
| "global_step": 96000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22737608003638018, | |
| "grad_norm": 5.168328285217285, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.8057, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22737608003638018, | |
| "eval_accuracy": 0.7673858936928836, | |
| "eval_loss": 1.3882914781570435, | |
| "eval_runtime": 57.4714, | |
| "eval_samples_per_second": 265.141, | |
| "eval_steps_per_second": 8.3, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.45475216007276037, | |
| "grad_norm": 6.386734485626221, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.2412, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45475216007276037, | |
| "eval_accuracy": 0.8023059017621376, | |
| "eval_loss": 1.1490817070007324, | |
| "eval_runtime": 57.9662, | |
| "eval_samples_per_second": 262.877, | |
| "eval_steps_per_second": 8.229, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6821282401091405, | |
| "grad_norm": 3.5194196701049805, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.0971, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6821282401091405, | |
| "eval_accuracy": 0.8191701818204763, | |
| "eval_loss": 1.0539811849594116, | |
| "eval_runtime": 58.0944, | |
| "eval_samples_per_second": 262.297, | |
| "eval_steps_per_second": 8.211, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9095043201455207, | |
| "grad_norm": 4.919663906097412, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.0083, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9095043201455207, | |
| "eval_accuracy": 0.8290244187626605, | |
| "eval_loss": 0.9915244579315186, | |
| "eval_runtime": 59.0201, | |
| "eval_samples_per_second": 258.183, | |
| "eval_steps_per_second": 8.082, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1368804001819008, | |
| "grad_norm": 3.4152417182922363, | |
| "learning_rate": 9.75e-05, | |
| "loss": 0.9343, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.1368804001819008, | |
| "eval_accuracy": 0.8386819891878221, | |
| "eval_loss": 0.9384229779243469, | |
| "eval_runtime": 58.1514, | |
| "eval_samples_per_second": 262.04, | |
| "eval_steps_per_second": 8.203, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.364256480218281, | |
| "grad_norm": 3.325963258743286, | |
| "learning_rate": 9.7e-05, | |
| "loss": 0.8966, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.364256480218281, | |
| "eval_accuracy": 0.8441958792162307, | |
| "eval_loss": 0.8955113887786865, | |
| "eval_runtime": 59.0626, | |
| "eval_samples_per_second": 257.997, | |
| "eval_steps_per_second": 8.076, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.591632560254661, | |
| "grad_norm": 4.558549880981445, | |
| "learning_rate": 9.65e-05, | |
| "loss": 0.8609, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.591632560254661, | |
| "eval_accuracy": 0.8482073460109512, | |
| "eval_loss": 0.8796423673629761, | |
| "eval_runtime": 58.1755, | |
| "eval_samples_per_second": 261.931, | |
| "eval_steps_per_second": 8.199, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8190086402910413, | |
| "grad_norm": 4.228474140167236, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8394, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8190086402910413, | |
| "eval_accuracy": 0.8522584062680361, | |
| "eval_loss": 0.8565191030502319, | |
| "eval_runtime": 59.0465, | |
| "eval_samples_per_second": 258.068, | |
| "eval_steps_per_second": 8.078, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0463847203274215, | |
| "grad_norm": 4.954216003417969, | |
| "learning_rate": 9.55e-05, | |
| "loss": 0.8082, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0463847203274215, | |
| "eval_accuracy": 0.8560422496354393, | |
| "eval_loss": 0.8301263451576233, | |
| "eval_runtime": 59.074, | |
| "eval_samples_per_second": 257.947, | |
| "eval_steps_per_second": 8.075, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.2737608003638017, | |
| "grad_norm": 3.8996498584747314, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.7751, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.2737608003638017, | |
| "eval_accuracy": 0.8575601290381725, | |
| "eval_loss": 0.8123583793640137, | |
| "eval_runtime": 59.0078, | |
| "eval_samples_per_second": 258.237, | |
| "eval_steps_per_second": 8.084, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.501136880400182, | |
| "grad_norm": 3.683563709259033, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 0.7746, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.501136880400182, | |
| "eval_accuracy": 0.8608798596484555, | |
| "eval_loss": 0.8004751801490784, | |
| "eval_runtime": 57.9526, | |
| "eval_samples_per_second": 262.939, | |
| "eval_steps_per_second": 8.231, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.728512960436562, | |
| "grad_norm": 3.8639516830444336, | |
| "learning_rate": 9.4e-05, | |
| "loss": 0.7467, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.728512960436562, | |
| "eval_accuracy": 0.8630581086069647, | |
| "eval_loss": 0.7939795255661011, | |
| "eval_runtime": 59.0043, | |
| "eval_samples_per_second": 258.252, | |
| "eval_steps_per_second": 8.084, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9558890404729423, | |
| "grad_norm": 3.31400990486145, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 0.7404, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.9558890404729423, | |
| "eval_accuracy": 0.8657109489879384, | |
| "eval_loss": 0.7746465802192688, | |
| "eval_runtime": 58.2316, | |
| "eval_samples_per_second": 261.679, | |
| "eval_steps_per_second": 8.191, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.1832651205093225, | |
| "grad_norm": 2.843773603439331, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 0.7193, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1832651205093225, | |
| "eval_accuracy": 0.8669591175218982, | |
| "eval_loss": 0.7662197947502136, | |
| "eval_runtime": 58.8578, | |
| "eval_samples_per_second": 258.895, | |
| "eval_steps_per_second": 8.104, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.4106412005457027, | |
| "grad_norm": 3.61684513092041, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 0.7127, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.4106412005457027, | |
| "eval_accuracy": 0.8696189271143282, | |
| "eval_loss": 0.7391215562820435, | |
| "eval_runtime": 58.0068, | |
| "eval_samples_per_second": 262.693, | |
| "eval_steps_per_second": 8.223, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.6380172805820825, | |
| "grad_norm": 2.982285737991333, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.7077, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.6380172805820825, | |
| "eval_accuracy": 0.8704082502343816, | |
| "eval_loss": 0.7502180933952332, | |
| "eval_runtime": 58.7428, | |
| "eval_samples_per_second": 259.402, | |
| "eval_steps_per_second": 8.12, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.865393360618463, | |
| "grad_norm": 3.2284772396087646, | |
| "learning_rate": 9.15e-05, | |
| "loss": 0.6954, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.865393360618463, | |
| "eval_accuracy": 0.8723014320863037, | |
| "eval_loss": 0.7357130646705627, | |
| "eval_runtime": 57.9037, | |
| "eval_samples_per_second": 263.161, | |
| "eval_steps_per_second": 8.238, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.092769440654843, | |
| "grad_norm": 3.141983985900879, | |
| "learning_rate": 9.1e-05, | |
| "loss": 0.6831, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.092769440654843, | |
| "eval_accuracy": 0.87450734136313, | |
| "eval_loss": 0.7237355709075928, | |
| "eval_runtime": 58.8298, | |
| "eval_samples_per_second": 259.018, | |
| "eval_steps_per_second": 8.108, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.320145520691224, | |
| "grad_norm": 3.805156707763672, | |
| "learning_rate": 9.05e-05, | |
| "loss": 0.6653, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.320145520691224, | |
| "eval_accuracy": 0.875826148405435, | |
| "eval_loss": 0.7154198288917542, | |
| "eval_runtime": 58.8099, | |
| "eval_samples_per_second": 259.106, | |
| "eval_steps_per_second": 8.111, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.547521600727603, | |
| "grad_norm": 2.943392753601074, | |
| "learning_rate": 9e-05, | |
| "loss": 0.6587, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.547521600727603, | |
| "eval_accuracy": 0.8750290297659766, | |
| "eval_loss": 0.7158774733543396, | |
| "eval_runtime": 58.8052, | |
| "eval_samples_per_second": 259.127, | |
| "eval_steps_per_second": 8.112, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.774897680763983, | |
| "grad_norm": 3.1016767024993896, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 0.657, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.774897680763983, | |
| "eval_accuracy": 0.876631092564109, | |
| "eval_loss": 0.7100118398666382, | |
| "eval_runtime": 57.9757, | |
| "eval_samples_per_second": 262.834, | |
| "eval_steps_per_second": 8.228, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.002273760800364, | |
| "grad_norm": 2.9359586238861084, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 0.6544, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.002273760800364, | |
| "eval_accuracy": 0.8772724264740639, | |
| "eval_loss": 0.6987695097923279, | |
| "eval_runtime": 58.0324, | |
| "eval_samples_per_second": 262.578, | |
| "eval_steps_per_second": 8.22, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.229649840836744, | |
| "grad_norm": 3.289794445037842, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 0.6315, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.229649840836744, | |
| "eval_accuracy": 0.8785936092935085, | |
| "eval_loss": 0.7098827362060547, | |
| "eval_runtime": 58.2042, | |
| "eval_samples_per_second": 261.802, | |
| "eval_steps_per_second": 8.195, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.457025920873124, | |
| "grad_norm": 3.1603569984436035, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.6406, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.457025920873124, | |
| "eval_accuracy": 0.8785365360431241, | |
| "eval_loss": 0.697100043296814, | |
| "eval_runtime": 58.956, | |
| "eval_samples_per_second": 258.464, | |
| "eval_steps_per_second": 8.091, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.684402000909504, | |
| "grad_norm": 3.9137370586395264, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.6296, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.684402000909504, | |
| "eval_accuracy": 0.8807072007268872, | |
| "eval_loss": 0.6913357377052307, | |
| "eval_runtime": 58.1403, | |
| "eval_samples_per_second": 262.09, | |
| "eval_steps_per_second": 8.204, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.911778080945885, | |
| "grad_norm": 3.703839063644409, | |
| "learning_rate": 8.7e-05, | |
| "loss": 0.626, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.911778080945885, | |
| "eval_accuracy": 0.8809903070027411, | |
| "eval_loss": 0.6919081807136536, | |
| "eval_runtime": 58.1489, | |
| "eval_samples_per_second": 262.051, | |
| "eval_steps_per_second": 8.203, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.139154160982264, | |
| "grad_norm": 3.5155723094940186, | |
| "learning_rate": 8.65e-05, | |
| "loss": 0.6198, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.139154160982264, | |
| "eval_accuracy": 0.8827777558822371, | |
| "eval_loss": 0.6714188456535339, | |
| "eval_runtime": 58.9629, | |
| "eval_samples_per_second": 258.434, | |
| "eval_steps_per_second": 8.09, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.366530241018645, | |
| "grad_norm": 3.117462635040283, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.6076, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.366530241018645, | |
| "eval_accuracy": 0.8834275999297155, | |
| "eval_loss": 0.6733196973800659, | |
| "eval_runtime": 59.0324, | |
| "eval_samples_per_second": 258.13, | |
| "eval_steps_per_second": 8.08, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.593906321055025, | |
| "grad_norm": 2.6322078704833984, | |
| "learning_rate": 8.55e-05, | |
| "loss": 0.6087, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.593906321055025, | |
| "eval_accuracy": 0.8831569539581932, | |
| "eval_loss": 0.6722173690795898, | |
| "eval_runtime": 58.2146, | |
| "eval_samples_per_second": 261.755, | |
| "eval_steps_per_second": 8.194, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.8212824010914055, | |
| "grad_norm": 2.875756025314331, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.6016, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.8212824010914055, | |
| "eval_accuracy": 0.8827436768449041, | |
| "eval_loss": 0.6779712438583374, | |
| "eval_runtime": 57.9717, | |
| "eval_samples_per_second": 262.852, | |
| "eval_steps_per_second": 8.228, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.048658481127785, | |
| "grad_norm": 2.797990560531616, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 0.5932, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.048658481127785, | |
| "eval_accuracy": 0.8849467695494039, | |
| "eval_loss": 0.6560626029968262, | |
| "eval_runtime": 59.2909, | |
| "eval_samples_per_second": 257.004, | |
| "eval_steps_per_second": 8.045, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.276034561164166, | |
| "grad_norm": 3.173975706100464, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.5877, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.276034561164166, | |
| "eval_accuracy": 0.8849921630094044, | |
| "eval_loss": 0.6600627303123474, | |
| "eval_runtime": 57.9674, | |
| "eval_samples_per_second": 262.872, | |
| "eval_steps_per_second": 8.229, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.503410641200546, | |
| "grad_norm": 2.7644402980804443, | |
| "learning_rate": 8.35e-05, | |
| "loss": 0.5909, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.503410641200546, | |
| "eval_accuracy": 0.885632879834746, | |
| "eval_loss": 0.6634506583213806, | |
| "eval_runtime": 58.1288, | |
| "eval_samples_per_second": 262.142, | |
| "eval_steps_per_second": 8.206, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.730786721236926, | |
| "grad_norm": 3.6716108322143555, | |
| "learning_rate": 8.3e-05, | |
| "loss": 0.5848, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.730786721236926, | |
| "eval_accuracy": 0.8870372975442135, | |
| "eval_loss": 0.6555737257003784, | |
| "eval_runtime": 58.0593, | |
| "eval_samples_per_second": 262.456, | |
| "eval_steps_per_second": 8.216, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.958162801273306, | |
| "grad_norm": 2.635899782180786, | |
| "learning_rate": 8.25e-05, | |
| "loss": 0.5806, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.958162801273306, | |
| "eval_accuracy": 0.8872147874120453, | |
| "eval_loss": 0.6475590467453003, | |
| "eval_runtime": 58.9688, | |
| "eval_samples_per_second": 258.408, | |
| "eval_steps_per_second": 8.089, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.185538881309686, | |
| "grad_norm": 3.155376434326172, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.5717, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.185538881309686, | |
| "eval_accuracy": 0.887904005764964, | |
| "eval_loss": 0.6570438146591187, | |
| "eval_runtime": 58.2291, | |
| "eval_samples_per_second": 261.69, | |
| "eval_steps_per_second": 8.192, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.412914961346067, | |
| "grad_norm": 2.7952346801757812, | |
| "learning_rate": 8.15e-05, | |
| "loss": 0.564, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 8.412914961346067, | |
| "eval_accuracy": 0.8873371327146311, | |
| "eval_loss": 0.6420606374740601, | |
| "eval_runtime": 58.022, | |
| "eval_samples_per_second": 262.624, | |
| "eval_steps_per_second": 8.221, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 8.640291041382447, | |
| "grad_norm": 2.901456832885742, | |
| "learning_rate": 8.1e-05, | |
| "loss": 0.5668, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.640291041382447, | |
| "eval_accuracy": 0.8888152172025493, | |
| "eval_loss": 0.6471173763275146, | |
| "eval_runtime": 58.0374, | |
| "eval_samples_per_second": 262.555, | |
| "eval_steps_per_second": 8.219, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.867667121418826, | |
| "grad_norm": 3.783108949661255, | |
| "learning_rate": 8.05e-05, | |
| "loss": 0.5661, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.867667121418826, | |
| "eval_accuracy": 0.8896724765068369, | |
| "eval_loss": 0.6358206272125244, | |
| "eval_runtime": 57.9826, | |
| "eval_samples_per_second": 262.803, | |
| "eval_steps_per_second": 8.227, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 9.095043201455207, | |
| "grad_norm": 3.0208489894866943, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5575, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 9.095043201455207, | |
| "eval_accuracy": 0.8890150798157612, | |
| "eval_loss": 0.6520903706550598, | |
| "eval_runtime": 58.9559, | |
| "eval_samples_per_second": 258.464, | |
| "eval_steps_per_second": 8.091, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 9.322419281491587, | |
| "grad_norm": 2.4719395637512207, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 0.5563, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 9.322419281491587, | |
| "eval_accuracy": 0.8897703404100196, | |
| "eval_loss": 0.6332861185073853, | |
| "eval_runtime": 58.0586, | |
| "eval_samples_per_second": 262.459, | |
| "eval_steps_per_second": 8.216, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 9.549795361527968, | |
| "grad_norm": 3.9511001110076904, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 0.5462, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.549795361527968, | |
| "eval_accuracy": 0.8907973350119666, | |
| "eval_loss": 0.6307789087295532, | |
| "eval_runtime": 58.9193, | |
| "eval_samples_per_second": 258.625, | |
| "eval_steps_per_second": 8.096, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.777171441564347, | |
| "grad_norm": 3.5722222328186035, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 0.554, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.777171441564347, | |
| "eval_accuracy": 0.892567522812507, | |
| "eval_loss": 0.6097805500030518, | |
| "eval_runtime": 58.065, | |
| "eval_samples_per_second": 262.43, | |
| "eval_steps_per_second": 8.215, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 10.004547521600728, | |
| "grad_norm": 2.951775074005127, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.5484, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 10.004547521600728, | |
| "eval_accuracy": 0.8916424877002649, | |
| "eval_loss": 0.6175746917724609, | |
| "eval_runtime": 58.4944, | |
| "eval_samples_per_second": 260.504, | |
| "eval_steps_per_second": 8.155, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 10.231923601637108, | |
| "grad_norm": 3.9985241889953613, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.5443, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.231923601637108, | |
| "eval_accuracy": 0.8924841760921375, | |
| "eval_loss": 0.6330751180648804, | |
| "eval_runtime": 58.5296, | |
| "eval_samples_per_second": 260.347, | |
| "eval_steps_per_second": 8.15, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.459299681673487, | |
| "grad_norm": 3.2177023887634277, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.5329, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 10.459299681673487, | |
| "eval_accuracy": 0.8919113818187556, | |
| "eval_loss": 0.6282722353935242, | |
| "eval_runtime": 58.5087, | |
| "eval_samples_per_second": 260.44, | |
| "eval_steps_per_second": 8.153, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 10.686675761709868, | |
| "grad_norm": 2.4907150268554688, | |
| "learning_rate": 7.65e-05, | |
| "loss": 0.5332, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.686675761709868, | |
| "eval_accuracy": 0.8929225353778606, | |
| "eval_loss": 0.6218137145042419, | |
| "eval_runtime": 58.6581, | |
| "eval_samples_per_second": 259.777, | |
| "eval_steps_per_second": 8.132, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.914051841746248, | |
| "grad_norm": 2.3024277687072754, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.5323, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.914051841746248, | |
| "eval_accuracy": 0.8931125650111673, | |
| "eval_loss": 0.6083813309669495, | |
| "eval_runtime": 57.6897, | |
| "eval_samples_per_second": 264.137, | |
| "eval_steps_per_second": 8.268, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 11.141427921782629, | |
| "grad_norm": 3.06503963470459, | |
| "learning_rate": 7.55e-05, | |
| "loss": 0.5239, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 11.141427921782629, | |
| "eval_accuracy": 0.8939010799940649, | |
| "eval_loss": 0.637162446975708, | |
| "eval_runtime": 57.7617, | |
| "eval_samples_per_second": 263.808, | |
| "eval_steps_per_second": 8.258, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 11.368804001819008, | |
| "grad_norm": 2.7696726322174072, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.5261, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 11.368804001819008, | |
| "eval_accuracy": 0.8937561222232755, | |
| "eval_loss": 0.6276402473449707, | |
| "eval_runtime": 57.6676, | |
| "eval_samples_per_second": 264.239, | |
| "eval_steps_per_second": 8.272, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 11.596180081855389, | |
| "grad_norm": 2.751497268676758, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.5175, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 11.596180081855389, | |
| "eval_accuracy": 0.8951661397774281, | |
| "eval_loss": 0.6050118207931519, | |
| "eval_runtime": 57.5452, | |
| "eval_samples_per_second": 264.8, | |
| "eval_steps_per_second": 8.289, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 11.82355616189177, | |
| "grad_norm": 2.554222822189331, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.5262, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.82355616189177, | |
| "eval_accuracy": 0.8952365458244355, | |
| "eval_loss": 0.6047869324684143, | |
| "eval_runtime": 58.525, | |
| "eval_samples_per_second": 260.367, | |
| "eval_steps_per_second": 8.15, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 12.05093224192815, | |
| "grad_norm": 2.832977294921875, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.5189, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 12.05093224192815, | |
| "eval_accuracy": 0.8951059547768438, | |
| "eval_loss": 0.6053177714347839, | |
| "eval_runtime": 57.7314, | |
| "eval_samples_per_second": 263.946, | |
| "eval_steps_per_second": 8.262, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 12.278308321964529, | |
| "grad_norm": 2.9270408153533936, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.5162, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 12.278308321964529, | |
| "eval_accuracy": 0.8956970707551482, | |
| "eval_loss": 0.600378692150116, | |
| "eval_runtime": 68.0954, | |
| "eval_samples_per_second": 223.774, | |
| "eval_steps_per_second": 7.005, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 12.50568440200091, | |
| "grad_norm": 4.0453782081604, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.5111, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 12.50568440200091, | |
| "eval_accuracy": 0.8950593575694928, | |
| "eval_loss": 0.601889431476593, | |
| "eval_runtime": 69.1002, | |
| "eval_samples_per_second": 220.52, | |
| "eval_steps_per_second": 6.903, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 12.73306048203729, | |
| "grad_norm": 1.8677308559417725, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.51, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 12.73306048203729, | |
| "eval_accuracy": 0.8962058115277431, | |
| "eval_loss": 0.6174491047859192, | |
| "eval_runtime": 69.7483, | |
| "eval_samples_per_second": 218.471, | |
| "eval_steps_per_second": 6.839, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 12.96043656207367, | |
| "grad_norm": 2.551996946334839, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.5109, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.96043656207367, | |
| "eval_accuracy": 0.8969674492010259, | |
| "eval_loss": 0.6083965301513672, | |
| "eval_runtime": 66.6755, | |
| "eval_samples_per_second": 228.54, | |
| "eval_steps_per_second": 7.154, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 13.18781264211005, | |
| "grad_norm": 2.474005699157715, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.502, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 13.18781264211005, | |
| "eval_accuracy": 0.8963457500841246, | |
| "eval_loss": 0.6135991215705872, | |
| "eval_runtime": 67.7791, | |
| "eval_samples_per_second": 224.818, | |
| "eval_steps_per_second": 7.038, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 13.41518872214643, | |
| "grad_norm": 3.0140531063079834, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.5, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 13.41518872214643, | |
| "eval_accuracy": 0.8979126349984058, | |
| "eval_loss": 0.6121774911880493, | |
| "eval_runtime": 69.2043, | |
| "eval_samples_per_second": 220.189, | |
| "eval_steps_per_second": 6.893, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 13.642564802182811, | |
| "grad_norm": 2.9221041202545166, | |
| "learning_rate": 7e-05, | |
| "loss": 0.5024, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 13.642564802182811, | |
| "eval_accuracy": 0.8981451519977776, | |
| "eval_loss": 0.59588623046875, | |
| "eval_runtime": 68.802, | |
| "eval_samples_per_second": 221.476, | |
| "eval_steps_per_second": 6.933, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 13.86994088221919, | |
| "grad_norm": 3.138761043548584, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.4998, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 13.86994088221919, | |
| "eval_accuracy": 0.8978226704469172, | |
| "eval_loss": 0.5944364666938782, | |
| "eval_runtime": 66.5107, | |
| "eval_samples_per_second": 229.106, | |
| "eval_steps_per_second": 7.172, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 14.09731696225557, | |
| "grad_norm": 2.7592108249664307, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.4854, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 14.09731696225557, | |
| "eval_accuracy": 0.898367703578009, | |
| "eval_loss": 0.5859193205833435, | |
| "eval_runtime": 66.7713, | |
| "eval_samples_per_second": 228.212, | |
| "eval_steps_per_second": 7.144, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 14.324693042291951, | |
| "grad_norm": 2.9050989151000977, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.4882, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 14.324693042291951, | |
| "eval_accuracy": 0.8975941653774508, | |
| "eval_loss": 0.60106360912323, | |
| "eval_runtime": 68.9335, | |
| "eval_samples_per_second": 221.054, | |
| "eval_steps_per_second": 6.92, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 14.552069122328332, | |
| "grad_norm": 2.3342740535736084, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.4841, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 14.552069122328332, | |
| "eval_accuracy": 0.8990730423650459, | |
| "eval_loss": 0.584600567817688, | |
| "eval_runtime": 67.2181, | |
| "eval_samples_per_second": 226.695, | |
| "eval_steps_per_second": 7.096, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 14.77944520236471, | |
| "grad_norm": 2.1228647232055664, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.4844, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 14.77944520236471, | |
| "eval_accuracy": 0.8990018801499072, | |
| "eval_loss": 0.5907680988311768, | |
| "eval_runtime": 59.205, | |
| "eval_samples_per_second": 257.377, | |
| "eval_steps_per_second": 8.057, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 15.006821282401091, | |
| "grad_norm": 3.1188831329345703, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.4831, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 15.006821282401091, | |
| "eval_accuracy": 0.8995549513632177, | |
| "eval_loss": 0.5859436392784119, | |
| "eval_runtime": 58.4781, | |
| "eval_samples_per_second": 260.576, | |
| "eval_steps_per_second": 8.157, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 15.234197362437472, | |
| "grad_norm": 3.4289638996124268, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.4783, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 15.234197362437472, | |
| "eval_accuracy": 0.8997874968645139, | |
| "eval_loss": 0.586015522480011, | |
| "eval_runtime": 58.2205, | |
| "eval_samples_per_second": 261.729, | |
| "eval_steps_per_second": 8.193, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 15.461573442473851, | |
| "grad_norm": 2.7651004791259766, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.4752, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 15.461573442473851, | |
| "eval_accuracy": 0.9008641609570693, | |
| "eval_loss": 0.5896708369255066, | |
| "eval_runtime": 59.224, | |
| "eval_samples_per_second": 257.294, | |
| "eval_steps_per_second": 8.054, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 15.688949522510232, | |
| "grad_norm": 2.334015130996704, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.4751, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 15.688949522510232, | |
| "eval_accuracy": 0.8998034774235811, | |
| "eval_loss": 0.5798426270484924, | |
| "eval_runtime": 58.5323, | |
| "eval_samples_per_second": 260.335, | |
| "eval_steps_per_second": 8.149, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 15.916325602546612, | |
| "grad_norm": 1.9268873929977417, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.4781, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 15.916325602546612, | |
| "eval_accuracy": 0.900376427009191, | |
| "eval_loss": 0.574535071849823, | |
| "eval_runtime": 59.3062, | |
| "eval_samples_per_second": 256.938, | |
| "eval_steps_per_second": 8.043, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 16.143701682582993, | |
| "grad_norm": 3.1282005310058594, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.4741, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 16.143701682582993, | |
| "eval_accuracy": 0.9008166774124221, | |
| "eval_loss": 0.5920200347900391, | |
| "eval_runtime": 67.8707, | |
| "eval_samples_per_second": 224.515, | |
| "eval_steps_per_second": 7.028, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 16.37107776261937, | |
| "grad_norm": 2.4809906482696533, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.4764, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 16.37107776261937, | |
| "eval_accuracy": 0.9007614317314762, | |
| "eval_loss": 0.5653102397918701, | |
| "eval_runtime": 67.6973, | |
| "eval_samples_per_second": 225.09, | |
| "eval_steps_per_second": 7.046, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 16.598453842655754, | |
| "grad_norm": 2.460899591445923, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.4693, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 16.598453842655754, | |
| "eval_accuracy": 0.9008501214346764, | |
| "eval_loss": 0.5879611968994141, | |
| "eval_runtime": 66.4301, | |
| "eval_samples_per_second": 229.384, | |
| "eval_steps_per_second": 7.18, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 16.825829922692133, | |
| "grad_norm": 2.389681816101074, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.4638, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 16.825829922692133, | |
| "eval_accuracy": 0.9024308529635798, | |
| "eval_loss": 0.5856647491455078, | |
| "eval_runtime": 66.0584, | |
| "eval_samples_per_second": 230.675, | |
| "eval_steps_per_second": 7.221, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 17.053206002728512, | |
| "grad_norm": 2.5536763668060303, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.4643, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 17.053206002728512, | |
| "eval_accuracy": 0.9017984039775945, | |
| "eval_loss": 0.5838255882263184, | |
| "eval_runtime": 66.4217, | |
| "eval_samples_per_second": 229.413, | |
| "eval_steps_per_second": 7.181, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 17.280582082764894, | |
| "grad_norm": 2.8841841220855713, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.4559, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 17.280582082764894, | |
| "eval_accuracy": 0.9016667691546006, | |
| "eval_loss": 0.5722949504852295, | |
| "eval_runtime": 67.5295, | |
| "eval_samples_per_second": 225.65, | |
| "eval_steps_per_second": 7.064, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 17.507958162801273, | |
| "grad_norm": 2.0763895511627197, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.4649, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 17.507958162801273, | |
| "eval_accuracy": 0.9028016315170558, | |
| "eval_loss": 0.5597889423370361, | |
| "eval_runtime": 68.1673, | |
| "eval_samples_per_second": 223.538, | |
| "eval_steps_per_second": 6.997, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 17.735334242837652, | |
| "grad_norm": 2.7098090648651123, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.4619, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 17.735334242837652, | |
| "eval_accuracy": 0.9032880942757541, | |
| "eval_loss": 0.5635744333267212, | |
| "eval_runtime": 65.7624, | |
| "eval_samples_per_second": 231.713, | |
| "eval_steps_per_second": 7.253, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 17.962710322874035, | |
| "grad_norm": 2.361464023590088, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.4554, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 17.962710322874035, | |
| "eval_accuracy": 0.9026038090276955, | |
| "eval_loss": 0.5572078227996826, | |
| "eval_runtime": 65.3403, | |
| "eval_samples_per_second": 233.21, | |
| "eval_steps_per_second": 7.3, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 18.190086402910413, | |
| "grad_norm": 2.2354624271392822, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4516, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 18.190086402910413, | |
| "eval_accuracy": 0.9029999052162649, | |
| "eval_loss": 0.5721431970596313, | |
| "eval_runtime": 66.8398, | |
| "eval_samples_per_second": 227.978, | |
| "eval_steps_per_second": 7.136, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 18.417462482946792, | |
| "grad_norm": 2.5873236656188965, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.4488, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 18.417462482946792, | |
| "eval_accuracy": 0.9031028868580921, | |
| "eval_loss": 0.577458381652832, | |
| "eval_runtime": 65.9075, | |
| "eval_samples_per_second": 231.203, | |
| "eval_steps_per_second": 7.237, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 18.644838562983175, | |
| "grad_norm": 2.14237642288208, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.4529, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 18.644838562983175, | |
| "eval_accuracy": 0.9034652567356332, | |
| "eval_loss": 0.5635027289390564, | |
| "eval_runtime": 57.1062, | |
| "eval_samples_per_second": 266.836, | |
| "eval_steps_per_second": 8.353, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 18.872214643019554, | |
| "grad_norm": 2.687331199645996, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.4457, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 18.872214643019554, | |
| "eval_accuracy": 0.9038571769197852, | |
| "eval_loss": 0.5578777194023132, | |
| "eval_runtime": 57.0961, | |
| "eval_samples_per_second": 266.883, | |
| "eval_steps_per_second": 8.354, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 19.099590723055936, | |
| "grad_norm": 2.4610774517059326, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.4456, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 19.099590723055936, | |
| "eval_accuracy": 0.9040196805071515, | |
| "eval_loss": 0.570646345615387, | |
| "eval_runtime": 57.0535, | |
| "eval_samples_per_second": 267.083, | |
| "eval_steps_per_second": 8.361, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 19.326966803092315, | |
| "grad_norm": 2.7358736991882324, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.4439, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 19.326966803092315, | |
| "eval_accuracy": 0.9047236917555364, | |
| "eval_loss": 0.5603917837142944, | |
| "eval_runtime": 57.9311, | |
| "eval_samples_per_second": 263.037, | |
| "eval_steps_per_second": 8.234, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 19.554342883128694, | |
| "grad_norm": 2.6887753009796143, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.4371, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 19.554342883128694, | |
| "eval_accuracy": 0.9050488743420761, | |
| "eval_loss": 0.5606555938720703, | |
| "eval_runtime": 57.2573, | |
| "eval_samples_per_second": 266.132, | |
| "eval_steps_per_second": 8.331, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 19.781718963165076, | |
| "grad_norm": 4.247918128967285, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.4419, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 19.781718963165076, | |
| "eval_accuracy": 0.9042590959223571, | |
| "eval_loss": 0.5585463643074036, | |
| "eval_runtime": 66.8598, | |
| "eval_samples_per_second": 227.91, | |
| "eval_steps_per_second": 7.134, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 20.009095043201455, | |
| "grad_norm": 2.268353223800659, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.4415, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 20.009095043201455, | |
| "eval_accuracy": 0.9043696805910753, | |
| "eval_loss": 0.5533380508422852, | |
| "eval_runtime": 75.8782, | |
| "eval_samples_per_second": 200.822, | |
| "eval_steps_per_second": 6.286, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 20.236471123237834, | |
| "grad_norm": 2.3621208667755127, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.4363, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 20.236471123237834, | |
| "eval_accuracy": 0.9052580054316933, | |
| "eval_loss": 0.561882734298706, | |
| "eval_runtime": 76.0871, | |
| "eval_samples_per_second": 200.27, | |
| "eval_steps_per_second": 6.269, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 20.463847203274216, | |
| "grad_norm": 2.7799289226531982, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.4327, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 20.463847203274216, | |
| "eval_accuracy": 0.9050759625981704, | |
| "eval_loss": 0.5621650815010071, | |
| "eval_runtime": 75.1467, | |
| "eval_samples_per_second": 202.777, | |
| "eval_steps_per_second": 6.348, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 20.691223283310595, | |
| "grad_norm": 2.6320040225982666, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.4341, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 20.691223283310595, | |
| "eval_accuracy": 0.90537806159975, | |
| "eval_loss": 0.5658455491065979, | |
| "eval_runtime": 75.0197, | |
| "eval_samples_per_second": 203.12, | |
| "eval_steps_per_second": 6.358, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 20.918599363346974, | |
| "grad_norm": 2.626370668411255, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.436, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 20.918599363346974, | |
| "eval_accuracy": 0.9057144525510582, | |
| "eval_loss": 0.5474947094917297, | |
| "eval_runtime": 74.8407, | |
| "eval_samples_per_second": 203.606, | |
| "eval_steps_per_second": 6.374, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 21.145975443383357, | |
| "grad_norm": 2.2103912830352783, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.43, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 21.145975443383357, | |
| "eval_accuracy": 0.9056834598326479, | |
| "eval_loss": 0.5598079562187195, | |
| "eval_runtime": 67.154, | |
| "eval_samples_per_second": 226.911, | |
| "eval_steps_per_second": 7.103, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 21.373351523419736, | |
| "grad_norm": 2.3939335346221924, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.4253, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 21.373351523419736, | |
| "eval_accuracy": 0.9061305526247623, | |
| "eval_loss": 0.5633291602134705, | |
| "eval_runtime": 66.5353, | |
| "eval_samples_per_second": 229.021, | |
| "eval_steps_per_second": 7.169, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 21.600727603456118, | |
| "grad_norm": 2.4314322471618652, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.4276, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 21.600727603456118, | |
| "eval_accuracy": 0.9062991778333528, | |
| "eval_loss": 0.5566443800926208, | |
| "eval_runtime": 68.0296, | |
| "eval_samples_per_second": 223.991, | |
| "eval_steps_per_second": 7.012, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 21.828103683492497, | |
| "grad_norm": 2.792711019515991, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.4235, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 21.828103683492497, | |
| "eval_accuracy": 0.9063092114708347, | |
| "eval_loss": 0.5603668093681335, | |
| "eval_runtime": 67.1839, | |
| "eval_samples_per_second": 226.81, | |
| "eval_steps_per_second": 7.1, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 22.055479763528876, | |
| "grad_norm": 3.2149298191070557, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.4293, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 22.055479763528876, | |
| "eval_accuracy": 0.9061439320028164, | |
| "eval_loss": 0.5632808804512024, | |
| "eval_runtime": 55.9195, | |
| "eval_samples_per_second": 272.499, | |
| "eval_steps_per_second": 8.53, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 22.282855843565258, | |
| "grad_norm": 2.1803324222564697, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.4226, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 22.282855843565258, | |
| "eval_accuracy": 0.9075216851954307, | |
| "eval_loss": 0.5528887510299683, | |
| "eval_runtime": 56.7606, | |
| "eval_samples_per_second": 268.461, | |
| "eval_steps_per_second": 8.404, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 22.510231923601637, | |
| "grad_norm": 2.293896436691284, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.4193, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 22.510231923601637, | |
| "eval_accuracy": 0.907387491792463, | |
| "eval_loss": 0.5482432246208191, | |
| "eval_runtime": 57.104, | |
| "eval_samples_per_second": 266.846, | |
| "eval_steps_per_second": 8.353, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 22.737608003638016, | |
| "grad_norm": 2.2139928340911865, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4158, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 22.737608003638016, | |
| "eval_accuracy": 0.9077673279064253, | |
| "eval_loss": 0.5403118133544922, | |
| "eval_runtime": 55.9516, | |
| "eval_samples_per_second": 272.342, | |
| "eval_steps_per_second": 8.525, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 22.9649840836744, | |
| "grad_norm": 2.241081476211548, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.4174, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 22.9649840836744, | |
| "eval_accuracy": 0.9083347677575133, | |
| "eval_loss": 0.5309577584266663, | |
| "eval_runtime": 56.8789, | |
| "eval_samples_per_second": 267.903, | |
| "eval_steps_per_second": 8.386, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 23.192360163710777, | |
| "grad_norm": 3.3256096839904785, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.4191, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 23.192360163710777, | |
| "eval_accuracy": 0.9088304562502367, | |
| "eval_loss": 0.5443009734153748, | |
| "eval_runtime": 56.8135, | |
| "eval_samples_per_second": 268.211, | |
| "eval_steps_per_second": 8.396, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 23.419736243747156, | |
| "grad_norm": 2.716857433319092, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.4071, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 23.419736243747156, | |
| "eval_accuracy": 0.9077237472285563, | |
| "eval_loss": 0.5535444617271423, | |
| "eval_runtime": 56.7929, | |
| "eval_samples_per_second": 268.308, | |
| "eval_steps_per_second": 8.399, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 23.64711232378354, | |
| "grad_norm": 2.466326951980591, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.4159, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 23.64711232378354, | |
| "eval_accuracy": 0.9093382373851798, | |
| "eval_loss": 0.5515927076339722, | |
| "eval_runtime": 56.8114, | |
| "eval_samples_per_second": 268.221, | |
| "eval_steps_per_second": 8.396, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 23.874488403819917, | |
| "grad_norm": 2.6443376541137695, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.4159, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 23.874488403819917, | |
| "eval_accuracy": 0.9080766240994744, | |
| "eval_loss": 0.5417291522026062, | |
| "eval_runtime": 56.0043, | |
| "eval_samples_per_second": 272.086, | |
| "eval_steps_per_second": 8.517, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 24.1018644838563, | |
| "grad_norm": 2.359405994415283, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4128, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 24.1018644838563, | |
| "eval_accuracy": 0.9082854247369369, | |
| "eval_loss": 0.546323299407959, | |
| "eval_runtime": 55.971, | |
| "eval_samples_per_second": 272.248, | |
| "eval_steps_per_second": 8.522, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 24.32924056389268, | |
| "grad_norm": 2.0880720615386963, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.4034, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 24.32924056389268, | |
| "eval_accuracy": 0.9087918433807937, | |
| "eval_loss": 0.5541105270385742, | |
| "eval_runtime": 58.891, | |
| "eval_samples_per_second": 258.749, | |
| "eval_steps_per_second": 8.1, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 24.556616643929058, | |
| "grad_norm": 2.977452039718628, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4051, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 24.556616643929058, | |
| "eval_accuracy": 0.908003587059478, | |
| "eval_loss": 0.5499656200408936, | |
| "eval_runtime": 57.0241, | |
| "eval_samples_per_second": 267.22, | |
| "eval_steps_per_second": 8.365, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 24.78399272396544, | |
| "grad_norm": 2.6565568447113037, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.3973, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 24.78399272396544, | |
| "eval_accuracy": 0.9084627466876093, | |
| "eval_loss": 0.5408248901367188, | |
| "eval_runtime": 56.052, | |
| "eval_samples_per_second": 271.855, | |
| "eval_steps_per_second": 8.51, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 25.01136880400182, | |
| "grad_norm": 2.017199993133545, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4038, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 25.01136880400182, | |
| "eval_accuracy": 0.9091495750162909, | |
| "eval_loss": 0.531546413898468, | |
| "eval_runtime": 56.6501, | |
| "eval_samples_per_second": 268.984, | |
| "eval_steps_per_second": 8.42, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 25.238744884038198, | |
| "grad_norm": 2.50418758392334, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.3993, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 25.238744884038198, | |
| "eval_accuracy": 0.9091283316273341, | |
| "eval_loss": 0.5372242331504822, | |
| "eval_runtime": 56.2979, | |
| "eval_samples_per_second": 270.667, | |
| "eval_steps_per_second": 8.473, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 25.46612096407458, | |
| "grad_norm": 2.4235289096832275, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3981, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 25.46612096407458, | |
| "eval_accuracy": 0.9090792662970008, | |
| "eval_loss": 0.5424542427062988, | |
| "eval_runtime": 56.0065, | |
| "eval_samples_per_second": 272.075, | |
| "eval_steps_per_second": 8.517, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 25.69349704411096, | |
| "grad_norm": 2.049229621887207, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.3965, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 25.69349704411096, | |
| "eval_accuracy": 0.9091030899558308, | |
| "eval_loss": 0.5371273159980774, | |
| "eval_runtime": 56.9032, | |
| "eval_samples_per_second": 267.788, | |
| "eval_steps_per_second": 8.383, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 25.920873124147338, | |
| "grad_norm": 3.2039127349853516, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.3956, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 25.920873124147338, | |
| "eval_accuracy": 0.9102188871200084, | |
| "eval_loss": 0.5293972492218018, | |
| "eval_runtime": 56.0727, | |
| "eval_samples_per_second": 271.754, | |
| "eval_steps_per_second": 8.507, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 26.14824920418372, | |
| "grad_norm": 2.5481934547424316, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3968, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 26.14824920418372, | |
| "eval_accuracy": 0.9097068103213896, | |
| "eval_loss": 0.5351966619491577, | |
| "eval_runtime": 55.9717, | |
| "eval_samples_per_second": 272.245, | |
| "eval_steps_per_second": 8.522, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 26.3756252842201, | |
| "grad_norm": 2.174415111541748, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.3902, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 26.3756252842201, | |
| "eval_accuracy": 0.9100280555809778, | |
| "eval_loss": 0.5361006855964661, | |
| "eval_runtime": 56.0896, | |
| "eval_samples_per_second": 271.672, | |
| "eval_steps_per_second": 8.504, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 26.60300136425648, | |
| "grad_norm": 2.4938621520996094, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.3919, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 26.60300136425648, | |
| "eval_accuracy": 0.9107032517948321, | |
| "eval_loss": 0.5426139831542969, | |
| "eval_runtime": 56.0504, | |
| "eval_samples_per_second": 271.863, | |
| "eval_steps_per_second": 8.51, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 26.83037744429286, | |
| "grad_norm": 2.729896306991577, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.3932, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 26.83037744429286, | |
| "eval_accuracy": 0.910556622486772, | |
| "eval_loss": 0.5372660756111145, | |
| "eval_runtime": 56.0573, | |
| "eval_samples_per_second": 271.829, | |
| "eval_steps_per_second": 8.509, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 27.05775352432924, | |
| "grad_norm": 2.150261163711548, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.3902, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 27.05775352432924, | |
| "eval_accuracy": 0.9114104771079771, | |
| "eval_loss": 0.5252653360366821, | |
| "eval_runtime": 56.1245, | |
| "eval_samples_per_second": 271.503, | |
| "eval_steps_per_second": 8.499, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 27.285129604365622, | |
| "grad_norm": 2.6134257316589355, | |
| "learning_rate": 4e-05, | |
| "loss": 0.393, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 27.285129604365622, | |
| "eval_accuracy": 0.910748714906454, | |
| "eval_loss": 0.5393661260604858, | |
| "eval_runtime": 56.1182, | |
| "eval_samples_per_second": 271.534, | |
| "eval_steps_per_second": 8.5, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 27.512505684402, | |
| "grad_norm": 3.204314947128296, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.3869, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 27.512505684402, | |
| "eval_accuracy": 0.9109071564105236, | |
| "eval_loss": 0.5372085571289062, | |
| "eval_runtime": 55.938, | |
| "eval_samples_per_second": 272.409, | |
| "eval_steps_per_second": 8.527, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 27.73988176443838, | |
| "grad_norm": 2.1017961502075195, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.3869, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 27.73988176443838, | |
| "eval_accuracy": 0.911014457937958, | |
| "eval_loss": 0.5260709524154663, | |
| "eval_runtime": 56.0593, | |
| "eval_samples_per_second": 271.819, | |
| "eval_steps_per_second": 8.509, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 27.967257844474762, | |
| "grad_norm": 1.7981553077697754, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.3817, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 27.967257844474762, | |
| "eval_accuracy": 0.9113206772639144, | |
| "eval_loss": 0.532031238079071, | |
| "eval_runtime": 56.0692, | |
| "eval_samples_per_second": 271.771, | |
| "eval_steps_per_second": 8.507, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 28.19463392451114, | |
| "grad_norm": 2.4235992431640625, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.3781, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 28.19463392451114, | |
| "eval_accuracy": 0.9117986222084103, | |
| "eval_loss": 0.5406020283699036, | |
| "eval_runtime": 56.0456, | |
| "eval_samples_per_second": 271.886, | |
| "eval_steps_per_second": 8.511, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 28.42201000454752, | |
| "grad_norm": 2.256941795349121, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3793, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 28.42201000454752, | |
| "eval_accuracy": 0.9126116944372022, | |
| "eval_loss": 0.5184915661811829, | |
| "eval_runtime": 56.0344, | |
| "eval_samples_per_second": 271.94, | |
| "eval_steps_per_second": 8.513, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 28.649386084583902, | |
| "grad_norm": 2.4860892295837402, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.3758, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 28.649386084583902, | |
| "eval_accuracy": 0.911845377066606, | |
| "eval_loss": 0.5190649628639221, | |
| "eval_runtime": 55.9764, | |
| "eval_samples_per_second": 272.222, | |
| "eval_steps_per_second": 8.521, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 28.87676216462028, | |
| "grad_norm": 2.1798510551452637, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.382, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 28.87676216462028, | |
| "eval_accuracy": 0.9118922815248525, | |
| "eval_loss": 0.5215730667114258, | |
| "eval_runtime": 56.0709, | |
| "eval_samples_per_second": 271.763, | |
| "eval_steps_per_second": 8.507, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 29.104138244656664, | |
| "grad_norm": 2.404370069503784, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.3763, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 29.104138244656664, | |
| "eval_accuracy": 0.912145037637229, | |
| "eval_loss": 0.5312708020210266, | |
| "eval_runtime": 56.2338, | |
| "eval_samples_per_second": 270.976, | |
| "eval_steps_per_second": 8.482, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 29.331514324693043, | |
| "grad_norm": 2.4613826274871826, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.3788, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 29.331514324693043, | |
| "eval_accuracy": 0.9128715433606646, | |
| "eval_loss": 0.5222127437591553, | |
| "eval_runtime": 56.8998, | |
| "eval_samples_per_second": 267.804, | |
| "eval_steps_per_second": 8.383, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 29.55889040472942, | |
| "grad_norm": 3.3356547355651855, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3755, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 29.55889040472942, | |
| "eval_accuracy": 0.913125950801356, | |
| "eval_loss": 0.5145973563194275, | |
| "eval_runtime": 56.8962, | |
| "eval_samples_per_second": 267.821, | |
| "eval_steps_per_second": 8.384, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 29.786266484765804, | |
| "grad_norm": 2.2212953567504883, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.3788, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 29.786266484765804, | |
| "eval_accuracy": 0.9125449385052034, | |
| "eval_loss": 0.531129002571106, | |
| "eval_runtime": 56.1683, | |
| "eval_samples_per_second": 271.292, | |
| "eval_steps_per_second": 8.492, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 30.013642564802183, | |
| "grad_norm": 2.8541479110717773, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.3737, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 30.013642564802183, | |
| "eval_accuracy": 0.9125566886698588, | |
| "eval_loss": 0.5407569408416748, | |
| "eval_runtime": 56.0905, | |
| "eval_samples_per_second": 271.668, | |
| "eval_steps_per_second": 8.504, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 30.24101864483856, | |
| "grad_norm": 2.438603162765503, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.3702, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 30.24101864483856, | |
| "eval_accuracy": 0.9129054862362733, | |
| "eval_loss": 0.5133882761001587, | |
| "eval_runtime": 57.1245, | |
| "eval_samples_per_second": 266.751, | |
| "eval_steps_per_second": 8.35, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 30.468394724874944, | |
| "grad_norm": 2.1143336296081543, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.3729, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 30.468394724874944, | |
| "eval_accuracy": 0.9124566054408887, | |
| "eval_loss": 0.5418105125427246, | |
| "eval_runtime": 56.0088, | |
| "eval_samples_per_second": 272.065, | |
| "eval_steps_per_second": 8.517, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 30.695770804911323, | |
| "grad_norm": 2.317859649658203, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3662, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 30.695770804911323, | |
| "eval_accuracy": 0.9135011845911662, | |
| "eval_loss": 0.5446010828018188, | |
| "eval_runtime": 56.0915, | |
| "eval_samples_per_second": 271.663, | |
| "eval_steps_per_second": 8.504, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 30.923146884947702, | |
| "grad_norm": 2.983668088912964, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.3647, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 30.923146884947702, | |
| "eval_accuracy": 0.9138354207606285, | |
| "eval_loss": 0.527925968170166, | |
| "eval_runtime": 56.0461, | |
| "eval_samples_per_second": 271.883, | |
| "eval_steps_per_second": 8.511, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 31.150522964984084, | |
| "grad_norm": 2.1927847862243652, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.3683, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 31.150522964984084, | |
| "eval_accuracy": 0.9147327240782173, | |
| "eval_loss": 0.5225592255592346, | |
| "eval_runtime": 57.0493, | |
| "eval_samples_per_second": 267.102, | |
| "eval_steps_per_second": 8.361, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 31.377899045020463, | |
| "grad_norm": 2.345428228378296, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.3628, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 31.377899045020463, | |
| "eval_accuracy": 0.913656255193207, | |
| "eval_loss": 0.5174685716629028, | |
| "eval_runtime": 55.9136, | |
| "eval_samples_per_second": 272.528, | |
| "eval_steps_per_second": 8.531, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 31.605275125056846, | |
| "grad_norm": 2.5066728591918945, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.3651, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 31.605275125056846, | |
| "eval_accuracy": 0.9143142595628372, | |
| "eval_loss": 0.5336447358131409, | |
| "eval_runtime": 56.0707, | |
| "eval_samples_per_second": 271.764, | |
| "eval_steps_per_second": 8.507, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 31.832651205093224, | |
| "grad_norm": 2.5718226432800293, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3621, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 31.832651205093224, | |
| "eval_accuracy": 0.914124758029471, | |
| "eval_loss": 0.5136735439300537, | |
| "eval_runtime": 56.0417, | |
| "eval_samples_per_second": 271.905, | |
| "eval_steps_per_second": 8.512, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 32.06002728512961, | |
| "grad_norm": 3.0197207927703857, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.3598, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 32.06002728512961, | |
| "eval_accuracy": 0.9143737680741543, | |
| "eval_loss": 0.5096654295921326, | |
| "eval_runtime": 56.0653, | |
| "eval_samples_per_second": 271.79, | |
| "eval_steps_per_second": 8.508, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 32.287403365165986, | |
| "grad_norm": 2.932882785797119, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.3583, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 32.287403365165986, | |
| "eval_accuracy": 0.9142125358692816, | |
| "eval_loss": 0.5173851251602173, | |
| "eval_runtime": 56.0814, | |
| "eval_samples_per_second": 271.712, | |
| "eval_steps_per_second": 8.505, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 32.514779445202365, | |
| "grad_norm": 1.9991718530654907, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.3542, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 32.514779445202365, | |
| "eval_accuracy": 0.9147026168203635, | |
| "eval_loss": 0.522916316986084, | |
| "eval_runtime": 55.8503, | |
| "eval_samples_per_second": 272.837, | |
| "eval_steps_per_second": 8.541, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 32.74215552523874, | |
| "grad_norm": 1.8940651416778564, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.356, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 32.74215552523874, | |
| "eval_accuracy": 0.9140601083755969, | |
| "eval_loss": 0.5267335176467896, | |
| "eval_runtime": 56.0286, | |
| "eval_samples_per_second": 271.968, | |
| "eval_steps_per_second": 8.514, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 32.96953160527512, | |
| "grad_norm": 2.228545665740967, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3554, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 32.96953160527512, | |
| "eval_accuracy": 0.9148385388009793, | |
| "eval_loss": 0.518826425075531, | |
| "eval_runtime": 57.0798, | |
| "eval_samples_per_second": 266.96, | |
| "eval_steps_per_second": 8.357, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 33.19690768531151, | |
| "grad_norm": 1.9578146934509277, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.3503, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 33.19690768531151, | |
| "eval_accuracy": 0.9152538892188704, | |
| "eval_loss": 0.5155122876167297, | |
| "eval_runtime": 56.9103, | |
| "eval_samples_per_second": 267.755, | |
| "eval_steps_per_second": 8.382, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 33.42428376534789, | |
| "grad_norm": 2.4094908237457275, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.3557, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 33.42428376534789, | |
| "eval_accuracy": 0.9150505399718056, | |
| "eval_loss": 0.5098891854286194, | |
| "eval_runtime": 55.8856, | |
| "eval_samples_per_second": 272.664, | |
| "eval_steps_per_second": 8.535, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 33.651659845384266, | |
| "grad_norm": 3.974923610687256, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.3504, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 33.651659845384266, | |
| "eval_accuracy": 0.9151609378276823, | |
| "eval_loss": 0.5117126703262329, | |
| "eval_runtime": 56.0457, | |
| "eval_samples_per_second": 271.885, | |
| "eval_steps_per_second": 8.511, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 33.879035925420645, | |
| "grad_norm": 2.490255355834961, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.3543, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 33.879035925420645, | |
| "eval_accuracy": 0.9146644769193912, | |
| "eval_loss": 0.5268692970275879, | |
| "eval_runtime": 55.9308, | |
| "eval_samples_per_second": 272.444, | |
| "eval_steps_per_second": 8.528, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 34.106412005457024, | |
| "grad_norm": 2.831305742263794, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.352, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 34.106412005457024, | |
| "eval_accuracy": 0.9151447226136769, | |
| "eval_loss": 0.5093286037445068, | |
| "eval_runtime": 56.9838, | |
| "eval_samples_per_second": 267.409, | |
| "eval_steps_per_second": 8.371, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 34.3337880854934, | |
| "grad_norm": 2.085205554962158, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.3511, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 34.3337880854934, | |
| "eval_accuracy": 0.9152598801148045, | |
| "eval_loss": 0.5099524259567261, | |
| "eval_runtime": 56.0735, | |
| "eval_samples_per_second": 271.75, | |
| "eval_steps_per_second": 8.507, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 34.56116416552979, | |
| "grad_norm": 2.5950026512145996, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.3477, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 34.56116416552979, | |
| "eval_accuracy": 0.9159398865939414, | |
| "eval_loss": 0.5132637023925781, | |
| "eval_runtime": 56.0447, | |
| "eval_samples_per_second": 271.89, | |
| "eval_steps_per_second": 8.511, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 34.78854024556617, | |
| "grad_norm": 3.0509707927703857, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.3487, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 34.78854024556617, | |
| "eval_accuracy": 0.9158777972417345, | |
| "eval_loss": 0.5157153010368347, | |
| "eval_runtime": 56.0477, | |
| "eval_samples_per_second": 271.876, | |
| "eval_steps_per_second": 8.511, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 35.01591632560255, | |
| "grad_norm": 3.29341983795166, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.3432, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 35.01591632560255, | |
| "eval_accuracy": 0.9159272981938116, | |
| "eval_loss": 0.503484308719635, | |
| "eval_runtime": 56.0514, | |
| "eval_samples_per_second": 271.857, | |
| "eval_steps_per_second": 8.51, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 35.243292405638925, | |
| "grad_norm": 2.7035512924194336, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3468, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 35.243292405638925, | |
| "eval_accuracy": 0.9168826617753377, | |
| "eval_loss": 0.504048228263855, | |
| "eval_runtime": 56.0964, | |
| "eval_samples_per_second": 271.639, | |
| "eval_steps_per_second": 8.503, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 35.470668485675304, | |
| "grad_norm": 4.089804649353027, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.3444, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 35.470668485675304, | |
| "eval_accuracy": 0.9165130415874907, | |
| "eval_loss": 0.5067149996757507, | |
| "eval_runtime": 56.0967, | |
| "eval_samples_per_second": 271.638, | |
| "eval_steps_per_second": 8.503, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 35.69804456571169, | |
| "grad_norm": 2.281663656234741, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.3391, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 35.69804456571169, | |
| "eval_accuracy": 0.9166254129979897, | |
| "eval_loss": 0.4899181127548218, | |
| "eval_runtime": 56.9423, | |
| "eval_samples_per_second": 267.604, | |
| "eval_steps_per_second": 8.377, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 35.92542064574807, | |
| "grad_norm": 2.6281392574310303, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.3395, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 35.92542064574807, | |
| "eval_accuracy": 0.9163551690154882, | |
| "eval_loss": 0.5145460367202759, | |
| "eval_runtime": 56.0529, | |
| "eval_samples_per_second": 271.85, | |
| "eval_steps_per_second": 8.51, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 36.15279672578445, | |
| "grad_norm": 2.2223994731903076, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.3405, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 36.15279672578445, | |
| "eval_accuracy": 0.9158962396841802, | |
| "eval_loss": 0.51621013879776, | |
| "eval_runtime": 56.078, | |
| "eval_samples_per_second": 271.729, | |
| "eval_steps_per_second": 8.506, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 36.38017280582083, | |
| "grad_norm": 2.688448190689087, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3415, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 36.38017280582083, | |
| "eval_accuracy": 0.9164056168752696, | |
| "eval_loss": 0.5185515284538269, | |
| "eval_runtime": 56.0615, | |
| "eval_samples_per_second": 271.809, | |
| "eval_steps_per_second": 8.509, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 36.607548885857206, | |
| "grad_norm": 1.9620684385299683, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.3375, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 36.607548885857206, | |
| "eval_accuracy": 0.9163537188934124, | |
| "eval_loss": 0.517730176448822, | |
| "eval_runtime": 56.9493, | |
| "eval_samples_per_second": 267.571, | |
| "eval_steps_per_second": 8.376, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 36.834924965893585, | |
| "grad_norm": 2.495645761489868, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.3398, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 36.834924965893585, | |
| "eval_accuracy": 0.9171486235060389, | |
| "eval_loss": 0.5006797313690186, | |
| "eval_runtime": 56.9369, | |
| "eval_samples_per_second": 267.63, | |
| "eval_steps_per_second": 8.378, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 37.06230104592997, | |
| "grad_norm": 2.4040660858154297, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.3393, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 37.06230104592997, | |
| "eval_accuracy": 0.9175943590821654, | |
| "eval_loss": 0.5150498151779175, | |
| "eval_runtime": 56.1005, | |
| "eval_samples_per_second": 271.62, | |
| "eval_steps_per_second": 8.503, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 37.28967712596635, | |
| "grad_norm": 2.7558135986328125, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.3344, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 37.28967712596635, | |
| "eval_accuracy": 0.9171837999528608, | |
| "eval_loss": 0.5036485195159912, | |
| "eval_runtime": 56.078, | |
| "eval_samples_per_second": 271.729, | |
| "eval_steps_per_second": 8.506, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 37.51705320600273, | |
| "grad_norm": 2.468700647354126, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.3352, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 37.51705320600273, | |
| "eval_accuracy": 0.9175697679179382, | |
| "eval_loss": 0.5030218958854675, | |
| "eval_runtime": 56.0873, | |
| "eval_samples_per_second": 271.684, | |
| "eval_steps_per_second": 8.505, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 37.74442928603911, | |
| "grad_norm": 2.5978448390960693, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.3329, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 37.74442928603911, | |
| "eval_accuracy": 0.9171495853383843, | |
| "eval_loss": 0.5056445598602295, | |
| "eval_runtime": 56.0025, | |
| "eval_samples_per_second": 272.095, | |
| "eval_steps_per_second": 8.517, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 37.971805366075486, | |
| "grad_norm": 2.1514439582824707, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.3298, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 37.971805366075486, | |
| "eval_accuracy": 0.917162721884158, | |
| "eval_loss": 0.5005716681480408, | |
| "eval_runtime": 57.1445, | |
| "eval_samples_per_second": 266.658, | |
| "eval_steps_per_second": 8.347, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 38.19918144611187, | |
| "grad_norm": 2.6213526725769043, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.3283, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 38.19918144611187, | |
| "eval_accuracy": 0.9176722555185908, | |
| "eval_loss": 0.5062026977539062, | |
| "eval_runtime": 56.1037, | |
| "eval_samples_per_second": 271.604, | |
| "eval_steps_per_second": 8.502, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 38.42655752614825, | |
| "grad_norm": 2.2421905994415283, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.327, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 38.42655752614825, | |
| "eval_accuracy": 0.918665843034967, | |
| "eval_loss": 0.4939974844455719, | |
| "eval_runtime": 56.9571, | |
| "eval_samples_per_second": 267.535, | |
| "eval_steps_per_second": 8.375, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 38.65393360618463, | |
| "grad_norm": 2.6543004512786865, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3335, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 38.65393360618463, | |
| "eval_accuracy": 0.9179265011087174, | |
| "eval_loss": 0.5135884284973145, | |
| "eval_runtime": 56.0886, | |
| "eval_samples_per_second": 271.677, | |
| "eval_steps_per_second": 8.504, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 38.88130968622101, | |
| "grad_norm": 2.2833940982818604, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.3323, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 38.88130968622101, | |
| "eval_accuracy": 0.9180402792823424, | |
| "eval_loss": 0.5044585466384888, | |
| "eval_runtime": 56.0736, | |
| "eval_samples_per_second": 271.75, | |
| "eval_steps_per_second": 8.507, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 39.10868576625739, | |
| "grad_norm": 2.157496213912964, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.3323, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 39.10868576625739, | |
| "eval_accuracy": 0.9178364524991991, | |
| "eval_loss": 0.5072239637374878, | |
| "eval_runtime": 56.0862, | |
| "eval_samples_per_second": 271.689, | |
| "eval_steps_per_second": 8.505, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 39.33606184629377, | |
| "grad_norm": 2.467801809310913, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.3209, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 39.33606184629377, | |
| "eval_accuracy": 0.9178805584191491, | |
| "eval_loss": 0.5000079870223999, | |
| "eval_runtime": 56.0688, | |
| "eval_samples_per_second": 271.773, | |
| "eval_steps_per_second": 8.507, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 39.56343792633015, | |
| "grad_norm": 1.764186143875122, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.318, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 39.56343792633015, | |
| "eval_accuracy": 0.9177274550237315, | |
| "eval_loss": 0.5174301862716675, | |
| "eval_runtime": 55.9254, | |
| "eval_samples_per_second": 272.47, | |
| "eval_steps_per_second": 8.529, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 39.79081400636653, | |
| "grad_norm": 1.9439367055892944, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.3304, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 39.79081400636653, | |
| "eval_accuracy": 0.9182047509597898, | |
| "eval_loss": 0.5170900821685791, | |
| "eval_runtime": 56.0765, | |
| "eval_samples_per_second": 271.736, | |
| "eval_steps_per_second": 8.506, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 40.01819008640291, | |
| "grad_norm": 2.7483394145965576, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.3269, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 40.01819008640291, | |
| "eval_accuracy": 0.9192084760524332, | |
| "eval_loss": 0.5051037669181824, | |
| "eval_runtime": 56.982, | |
| "eval_samples_per_second": 267.418, | |
| "eval_steps_per_second": 8.371, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 40.24556616643929, | |
| "grad_norm": 1.9884289503097534, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.3226, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 40.24556616643929, | |
| "eval_accuracy": 0.9179925787853008, | |
| "eval_loss": 0.4963218867778778, | |
| "eval_runtime": 56.9742, | |
| "eval_samples_per_second": 267.454, | |
| "eval_steps_per_second": 8.372, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 40.47294224647567, | |
| "grad_norm": 2.4133260250091553, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.3193, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 40.47294224647567, | |
| "eval_accuracy": 0.9186944890989619, | |
| "eval_loss": 0.5157626867294312, | |
| "eval_runtime": 56.072, | |
| "eval_samples_per_second": 271.758, | |
| "eval_steps_per_second": 8.507, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 40.700318326512054, | |
| "grad_norm": 2.426737070083618, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.3204, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 40.700318326512054, | |
| "eval_accuracy": 0.918837528271839, | |
| "eval_loss": 0.49108728766441345, | |
| "eval_runtime": 56.0958, | |
| "eval_samples_per_second": 271.643, | |
| "eval_steps_per_second": 8.503, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 40.92769440654843, | |
| "grad_norm": 2.0523056983947754, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3234, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 40.92769440654843, | |
| "eval_accuracy": 0.9187695504636845, | |
| "eval_loss": 0.497799277305603, | |
| "eval_runtime": 56.0896, | |
| "eval_samples_per_second": 271.673, | |
| "eval_steps_per_second": 8.504, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 41.15507048658481, | |
| "grad_norm": 2.7144970893859863, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.3211, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 41.15507048658481, | |
| "eval_accuracy": 0.9191388218103149, | |
| "eval_loss": 0.4986066222190857, | |
| "eval_runtime": 56.0597, | |
| "eval_samples_per_second": 271.818, | |
| "eval_steps_per_second": 8.509, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 41.38244656662119, | |
| "grad_norm": 2.1701807975769043, | |
| "learning_rate": 9e-06, | |
| "loss": 0.3202, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 41.38244656662119, | |
| "eval_accuracy": 0.9192327856840563, | |
| "eval_loss": 0.5045046210289001, | |
| "eval_runtime": 56.0657, | |
| "eval_samples_per_second": 271.788, | |
| "eval_steps_per_second": 8.508, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 41.60982264665757, | |
| "grad_norm": 1.8716968297958374, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.3178, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 41.60982264665757, | |
| "eval_accuracy": 0.918940321981522, | |
| "eval_loss": 0.5036594867706299, | |
| "eval_runtime": 57.7478, | |
| "eval_samples_per_second": 263.872, | |
| "eval_steps_per_second": 8.26, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 41.83719872669395, | |
| "grad_norm": 2.7744719982147217, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.3181, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 41.83719872669395, | |
| "eval_accuracy": 0.9190958139064731, | |
| "eval_loss": 0.49405789375305176, | |
| "eval_runtime": 55.9547, | |
| "eval_samples_per_second": 272.327, | |
| "eval_steps_per_second": 8.525, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 42.064574806730334, | |
| "grad_norm": 1.9463552236557007, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.3139, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 42.064574806730334, | |
| "eval_accuracy": 0.9185921615457032, | |
| "eval_loss": 0.5153664946556091, | |
| "eval_runtime": 55.9616, | |
| "eval_samples_per_second": 272.294, | |
| "eval_steps_per_second": 8.524, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 42.29195088676671, | |
| "grad_norm": 1.7864753007888794, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.3155, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 42.29195088676671, | |
| "eval_accuracy": 0.9193213422174352, | |
| "eval_loss": 0.4968840777873993, | |
| "eval_runtime": 56.9973, | |
| "eval_samples_per_second": 267.346, | |
| "eval_steps_per_second": 8.369, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 42.51932696680309, | |
| "grad_norm": 2.8307926654815674, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.3225, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 42.51932696680309, | |
| "eval_accuracy": 0.9194418625202981, | |
| "eval_loss": 0.49802207946777344, | |
| "eval_runtime": 57.0104, | |
| "eval_samples_per_second": 267.284, | |
| "eval_steps_per_second": 8.367, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 42.74670304683947, | |
| "grad_norm": 2.3244693279266357, | |
| "learning_rate": 6e-06, | |
| "loss": 0.3166, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 42.74670304683947, | |
| "eval_accuracy": 0.9192593340761245, | |
| "eval_loss": 0.5010645985603333, | |
| "eval_runtime": 55.9693, | |
| "eval_samples_per_second": 272.256, | |
| "eval_steps_per_second": 8.523, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 42.97407912687585, | |
| "grad_norm": 1.967890977859497, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.3146, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 42.97407912687585, | |
| "eval_accuracy": 0.9196343334604102, | |
| "eval_loss": 0.48918718099594116, | |
| "eval_runtime": 56.094, | |
| "eval_samples_per_second": 271.651, | |
| "eval_steps_per_second": 8.504, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 43.201455206912236, | |
| "grad_norm": 2.064098596572876, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3152, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 43.201455206912236, | |
| "eval_accuracy": 0.9197978271795374, | |
| "eval_loss": 0.4905773103237152, | |
| "eval_runtime": 56.0802, | |
| "eval_samples_per_second": 271.718, | |
| "eval_steps_per_second": 8.506, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 43.428831286948615, | |
| "grad_norm": 2.1124424934387207, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.3077, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 43.428831286948615, | |
| "eval_accuracy": 0.9199429121130404, | |
| "eval_loss": 0.4860183894634247, | |
| "eval_runtime": 55.9425, | |
| "eval_samples_per_second": 272.387, | |
| "eval_steps_per_second": 8.527, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 43.656207366984994, | |
| "grad_norm": 1.9392900466918945, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.3185, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 43.656207366984994, | |
| "eval_accuracy": 0.919319351233956, | |
| "eval_loss": 0.4847618043422699, | |
| "eval_runtime": 56.9968, | |
| "eval_samples_per_second": 267.348, | |
| "eval_steps_per_second": 8.369, | |
| "step": 96000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 46, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.105977491866255e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |