| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.24509803921568626, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004456327985739751, | |
| "grad_norm": 0.49718141555786133, | |
| "learning_rate": 0.00039272727272727273, | |
| "loss": 0.3714, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004456327985739751, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.4125003516674042, | |
| "eval_runtime": 535.6711, | |
| "eval_samples_per_second": 8.378, | |
| "eval_steps_per_second": 2.095, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "grad_norm": 0.15621569752693176, | |
| "learning_rate": 0.0003854545454545455, | |
| "loss": 0.3392, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.445027619600296, | |
| "eval_runtime": 527.1231, | |
| "eval_samples_per_second": 8.514, | |
| "eval_steps_per_second": 2.129, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "grad_norm": 0.836126446723938, | |
| "learning_rate": 0.0003781818181818182, | |
| "loss": 0.3419, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3319561779499054, | |
| "eval_runtime": 533.3212, | |
| "eval_samples_per_second": 8.415, | |
| "eval_steps_per_second": 2.104, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "grad_norm": 0.5524694919586182, | |
| "learning_rate": 0.0003709090909090909, | |
| "loss": 0.2675, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3492385745048523, | |
| "eval_runtime": 534.7065, | |
| "eval_samples_per_second": 8.393, | |
| "eval_steps_per_second": 2.098, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "grad_norm": 3.021937847137451, | |
| "learning_rate": 0.00036363636363636367, | |
| "loss": 0.3912, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3363156020641327, | |
| "eval_runtime": 537.4904, | |
| "eval_samples_per_second": 8.35, | |
| "eval_steps_per_second": 2.087, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "grad_norm": 0.8108776807785034, | |
| "learning_rate": 0.0003563636363636364, | |
| "loss": 0.3462, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3350915014743805, | |
| "eval_runtime": 531.3591, | |
| "eval_samples_per_second": 8.446, | |
| "eval_steps_per_second": 2.112, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "grad_norm": 0.2669970989227295, | |
| "learning_rate": 0.0003490909090909091, | |
| "loss": 0.3693, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3409470319747925, | |
| "eval_runtime": 529.3629, | |
| "eval_samples_per_second": 8.478, | |
| "eval_steps_per_second": 2.12, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "grad_norm": 0.12394805252552032, | |
| "learning_rate": 0.0003418181818181818, | |
| "loss": 0.4736, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3342541754245758, | |
| "eval_runtime": 532.7547, | |
| "eval_samples_per_second": 8.424, | |
| "eval_steps_per_second": 2.106, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "grad_norm": 2.5240843296051025, | |
| "learning_rate": 0.00033454545454545456, | |
| "loss": 0.3994, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.30716776847839355, | |
| "eval_runtime": 533.0712, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 2.105, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "grad_norm": 0.5875090956687927, | |
| "learning_rate": 0.0003272727272727273, | |
| "loss": 0.3118, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3242183029651642, | |
| "eval_runtime": 538.0823, | |
| "eval_samples_per_second": 8.341, | |
| "eval_steps_per_second": 2.085, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "grad_norm": 1.1133062839508057, | |
| "learning_rate": 0.00032, | |
| "loss": 0.4577, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3053763210773468, | |
| "eval_runtime": 532.813, | |
| "eval_samples_per_second": 8.423, | |
| "eval_steps_per_second": 2.106, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "grad_norm": 1.3879055976867676, | |
| "learning_rate": 0.00031272727272727273, | |
| "loss": 0.3927, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.29549089074134827, | |
| "eval_runtime": 542.0362, | |
| "eval_samples_per_second": 8.28, | |
| "eval_steps_per_second": 2.07, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "grad_norm": 0.12985749542713165, | |
| "learning_rate": 0.0003054545454545455, | |
| "loss": 0.229, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "eval_accuracy": 0.8972816467285156, | |
| "eval_loss": 0.26039257645606995, | |
| "eval_runtime": 536.2206, | |
| "eval_samples_per_second": 8.37, | |
| "eval_steps_per_second": 2.092, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "grad_norm": 1.7885103225708008, | |
| "learning_rate": 0.0002981818181818182, | |
| "loss": 0.2635, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3494693636894226, | |
| "eval_runtime": 538.5549, | |
| "eval_samples_per_second": 8.333, | |
| "eval_steps_per_second": 2.083, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "grad_norm": 0.5339781641960144, | |
| "learning_rate": 0.0002909090909090909, | |
| "loss": 0.4063, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2612840533256531, | |
| "eval_runtime": 538.3622, | |
| "eval_samples_per_second": 8.336, | |
| "eval_steps_per_second": 2.084, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "grad_norm": 1.367612600326538, | |
| "learning_rate": 0.0002836363636363637, | |
| "loss": 0.2946, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2859351336956024, | |
| "eval_runtime": 537.0937, | |
| "eval_samples_per_second": 8.356, | |
| "eval_steps_per_second": 2.089, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "grad_norm": 1.688023567199707, | |
| "learning_rate": 0.0002763636363636364, | |
| "loss": 0.2653, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "eval_accuracy": 0.89371657371521, | |
| "eval_loss": 0.20645655691623688, | |
| "eval_runtime": 544.7393, | |
| "eval_samples_per_second": 8.239, | |
| "eval_steps_per_second": 2.06, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "grad_norm": 1.5258842706680298, | |
| "learning_rate": 0.0002690909090909091, | |
| "loss": 0.2861, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "eval_accuracy": 0.8636363744735718, | |
| "eval_loss": 0.2519168555736542, | |
| "eval_runtime": 536.5715, | |
| "eval_samples_per_second": 8.364, | |
| "eval_steps_per_second": 2.091, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "grad_norm": 0.10538855940103531, | |
| "learning_rate": 0.00026181818181818185, | |
| "loss": 0.2971, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.4605836868286133, | |
| "eval_runtime": 537.7296, | |
| "eval_samples_per_second": 8.346, | |
| "eval_steps_per_second": 2.087, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "grad_norm": 1.2356140613555908, | |
| "learning_rate": 0.00025454545454545456, | |
| "loss": 0.3759, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2962367832660675, | |
| "eval_runtime": 538.9507, | |
| "eval_samples_per_second": 8.327, | |
| "eval_steps_per_second": 2.082, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "grad_norm": 0.343402624130249, | |
| "learning_rate": 0.00024727272727272727, | |
| "loss": 0.2156, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2896555960178375, | |
| "eval_runtime": 537.3494, | |
| "eval_samples_per_second": 8.352, | |
| "eval_steps_per_second": 2.088, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "grad_norm": 0.3443286120891571, | |
| "learning_rate": 0.00024, | |
| "loss": 0.2512, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.25072118639945984, | |
| "eval_runtime": 534.4338, | |
| "eval_samples_per_second": 8.398, | |
| "eval_steps_per_second": 2.099, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "grad_norm": 1.4869568347930908, | |
| "learning_rate": 0.00023272727272727271, | |
| "loss": 0.2509, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "eval_accuracy": 0.8954991102218628, | |
| "eval_loss": 0.19357898831367493, | |
| "eval_runtime": 535.513, | |
| "eval_samples_per_second": 8.381, | |
| "eval_steps_per_second": 2.095, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "grad_norm": 2.3914287090301514, | |
| "learning_rate": 0.00022545454545454545, | |
| "loss": 0.2218, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "eval_accuracy": 0.9015151262283325, | |
| "eval_loss": 0.21406562626361847, | |
| "eval_runtime": 534.6792, | |
| "eval_samples_per_second": 8.394, | |
| "eval_steps_per_second": 2.098, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "grad_norm": 2.300201416015625, | |
| "learning_rate": 0.00021818181818181818, | |
| "loss": 0.3693, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "eval_accuracy": 0.8792335391044617, | |
| "eval_loss": 0.21343877911567688, | |
| "eval_runtime": 530.968, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 2.113, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "grad_norm": 2.3365836143493652, | |
| "learning_rate": 0.0002109090909090909, | |
| "loss": 0.3644, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.22058773040771484, | |
| "eval_runtime": 530.143, | |
| "eval_samples_per_second": 8.466, | |
| "eval_steps_per_second": 2.116, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "grad_norm": 0.33425235748291016, | |
| "learning_rate": 0.00020363636363636363, | |
| "loss": 0.167, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.1814015805721283, | |
| "eval_runtime": 531.0237, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 2.113, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "grad_norm": 1.2170227766036987, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 0.1849, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "eval_accuracy": 0.9012923240661621, | |
| "eval_loss": 0.21576102077960968, | |
| "eval_runtime": 530.0384, | |
| "eval_samples_per_second": 8.467, | |
| "eval_steps_per_second": 2.117, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "grad_norm": 0.8702998161315918, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 0.2305, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "eval_accuracy": 0.9001782536506653, | |
| "eval_loss": 0.16198058426380157, | |
| "eval_runtime": 536.0043, | |
| "eval_samples_per_second": 8.373, | |
| "eval_steps_per_second": 2.093, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "grad_norm": 2.6873793601989746, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.2787, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "eval_accuracy": 0.8970588445663452, | |
| "eval_loss": 0.4286949932575226, | |
| "eval_runtime": 533.2723, | |
| "eval_samples_per_second": 8.416, | |
| "eval_steps_per_second": 2.104, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "grad_norm": 2.6352877616882324, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.4769, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2692972421646118, | |
| "eval_runtime": 537.214, | |
| "eval_samples_per_second": 8.354, | |
| "eval_steps_per_second": 2.089, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "grad_norm": 7.490699291229248, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.2615, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "eval_accuracy": 0.8977272510528564, | |
| "eval_loss": 0.17833727598190308, | |
| "eval_runtime": 530.3846, | |
| "eval_samples_per_second": 8.462, | |
| "eval_steps_per_second": 2.115, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.2769690454006195, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1976, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "eval_accuracy": 0.897504448890686, | |
| "eval_loss": 0.1791936755180359, | |
| "eval_runtime": 529.6531, | |
| "eval_samples_per_second": 8.473, | |
| "eval_steps_per_second": 2.118, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.1449800729751587, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.1132, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "eval_accuracy": 0.8979501128196716, | |
| "eval_loss": 0.17950215935707092, | |
| "eval_runtime": 530.325, | |
| "eval_samples_per_second": 8.463, | |
| "eval_steps_per_second": 2.116, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "grad_norm": 1.8357865810394287, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.2649, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "eval_accuracy": 0.8990641832351685, | |
| "eval_loss": 0.17018087208271027, | |
| "eval_runtime": 528.0517, | |
| "eval_samples_per_second": 8.499, | |
| "eval_steps_per_second": 2.125, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "grad_norm": 1.703097939491272, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.1975, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "eval_accuracy": 0.8992869853973389, | |
| "eval_loss": 0.18035584688186646, | |
| "eval_runtime": 539.3037, | |
| "eval_samples_per_second": 8.322, | |
| "eval_steps_per_second": 2.08, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "grad_norm": 1.1701892614364624, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.143, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "eval_accuracy": 0.897504448890686, | |
| "eval_loss": 0.1814272254705429, | |
| "eval_runtime": 531.4635, | |
| "eval_samples_per_second": 8.445, | |
| "eval_steps_per_second": 2.111, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "grad_norm": 0.24161402881145477, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.1982, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.17482873797416687, | |
| "eval_runtime": 532.8609, | |
| "eval_samples_per_second": 8.422, | |
| "eval_steps_per_second": 2.106, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "grad_norm": 0.49488192796707153, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.1935, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.18158306181430817, | |
| "eval_runtime": 531.1778, | |
| "eval_samples_per_second": 8.449, | |
| "eval_steps_per_second": 2.112, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "grad_norm": 1.5082249641418457, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.1893, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "eval_accuracy": 0.8977272510528564, | |
| "eval_loss": 0.16474968194961548, | |
| "eval_runtime": 531.3613, | |
| "eval_samples_per_second": 8.446, | |
| "eval_steps_per_second": 2.112, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "grad_norm": 0.7104145288467407, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.1793, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "eval_accuracy": 0.9039661288261414, | |
| "eval_loss": 0.16507278382778168, | |
| "eval_runtime": 532.5438, | |
| "eval_samples_per_second": 8.427, | |
| "eval_steps_per_second": 2.107, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "grad_norm": 2.474039316177368, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.1832, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "eval_accuracy": 0.9053030014038086, | |
| "eval_loss": 0.15766561031341553, | |
| "eval_runtime": 530.4857, | |
| "eval_samples_per_second": 8.46, | |
| "eval_steps_per_second": 2.115, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "grad_norm": 0.4223766028881073, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.1833, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "eval_accuracy": 0.9099822044372559, | |
| "eval_loss": 0.16656117141246796, | |
| "eval_runtime": 538.6394, | |
| "eval_samples_per_second": 8.332, | |
| "eval_steps_per_second": 2.083, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "grad_norm": 1.0373200178146362, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1436, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "eval_accuracy": 0.9119875431060791, | |
| "eval_loss": 0.1528429239988327, | |
| "eval_runtime": 536.2551, | |
| "eval_samples_per_second": 8.369, | |
| "eval_steps_per_second": 2.092, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "grad_norm": 2.6555793285369873, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.1856, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "eval_accuracy": 0.9115418791770935, | |
| "eval_loss": 0.14956268668174744, | |
| "eval_runtime": 534.8893, | |
| "eval_samples_per_second": 8.391, | |
| "eval_steps_per_second": 2.098, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "grad_norm": 0.36912649869918823, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.1281, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "eval_accuracy": 0.9144384860992432, | |
| "eval_loss": 0.14639925956726074, | |
| "eval_runtime": 534.7579, | |
| "eval_samples_per_second": 8.393, | |
| "eval_steps_per_second": 2.098, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "grad_norm": 0.9557098150253296, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.1894, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "eval_accuracy": 0.9215686321258545, | |
| "eval_loss": 0.1419779509305954, | |
| "eval_runtime": 535.2427, | |
| "eval_samples_per_second": 8.385, | |
| "eval_steps_per_second": 2.096, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 0.6485058069229126, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.222, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "eval_accuracy": 0.9057486653327942, | |
| "eval_loss": 0.16202405095100403, | |
| "eval_runtime": 539.6942, | |
| "eval_samples_per_second": 8.316, | |
| "eval_steps_per_second": 2.079, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "grad_norm": 0.5341868996620178, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.129, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "eval_accuracy": 0.9226827025413513, | |
| "eval_loss": 0.1420886367559433, | |
| "eval_runtime": 539.6845, | |
| "eval_samples_per_second": 8.316, | |
| "eval_steps_per_second": 2.079, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "grad_norm": 0.11955850571393967, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0887, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "eval_accuracy": 0.9318181872367859, | |
| "eval_loss": 0.13754913210868835, | |
| "eval_runtime": 539.9321, | |
| "eval_samples_per_second": 8.312, | |
| "eval_steps_per_second": 2.078, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 0.711179256439209, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.0965, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "eval_accuracy": 0.9202316999435425, | |
| "eval_loss": 0.16527850925922394, | |
| "eval_runtime": 536.1841, | |
| "eval_samples_per_second": 8.37, | |
| "eval_steps_per_second": 2.093, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "grad_norm": 0.823353111743927, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.2106, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "eval_accuracy": 0.9280303120613098, | |
| "eval_loss": 0.14900797605514526, | |
| "eval_runtime": 535.1968, | |
| "eval_samples_per_second": 8.386, | |
| "eval_steps_per_second": 2.096, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "grad_norm": 3.12156343460083, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.2153, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "eval_accuracy": 0.9347147941589355, | |
| "eval_loss": 0.13268369436264038, | |
| "eval_runtime": 533.2781, | |
| "eval_samples_per_second": 8.416, | |
| "eval_steps_per_second": 2.104, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "grad_norm": 5.130224227905273, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.157, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "eval_accuracy": 0.9318181872367859, | |
| "eval_loss": 0.13205085694789886, | |
| "eval_runtime": 537.2184, | |
| "eval_samples_per_second": 8.354, | |
| "eval_steps_per_second": 2.089, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "grad_norm": 0.49416056275367737, | |
| "learning_rate": 0.0, | |
| "loss": 0.1583, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "eval_accuracy": 0.9304812550544739, | |
| "eval_loss": 0.13242210447788239, | |
| "eval_runtime": 549.6193, | |
| "eval_samples_per_second": 8.166, | |
| "eval_steps_per_second": 2.041, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.326400520422712e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |