| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5868060330995278, | |
| "eval_steps": 200, | |
| "global_step": 6400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "eval_bertscore": 0.7401605248451233, | |
| "eval_loss": 1.9530484676361084, | |
| "eval_rouge1": 0.6562857460474375, | |
| "eval_rouge2": 0.3640670727106235, | |
| "eval_rougeL": 0.5655212336424695, | |
| "eval_rougeLsum": 0.6414840198810386, | |
| "eval_runtime": 21.7196, | |
| "eval_samples_per_second": 1.381, | |
| "eval_steps_per_second": 0.691, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.25105270743370056, | |
| "learning_rate": 0.00019771674842969145, | |
| "loss": 1.7353, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_bertscore": 0.7432050108909607, | |
| "eval_loss": 1.9583823680877686, | |
| "eval_rouge1": 0.6554226269617707, | |
| "eval_rouge2": 0.36661086995296877, | |
| "eval_rougeL": 0.5637448790342183, | |
| "eval_rougeLsum": 0.6419796784912521, | |
| "eval_runtime": 21.9623, | |
| "eval_samples_per_second": 1.366, | |
| "eval_steps_per_second": 0.683, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.26550447940826416, | |
| "learning_rate": 0.00019542432717436156, | |
| "loss": 1.7786, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_bertscore": 0.7469045519828796, | |
| "eval_loss": 1.9245686531066895, | |
| "eval_rouge1": 0.6662431635890791, | |
| "eval_rouge2": 0.3735263724826765, | |
| "eval_rougeL": 0.5755071616151013, | |
| "eval_rougeLsum": 0.6538383087686117, | |
| "eval_runtime": 21.5302, | |
| "eval_samples_per_second": 1.393, | |
| "eval_steps_per_second": 0.697, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.1538015753030777, | |
| "learning_rate": 0.0001931319059190317, | |
| "loss": 1.8851, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_bertscore": 0.7442477941513062, | |
| "eval_loss": 1.9187489748001099, | |
| "eval_rouge1": 0.6606221897489035, | |
| "eval_rouge2": 0.368654563659435, | |
| "eval_rougeL": 0.5731546210408094, | |
| "eval_rougeLsum": 0.6470590823125606, | |
| "eval_runtime": 21.9831, | |
| "eval_samples_per_second": 1.365, | |
| "eval_steps_per_second": 0.682, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.1681252270936966, | |
| "learning_rate": 0.0001908394846637018, | |
| "loss": 1.8919, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_bertscore": 0.7458053231239319, | |
| "eval_loss": 1.9159075021743774, | |
| "eval_rouge1": 0.6621259186456026, | |
| "eval_rouge2": 0.372024043683234, | |
| "eval_rougeL": 0.5743354509339939, | |
| "eval_rougeLsum": 0.6491550893780276, | |
| "eval_runtime": 21.7159, | |
| "eval_samples_per_second": 1.381, | |
| "eval_steps_per_second": 0.691, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_bertscore": 0.7468854784965515, | |
| "eval_loss": 1.9140182733535767, | |
| "eval_rouge1": 0.6626581781149132, | |
| "eval_rouge2": 0.37318557504782157, | |
| "eval_rougeL": 0.5759264203594217, | |
| "eval_rougeLsum": 0.6490702446275723, | |
| "eval_runtime": 21.6486, | |
| "eval_samples_per_second": 1.386, | |
| "eval_steps_per_second": 0.693, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.1552441120147705, | |
| "learning_rate": 0.00018854706340837193, | |
| "loss": 1.9052, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_bertscore": 0.7475314736366272, | |
| "eval_loss": 1.913794755935669, | |
| "eval_rouge1": 0.6648687174353192, | |
| "eval_rouge2": 0.3760379232448734, | |
| "eval_rougeL": 0.5784915488164926, | |
| "eval_rougeLsum": 0.6513864520108938, | |
| "eval_runtime": 21.664, | |
| "eval_samples_per_second": 1.385, | |
| "eval_steps_per_second": 0.692, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.14638397097587585, | |
| "learning_rate": 0.00018625464215304204, | |
| "loss": 1.8843, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_bertscore": 0.747238039970398, | |
| "eval_loss": 1.9117029905319214, | |
| "eval_rouge1": 0.6638085237198453, | |
| "eval_rouge2": 0.3742779818055127, | |
| "eval_rougeL": 0.5754209460423059, | |
| "eval_rougeLsum": 0.6506476155592722, | |
| "eval_runtime": 21.9308, | |
| "eval_samples_per_second": 1.368, | |
| "eval_steps_per_second": 0.684, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.15738993883132935, | |
| "learning_rate": 0.00018396222089771218, | |
| "loss": 1.8964, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_bertscore": 0.7473016381263733, | |
| "eval_loss": 1.9117563962936401, | |
| "eval_rouge1": 0.6620053151663765, | |
| "eval_rouge2": 0.37406692119411245, | |
| "eval_rougeL": 0.5758911607323577, | |
| "eval_rougeLsum": 0.6494070575604445, | |
| "eval_runtime": 21.6727, | |
| "eval_samples_per_second": 1.384, | |
| "eval_steps_per_second": 0.692, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.1588907092809677, | |
| "learning_rate": 0.00018166979964238228, | |
| "loss": 1.8827, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_bertscore": 0.7485987544059753, | |
| "eval_loss": 1.9126006364822388, | |
| "eval_rouge1": 0.6641836156334741, | |
| "eval_rouge2": 0.37320215574735827, | |
| "eval_rougeL": 0.5783015040447993, | |
| "eval_rougeLsum": 0.6522235940423647, | |
| "eval_runtime": 21.9759, | |
| "eval_samples_per_second": 1.365, | |
| "eval_steps_per_second": 0.683, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bertscore": 0.7482583522796631, | |
| "eval_loss": 1.9075205326080322, | |
| "eval_rouge1": 0.6658219484766166, | |
| "eval_rouge2": 0.37723364952258465, | |
| "eval_rougeL": 0.5769040785174693, | |
| "eval_rougeLsum": 0.6511328888044219, | |
| "eval_runtime": 21.5892, | |
| "eval_samples_per_second": 1.39, | |
| "eval_steps_per_second": 0.695, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.15247465670108795, | |
| "learning_rate": 0.00017937737838705242, | |
| "loss": 1.8831, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_bertscore": 0.7460805177688599, | |
| "eval_loss": 1.9088668823242188, | |
| "eval_rouge1": 0.6627321043292516, | |
| "eval_rouge2": 0.3696581195003696, | |
| "eval_rougeL": 0.5740988544467178, | |
| "eval_rougeLsum": 0.6478729042661874, | |
| "eval_runtime": 21.9221, | |
| "eval_samples_per_second": 1.368, | |
| "eval_steps_per_second": 0.684, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.1587379276752472, | |
| "learning_rate": 0.00017708495713172253, | |
| "loss": 1.8829, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_bertscore": 0.7472203373908997, | |
| "eval_loss": 1.906219482421875, | |
| "eval_rouge1": 0.6637415370426804, | |
| "eval_rouge2": 0.37565276875837994, | |
| "eval_rougeL": 0.5773879369079004, | |
| "eval_rougeLsum": 0.6488719947518645, | |
| "eval_runtime": 21.8112, | |
| "eval_samples_per_second": 1.375, | |
| "eval_steps_per_second": 0.688, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.1558646410703659, | |
| "learning_rate": 0.00017479253587639266, | |
| "loss": 1.8978, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_bertscore": 0.7466126680374146, | |
| "eval_loss": 1.9045982360839844, | |
| "eval_rouge1": 0.6616225540296956, | |
| "eval_rouge2": 0.37370762164745913, | |
| "eval_rougeL": 0.5759418528371097, | |
| "eval_rougeLsum": 0.6479977636906877, | |
| "eval_runtime": 21.8772, | |
| "eval_samples_per_second": 1.371, | |
| "eval_steps_per_second": 0.686, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.14783035218715668, | |
| "learning_rate": 0.00017250011462106277, | |
| "loss": 1.8978, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_bertscore": 0.7485571503639221, | |
| "eval_loss": 1.9035439491271973, | |
| "eval_rouge1": 0.6664050030501707, | |
| "eval_rouge2": 0.379492440917784, | |
| "eval_rougeL": 0.5806973731221475, | |
| "eval_rougeLsum": 0.6524346156604702, | |
| "eval_runtime": 21.9217, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.684, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_bertscore": 0.7483461499214172, | |
| "eval_loss": 1.9022458791732788, | |
| "eval_rouge1": 0.6618989733136488, | |
| "eval_rouge2": 0.37377379177271053, | |
| "eval_rougeL": 0.5780989082173933, | |
| "eval_rougeLsum": 0.6490379362631586, | |
| "eval_runtime": 21.7847, | |
| "eval_samples_per_second": 1.377, | |
| "eval_steps_per_second": 0.689, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.16484151780605316, | |
| "learning_rate": 0.0001702076933657329, | |
| "loss": 1.8715, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_bertscore": 0.7490711212158203, | |
| "eval_loss": 1.9013088941574097, | |
| "eval_rouge1": 0.6638141306545007, | |
| "eval_rouge2": 0.37356255553691553, | |
| "eval_rougeL": 0.577975450251653, | |
| "eval_rougeLsum": 0.6492478632295806, | |
| "eval_runtime": 21.8807, | |
| "eval_samples_per_second": 1.371, | |
| "eval_steps_per_second": 0.686, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.14130128920078278, | |
| "learning_rate": 0.000167915272110403, | |
| "loss": 1.8819, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_bertscore": 0.7475283741950989, | |
| "eval_loss": 1.9002223014831543, | |
| "eval_rouge1": 0.6628836314413511, | |
| "eval_rouge2": 0.37179988805094977, | |
| "eval_rougeL": 0.5764222388923268, | |
| "eval_rougeLsum": 0.649864229310889, | |
| "eval_runtime": 22.124, | |
| "eval_samples_per_second": 1.356, | |
| "eval_steps_per_second": 0.678, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.1494186818599701, | |
| "learning_rate": 0.00016562285085507315, | |
| "loss": 1.8828, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_bertscore": 0.7486498951911926, | |
| "eval_loss": 1.9011151790618896, | |
| "eval_rouge1": 0.6669673680023924, | |
| "eval_rouge2": 0.3771780440183751, | |
| "eval_rougeL": 0.5792518624130161, | |
| "eval_rougeLsum": 0.6534484242953056, | |
| "eval_runtime": 21.813, | |
| "eval_samples_per_second": 1.375, | |
| "eval_steps_per_second": 0.688, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.14803479611873627, | |
| "learning_rate": 0.00016333042959974325, | |
| "loss": 1.8761, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_bertscore": 0.7471507787704468, | |
| "eval_loss": 1.9001713991165161, | |
| "eval_rouge1": 0.6651735220672027, | |
| "eval_rouge2": 0.3736698451416937, | |
| "eval_rougeL": 0.5779938808281732, | |
| "eval_rougeLsum": 0.6509815118131576, | |
| "eval_runtime": 21.5004, | |
| "eval_samples_per_second": 1.395, | |
| "eval_steps_per_second": 0.698, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bertscore": 0.7485501766204834, | |
| "eval_loss": 1.8993827104568481, | |
| "eval_rouge1": 0.6646424082737133, | |
| "eval_rouge2": 0.37318485364862475, | |
| "eval_rougeL": 0.5773338159759467, | |
| "eval_rougeLsum": 0.6507594353103527, | |
| "eval_runtime": 21.2963, | |
| "eval_samples_per_second": 1.409, | |
| "eval_steps_per_second": 0.704, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.15562959015369415, | |
| "learning_rate": 0.0001610380083444134, | |
| "loss": 1.8672, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_bertscore": 0.7469989061355591, | |
| "eval_loss": 1.900540828704834, | |
| "eval_rouge1": 0.6620664558691891, | |
| "eval_rouge2": 0.37299419371215703, | |
| "eval_rougeL": 0.5765442194831125, | |
| "eval_rougeLsum": 0.6472642385429858, | |
| "eval_runtime": 21.9086, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.685, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.15420928597450256, | |
| "learning_rate": 0.0001587455870890835, | |
| "loss": 1.8754, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_bertscore": 0.7475299835205078, | |
| "eval_loss": 1.8988685607910156, | |
| "eval_rouge1": 0.6656661780424216, | |
| "eval_rouge2": 0.37467258880478527, | |
| "eval_rougeL": 0.5770800519970718, | |
| "eval_rougeLsum": 0.6522703864288166, | |
| "eval_runtime": 22.063, | |
| "eval_samples_per_second": 1.36, | |
| "eval_steps_per_second": 0.68, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.15809176862239838, | |
| "learning_rate": 0.00015645316583375363, | |
| "loss": 1.8848, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_bertscore": 0.7490234375, | |
| "eval_loss": 1.8991097211837769, | |
| "eval_rouge1": 0.6651730257289085, | |
| "eval_rouge2": 0.3778893043274054, | |
| "eval_rougeL": 0.5782673838033503, | |
| "eval_rougeLsum": 0.6516865674488727, | |
| "eval_runtime": 22.0202, | |
| "eval_samples_per_second": 1.362, | |
| "eval_steps_per_second": 0.681, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.17979757487773895, | |
| "learning_rate": 0.00015416074457842374, | |
| "loss": 1.8851, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_bertscore": 0.7492111325263977, | |
| "eval_loss": 1.897339940071106, | |
| "eval_rouge1": 0.665920573890169, | |
| "eval_rouge2": 0.37917993898535385, | |
| "eval_rougeL": 0.5800236892888617, | |
| "eval_rougeLsum": 0.6529131688355863, | |
| "eval_runtime": 21.6103, | |
| "eval_samples_per_second": 1.388, | |
| "eval_steps_per_second": 0.694, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_bertscore": 0.7491253614425659, | |
| "eval_loss": 1.897528052330017, | |
| "eval_rouge1": 0.6653452054219615, | |
| "eval_rouge2": 0.3759208437918665, | |
| "eval_rougeL": 0.5776757077854651, | |
| "eval_rougeLsum": 0.6511876484723524, | |
| "eval_runtime": 21.3101, | |
| "eval_samples_per_second": 1.408, | |
| "eval_steps_per_second": 0.704, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.16869671642780304, | |
| "learning_rate": 0.00015186832332309387, | |
| "loss": 1.8783, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_bertscore": 0.7494469881057739, | |
| "eval_loss": 1.895969271659851, | |
| "eval_rouge1": 0.6660951369469854, | |
| "eval_rouge2": 0.3764077134133328, | |
| "eval_rougeL": 0.578785826234568, | |
| "eval_rougeLsum": 0.6525967284041656, | |
| "eval_runtime": 21.7955, | |
| "eval_samples_per_second": 1.376, | |
| "eval_steps_per_second": 0.688, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.15996231138706207, | |
| "learning_rate": 0.00014957590206776398, | |
| "loss": 1.8805, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_bertscore": 0.7486470341682434, | |
| "eval_loss": 1.8955131769180298, | |
| "eval_rouge1": 0.6670292173522965, | |
| "eval_rouge2": 0.37457018529010144, | |
| "eval_rougeL": 0.5775243235432015, | |
| "eval_rougeLsum": 0.652574079807632, | |
| "eval_runtime": 21.7576, | |
| "eval_samples_per_second": 1.379, | |
| "eval_steps_per_second": 0.689, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.17192547023296356, | |
| "learning_rate": 0.00014728348081243412, | |
| "loss": 1.8884, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_bertscore": 0.7483081817626953, | |
| "eval_loss": 1.895763874053955, | |
| "eval_rouge1": 0.6659275328276997, | |
| "eval_rouge2": 0.3778666475350364, | |
| "eval_rougeL": 0.579425140056643, | |
| "eval_rougeLsum": 0.6515870828784887, | |
| "eval_runtime": 21.6648, | |
| "eval_samples_per_second": 1.385, | |
| "eval_steps_per_second": 0.692, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.15838442742824554, | |
| "learning_rate": 0.00014499105955710422, | |
| "loss": 1.8913, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_bertscore": 0.7493732571601868, | |
| "eval_loss": 1.8914682865142822, | |
| "eval_rouge1": 0.6669695240447069, | |
| "eval_rouge2": 0.3769441114214874, | |
| "eval_rougeL": 0.5798986667152066, | |
| "eval_rougeLsum": 0.6534527583592111, | |
| "eval_runtime": 21.4686, | |
| "eval_samples_per_second": 1.397, | |
| "eval_steps_per_second": 0.699, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_bertscore": 0.7510559558868408, | |
| "eval_loss": 1.8923884630203247, | |
| "eval_rouge1": 0.6677938121282943, | |
| "eval_rouge2": 0.37854575387307554, | |
| "eval_rougeL": 0.5817052753830161, | |
| "eval_rougeLsum": 0.6534737907551461, | |
| "eval_runtime": 21.593, | |
| "eval_samples_per_second": 1.389, | |
| "eval_steps_per_second": 0.695, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.15312573313713074, | |
| "learning_rate": 0.00014269863830177433, | |
| "loss": 1.8705, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_bertscore": 0.7479371428489685, | |
| "eval_loss": 1.891802430152893, | |
| "eval_rouge1": 0.6658674357402252, | |
| "eval_rouge2": 0.3757712649269345, | |
| "eval_rougeL": 0.5791817270712349, | |
| "eval_rougeLsum": 0.6509960265397259, | |
| "eval_runtime": 21.8726, | |
| "eval_samples_per_second": 1.372, | |
| "eval_steps_per_second": 0.686, | |
| "step": 6400 | |
| } | |
| ], | |
| "logging_steps": 250, | |
| "max_steps": 21812, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 800, | |
| "total_flos": 4.31434028285952e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |