| { | |
| "best_metric": 0.9261695691084951, | |
| "best_model_checkpoint": "cls_comment-phobert-base-v2-v3.2.1/checkpoint-2000", | |
| "epoch": 26.08695652173913, | |
| "eval_steps": 100, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 2.4986989498138428, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.8947, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_accuracy": 0.4001087251970644, | |
| "eval_f1_score": 0.08320411950694513, | |
| "eval_loss": 1.68748140335083, | |
| "eval_precision": 0.14637409036074248, | |
| "eval_recall": 0.14367816091954025, | |
| "eval_runtime": 6.6876, | |
| "eval_samples_per_second": 550.125, | |
| "eval_steps_per_second": 8.673, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.596021294593811, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5395, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_accuracy": 0.5849415602065778, | |
| "eval_f1_score": 0.2355807809182458, | |
| "eval_loss": 1.2897096872329712, | |
| "eval_precision": 0.27516139357553443, | |
| "eval_recall": 0.26320585050663253, | |
| "eval_runtime": 6.7112, | |
| "eval_samples_per_second": 548.189, | |
| "eval_steps_per_second": 8.642, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 5.161496162414551, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.1205, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "eval_accuracy": 0.7999456374014677, | |
| "eval_f1_score": 0.5833178086765388, | |
| "eval_loss": 0.8468331098556519, | |
| "eval_precision": 0.5889764394952819, | |
| "eval_recall": 0.5810488238671468, | |
| "eval_runtime": 6.6686, | |
| "eval_samples_per_second": 551.694, | |
| "eval_steps_per_second": 8.698, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 5.227330207824707, | |
| "learning_rate": 1e-05, | |
| "loss": 0.82, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "eval_accuracy": 0.8369122044033704, | |
| "eval_f1_score": 0.6179371343772609, | |
| "eval_loss": 0.6537477374076843, | |
| "eval_precision": 0.6062100200393906, | |
| "eval_recall": 0.6355302315827523, | |
| "eval_runtime": 6.728, | |
| "eval_samples_per_second": 546.823, | |
| "eval_steps_per_second": 8.621, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 5.6816534996032715, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.6232, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "eval_accuracy": 0.8537646099483556, | |
| "eval_f1_score": 0.633743239294036, | |
| "eval_loss": 0.537100613117218, | |
| "eval_precision": 0.7525070200257705, | |
| "eval_recall": 0.6518017678843925, | |
| "eval_runtime": 6.7932, | |
| "eval_samples_per_second": 541.573, | |
| "eval_steps_per_second": 8.538, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 5.096814155578613, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.5148, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "eval_accuracy": 0.872791519434629, | |
| "eval_f1_score": 0.7299293979398146, | |
| "eval_loss": 0.46505650877952576, | |
| "eval_precision": 0.7548552896750885, | |
| "eval_recall": 0.7210618976649555, | |
| "eval_runtime": 6.7028, | |
| "eval_samples_per_second": 548.875, | |
| "eval_steps_per_second": 8.653, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.086956521739131, | |
| "grad_norm": 5.458530902862549, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.4204, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.086956521739131, | |
| "eval_accuracy": 0.8869257950530035, | |
| "eval_f1_score": 0.7654329783869755, | |
| "eval_loss": 0.40097591280937195, | |
| "eval_precision": 0.8914471413846636, | |
| "eval_recall": 0.7712133932759179, | |
| "eval_runtime": 6.7443, | |
| "eval_samples_per_second": 545.497, | |
| "eval_steps_per_second": 8.6, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 6.144416809082031, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.3421, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_accuracy": 0.9051372655612938, | |
| "eval_f1_score": 0.8713582894968701, | |
| "eval_loss": 0.3648029565811157, | |
| "eval_precision": 0.8940734807154502, | |
| "eval_recall": 0.8588405388993653, | |
| "eval_runtime": 6.7352, | |
| "eval_samples_per_second": 546.232, | |
| "eval_steps_per_second": 8.611, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.826086956521739, | |
| "grad_norm": 9.907292366027832, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.2841, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.826086956521739, | |
| "eval_accuracy": 0.9181842892090242, | |
| "eval_f1_score": 0.9006880118200489, | |
| "eval_loss": 0.3239505887031555, | |
| "eval_precision": 0.8978148514278343, | |
| "eval_recall": 0.9038252102525616, | |
| "eval_runtime": 6.7315, | |
| "eval_samples_per_second": 546.532, | |
| "eval_steps_per_second": 8.616, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 6.941843032836914, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.2319, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_accuracy": 0.9203587931503125, | |
| "eval_f1_score": 0.9060754755748909, | |
| "eval_loss": 0.3025033473968506, | |
| "eval_precision": 0.9175362378163865, | |
| "eval_recall": 0.8975903509513042, | |
| "eval_runtime": 6.6908, | |
| "eval_samples_per_second": 549.861, | |
| "eval_steps_per_second": 8.669, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.565217391304348, | |
| "grad_norm": 4.255012035369873, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.205, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.565217391304348, | |
| "eval_accuracy": 0.9209024191356346, | |
| "eval_f1_score": 0.9098640550303895, | |
| "eval_loss": 0.29862046241760254, | |
| "eval_precision": 0.9123097696068861, | |
| "eval_recall": 0.9086287269577242, | |
| "eval_runtime": 6.7134, | |
| "eval_samples_per_second": 548.01, | |
| "eval_steps_per_second": 8.639, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 5.848569393157959, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.1783, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "eval_accuracy": 0.9206306061429737, | |
| "eval_f1_score": 0.9104384776037051, | |
| "eval_loss": 0.3047122657299042, | |
| "eval_precision": 0.9024848857165658, | |
| "eval_recall": 0.9207396220750284, | |
| "eval_runtime": 6.6561, | |
| "eval_samples_per_second": 552.726, | |
| "eval_steps_per_second": 8.714, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.304347826086957, | |
| "grad_norm": 7.340043544769287, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1587, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.304347826086957, | |
| "eval_accuracy": 0.9296004349007883, | |
| "eval_f1_score": 0.9202832724978299, | |
| "eval_loss": 0.2757803201675415, | |
| "eval_precision": 0.9233347498988893, | |
| "eval_recall": 0.917658614989255, | |
| "eval_runtime": 6.6787, | |
| "eval_samples_per_second": 550.859, | |
| "eval_steps_per_second": 8.684, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 5.315700054168701, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.1286, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "eval_accuracy": 0.9266104919815167, | |
| "eval_f1_score": 0.9144278995332229, | |
| "eval_loss": 0.29267847537994385, | |
| "eval_precision": 0.9100638576136009, | |
| "eval_recall": 0.9198715425139269, | |
| "eval_runtime": 6.7676, | |
| "eval_samples_per_second": 543.619, | |
| "eval_steps_per_second": 8.57, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "grad_norm": 5.173799514770508, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.1221, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "eval_accuracy": 0.9317749388420766, | |
| "eval_f1_score": 0.9245023460604546, | |
| "eval_loss": 0.28211963176727295, | |
| "eval_precision": 0.9309417300478454, | |
| "eval_recall": 0.9193579289135766, | |
| "eval_runtime": 6.7091, | |
| "eval_samples_per_second": 548.359, | |
| "eval_steps_per_second": 8.645, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 8.639619827270508, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.1087, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_accuracy": 0.9293286219081273, | |
| "eval_f1_score": 0.9159607873769989, | |
| "eval_loss": 0.27890825271606445, | |
| "eval_precision": 0.9090390134661626, | |
| "eval_recall": 0.9236924050215896, | |
| "eval_runtime": 6.7017, | |
| "eval_samples_per_second": 548.966, | |
| "eval_steps_per_second": 8.655, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.782608695652174, | |
| "grad_norm": 6.424872398376465, | |
| "learning_rate": 6.3888888888888885e-06, | |
| "loss": 0.0982, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 14.782608695652174, | |
| "eval_accuracy": 0.9290568089154662, | |
| "eval_f1_score": 0.9196461825352048, | |
| "eval_loss": 0.2833573520183563, | |
| "eval_precision": 0.9187836187318232, | |
| "eval_recall": 0.9213402050339831, | |
| "eval_runtime": 6.7096, | |
| "eval_samples_per_second": 548.32, | |
| "eval_steps_per_second": 8.644, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "grad_norm": 4.618613243103027, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.089, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "eval_accuracy": 0.9298722478934494, | |
| "eval_f1_score": 0.9202166850732406, | |
| "eval_loss": 0.28280356526374817, | |
| "eval_precision": 0.9151663252588741, | |
| "eval_recall": 0.9260674008256092, | |
| "eval_runtime": 6.7345, | |
| "eval_samples_per_second": 546.292, | |
| "eval_steps_per_second": 8.612, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.52173913043478, | |
| "grad_norm": 1.9568698406219482, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0795, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 16.52173913043478, | |
| "eval_accuracy": 0.9331340038053819, | |
| "eval_f1_score": 0.9244095368032713, | |
| "eval_loss": 0.273701936006546, | |
| "eval_precision": 0.925343846727414, | |
| "eval_recall": 0.9238732382441093, | |
| "eval_runtime": 6.7425, | |
| "eval_samples_per_second": 545.641, | |
| "eval_steps_per_second": 8.602, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 2.161759614944458, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.0684, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "eval_accuracy": 0.9323185648273987, | |
| "eval_f1_score": 0.9261695691084951, | |
| "eval_loss": 0.2873239815235138, | |
| "eval_precision": 0.9319834922740233, | |
| "eval_recall": 0.9216726996777184, | |
| "eval_runtime": 6.7577, | |
| "eval_samples_per_second": 544.415, | |
| "eval_steps_per_second": 8.583, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.26086956521739, | |
| "grad_norm": 4.607916355133057, | |
| "learning_rate": 5.2777777777777785e-06, | |
| "loss": 0.0673, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 18.26086956521739, | |
| "eval_accuracy": 0.9320467518347377, | |
| "eval_f1_score": 0.925184613434992, | |
| "eval_loss": 0.2904324531555176, | |
| "eval_precision": 0.9332741752610002, | |
| "eval_recall": 0.9184445089519725, | |
| "eval_runtime": 6.7294, | |
| "eval_samples_per_second": 546.702, | |
| "eval_steps_per_second": 8.619, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.130434782608695, | |
| "grad_norm": 6.327251434326172, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0571, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 19.130434782608695, | |
| "eval_accuracy": 0.9293286219081273, | |
| "eval_f1_score": 0.9221668516434853, | |
| "eval_loss": 0.3166205883026123, | |
| "eval_precision": 0.925137476734381, | |
| "eval_recall": 0.920952609526737, | |
| "eval_runtime": 6.7037, | |
| "eval_samples_per_second": 548.799, | |
| "eval_steps_per_second": 8.652, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 9.082805633544922, | |
| "learning_rate": 4.722222222222222e-06, | |
| "loss": 0.0561, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9317749388420766, | |
| "eval_f1_score": 0.9221280725480369, | |
| "eval_loss": 0.2922111749649048, | |
| "eval_precision": 0.9150274852978553, | |
| "eval_recall": 0.9297539237688469, | |
| "eval_runtime": 6.7503, | |
| "eval_samples_per_second": 545.017, | |
| "eval_steps_per_second": 8.592, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 5.283856391906738, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.0511, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_accuracy": 0.9315031258494156, | |
| "eval_f1_score": 0.9190655007648246, | |
| "eval_loss": 0.29927295446395874, | |
| "eval_precision": 0.9088064828335735, | |
| "eval_recall": 0.9303236730969998, | |
| "eval_runtime": 6.7281, | |
| "eval_samples_per_second": 546.809, | |
| "eval_steps_per_second": 8.621, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "grad_norm": 6.0074896812438965, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0442, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "eval_accuracy": 0.9266104919815167, | |
| "eval_f1_score": 0.9161795338292905, | |
| "eval_loss": 0.32011494040489197, | |
| "eval_precision": 0.9060451440252857, | |
| "eval_recall": 0.9280493422296427, | |
| "eval_runtime": 6.7127, | |
| "eval_samples_per_second": 548.067, | |
| "eval_steps_per_second": 8.64, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.608695652173914, | |
| "grad_norm": 3.1078407764434814, | |
| "learning_rate": 3.88888888888889e-06, | |
| "loss": 0.0447, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 22.608695652173914, | |
| "eval_accuracy": 0.928241369937483, | |
| "eval_f1_score": 0.9137497551284842, | |
| "eval_loss": 0.3155056834220886, | |
| "eval_precision": 0.9009580466238951, | |
| "eval_recall": 0.9281730038314259, | |
| "eval_runtime": 6.7337, | |
| "eval_samples_per_second": 546.354, | |
| "eval_steps_per_second": 8.613, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 23.47826086956522, | |
| "grad_norm": 2.9584195613861084, | |
| "learning_rate": 3.6111111111111115e-06, | |
| "loss": 0.0415, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 23.47826086956522, | |
| "eval_accuracy": 0.9334058167980429, | |
| "eval_f1_score": 0.9226018260362496, | |
| "eval_loss": 0.30177852511405945, | |
| "eval_precision": 0.9185179495480513, | |
| "eval_recall": 0.9269833265460256, | |
| "eval_runtime": 6.7411, | |
| "eval_samples_per_second": 545.757, | |
| "eval_steps_per_second": 8.604, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 12.190321922302246, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0359, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "eval_accuracy": 0.9298722478934494, | |
| "eval_f1_score": 0.9177278989948806, | |
| "eval_loss": 0.31918126344680786, | |
| "eval_precision": 0.9062664068560837, | |
| "eval_recall": 0.9308234663752396, | |
| "eval_runtime": 6.7802, | |
| "eval_samples_per_second": 542.606, | |
| "eval_steps_per_second": 8.554, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 25.217391304347824, | |
| "grad_norm": 0.2598835527896881, | |
| "learning_rate": 3.055555555555556e-06, | |
| "loss": 0.0369, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 25.217391304347824, | |
| "eval_accuracy": 0.933677629790704, | |
| "eval_f1_score": 0.9210521238209074, | |
| "eval_loss": 0.3063570559024811, | |
| "eval_precision": 0.9140578271273506, | |
| "eval_recall": 0.9285610502121662, | |
| "eval_runtime": 6.7729, | |
| "eval_samples_per_second": 543.197, | |
| "eval_steps_per_second": 8.564, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.24433408677577972, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 0.0296, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "eval_accuracy": 0.9328621908127208, | |
| "eval_f1_score": 0.9237047805131925, | |
| "eval_loss": 0.311038613319397, | |
| "eval_precision": 0.9198460229141495, | |
| "eval_recall": 0.9279424126946928, | |
| "eval_runtime": 6.8161, | |
| "eval_samples_per_second": 539.754, | |
| "eval_steps_per_second": 8.509, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "step": 3000, | |
| "total_flos": 6579999363349350.0, | |
| "train_loss": 0.31640464369455973, | |
| "train_runtime": 3238.6183, | |
| "train_samples_per_second": 158.092, | |
| "train_steps_per_second": 1.235 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 35, | |
| "save_steps": 100, | |
| "total_flos": 6579999363349350.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |