| { |
| "best_global_step": 3500, |
| "best_metric": 134.6686553955078, |
| "best_model_checkpoint": "/kaggle/working/checkpoints/checkpoint-3500", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 3678, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.04079967360261118, |
| "grad_norm": 43363.63671875, |
| "learning_rate": 4.9000000000000005e-06, |
| "loss": 9747.83125, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08159934720522236, |
| "grad_norm": 86221.15625, |
| "learning_rate": 9.900000000000002e-06, |
| "loss": 9771.861875, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.12239902080783353, |
| "grad_norm": 60229.48046875, |
| "learning_rate": 1.49e-05, |
| "loss": 9744.386875, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.1631986944104447, |
| "grad_norm": 37425.95703125, |
| "learning_rate": 1.9900000000000003e-05, |
| "loss": 9638.84125, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.2039983680130559, |
| "grad_norm": 33435.8125, |
| "learning_rate": 2.4900000000000002e-05, |
| "loss": 9571.8575, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.24479804161566707, |
| "grad_norm": 49625.01171875, |
| "learning_rate": 2.9900000000000002e-05, |
| "loss": 9435.87875, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.28559771521827826, |
| "grad_norm": 29603.341796875, |
| "learning_rate": 3.49e-05, |
| "loss": 9189.1225, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.3263973888208894, |
| "grad_norm": 21404.0546875, |
| "learning_rate": 3.99e-05, |
| "loss": 8747.344375, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.3671970624235006, |
| "grad_norm": 51160.33984375, |
| "learning_rate": 4.49e-05, |
| "loss": 8256.139375, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.4079967360261118, |
| "grad_norm": 24317.361328125, |
| "learning_rate": 4.99e-05, |
| "loss": 7713.749375, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4079967360261118, |
| "eval_loss": 1422.8065185546875, |
| "eval_runtime": 12.9348, |
| "eval_samples_per_second": 61.308, |
| "eval_steps_per_second": 7.731, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.44879640962872297, |
| "grad_norm": 27936.181640625, |
| "learning_rate": 4.9229074889867845e-05, |
| "loss": 7073.478125, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.48959608323133413, |
| "grad_norm": 13936.109375, |
| "learning_rate": 4.844241661422278e-05, |
| "loss": 6461.896875, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.5303957568339454, |
| "grad_norm": 15602.5537109375, |
| "learning_rate": 4.765575833857772e-05, |
| "loss": 5912.413125, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.5711954304365565, |
| "grad_norm": 11296.9208984375, |
| "learning_rate": 4.6869100062932666e-05, |
| "loss": 5414.915625, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.6119951040391677, |
| "grad_norm": 6740.5146484375, |
| "learning_rate": 4.6082441787287604e-05, |
| "loss": 4977.4490625, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.6527947776417788, |
| "grad_norm": 8457.17578125, |
| "learning_rate": 4.529578351164254e-05, |
| "loss": 4560.69625, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.69359445124439, |
| "grad_norm": 4019.813720703125, |
| "learning_rate": 4.4509125235997486e-05, |
| "loss": 4228.5184375, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.7343941248470012, |
| "grad_norm": 4167.255859375, |
| "learning_rate": 4.3722466960352424e-05, |
| "loss": 3887.0775, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.7751937984496124, |
| "grad_norm": 4534.84765625, |
| "learning_rate": 4.293580868470736e-05, |
| "loss": 3555.156875, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.8159934720522236, |
| "grad_norm": 2833.71875, |
| "learning_rate": 4.2149150409062306e-05, |
| "loss": 3212.7846875, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8159934720522236, |
| "eval_loss": 534.9082641601562, |
| "eval_runtime": 12.9328, |
| "eval_samples_per_second": 61.317, |
| "eval_steps_per_second": 7.732, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8567931456548348, |
| "grad_norm": 987.5806274414062, |
| "learning_rate": 4.1362492133417244e-05, |
| "loss": 2972.9928125, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.8975928192574459, |
| "grad_norm": 1090.98583984375, |
| "learning_rate": 4.057583385777218e-05, |
| "loss": 2711.8125, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9383924928600571, |
| "grad_norm": 821.1162109375, |
| "learning_rate": 3.978917558212713e-05, |
| "loss": 2529.5009375, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.9791921664626683, |
| "grad_norm": 1336.107666015625, |
| "learning_rate": 3.9002517306482065e-05, |
| "loss": 2375.62984375, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.0195838433292534, |
| "grad_norm": 349.67083740234375, |
| "learning_rate": 3.8215859030837e-05, |
| "loss": 2197.37, |
| "step": 1250 |
| }, |
| { |
| "epoch": 1.0603835169318645, |
| "grad_norm": 352.9945373535156, |
| "learning_rate": 3.742920075519195e-05, |
| "loss": 2055.79296875, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.1011831905344758, |
| "grad_norm": 140.80178833007812, |
| "learning_rate": 3.6642542479546885e-05, |
| "loss": 1932.67640625, |
| "step": 1350 |
| }, |
| { |
| "epoch": 1.1419828641370868, |
| "grad_norm": 157.03147888183594, |
| "learning_rate": 3.585588420390183e-05, |
| "loss": 1804.7475, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.182782537739698, |
| "grad_norm": 395.89825439453125, |
| "learning_rate": 3.506922592825677e-05, |
| "loss": 1648.63546875, |
| "step": 1450 |
| }, |
| { |
| "epoch": 1.2235822113423094, |
| "grad_norm": 237.44638061523438, |
| "learning_rate": 3.4282567652611705e-05, |
| "loss": 1562.99953125, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2235822113423094, |
| "eval_loss": 245.6854705810547, |
| "eval_runtime": 12.9936, |
| "eval_samples_per_second": 61.03, |
| "eval_steps_per_second": 7.696, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2643818849449204, |
| "grad_norm": 631.6441650390625, |
| "learning_rate": 3.349590937696665e-05, |
| "loss": 1483.0425, |
| "step": 1550 |
| }, |
| { |
| "epoch": 1.3051815585475315, |
| "grad_norm": 384.0088195800781, |
| "learning_rate": 3.270925110132159e-05, |
| "loss": 1418.34265625, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.3459812321501428, |
| "grad_norm": 693.582763671875, |
| "learning_rate": 3.1922592825676526e-05, |
| "loss": 1389.78703125, |
| "step": 1650 |
| }, |
| { |
| "epoch": 1.386780905752754, |
| "grad_norm": 96.37037658691406, |
| "learning_rate": 3.113593455003147e-05, |
| "loss": 1321.4690625, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.427580579355365, |
| "grad_norm": 189.0079345703125, |
| "learning_rate": 3.0349276274386408e-05, |
| "loss": 1287.67078125, |
| "step": 1750 |
| }, |
| { |
| "epoch": 1.4683802529579764, |
| "grad_norm": 242.67160034179688, |
| "learning_rate": 2.956261799874135e-05, |
| "loss": 1258.1340625, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.5091799265605874, |
| "grad_norm": 289.3849182128906, |
| "learning_rate": 2.877595972309629e-05, |
| "loss": 1227.222265625, |
| "step": 1850 |
| }, |
| { |
| "epoch": 1.5499796001631987, |
| "grad_norm": 165.50555419921875, |
| "learning_rate": 2.798930144745123e-05, |
| "loss": 1198.26421875, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.59077927376581, |
| "grad_norm": 216.09091186523438, |
| "learning_rate": 2.720264317180617e-05, |
| "loss": 1155.31796875, |
| "step": 1950 |
| }, |
| { |
| "epoch": 1.631578947368421, |
| "grad_norm": 185.73233032226562, |
| "learning_rate": 2.641598489616111e-05, |
| "loss": 1141.037890625, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.631578947368421, |
| "eval_loss": 171.78634643554688, |
| "eval_runtime": 12.946, |
| "eval_samples_per_second": 61.255, |
| "eval_steps_per_second": 7.724, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.672378620971032, |
| "grad_norm": 146.154296875, |
| "learning_rate": 2.562932662051605e-05, |
| "loss": 1107.244921875, |
| "step": 2050 |
| }, |
| { |
| "epoch": 1.7131782945736433, |
| "grad_norm": 273.62255859375, |
| "learning_rate": 2.484266834487099e-05, |
| "loss": 1082.53859375, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.7539779681762546, |
| "grad_norm": 109.22996520996094, |
| "learning_rate": 2.4056010069225928e-05, |
| "loss": 1073.700703125, |
| "step": 2150 |
| }, |
| { |
| "epoch": 1.794777641778866, |
| "grad_norm": 219.9355926513672, |
| "learning_rate": 2.326935179358087e-05, |
| "loss": 1060.901171875, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.835577315381477, |
| "grad_norm": 304.3589782714844, |
| "learning_rate": 2.248269351793581e-05, |
| "loss": 1043.506640625, |
| "step": 2250 |
| }, |
| { |
| "epoch": 1.876376988984088, |
| "grad_norm": 276.713623046875, |
| "learning_rate": 2.169603524229075e-05, |
| "loss": 1031.16875, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.9171766625866993, |
| "grad_norm": 836.3171997070312, |
| "learning_rate": 2.090937696664569e-05, |
| "loss": 1007.080625, |
| "step": 2350 |
| }, |
| { |
| "epoch": 1.9579763361893106, |
| "grad_norm": 79.88094329833984, |
| "learning_rate": 2.012271869100063e-05, |
| "loss": 993.526953125, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.9987760097919218, |
| "grad_norm": 172.98716735839844, |
| "learning_rate": 1.933606041535557e-05, |
| "loss": 986.5153125, |
| "step": 2450 |
| }, |
| { |
| "epoch": 2.039167686658507, |
| "grad_norm": 50.125850677490234, |
| "learning_rate": 1.8549402139710513e-05, |
| "loss": 961.28171875, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.039167686658507, |
| "eval_loss": 148.80242919921875, |
| "eval_runtime": 13.0561, |
| "eval_samples_per_second": 60.738, |
| "eval_steps_per_second": 7.659, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0799673602611177, |
| "grad_norm": 516.65087890625, |
| "learning_rate": 1.776274386406545e-05, |
| "loss": 958.1346875, |
| "step": 2550 |
| }, |
| { |
| "epoch": 2.120767033863729, |
| "grad_norm": 82.28546905517578, |
| "learning_rate": 1.697608558842039e-05, |
| "loss": 939.61703125, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.1615667074663403, |
| "grad_norm": 98.05081939697266, |
| "learning_rate": 1.6189427312775334e-05, |
| "loss": 940.276171875, |
| "step": 2650 |
| }, |
| { |
| "epoch": 2.2023663810689516, |
| "grad_norm": 100.78054809570312, |
| "learning_rate": 1.540276903713027e-05, |
| "loss": 929.987734375, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.243166054671563, |
| "grad_norm": 117.33869934082031, |
| "learning_rate": 1.4616110761485211e-05, |
| "loss": 919.060859375, |
| "step": 2750 |
| }, |
| { |
| "epoch": 2.2839657282741737, |
| "grad_norm": 87.76734161376953, |
| "learning_rate": 1.3829452485840152e-05, |
| "loss": 912.472734375, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.324765401876785, |
| "grad_norm": 145.8767852783203, |
| "learning_rate": 1.3042794210195092e-05, |
| "loss": 898.82875, |
| "step": 2850 |
| }, |
| { |
| "epoch": 2.365565075479396, |
| "grad_norm": 82.79586791992188, |
| "learning_rate": 1.2256135934550033e-05, |
| "loss": 897.908828125, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.4063647490820075, |
| "grad_norm": 45.99494552612305, |
| "learning_rate": 1.1469477658904971e-05, |
| "loss": 895.566484375, |
| "step": 2950 |
| }, |
| { |
| "epoch": 2.4471644226846188, |
| "grad_norm": 148.60716247558594, |
| "learning_rate": 1.0682819383259912e-05, |
| "loss": 877.555078125, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4471644226846188, |
| "eval_loss": 137.87533569335938, |
| "eval_runtime": 12.9383, |
| "eval_samples_per_second": 61.291, |
| "eval_steps_per_second": 7.729, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4879640962872296, |
| "grad_norm": 71.56707763671875, |
| "learning_rate": 9.896161107614853e-06, |
| "loss": 883.4053125, |
| "step": 3050 |
| }, |
| { |
| "epoch": 2.528763769889841, |
| "grad_norm": 386.5635070800781, |
| "learning_rate": 9.109502831969793e-06, |
| "loss": 881.8925, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.569563443492452, |
| "grad_norm": 83.39624786376953, |
| "learning_rate": 8.322844556324733e-06, |
| "loss": 868.849765625, |
| "step": 3150 |
| }, |
| { |
| "epoch": 2.610363117095063, |
| "grad_norm": 89.16000366210938, |
| "learning_rate": 7.536186280679673e-06, |
| "loss": 874.761953125, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.6511627906976747, |
| "grad_norm": 75.8927001953125, |
| "learning_rate": 6.749528005034614e-06, |
| "loss": 862.015, |
| "step": 3250 |
| }, |
| { |
| "epoch": 2.6919624643002855, |
| "grad_norm": 41.298606872558594, |
| "learning_rate": 5.962869729389554e-06, |
| "loss": 858.34359375, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.732762137902897, |
| "grad_norm": 92.17044067382812, |
| "learning_rate": 5.176211453744493e-06, |
| "loss": 847.730546875, |
| "step": 3350 |
| }, |
| { |
| "epoch": 2.773561811505508, |
| "grad_norm": 95.68132781982422, |
| "learning_rate": 4.389553178099434e-06, |
| "loss": 849.26859375, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.814361485108119, |
| "grad_norm": 73.86445617675781, |
| "learning_rate": 3.602894902454374e-06, |
| "loss": 849.22, |
| "step": 3450 |
| }, |
| { |
| "epoch": 2.85516115871073, |
| "grad_norm": 100.50775146484375, |
| "learning_rate": 2.816236626809314e-06, |
| "loss": 851.040078125, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.85516115871073, |
| "eval_loss": 134.6686553955078, |
| "eval_runtime": 13.0076, |
| "eval_samples_per_second": 60.964, |
| "eval_steps_per_second": 7.688, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8959608323133414, |
| "grad_norm": 72.54052734375, |
| "learning_rate": 2.0295783511642545e-06, |
| "loss": 850.704375, |
| "step": 3550 |
| }, |
| { |
| "epoch": 2.9367605059159527, |
| "grad_norm": 172.28179931640625, |
| "learning_rate": 1.2429200755191946e-06, |
| "loss": 850.34234375, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.977560179518564, |
| "grad_norm": 63.676513671875, |
| "learning_rate": 4.562617998741347e-07, |
| "loss": 851.570390625, |
| "step": 3650 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 3678, |
| "total_flos": 8085383880376320.0, |
| "train_loss": 2813.158582810801, |
| "train_runtime": 8544.3156, |
| "train_samples_per_second": 27.534, |
| "train_steps_per_second": 0.43 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 3678, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8085383880376320.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
