| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 212, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 56.48654000341791, | |
| "learning_rate": 2.2727272727272729e-07, | |
| "loss": 1.4277, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 32.7318603335361, | |
| "learning_rate": 1.1363636363636364e-06, | |
| "loss": 1.3555, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 18.95719710067938, | |
| "learning_rate": 2.2727272727272728e-06, | |
| "loss": 1.1538, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 6.971441284336922, | |
| "learning_rate": 3.409090909090909e-06, | |
| "loss": 1.0874, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 5.3823086685656945, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 1.0206, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 4.089694965230436, | |
| "learning_rate": 4.9969249228707625e-06, | |
| "loss": 1.003, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.6458232095508354, | |
| "learning_rate": 4.978160173317439e-06, | |
| "loss": 0.9657, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 3.2176159938800266, | |
| "learning_rate": 4.942467076958999e-06, | |
| "loss": 0.9435, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 2.977792325165127, | |
| "learning_rate": 4.890089453835894e-06, | |
| "loss": 0.9368, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 3.1118165842382504, | |
| "learning_rate": 4.821385096224268e-06, | |
| "loss": 0.9261, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 3.3730430715720994, | |
| "learning_rate": 4.736823324551909e-06, | |
| "loss": 0.9112, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 3.1574351012684474, | |
| "learning_rate": 4.636981781463848e-06, | |
| "loss": 0.9145, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 3.0769386301276183, | |
| "learning_rate": 4.522542485937369e-06, | |
| "loss": 0.8814, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 3.055053060479708, | |
| "learning_rate": 4.394287174400838e-06, | |
| "loss": 0.8751, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 5.4232076990901, | |
| "learning_rate": 4.253091960681222e-06, | |
| "loss": 0.8909, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 3.2020191326645273, | |
| "learning_rate": 4.099921351258292e-06, | |
| "loss": 0.8597, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.290349939143738, | |
| "learning_rate": 3.935821656707359e-06, | |
| "loss": 0.8913, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.4145364641074396, | |
| "learning_rate": 3.76191384433711e-06, | |
| "loss": 0.8613, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 3.263977821412841, | |
| "learning_rate": 3.579385880846232e-06, | |
| "loss": 0.8759, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 10.655125505241807, | |
| "learning_rate": 3.3894846173062917e-06, | |
| "loss": 0.8519, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 250.56147092672978, | |
| "learning_rate": 3.193507271904612e-06, | |
| "loss": 0.8517, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.419289400374, | |
| "learning_rate": 2.9927925686287006e-06, | |
| "loss": 0.8594, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 3.263295130566819, | |
| "learning_rate": 2.788711592423966e-06, | |
| "loss": 0.849, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 3.466296576337294, | |
| "learning_rate": 2.5826584232932707e-06, | |
| "loss": 0.8876, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 3.464842972592367, | |
| "learning_rate": 2.376040613316944e-06, | |
| "loss": 0.8605, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 3.301898509073216, | |
| "learning_rate": 2.1702695716448276e-06, | |
| "loss": 0.8485, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 3.3264276395542414, | |
| "learning_rate": 1.9667509231406332e-06, | |
| "loss": 0.84, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 3.2310503156717245, | |
| "learning_rate": 1.7668749065388385e-06, | |
| "loss": 0.835, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 3.361233531250762, | |
| "learning_rate": 1.5720068777044479e-06, | |
| "loss": 0.8413, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 3.229898438412517, | |
| "learning_rate": 1.383477982867984e-06, | |
| "loss": 0.8471, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 3.082051479835871, | |
| "learning_rate": 1.2025760655469629e-06, | |
| "loss": 0.8435, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 3.392093980236051, | |
| "learning_rate": 1.0305368692688175e-06, | |
| "loss": 0.8405, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.3593945187641316, | |
| "learning_rate": 8.685355961895783e-07, | |
| "loss": 0.8475, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 3.183216674509743, | |
| "learning_rate": 7.176788792715076e-07, | |
| "loss": 0.8551, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.1059443889529974, | |
| "learning_rate": 5.78997222857853e-07, | |
| "loss": 0.8347, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 3.1090745975048404, | |
| "learning_rate": 4.534379632832692e-07, | |
| "loss": 0.8251, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 3.1186904842842007, | |
| "learning_rate": 3.4185879760606525e-07, | |
| "loss": 0.816, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 13.97461867271608, | |
| "learning_rate": 2.450219246676028e-07, | |
| "loss": 0.8235, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 3.0246555374160558, | |
| "learning_rate": 1.6358883850134815e-07, | |
| "loss": 0.8356, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 2.969712888409742, | |
| "learning_rate": 9.811580965787965e-08, | |
| "loss": 0.8327, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 2.9445398637597826, | |
| "learning_rate": 4.905008531297661e-08, | |
| "loss": 0.86, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.096896769727839, | |
| "learning_rate": 1.6726834115904645e-08, | |
| "loss": 0.8216, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 2.9776364501353694, | |
| "learning_rate": 1.3668566476848777e-09, | |
| "loss": 0.8468, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.3747081756591797, | |
| "eval_runtime": 1.4843, | |
| "eval_samples_per_second": 9.432, | |
| "eval_steps_per_second": 0.674, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 212, | |
| "total_flos": 88776974008320.0, | |
| "train_loss": 0.8946974671111917, | |
| "train_runtime": 2355.8103, | |
| "train_samples_per_second": 5.733, | |
| "train_steps_per_second": 0.09 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 212, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 88776974008320.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |