| { | |
| "best_metric": 0.6242462992668152, | |
| "best_model_checkpoint": "saves/granite-3.1-3b-a800m-instruct\\checkpoint-400", | |
| "epoch": 0.6676403087836428, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01669100771959107, | |
| "grad_norm": 2.9202163219451904, | |
| "learning_rate": 0.0004, | |
| "loss": 1.278, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03338201543918214, | |
| "grad_norm": 4.189798355102539, | |
| "learning_rate": 0.0004, | |
| "loss": 0.9626, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05007302315877321, | |
| "grad_norm": 2.869903087615967, | |
| "learning_rate": 0.0004, | |
| "loss": 0.993, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06676403087836429, | |
| "grad_norm": 2.438555955886841, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8561, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08345503859795535, | |
| "grad_norm": 2.5792276859283447, | |
| "learning_rate": 0.0004, | |
| "loss": 1.008, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10014604631754642, | |
| "grad_norm": 2.982292890548706, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7934, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1168370540371375, | |
| "grad_norm": 1.8241057395935059, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7811, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13352806175672857, | |
| "grad_norm": 1.7634323835372925, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8709, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15021906947631963, | |
| "grad_norm": 2.9324874877929688, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8429, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1669100771959107, | |
| "grad_norm": 1.6204209327697754, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7855, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1669100771959107, | |
| "eval_loss": 0.6945974230766296, | |
| "eval_runtime": 172.0416, | |
| "eval_samples_per_second": 6.196, | |
| "eval_steps_per_second": 3.098, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18360108491550178, | |
| "grad_norm": 3.2142117023468018, | |
| "learning_rate": 0.0004, | |
| "loss": 0.802, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.20029209263509284, | |
| "grad_norm": 2.2638301849365234, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7556, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2169831003546839, | |
| "grad_norm": 1.815718412399292, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8419, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.233674108074275, | |
| "grad_norm": 2.0672075748443604, | |
| "learning_rate": 0.0004, | |
| "loss": 0.9097, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25036511579386606, | |
| "grad_norm": 1.7800904512405396, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8887, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.26705612351345714, | |
| "grad_norm": 2.7289271354675293, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8078, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2837471312330482, | |
| "grad_norm": 1.8588060140609741, | |
| "learning_rate": 0.0004, | |
| "loss": 0.824, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.30043813895263927, | |
| "grad_norm": 3.1786084175109863, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7997, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.31712914667223036, | |
| "grad_norm": 1.992241382598877, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7627, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3338201543918214, | |
| "grad_norm": 1.9906195402145386, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7525, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3338201543918214, | |
| "eval_loss": 0.6729084849357605, | |
| "eval_runtime": 168.6462, | |
| "eval_samples_per_second": 6.321, | |
| "eval_steps_per_second": 3.16, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3505111621114125, | |
| "grad_norm": 1.408159613609314, | |
| "learning_rate": 0.0004, | |
| "loss": 0.803, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.36720216983100357, | |
| "grad_norm": 2.2278130054473877, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8844, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3838931775505946, | |
| "grad_norm": 2.3945512771606445, | |
| "learning_rate": 0.0004, | |
| "loss": 0.9442, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4005841852701857, | |
| "grad_norm": 1.1758439540863037, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8379, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4172751929897768, | |
| "grad_norm": 2.483109951019287, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7718, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4339662007093678, | |
| "grad_norm": 1.4695591926574707, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7664, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4506572084289589, | |
| "grad_norm": 1.5021220445632935, | |
| "learning_rate": 0.0004, | |
| "loss": 0.92, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.46734821614855, | |
| "grad_norm": 1.426329255104065, | |
| "learning_rate": 0.0004, | |
| "loss": 0.9022, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.484039223868141, | |
| "grad_norm": 1.4195940494537354, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7857, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5007302315877321, | |
| "grad_norm": 1.7680014371871948, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8381, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5007302315877321, | |
| "eval_loss": 0.6377778649330139, | |
| "eval_runtime": 168.5375, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 3.163, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5174212393073232, | |
| "grad_norm": 1.8080698251724243, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8346, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5341122470269143, | |
| "grad_norm": 7.522341728210449, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7861, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5508032547465053, | |
| "grad_norm": 1.397352933883667, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8113, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5674942624660964, | |
| "grad_norm": 1.4032018184661865, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7936, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5841852701856874, | |
| "grad_norm": 1.6604522466659546, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7278, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6008762779052785, | |
| "grad_norm": 1.8743423223495483, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8652, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6175672856248696, | |
| "grad_norm": 3.0662145614624023, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8746, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6342582933444607, | |
| "grad_norm": 1.4264730215072632, | |
| "learning_rate": 0.0004, | |
| "loss": 0.7431, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6509493010640517, | |
| "grad_norm": 1.9815279245376587, | |
| "learning_rate": 0.0004, | |
| "loss": 0.828, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6676403087836428, | |
| "grad_norm": 1.3689788579940796, | |
| "learning_rate": 0.0004, | |
| "loss": 0.8564, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6676403087836428, | |
| "eval_loss": 0.6242462992668152, | |
| "eval_runtime": 168.5455, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 3.162, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 17, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.420426751907922e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |