| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.1646223369916076, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.3856842517852783, | |
| "learning_rate": 1.9868055555555558e-05, | |
| "loss": 1.1287, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.4272608458995819, | |
| "learning_rate": 1.9736111111111115e-05, | |
| "loss": 1.0586, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.4174883961677551, | |
| "learning_rate": 1.9597222222222224e-05, | |
| "loss": 1.0166, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.35379958152770996, | |
| "learning_rate": 1.9458333333333333e-05, | |
| "loss": 0.8919, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.3820742964744568, | |
| "learning_rate": 1.9319444444444446e-05, | |
| "loss": 0.7971, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.34241142868995667, | |
| "learning_rate": 1.918055555555556e-05, | |
| "loss": 0.7466, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.48230335116386414, | |
| "learning_rate": 1.9041666666666668e-05, | |
| "loss": 0.7468, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.45057639479637146, | |
| "learning_rate": 1.890277777777778e-05, | |
| "loss": 0.7373, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 0.43228989839553833, | |
| "learning_rate": 1.876388888888889e-05, | |
| "loss": 0.756, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.460918664932251, | |
| "learning_rate": 1.8625000000000002e-05, | |
| "loss": 0.7186, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 0.8221462368965149, | |
| "learning_rate": 1.8486111111111115e-05, | |
| "loss": 0.6772, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.547721266746521, | |
| "learning_rate": 1.8347222222222224e-05, | |
| "loss": 0.6809, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 0.6005927920341492, | |
| "learning_rate": 1.8208333333333333e-05, | |
| "loss": 0.6651, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 0.7613247632980347, | |
| "learning_rate": 1.8069444444444446e-05, | |
| "loss": 0.641, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.668274462223053, | |
| "learning_rate": 1.793055555555556e-05, | |
| "loss": 0.6296, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "grad_norm": 0.6101451516151428, | |
| "learning_rate": 1.7791666666666668e-05, | |
| "loss": 0.6033, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 0.8970944285392761, | |
| "learning_rate": 1.765277777777778e-05, | |
| "loss": 0.5939, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "grad_norm": 0.9417675137519836, | |
| "learning_rate": 1.751388888888889e-05, | |
| "loss": 0.5966, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "grad_norm": 1.3364688158035278, | |
| "learning_rate": 1.7375000000000002e-05, | |
| "loss": 0.5771, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 0.7763314843177795, | |
| "learning_rate": 1.7236111111111115e-05, | |
| "loss": 0.5588, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "grad_norm": 1.156174659729004, | |
| "learning_rate": 1.7097222222222224e-05, | |
| "loss": 0.5571, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "grad_norm": 0.8457236886024475, | |
| "learning_rate": 1.6958333333333333e-05, | |
| "loss": 0.5431, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 1.1372079849243164, | |
| "learning_rate": 1.6819444444444446e-05, | |
| "loss": 0.4839, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 1.0457851886749268, | |
| "learning_rate": 1.668055555555556e-05, | |
| "loss": 0.5041, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "grad_norm": 1.1884819269180298, | |
| "learning_rate": 1.6541666666666668e-05, | |
| "loss": 0.468, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 2880, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 1.1559398685499392e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |