| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5460336720764447, | |
| "eval_steps": 100, | |
| "global_step": 1800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.3099820017814636, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8122, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 1.7837988138198853, | |
| "eval_runtime": 208.895, | |
| "eval_samples_per_second": 6.3, | |
| "eval_steps_per_second": 1.575, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.17340590059757233, | |
| "learning_rate": 9.655172413793105e-05, | |
| "loss": 1.6719, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.7731986045837402, | |
| "eval_runtime": 208.8797, | |
| "eval_samples_per_second": 6.3, | |
| "eval_steps_per_second": 1.575, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.16290397942066193, | |
| "learning_rate": 9.310344827586207e-05, | |
| "loss": 1.6564, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 1.76987886428833, | |
| "eval_runtime": 208.9321, | |
| "eval_samples_per_second": 6.299, | |
| "eval_steps_per_second": 1.575, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1616089642047882, | |
| "learning_rate": 8.96551724137931e-05, | |
| "loss": 1.6781, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.7673217058181763, | |
| "eval_runtime": 208.8655, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.21069930493831635, | |
| "learning_rate": 8.620689655172413e-05, | |
| "loss": 1.6603, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.7704393863677979, | |
| "eval_runtime": 208.9211, | |
| "eval_samples_per_second": 6.299, | |
| "eval_steps_per_second": 1.575, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.19658587872982025, | |
| "learning_rate": 8.275862068965517e-05, | |
| "loss": 1.6579, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 1.7655415534973145, | |
| "eval_runtime": 208.8518, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.17741167545318604, | |
| "learning_rate": 7.931034482758621e-05, | |
| "loss": 1.6573, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 1.7669495344161987, | |
| "eval_runtime": 208.8515, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.17904286086559296, | |
| "learning_rate": 7.586206896551724e-05, | |
| "loss": 1.6398, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.7682162523269653, | |
| "eval_runtime": 208.8599, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.17510022222995758, | |
| "learning_rate": 7.241379310344828e-05, | |
| "loss": 1.6464, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.7680007219314575, | |
| "eval_runtime": 208.9236, | |
| "eval_samples_per_second": 6.299, | |
| "eval_steps_per_second": 1.575, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.16618536412715912, | |
| "learning_rate": 6.896551724137931e-05, | |
| "loss": 1.6402, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.7689732313156128, | |
| "eval_runtime": 208.9404, | |
| "eval_samples_per_second": 6.298, | |
| "eval_steps_per_second": 1.575, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.1838076114654541, | |
| "learning_rate": 6.551724137931034e-05, | |
| "loss": 1.6602, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.7677079439163208, | |
| "eval_runtime": 208.8621, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.1798967868089676, | |
| "learning_rate": 6.206896551724138e-05, | |
| "loss": 1.6399, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.7679466009140015, | |
| "eval_runtime": 208.7864, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.19117586314678192, | |
| "learning_rate": 5.862068965517241e-05, | |
| "loss": 1.6428, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 1.7693601846694946, | |
| "eval_runtime": 208.7845, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.18876388669013977, | |
| "learning_rate": 5.517241379310345e-05, | |
| "loss": 1.6403, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 1.7661293745040894, | |
| "eval_runtime": 208.7761, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.17072780430316925, | |
| "learning_rate": 5.172413793103449e-05, | |
| "loss": 1.6403, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 1.7662379741668701, | |
| "eval_runtime": 208.8393, | |
| "eval_samples_per_second": 6.301, | |
| "eval_steps_per_second": 1.575, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.20399431884288788, | |
| "learning_rate": 4.827586206896552e-05, | |
| "loss": 1.637, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 1.7682374715805054, | |
| "eval_runtime": 208.7744, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.2173861563205719, | |
| "learning_rate": 4.482758620689655e-05, | |
| "loss": 1.6366, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 1.7696117162704468, | |
| "eval_runtime": 208.7744, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.19538724422454834, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 1.6377, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 1.7682461738586426, | |
| "eval_runtime": 208.7775, | |
| "eval_samples_per_second": 6.303, | |
| "eval_steps_per_second": 1.576, | |
| "step": 1800 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "total_flos": 4.823467998404936e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |