| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9999572667834707, | |
| "eval_steps": 200, | |
| "global_step": 2925, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.942542412504543e-05, | |
| "loss": 1.0325, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.8921748995780945, | |
| "eval_runtime": 119.3845, | |
| "eval_samples_per_second": 43.565, | |
| "eval_steps_per_second": 21.787, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.77281074950681e-05, | |
| "loss": 0.9136, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 0.880075991153717, | |
| "eval_runtime": 119.7923, | |
| "eval_samples_per_second": 43.417, | |
| "eval_steps_per_second": 21.713, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.498606908508754e-05, | |
| "loss": 0.9043, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.8752565979957581, | |
| "eval_runtime": 120.3536, | |
| "eval_samples_per_second": 43.214, | |
| "eval_steps_per_second": 21.611, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.132534962458962e-05, | |
| "loss": 0.8903, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.8710557222366333, | |
| "eval_runtime": 120.031, | |
| "eval_samples_per_second": 43.33, | |
| "eval_steps_per_second": 21.669, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.69142180005327e-05, | |
| "loss": 0.9056, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 0.8682616949081421, | |
| "eval_runtime": 119.9746, | |
| "eval_samples_per_second": 43.351, | |
| "eval_steps_per_second": 21.68, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.195543659791132e-05, | |
| "loss": 0.9001, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 0.867277204990387, | |
| "eval_runtime": 119.7846, | |
| "eval_samples_per_second": 43.42, | |
| "eval_steps_per_second": 21.714, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.667694110977506e-05, | |
| "loss": 0.8949, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.8667991161346436, | |
| "eval_runtime": 120.2926, | |
| "eval_samples_per_second": 43.236, | |
| "eval_steps_per_second": 21.622, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.1321363229207096e-05, | |
| "loss": 0.8917, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.8655699491500854, | |
| "eval_runtime": 120.5955, | |
| "eval_samples_per_second": 43.128, | |
| "eval_steps_per_second": 21.568, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.613487782393661e-05, | |
| "loss": 0.8951, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.8647685050964355, | |
| "eval_runtime": 120.2019, | |
| "eval_samples_per_second": 43.269, | |
| "eval_steps_per_second": 21.639, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.1355887245137383e-05, | |
| "loss": 0.9014, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.8637479543685913, | |
| "eval_runtime": 119.9752, | |
| "eval_samples_per_second": 43.351, | |
| "eval_steps_per_second": 21.679, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.20406290826649e-06, | |
| "loss": 0.8874, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.8629816174507141, | |
| "eval_runtime": 120.3564, | |
| "eval_samples_per_second": 43.213, | |
| "eval_steps_per_second": 21.611, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.8702478614051355e-06, | |
| "loss": 0.8968, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.8628040552139282, | |
| "eval_runtime": 120.5881, | |
| "eval_samples_per_second": 43.13, | |
| "eval_steps_per_second": 21.569, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.8755, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.8626384735107422, | |
| "eval_runtime": 120.166, | |
| "eval_samples_per_second": 43.282, | |
| "eval_steps_per_second": 21.645, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.2497051885228827e-07, | |
| "loss": 0.9029, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.8625245094299316, | |
| "eval_runtime": 120.5975, | |
| "eval_samples_per_second": 43.127, | |
| "eval_steps_per_second": 21.568, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 2925, | |
| "total_flos": 1.4966547015951974e+17, | |
| "train_loss": 0.9060284554245126, | |
| "train_runtime": 4147.2863, | |
| "train_samples_per_second": 11.285, | |
| "train_steps_per_second": 0.705 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 2925, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "total_flos": 1.4966547015951974e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |