| { | |
| "best_metric": 0.561149001121521, | |
| "best_model_checkpoint": "/content/drive/MyDrive/c4_200m/weights/checkpoint-5000", | |
| "epoch": 0.969681298080031, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.8072148952676496e-05, | |
| "loss": 0.763, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_gen_len": 17.3073, | |
| "eval_loss": 0.6263013482093811, | |
| "eval_rouge1": 71.3829, | |
| "eval_rouge2": 60.9575, | |
| "eval_rougeL": 70.617, | |
| "eval_rougeLsum": 70.6508, | |
| "eval_runtime": 4434.8322, | |
| "eval_samples_per_second": 12.402, | |
| "eval_steps_per_second": 0.775, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.6132660977501943e-05, | |
| "loss": 0.6682, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_gen_len": 17.2827, | |
| "eval_loss": 0.5980147123336792, | |
| "eval_rouge1": 71.6839, | |
| "eval_rouge2": 61.5115, | |
| "eval_rougeL": 70.9369, | |
| "eval_rougeLsum": 70.9762, | |
| "eval_runtime": 4421.7138, | |
| "eval_samples_per_second": 12.439, | |
| "eval_steps_per_second": 0.778, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.4193173002327387e-05, | |
| "loss": 0.6547, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_gen_len": 17.2674, | |
| "eval_loss": 0.5852676630020142, | |
| "eval_rouge1": 71.8716, | |
| "eval_rouge2": 61.8264, | |
| "eval_rougeL": 71.1345, | |
| "eval_rougeLsum": 71.1741, | |
| "eval_runtime": 4432.6842, | |
| "eval_samples_per_second": 12.408, | |
| "eval_steps_per_second": 0.776, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.2257564003103182e-05, | |
| "loss": 0.6423, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_gen_len": 17.2636, | |
| "eval_loss": 0.5758475065231323, | |
| "eval_rouge1": 71.9319, | |
| "eval_rouge2": 61.9431, | |
| "eval_rougeL": 71.1965, | |
| "eval_rougeLsum": 71.2364, | |
| "eval_runtime": 4426.8507, | |
| "eval_samples_per_second": 12.424, | |
| "eval_steps_per_second": 0.777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.0318076027928628e-05, | |
| "loss": 0.6319, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_gen_len": 17.2555, | |
| "eval_loss": 0.5706557631492615, | |
| "eval_rouge1": 72.004, | |
| "eval_rouge2": 62.0721, | |
| "eval_rougeL": 71.2739, | |
| "eval_rougeLsum": 71.3176, | |
| "eval_runtime": 4455.9486, | |
| "eval_samples_per_second": 12.343, | |
| "eval_steps_per_second": 0.772, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.378588052754074e-06, | |
| "loss": 0.6239, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_gen_len": 17.251, | |
| "eval_loss": 0.5678849220275879, | |
| "eval_rouge1": 72.0655, | |
| "eval_rouge2": 62.1749, | |
| "eval_rougeL": 71.3432, | |
| "eval_rougeLsum": 71.3854, | |
| "eval_runtime": 4483.3834, | |
| "eval_samples_per_second": 12.268, | |
| "eval_steps_per_second": 0.767, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.43910007757952e-06, | |
| "loss": 0.619, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_gen_len": 17.25, | |
| "eval_loss": 0.5634791851043701, | |
| "eval_rouge1": 72.0893, | |
| "eval_rouge2": 62.2244, | |
| "eval_rougeL": 71.3669, | |
| "eval_rougeLsum": 71.4087, | |
| "eval_runtime": 4466.0069, | |
| "eval_samples_per_second": 12.315, | |
| "eval_steps_per_second": 0.77, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.499612102404966e-06, | |
| "loss": 0.6248, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_gen_len": 17.246, | |
| "eval_loss": 0.5618749260902405, | |
| "eval_rouge1": 72.1096, | |
| "eval_rouge2": 62.261, | |
| "eval_rougeL": 71.3877, | |
| "eval_rougeLsum": 71.4304, | |
| "eval_runtime": 4436.9366, | |
| "eval_samples_per_second": 12.396, | |
| "eval_steps_per_second": 0.775, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.560124127230411e-06, | |
| "loss": 0.6159, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_gen_len": 17.2457, | |
| "eval_loss": 0.5612673163414001, | |
| "eval_rouge1": 72.1232, | |
| "eval_rouge2": 62.2834, | |
| "eval_rougeL": 71.4006, | |
| "eval_rougeLsum": 71.4433, | |
| "eval_runtime": 4436.0596, | |
| "eval_samples_per_second": 12.398, | |
| "eval_steps_per_second": 0.775, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 6.206361520558573e-07, | |
| "loss": 0.6118, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_gen_len": 17.2461, | |
| "eval_loss": 0.561149001121521, | |
| "eval_rouge1": 72.1272, | |
| "eval_rouge2": 62.2947, | |
| "eval_rougeL": 71.4073, | |
| "eval_rougeLsum": 71.4508, | |
| "eval_runtime": 4447.3157, | |
| "eval_samples_per_second": 12.367, | |
| "eval_steps_per_second": 0.773, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 5156, | |
| "num_train_epochs": 1, | |
| "total_flos": 7.061289205579776e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |