| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.99630314232902, | |
| "eval_steps": 500, | |
| "global_step": 6760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.260355029585799e-05, | |
| "loss": 0.4297, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 58.4175, | |
| "eval_gen_len": 59.5692, | |
| "eval_loss": 0.5057221055030823, | |
| "eval_runtime": 11122.6762, | |
| "eval_samples_per_second": 22.228, | |
| "eval_steps_per_second": 0.695, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 8.520710059171599e-05, | |
| "loss": 0.4242, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 58.6857, | |
| "eval_gen_len": 59.5601, | |
| "eval_loss": 0.5072483420372009, | |
| "eval_runtime": 11023.2288, | |
| "eval_samples_per_second": 22.429, | |
| "eval_steps_per_second": 0.701, | |
| "step": 1352 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 7.781065088757397e-05, | |
| "loss": 0.4154, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 7.041420118343195e-05, | |
| "loss": 0.4075, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 58.5394, | |
| "eval_gen_len": 59.5163, | |
| "eval_loss": 0.5111977458000183, | |
| "eval_runtime": 11021.212, | |
| "eval_samples_per_second": 22.433, | |
| "eval_steps_per_second": 0.701, | |
| "step": 2028 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 6.301775147928994e-05, | |
| "loss": 0.3925, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 58.5548, | |
| "eval_gen_len": 59.4982, | |
| "eval_loss": 0.5156235694885254, | |
| "eval_runtime": 10993.763, | |
| "eval_samples_per_second": 22.489, | |
| "eval_steps_per_second": 0.703, | |
| "step": 2705 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 5.562130177514793e-05, | |
| "loss": 0.3854, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 58.6457, | |
| "eval_gen_len": 59.5187, | |
| "eval_loss": 0.5196160078048706, | |
| "eval_runtime": 11023.9666, | |
| "eval_samples_per_second": 22.427, | |
| "eval_steps_per_second": 0.701, | |
| "step": 3381 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 4.822485207100592e-05, | |
| "loss": 0.3791, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 4.0828402366863904e-05, | |
| "loss": 0.3725, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 58.7468, | |
| "eval_gen_len": 59.4776, | |
| "eval_loss": 0.5207210779190063, | |
| "eval_runtime": 11083.6264, | |
| "eval_samples_per_second": 22.307, | |
| "eval_steps_per_second": 0.697, | |
| "step": 4057 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 3.3431952662721895e-05, | |
| "loss": 0.3629, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 58.6738, | |
| "eval_gen_len": 59.4889, | |
| "eval_loss": 0.5233368277549744, | |
| "eval_runtime": 11457.3857, | |
| "eval_samples_per_second": 21.579, | |
| "eval_steps_per_second": 0.674, | |
| "step": 4733 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 2.6035502958579882e-05, | |
| "loss": 0.3606, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 58.7457, | |
| "eval_gen_len": 59.4951, | |
| "eval_loss": 0.5246109366416931, | |
| "eval_runtime": 11473.7808, | |
| "eval_samples_per_second": 21.548, | |
| "eval_steps_per_second": 0.673, | |
| "step": 5410 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "learning_rate": 1.8639053254437873e-05, | |
| "loss": 0.3596, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 1.1242603550295859e-05, | |
| "loss": 0.3563, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 58.7607, | |
| "eval_gen_len": 59.4623, | |
| "eval_loss": 0.5246689915657043, | |
| "eval_runtime": 10955.4266, | |
| "eval_samples_per_second": 22.568, | |
| "eval_steps_per_second": 0.705, | |
| "step": 6086 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 3.86094674556213e-06, | |
| "loss": 0.3556, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 58.769, | |
| "eval_gen_len": 59.4737, | |
| "eval_loss": 0.5249367952346802, | |
| "eval_runtime": 12113.7588, | |
| "eval_samples_per_second": 20.41, | |
| "eval_steps_per_second": 0.638, | |
| "step": 6760 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 6760, | |
| "total_flos": 6350480490627072.0, | |
| "train_loss": 0.383708986587073, | |
| "train_runtime": 114051.6958, | |
| "train_samples_per_second": 7.587, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6760, | |
| "num_train_epochs": 10, | |
| "save_steps": 50, | |
| "total_flos": 6350480490627072.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |