| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9834710743801653, | |
| "eval_steps": 5, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06611570247933884, | |
| "grad_norm": 1.9830907583236694, | |
| "learning_rate": 0.00019333333333333333, | |
| "loss": 10.1125, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1322314049586777, | |
| "grad_norm": 2.0950465202331543, | |
| "learning_rate": 0.0001866666666666667, | |
| "loss": 9.9005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.19834710743801653, | |
| "grad_norm": 2.9728968143463135, | |
| "learning_rate": 0.00018, | |
| "loss": 9.7099, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2644628099173554, | |
| "grad_norm": 3.350935935974121, | |
| "learning_rate": 0.00017333333333333334, | |
| "loss": 8.944, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3305785123966942, | |
| "grad_norm": 3.183183431625366, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 8.2807, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3305785123966942, | |
| "eval_clap": 0.016216065734624863, | |
| "eval_loss": 3.0157957077026367, | |
| "eval_runtime": 243.8548, | |
| "eval_samples_per_second": 0.033, | |
| "eval_steps_per_second": 0.033, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.39669421487603307, | |
| "grad_norm": 4.646206855773926, | |
| "learning_rate": 0.00016, | |
| "loss": 7.7738, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4628099173553719, | |
| "grad_norm": 7.790807723999023, | |
| "learning_rate": 0.00015333333333333334, | |
| "loss": 7.5699, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5289256198347108, | |
| "grad_norm": 5.643434047698975, | |
| "learning_rate": 0.00014666666666666666, | |
| "loss": 7.139, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5950413223140496, | |
| "grad_norm": 3.7166173458099365, | |
| "learning_rate": 0.00014, | |
| "loss": 7.129, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.6611570247933884, | |
| "grad_norm": 2.310825824737549, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 6.536, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6611570247933884, | |
| "eval_clap": 0.007407550700008869, | |
| "eval_loss": 3.2537384033203125, | |
| "eval_runtime": 229.6949, | |
| "eval_samples_per_second": 0.035, | |
| "eval_steps_per_second": 0.035, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 3.1138978004455566, | |
| "learning_rate": 0.00012666666666666666, | |
| "loss": 7.0649, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.7933884297520661, | |
| "grad_norm": 1.9888191223144531, | |
| "learning_rate": 0.00012, | |
| "loss": 6.579, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.859504132231405, | |
| "grad_norm": 2.7568395137786865, | |
| "learning_rate": 0.00011333333333333334, | |
| "loss": 6.927, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9256198347107438, | |
| "grad_norm": 1.5047709941864014, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 6.4366, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9917355371900827, | |
| "grad_norm": 1.930873155593872, | |
| "learning_rate": 0.0001, | |
| "loss": 6.1596, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.9917355371900827, | |
| "eval_clap": -0.039488501846790314, | |
| "eval_loss": 4.296116828918457, | |
| "eval_runtime": 237.1954, | |
| "eval_samples_per_second": 0.034, | |
| "eval_steps_per_second": 0.034, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0578512396694215, | |
| "grad_norm": 1.186317801475525, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 6.3429, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1239669421487604, | |
| "grad_norm": 1.2925118207931519, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 6.1757, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.1900826446280992, | |
| "grad_norm": 0.9777997136116028, | |
| "learning_rate": 8e-05, | |
| "loss": 6.2497, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.256198347107438, | |
| "grad_norm": 1.5951266288757324, | |
| "learning_rate": 7.333333333333333e-05, | |
| "loss": 6.3316, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.322314049586777, | |
| "grad_norm": 1.7874397039413452, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 6.44, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.322314049586777, | |
| "eval_clap": -0.038016706705093384, | |
| "eval_loss": 4.876384258270264, | |
| "eval_runtime": 233.9024, | |
| "eval_samples_per_second": 0.034, | |
| "eval_steps_per_second": 0.034, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3884297520661157, | |
| "grad_norm": 1.0251847505569458, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0348, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.8240154981613159, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 6.1891, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.5206611570247934, | |
| "grad_norm": 1.571574091911316, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 5.89, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.5867768595041323, | |
| "grad_norm": 0.9506106376647949, | |
| "learning_rate": 4e-05, | |
| "loss": 6.2511, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.6528925619834711, | |
| "grad_norm": 1.503645420074463, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 6.4944, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.6528925619834711, | |
| "eval_clap": -0.05864603444933891, | |
| "eval_loss": 5.022668361663818, | |
| "eval_runtime": 237.1887, | |
| "eval_samples_per_second": 0.034, | |
| "eval_steps_per_second": 0.034, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.71900826446281, | |
| "grad_norm": 1.0225498676300049, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 6.2162, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.7851239669421488, | |
| "grad_norm": 1.566749930381775, | |
| "learning_rate": 2e-05, | |
| "loss": 6.2582, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.8512396694214877, | |
| "grad_norm": 1.3055825233459473, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 6.1079, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.9173553719008265, | |
| "grad_norm": 0.9658945798873901, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 6.1765, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.9834710743801653, | |
| "grad_norm": 1.0808743238449097, | |
| "learning_rate": 0.0, | |
| "loss": 6.3776, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9834710743801653, | |
| "eval_clap": -0.05411681905388832, | |
| "eval_loss": 5.031132698059082, | |
| "eval_runtime": 232.4736, | |
| "eval_samples_per_second": 0.034, | |
| "eval_steps_per_second": 0.034, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9834710743801653, | |
| "step": 30, | |
| "total_flos": 26305190573760.0, | |
| "train_loss": 6.993270270029703, | |
| "train_runtime": 2006.6605, | |
| "train_samples_per_second": 0.121, | |
| "train_steps_per_second": 0.015 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 26305190573760.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |