| { | |
| "best_metric": 26.5258, | |
| "best_model_checkpoint": "/content/drive/My Drive/ArabartModel/checkpoint-6112", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 6112, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32722513089005234, | |
| "grad_norm": 1.563672423362732, | |
| "learning_rate": 4.6727748691099475e-05, | |
| "loss": 0.0195, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6544502617801047, | |
| "grad_norm": 1.7397737503051758, | |
| "learning_rate": 4.3455497382198955e-05, | |
| "loss": 0.0244, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.981675392670157, | |
| "grad_norm": 3.006633758544922, | |
| "learning_rate": 4.018324607329843e-05, | |
| "loss": 0.0293, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 5.626201629638672, | |
| "eval_rouge1": 25.2705, | |
| "eval_rouge2": 9.9865, | |
| "eval_rougeL": 21.8678, | |
| "eval_runtime": 372.5145, | |
| "eval_samples_per_second": 5.892, | |
| "eval_steps_per_second": 0.591, | |
| "step": 1528 | |
| }, | |
| { | |
| "epoch": 1.3089005235602094, | |
| "grad_norm": 3.4732136726379395, | |
| "learning_rate": 4.3455497382198955e-05, | |
| "loss": 0.0456, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6361256544502618, | |
| "grad_norm": 2.2575156688690186, | |
| "learning_rate": 4.181937172774869e-05, | |
| "loss": 0.0619, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.9633507853403143, | |
| "grad_norm": 6.030032157897949, | |
| "learning_rate": 4.018324607329843e-05, | |
| "loss": 0.0993, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 5.542174339294434, | |
| "eval_rouge1": 25.0744, | |
| "eval_rouge2": 9.8534, | |
| "eval_rougeL": 21.6992, | |
| "eval_runtime": 411.9676, | |
| "eval_samples_per_second": 5.328, | |
| "eval_steps_per_second": 0.534, | |
| "step": 3056 | |
| }, | |
| { | |
| "epoch": 2.2905759162303667, | |
| "grad_norm": 6.401626110076904, | |
| "learning_rate": 3.8547120418848174e-05, | |
| "loss": 0.21, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.6178010471204187, | |
| "grad_norm": 5.806931495666504, | |
| "learning_rate": 3.691099476439791e-05, | |
| "loss": 0.3595, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.945026178010471, | |
| "grad_norm": 5.840355396270752, | |
| "learning_rate": 3.5274869109947647e-05, | |
| "loss": 0.6436, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.927168369293213, | |
| "eval_rouge1": 25.5914, | |
| "eval_rouge2": 10.2904, | |
| "eval_rougeL": 22.3089, | |
| "eval_runtime": 412.3529, | |
| "eval_samples_per_second": 5.323, | |
| "eval_steps_per_second": 0.534, | |
| "step": 4584 | |
| }, | |
| { | |
| "epoch": 3.2722513089005236, | |
| "grad_norm": 5.320337772369385, | |
| "learning_rate": 3.3638743455497386e-05, | |
| "loss": 0.7617, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.599476439790576, | |
| "grad_norm": 5.708410739898682, | |
| "learning_rate": 3.2002617801047126e-05, | |
| "loss": 0.9186, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.9267015706806285, | |
| "grad_norm": 6.043923377990723, | |
| "learning_rate": 3.036649214659686e-05, | |
| "loss": 1.112, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.2077620029449463, | |
| "eval_rouge1": 26.5258, | |
| "eval_rouge2": 11.116, | |
| "eval_rougeL": 23.3381, | |
| "eval_runtime": 397.8337, | |
| "eval_samples_per_second": 5.517, | |
| "eval_steps_per_second": 0.553, | |
| "step": 6112 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15280, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.930549272296653e+16, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |