| { | |
| "best_metric": 35.2849, | |
| "best_model_checkpoint": "large_ox-wn_cod_15ep_eap/checkpoint-38360", | |
| "epoch": 15.0, | |
| "global_step": 41100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 2.1769, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 11.342940924045202, | |
| "eval_loss": 1.905047059059143, | |
| "eval_rouge1": 28.7222, | |
| "eval_rouge2": 9.1873, | |
| "eval_rougeL": 26.6888, | |
| "eval_rougeLsum": 26.6937, | |
| "eval_runtime": 86.0596, | |
| "eval_samples_per_second": 162.469, | |
| "eval_steps_per_second": 1.278, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 1.9408, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 11.416464025175225, | |
| "eval_loss": 1.8151417970657349, | |
| "eval_rouge1": 29.8799, | |
| "eval_rouge2": 10.2327, | |
| "eval_rougeL": 27.7947, | |
| "eval_rougeLsum": 27.8044, | |
| "eval_runtime": 98.7019, | |
| "eval_samples_per_second": 141.659, | |
| "eval_steps_per_second": 1.114, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8124, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 11.531039908453726, | |
| "eval_loss": 1.7607892751693726, | |
| "eval_rouge1": 30.9845, | |
| "eval_rouge2": 10.9982, | |
| "eval_rougeL": 28.8059, | |
| "eval_rougeLsum": 28.8131, | |
| "eval_runtime": 96.5011, | |
| "eval_samples_per_second": 144.889, | |
| "eval_steps_per_second": 1.14, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 1.7118, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 11.703690459161779, | |
| "eval_loss": 1.7228699922561646, | |
| "eval_rouge1": 31.6943, | |
| "eval_rouge2": 11.7412, | |
| "eval_rougeL": 29.4967, | |
| "eval_rougeLsum": 29.5319, | |
| "eval_runtime": 87.7321, | |
| "eval_samples_per_second": 159.372, | |
| "eval_steps_per_second": 1.254, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.6286, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 11.77835788871406, | |
| "eval_loss": 1.6936795711517334, | |
| "eval_rouge1": 32.5839, | |
| "eval_rouge2": 12.2431, | |
| "eval_rougeL": 30.1799, | |
| "eval_rougeLsum": 30.206, | |
| "eval_runtime": 84.5028, | |
| "eval_samples_per_second": 165.462, | |
| "eval_steps_per_second": 1.302, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5597, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 11.597410956944643, | |
| "eval_loss": 1.674757480621338, | |
| "eval_rouge1": 32.9915, | |
| "eval_rouge2": 12.8514, | |
| "eval_rougeL": 30.7016, | |
| "eval_rougeLsum": 30.7145, | |
| "eval_runtime": 87.802, | |
| "eval_samples_per_second": 159.245, | |
| "eval_steps_per_second": 1.253, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 1.4982, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 11.358031755113718, | |
| "eval_loss": 1.6578471660614014, | |
| "eval_rouge1": 33.2157, | |
| "eval_rouge2": 13.1389, | |
| "eval_rougeL": 30.9428, | |
| "eval_rougeLsum": 30.9519, | |
| "eval_runtime": 89.406, | |
| "eval_samples_per_second": 156.388, | |
| "eval_steps_per_second": 1.23, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 1.4468, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 11.572378772707767, | |
| "eval_loss": 1.6473166942596436, | |
| "eval_rouge1": 33.6146, | |
| "eval_rouge2": 13.5922, | |
| "eval_rougeL": 31.3001, | |
| "eval_rougeLsum": 31.3235, | |
| "eval_runtime": 98.6248, | |
| "eval_samples_per_second": 141.77, | |
| "eval_steps_per_second": 1.115, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4022, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 11.738878558146189, | |
| "eval_loss": 1.6383947134017944, | |
| "eval_rouge1": 34.1711, | |
| "eval_rouge2": 14.1117, | |
| "eval_rougeL": 31.7951, | |
| "eval_rougeLsum": 31.8066, | |
| "eval_runtime": 89.5372, | |
| "eval_samples_per_second": 156.159, | |
| "eval_steps_per_second": 1.229, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.364, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 11.665856100700902, | |
| "eval_loss": 1.6336920261383057, | |
| "eval_rouge1": 34.5489, | |
| "eval_rouge2": 14.5012, | |
| "eval_rougeL": 32.1329, | |
| "eval_rougeLsum": 32.1446, | |
| "eval_runtime": 103.7766, | |
| "eval_samples_per_second": 134.732, | |
| "eval_steps_per_second": 1.06, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 1.3321, | |
| "step": 30140 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 11.800314690316121, | |
| "eval_loss": 1.6291483640670776, | |
| "eval_rouge1": 34.7133, | |
| "eval_rouge2": 14.7297, | |
| "eval_rougeL": 32.3042, | |
| "eval_rougeLsum": 32.314, | |
| "eval_runtime": 91.3961, | |
| "eval_samples_per_second": 152.982, | |
| "eval_steps_per_second": 1.204, | |
| "step": 30140 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3054, | |
| "step": 32880 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 11.761908167644114, | |
| "eval_loss": 1.6267131567001343, | |
| "eval_rouge1": 34.9411, | |
| "eval_rouge2": 15.0282, | |
| "eval_rougeL": 32.5335, | |
| "eval_rougeLsum": 32.5451, | |
| "eval_runtime": 98.5092, | |
| "eval_samples_per_second": 141.936, | |
| "eval_steps_per_second": 1.117, | |
| "step": 32880 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.2845, | |
| "step": 35620 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_gen_len": 11.831712201401801, | |
| "eval_loss": 1.626239538192749, | |
| "eval_rouge1": 35.1648, | |
| "eval_rouge2": 15.2154, | |
| "eval_rougeL": 32.7387, | |
| "eval_rougeLsum": 32.742, | |
| "eval_runtime": 85.528, | |
| "eval_samples_per_second": 163.479, | |
| "eval_steps_per_second": 1.286, | |
| "step": 35620 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.2699, | |
| "step": 38360 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_gen_len": 11.816764411386067, | |
| "eval_loss": 1.6257190704345703, | |
| "eval_rouge1": 35.2849, | |
| "eval_rouge2": 15.3109, | |
| "eval_rougeL": 32.8508, | |
| "eval_rougeLsum": 32.853, | |
| "eval_runtime": 84.6116, | |
| "eval_samples_per_second": 165.249, | |
| "eval_steps_per_second": 1.3, | |
| "step": 38360 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.2595, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_gen_len": 11.797096266628522, | |
| "eval_loss": 1.6273423433303833, | |
| "eval_rouge1": 35.2224, | |
| "eval_rouge2": 15.2781, | |
| "eval_rougeL": 32.7718, | |
| "eval_rougeLsum": 32.7826, | |
| "eval_runtime": 95.1523, | |
| "eval_samples_per_second": 146.943, | |
| "eval_steps_per_second": 1.156, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 41100, | |
| "total_flos": 9.049973435337277e+17, | |
| "train_loss": 1.5328590292826185, | |
| "train_runtime": 18417.6233, | |
| "train_samples_per_second": 142.797, | |
| "train_steps_per_second": 2.232 | |
| } | |
| ], | |
| "max_steps": 41100, | |
| "num_train_epochs": 15, | |
| "total_flos": 9.049973435337277e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |