| { | |
| "best_metric": 28.6435, | |
| "best_model_checkpoint": "base_ox-wn_cod_15ep/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": 2.5645, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 11.496853096838793, | |
| "eval_loss": 2.2535109519958496, | |
| "eval_rouge1": 24.4437, | |
| "eval_rouge2": 6.4189, | |
| "eval_rougeL": 22.7949, | |
| "eval_rougeLsum": 22.7909, | |
| "eval_runtime": 92.2279, | |
| "eval_samples_per_second": 151.603, | |
| "eval_steps_per_second": 2.375, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 2.3501, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 11.71499070233157, | |
| "eval_loss": 2.1641664505004883, | |
| "eval_rouge1": 25.6642, | |
| "eval_rouge2": 7.289, | |
| "eval_rougeL": 23.8689, | |
| "eval_rougeLsum": 23.8749, | |
| "eval_runtime": 92.3758, | |
| "eval_samples_per_second": 151.36, | |
| "eval_steps_per_second": 2.371, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4e-05, | |
| "loss": 2.2516, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 11.744099556572737, | |
| "eval_loss": 2.1115779876708984, | |
| "eval_rouge1": 26.4562, | |
| "eval_rouge2": 7.8955, | |
| "eval_rougeL": 24.6275, | |
| "eval_rougeLsum": 24.6376, | |
| "eval_runtime": 91.9643, | |
| "eval_samples_per_second": 152.037, | |
| "eval_steps_per_second": 2.381, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 2.1806, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 11.793019596624232, | |
| "eval_loss": 2.073715925216675, | |
| "eval_rouge1": 27.0392, | |
| "eval_rouge2": 8.2393, | |
| "eval_rougeL": 25.1555, | |
| "eval_rougeLsum": 25.1641, | |
| "eval_runtime": 92.1881, | |
| "eval_samples_per_second": 151.668, | |
| "eval_steps_per_second": 2.376, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.1233, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 11.901444714633099, | |
| "eval_loss": 2.0459506511688232, | |
| "eval_rouge1": 27.2709, | |
| "eval_rouge2": 8.4244, | |
| "eval_rougeL": 25.3847, | |
| "eval_rougeLsum": 25.4003, | |
| "eval_runtime": 92.1849, | |
| "eval_samples_per_second": 151.674, | |
| "eval_steps_per_second": 2.376, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0765, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 11.811257330853955, | |
| "eval_loss": 2.0235939025878906, | |
| "eval_rouge1": 27.5456, | |
| "eval_rouge2": 8.6096, | |
| "eval_rougeL": 25.6321, | |
| "eval_rougeLsum": 25.6462, | |
| "eval_runtime": 92.2077, | |
| "eval_samples_per_second": 151.636, | |
| "eval_steps_per_second": 2.375, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 2.0371, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 11.687455299671006, | |
| "eval_loss": 2.004668951034546, | |
| "eval_rouge1": 27.7209, | |
| "eval_rouge2": 8.7277, | |
| "eval_rougeL": 25.7871, | |
| "eval_rougeLsum": 25.8084, | |
| "eval_runtime": 92.3796, | |
| "eval_samples_per_second": 151.354, | |
| "eval_steps_per_second": 2.371, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 2.0036, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 11.548705478472321, | |
| "eval_loss": 1.9917967319488525, | |
| "eval_rouge1": 28.0431, | |
| "eval_rouge2": 8.9863, | |
| "eval_rougeL": 26.1072, | |
| "eval_rougeLsum": 26.1198, | |
| "eval_runtime": 92.2225, | |
| "eval_samples_per_second": 151.612, | |
| "eval_steps_per_second": 2.375, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.9752, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 11.793949363467314, | |
| "eval_loss": 1.978798747062683, | |
| "eval_rouge1": 28.1807, | |
| "eval_rouge2": 9.0219, | |
| "eval_rougeL": 26.1692, | |
| "eval_rougeLsum": 26.1886, | |
| "eval_runtime": 92.3313, | |
| "eval_samples_per_second": 151.433, | |
| "eval_steps_per_second": 2.372, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.9513, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 11.593620369045917, | |
| "eval_loss": 1.9701664447784424, | |
| "eval_rouge1": 28.3204, | |
| "eval_rouge2": 9.1572, | |
| "eval_rougeL": 26.2955, | |
| "eval_rougeLsum": 26.3029, | |
| "eval_runtime": 92.353, | |
| "eval_samples_per_second": 151.397, | |
| "eval_steps_per_second": 2.371, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 1.9309, | |
| "step": 30140 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 11.837076240881133, | |
| "eval_loss": 1.9640249013900757, | |
| "eval_rouge1": 28.4289, | |
| "eval_rouge2": 9.2845, | |
| "eval_rougeL": 26.4006, | |
| "eval_rougeLsum": 26.418, | |
| "eval_runtime": 92.2786, | |
| "eval_samples_per_second": 151.519, | |
| "eval_steps_per_second": 2.373, | |
| "step": 30140 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9144, | |
| "step": 32880 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 11.620082963810614, | |
| "eval_loss": 1.9570931196212769, | |
| "eval_rouge1": 28.4504, | |
| "eval_rouge2": 9.3406, | |
| "eval_rougeL": 26.4273, | |
| "eval_rougeLsum": 26.4384, | |
| "eval_runtime": 143.3751, | |
| "eval_samples_per_second": 97.52, | |
| "eval_steps_per_second": 1.527, | |
| "step": 32880 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.9013, | |
| "step": 35620 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_gen_len": 11.706694321270204, | |
| "eval_loss": 1.954399824142456, | |
| "eval_rouge1": 28.6319, | |
| "eval_rouge2": 9.3682, | |
| "eval_rougeL": 26.605, | |
| "eval_rougeLsum": 26.613, | |
| "eval_runtime": 92.2917, | |
| "eval_samples_per_second": 151.498, | |
| "eval_steps_per_second": 2.373, | |
| "step": 35620 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.8914, | |
| "step": 38360 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_gen_len": 11.730653697611215, | |
| "eval_loss": 1.9511938095092773, | |
| "eval_rouge1": 28.6435, | |
| "eval_rouge2": 9.3976, | |
| "eval_rougeL": 26.5839, | |
| "eval_rougeLsum": 26.5918, | |
| "eval_runtime": 92.0528, | |
| "eval_samples_per_second": 151.891, | |
| "eval_steps_per_second": 2.379, | |
| "step": 38360 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.8866, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_gen_len": 11.74696037762838, | |
| "eval_loss": 1.9508894681930542, | |
| "eval_rouge1": 28.6111, | |
| "eval_rouge2": 9.3857, | |
| "eval_rougeL": 26.551, | |
| "eval_rougeLsum": 26.5648, | |
| "eval_runtime": 91.9698, | |
| "eval_samples_per_second": 152.028, | |
| "eval_steps_per_second": 2.381, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 41100, | |
| "total_flos": 4.436322961707786e+17, | |
| "train_loss": 2.069233327155566, | |
| "train_runtime": 14173.3385, | |
| "train_samples_per_second": 185.558, | |
| "train_steps_per_second": 2.9 | |
| } | |
| ], | |
| "max_steps": 41100, | |
| "num_train_epochs": 15, | |
| "total_flos": 4.436322961707786e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |