| { | |
| "best_metric": 0.8900671601295471, | |
| "best_model_checkpoint": "./outputs/checkpoint-3600", | |
| "epoch": 4.8, | |
| "eval_steps": 100, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.001, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.0910990238189697, | |
| "eval_runtime": 728.6528, | |
| "eval_samples_per_second": 2.649, | |
| "eval_steps_per_second": 0.332, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8971, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.0677164793014526, | |
| "eval_runtime": 727.2702, | |
| "eval_samples_per_second": 2.654, | |
| "eval_steps_per_second": 0.333, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8815, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.052659273147583, | |
| "eval_runtime": 730.164, | |
| "eval_samples_per_second": 2.643, | |
| "eval_steps_per_second": 0.331, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8695, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.0382879972457886, | |
| "eval_runtime": 732.8796, | |
| "eval_samples_per_second": 2.633, | |
| "eval_steps_per_second": 0.33, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8548, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.0260452032089233, | |
| "eval_runtime": 730.1971, | |
| "eval_samples_per_second": 2.643, | |
| "eval_steps_per_second": 0.331, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8496, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.0152662992477417, | |
| "eval_runtime": 732.4802, | |
| "eval_samples_per_second": 2.635, | |
| "eval_steps_per_second": 0.33, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8445, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.0105178356170654, | |
| "eval_runtime": 802.0691, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 0.302, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8346, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.0066964626312256, | |
| "eval_runtime": 789.7964, | |
| "eval_samples_per_second": 2.444, | |
| "eval_steps_per_second": 0.306, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8199, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.9989904165267944, | |
| "eval_runtime": 786.9058, | |
| "eval_samples_per_second": 2.453, | |
| "eval_steps_per_second": 0.308, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8133, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.9895688891410828, | |
| "eval_runtime": 782.3169, | |
| "eval_samples_per_second": 2.467, | |
| "eval_steps_per_second": 0.309, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8193, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.9853964447975159, | |
| "eval_runtime": 777.984, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 0.311, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8091, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.9784607887268066, | |
| "eval_runtime": 780.7522, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 0.31, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7983, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.9710213541984558, | |
| "eval_runtime": 787.7479, | |
| "eval_samples_per_second": 2.45, | |
| "eval_steps_per_second": 0.307, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7971, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 0.9653750658035278, | |
| "eval_runtime": 777.1907, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 0.311, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7899, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9598689675331116, | |
| "eval_runtime": 788.3937, | |
| "eval_samples_per_second": 2.448, | |
| "eval_steps_per_second": 0.307, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7689, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.9570510983467102, | |
| "eval_runtime": 782.2868, | |
| "eval_samples_per_second": 2.467, | |
| "eval_steps_per_second": 0.309, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7676, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.9513885378837585, | |
| "eval_runtime": 768.2363, | |
| "eval_samples_per_second": 2.512, | |
| "eval_steps_per_second": 0.315, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.766, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.9479925036430359, | |
| "eval_runtime": 792.9294, | |
| "eval_samples_per_second": 2.434, | |
| "eval_steps_per_second": 0.305, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7587, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.9420929551124573, | |
| "eval_runtime": 787.6767, | |
| "eval_samples_per_second": 2.45, | |
| "eval_steps_per_second": 0.307, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7546, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.9371617436408997, | |
| "eval_runtime": 785.7561, | |
| "eval_samples_per_second": 2.456, | |
| "eval_steps_per_second": 0.308, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7635, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 0.9326320290565491, | |
| "eval_runtime": 789.0027, | |
| "eval_samples_per_second": 2.446, | |
| "eval_steps_per_second": 0.307, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7632, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_loss": 0.9287375211715698, | |
| "eval_runtime": 785.4978, | |
| "eval_samples_per_second": 2.457, | |
| "eval_steps_per_second": 0.308, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7409, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_loss": 0.9274052381515503, | |
| "eval_runtime": 779.2439, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 0.311, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7247, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.9260075688362122, | |
| "eval_runtime": 775.8037, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 0.312, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7277, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.921540379524231, | |
| "eval_runtime": 774.2681, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 0.313, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7242, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "eval_loss": 0.9189460277557373, | |
| "eval_runtime": 777.6576, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 0.311, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7246, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_loss": 0.9128983020782471, | |
| "eval_runtime": 776.5971, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 0.312, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7224, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": 0.9102813601493835, | |
| "eval_runtime": 777.5262, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 0.311, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7222, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.9069462418556213, | |
| "eval_runtime": 778.8375, | |
| "eval_samples_per_second": 2.478, | |
| "eval_steps_per_second": 0.311, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7253, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9039433002471924, | |
| "eval_runtime": 773.591, | |
| "eval_samples_per_second": 2.495, | |
| "eval_steps_per_second": 0.313, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6845, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "eval_loss": 0.906529426574707, | |
| "eval_runtime": 784.1515, | |
| "eval_samples_per_second": 2.461, | |
| "eval_steps_per_second": 0.309, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6886, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_loss": 0.9038098454475403, | |
| "eval_runtime": 786.1544, | |
| "eval_samples_per_second": 2.455, | |
| "eval_steps_per_second": 0.308, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7049, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "eval_loss": 0.9004936218261719, | |
| "eval_runtime": 777.3844, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 0.311, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6887, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_loss": 0.8973463177680969, | |
| "eval_runtime": 777.6425, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 0.311, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6963, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 0.8955456614494324, | |
| "eval_runtime": 796.2741, | |
| "eval_samples_per_second": 2.424, | |
| "eval_steps_per_second": 0.304, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6882, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.8900671601295471, | |
| "eval_runtime": 776.8193, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 0.312, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "total_flos": 9.323259516653568e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |