| { | |
| "best_metric": 1.9820196628570557, | |
| "best_model_checkpoint": "./outputs/checkpoint-4100", | |
| "epoch": 2.987249544626594, | |
| "eval_steps": 100, | |
| "global_step": 4100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7398, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.641923666000366, | |
| "eval_runtime": 206.3681, | |
| "eval_samples_per_second": 30.402, | |
| "eval_steps_per_second": 3.804, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6046, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.592717409133911, | |
| "eval_runtime": 206.4891, | |
| "eval_samples_per_second": 30.384, | |
| "eval_steps_per_second": 3.802, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5628, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.5575265884399414, | |
| "eval_runtime": 206.1916, | |
| "eval_samples_per_second": 30.428, | |
| "eval_steps_per_second": 3.807, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5372, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.525071859359741, | |
| "eval_runtime": 205.8756, | |
| "eval_samples_per_second": 30.475, | |
| "eval_steps_per_second": 3.813, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4945, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4999899864196777, | |
| "eval_runtime": 206.061, | |
| "eval_samples_per_second": 30.447, | |
| "eval_steps_per_second": 3.81, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4681, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.474062919616699, | |
| "eval_runtime": 206.3754, | |
| "eval_samples_per_second": 30.401, | |
| "eval_steps_per_second": 3.804, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4511, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4490628242492676, | |
| "eval_runtime": 205.613, | |
| "eval_samples_per_second": 30.514, | |
| "eval_steps_per_second": 3.818, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.428, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.4292285442352295, | |
| "eval_runtime": 206.1663, | |
| "eval_samples_per_second": 30.432, | |
| "eval_steps_per_second": 3.808, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3917, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.4080374240875244, | |
| "eval_runtime": 206.1066, | |
| "eval_samples_per_second": 30.441, | |
| "eval_steps_per_second": 3.809, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3881, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.3876078128814697, | |
| "eval_runtime": 205.7165, | |
| "eval_samples_per_second": 30.498, | |
| "eval_steps_per_second": 3.816, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3815, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.3660354614257812, | |
| "eval_runtime": 206.3111, | |
| "eval_samples_per_second": 30.41, | |
| "eval_steps_per_second": 3.805, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3383, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.3476908206939697, | |
| "eval_runtime": 206.414, | |
| "eval_samples_per_second": 30.395, | |
| "eval_steps_per_second": 3.803, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3498, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.3309385776519775, | |
| "eval_runtime": 206.425, | |
| "eval_samples_per_second": 30.394, | |
| "eval_steps_per_second": 3.803, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2994, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.3077964782714844, | |
| "eval_runtime": 206.5558, | |
| "eval_samples_per_second": 30.374, | |
| "eval_steps_per_second": 3.8, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2611, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.290616989135742, | |
| "eval_runtime": 206.2977, | |
| "eval_samples_per_second": 30.412, | |
| "eval_steps_per_second": 3.805, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2529, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.2742502689361572, | |
| "eval_runtime": 206.4691, | |
| "eval_samples_per_second": 30.387, | |
| "eval_steps_per_second": 3.802, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2533, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 2.2574470043182373, | |
| "eval_runtime": 205.8893, | |
| "eval_samples_per_second": 30.473, | |
| "eval_steps_per_second": 3.813, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 2.214, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.241929769515991, | |
| "eval_runtime": 205.7921, | |
| "eval_samples_per_second": 30.487, | |
| "eval_steps_per_second": 3.815, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.199, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.2262773513793945, | |
| "eval_runtime": 206.0779, | |
| "eval_samples_per_second": 30.445, | |
| "eval_steps_per_second": 3.809, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1938, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 2.2117421627044678, | |
| "eval_runtime": 206.4818, | |
| "eval_samples_per_second": 30.385, | |
| "eval_steps_per_second": 3.802, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1761, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 2.1993796825408936, | |
| "eval_runtime": 206.3625, | |
| "eval_samples_per_second": 30.403, | |
| "eval_steps_per_second": 3.804, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 2.159, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 2.1830873489379883, | |
| "eval_runtime": 206.5502, | |
| "eval_samples_per_second": 30.375, | |
| "eval_steps_per_second": 3.801, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1497, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 2.1709437370300293, | |
| "eval_runtime": 206.5171, | |
| "eval_samples_per_second": 30.38, | |
| "eval_steps_per_second": 3.801, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1548, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 2.1570258140563965, | |
| "eval_runtime": 206.4592, | |
| "eval_samples_per_second": 30.389, | |
| "eval_steps_per_second": 3.802, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1218, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 2.1442573070526123, | |
| "eval_runtime": 206.6982, | |
| "eval_samples_per_second": 30.353, | |
| "eval_steps_per_second": 3.798, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1246, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 2.1329336166381836, | |
| "eval_runtime": 206.0093, | |
| "eval_samples_per_second": 30.455, | |
| "eval_steps_per_second": 3.811, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1067, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 2.117797374725342, | |
| "eval_runtime": 206.8081, | |
| "eval_samples_per_second": 30.337, | |
| "eval_steps_per_second": 3.796, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0643, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 2.111072540283203, | |
| "eval_runtime": 312.0262, | |
| "eval_samples_per_second": 20.107, | |
| "eval_steps_per_second": 2.516, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0379, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 2.097459554672241, | |
| "eval_runtime": 206.2193, | |
| "eval_samples_per_second": 30.424, | |
| "eval_steps_per_second": 3.807, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0567, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 2.0867855548858643, | |
| "eval_runtime": 206.4732, | |
| "eval_samples_per_second": 30.387, | |
| "eval_steps_per_second": 3.802, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0243, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.0787196159362793, | |
| "eval_runtime": 206.3044, | |
| "eval_samples_per_second": 30.411, | |
| "eval_steps_per_second": 3.805, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0407, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 2.066490888595581, | |
| "eval_runtime": 206.5529, | |
| "eval_samples_per_second": 30.375, | |
| "eval_steps_per_second": 3.8, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 2.009, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 2.0566089153289795, | |
| "eval_runtime": 205.9821, | |
| "eval_samples_per_second": 30.459, | |
| "eval_steps_per_second": 3.811, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 2.002, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 2.0491139888763428, | |
| "eval_runtime": 206.3719, | |
| "eval_samples_per_second": 30.401, | |
| "eval_steps_per_second": 3.804, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0002, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 2.0374956130981445, | |
| "eval_runtime": 206.2009, | |
| "eval_samples_per_second": 30.427, | |
| "eval_steps_per_second": 3.807, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9945, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 2.027383804321289, | |
| "eval_runtime": 206.0279, | |
| "eval_samples_per_second": 30.452, | |
| "eval_steps_per_second": 3.81, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9759, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 2.0199601650238037, | |
| "eval_runtime": 206.7188, | |
| "eval_samples_per_second": 30.35, | |
| "eval_steps_per_second": 3.797, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9675, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 2.0125296115875244, | |
| "eval_runtime": 206.1247, | |
| "eval_samples_per_second": 30.438, | |
| "eval_steps_per_second": 3.808, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9542, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.9999363422393799, | |
| "eval_runtime": 206.6362, | |
| "eval_samples_per_second": 30.363, | |
| "eval_steps_per_second": 3.799, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9555, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 1.9915796518325806, | |
| "eval_runtime": 206.7673, | |
| "eval_samples_per_second": 30.343, | |
| "eval_steps_per_second": 3.797, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9575, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 1.9820196628570557, | |
| "eval_runtime": 207.4963, | |
| "eval_samples_per_second": 30.237, | |
| "eval_steps_per_second": 3.783, | |
| "step": 4100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.1959366338920448e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |