| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.946524064171124, | |
| "eval_steps": 10, | |
| "global_step": 930, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.892473118279571e-05, | |
| "loss": 1.5206, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.292553186416626, | |
| "eval_loss": 2.0217974185943604, | |
| "eval_runtime": 19.3615, | |
| "eval_samples_per_second": 9.71, | |
| "eval_steps_per_second": 2.428, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.78494623655914e-05, | |
| "loss": 1.2661, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_accuracy": 0.3404255211353302, | |
| "eval_loss": 1.633784532546997, | |
| "eval_runtime": 22.8427, | |
| "eval_samples_per_second": 8.23, | |
| "eval_steps_per_second": 2.058, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 9.677419354838711e-05, | |
| "loss": 1.2961, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.4680851101875305, | |
| "eval_loss": 1.4141530990600586, | |
| "eval_runtime": 26.4217, | |
| "eval_samples_per_second": 7.115, | |
| "eval_steps_per_second": 1.779, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.56989247311828e-05, | |
| "loss": 1.1222, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.5159574747085571, | |
| "eval_loss": 1.227504014968872, | |
| "eval_runtime": 20.4617, | |
| "eval_samples_per_second": 9.188, | |
| "eval_steps_per_second": 2.297, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.46236559139785e-05, | |
| "loss": 0.9449, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.5531914830207825, | |
| "eval_loss": 1.2353336811065674, | |
| "eval_runtime": 19.3865, | |
| "eval_samples_per_second": 9.697, | |
| "eval_steps_per_second": 2.424, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.35483870967742e-05, | |
| "loss": 0.8884, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.5531914830207825, | |
| "eval_loss": 1.2557822465896606, | |
| "eval_runtime": 19.3788, | |
| "eval_samples_per_second": 9.701, | |
| "eval_steps_per_second": 2.425, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.247311827956989e-05, | |
| "loss": 1.1132, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.6010638475418091, | |
| "eval_loss": 1.172572374343872, | |
| "eval_runtime": 19.3951, | |
| "eval_samples_per_second": 9.693, | |
| "eval_steps_per_second": 2.423, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 9.13978494623656e-05, | |
| "loss": 0.9073, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.5744680762290955, | |
| "eval_loss": 1.2864995002746582, | |
| "eval_runtime": 20.2133, | |
| "eval_samples_per_second": 9.301, | |
| "eval_steps_per_second": 2.325, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 9.032258064516129e-05, | |
| "loss": 0.8923, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.6808510422706604, | |
| "eval_loss": 0.8788278698921204, | |
| "eval_runtime": 19.3337, | |
| "eval_samples_per_second": 9.724, | |
| "eval_steps_per_second": 2.431, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.924731182795699e-05, | |
| "loss": 0.8702, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.6808510422706604, | |
| "eval_loss": 0.9209534525871277, | |
| "eval_runtime": 19.7461, | |
| "eval_samples_per_second": 9.521, | |
| "eval_steps_per_second": 2.38, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 8.81720430107527e-05, | |
| "loss": 0.5832, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.6010638475418091, | |
| "eval_loss": 0.9852186441421509, | |
| "eval_runtime": 20.6843, | |
| "eval_samples_per_second": 9.089, | |
| "eval_steps_per_second": 2.272, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 8.709677419354839e-05, | |
| "loss": 0.776, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_accuracy": 0.664893627166748, | |
| "eval_loss": 1.0170938968658447, | |
| "eval_runtime": 19.2567, | |
| "eval_samples_per_second": 9.763, | |
| "eval_steps_per_second": 2.441, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 8.60215053763441e-05, | |
| "loss": 0.5927, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.6382978558540344, | |
| "eval_loss": 1.101302146911621, | |
| "eval_runtime": 20.3489, | |
| "eval_samples_per_second": 9.239, | |
| "eval_steps_per_second": 2.31, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 8.494623655913979e-05, | |
| "loss": 0.54, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.6808510422706604, | |
| "eval_loss": 0.9849129319190979, | |
| "eval_runtime": 19.1842, | |
| "eval_samples_per_second": 9.8, | |
| "eval_steps_per_second": 2.45, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.387096774193549e-05, | |
| "loss": 0.6341, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.6968085169792175, | |
| "eval_loss": 0.8642140626907349, | |
| "eval_runtime": 19.8877, | |
| "eval_samples_per_second": 9.453, | |
| "eval_steps_per_second": 2.363, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.27956989247312e-05, | |
| "loss": 0.8108, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_accuracy": 0.7340425252914429, | |
| "eval_loss": 0.7330851554870605, | |
| "eval_runtime": 20.2573, | |
| "eval_samples_per_second": 9.281, | |
| "eval_steps_per_second": 2.32, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 8.172043010752689e-05, | |
| "loss": 0.5861, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_accuracy": 0.7127659320831299, | |
| "eval_loss": 0.766554594039917, | |
| "eval_runtime": 19.2481, | |
| "eval_samples_per_second": 9.767, | |
| "eval_steps_per_second": 2.442, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 8.064516129032258e-05, | |
| "loss": 0.4829, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_accuracy": 0.7180851101875305, | |
| "eval_loss": 0.7813696265220642, | |
| "eval_runtime": 20.2831, | |
| "eval_samples_per_second": 9.269, | |
| "eval_steps_per_second": 2.317, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 7.956989247311829e-05, | |
| "loss": 0.6267, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.7234042286872864, | |
| "eval_loss": 0.7261902093887329, | |
| "eval_runtime": 19.8614, | |
| "eval_samples_per_second": 9.466, | |
| "eval_steps_per_second": 2.366, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 7.849462365591398e-05, | |
| "loss": 0.4318, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.7659574747085571, | |
| "eval_loss": 0.732907235622406, | |
| "eval_runtime": 20.4384, | |
| "eval_samples_per_second": 9.198, | |
| "eval_steps_per_second": 2.3, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.741935483870968e-05, | |
| "loss": 0.3761, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 0.7021276354789734, | |
| "eval_loss": 1.0907269716262817, | |
| "eval_runtime": 19.6052, | |
| "eval_samples_per_second": 9.589, | |
| "eval_steps_per_second": 2.397, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 7.634408602150538e-05, | |
| "loss": 0.324, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_accuracy": 0.7872340679168701, | |
| "eval_loss": 0.7849129438400269, | |
| "eval_runtime": 19.7685, | |
| "eval_samples_per_second": 9.51, | |
| "eval_steps_per_second": 2.378, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.526881720430108e-05, | |
| "loss": 0.3168, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_accuracy": 0.7553191781044006, | |
| "eval_loss": 0.7544443607330322, | |
| "eval_runtime": 19.9734, | |
| "eval_samples_per_second": 9.413, | |
| "eval_steps_per_second": 2.353, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.419354838709677e-05, | |
| "loss": 0.31, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.6808510422706604, | |
| "eval_loss": 1.3076503276824951, | |
| "eval_runtime": 19.5097, | |
| "eval_samples_per_second": 9.636, | |
| "eval_steps_per_second": 2.409, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 7.311827956989248e-05, | |
| "loss": 0.2488, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_accuracy": 0.7393617033958435, | |
| "eval_loss": 1.1082981824874878, | |
| "eval_runtime": 19.6472, | |
| "eval_samples_per_second": 9.569, | |
| "eval_steps_per_second": 2.392, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 7.204301075268818e-05, | |
| "loss": 0.2774, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_accuracy": 0.7393617033958435, | |
| "eval_loss": 1.0004332065582275, | |
| "eval_runtime": 19.5383, | |
| "eval_samples_per_second": 9.622, | |
| "eval_steps_per_second": 2.406, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.096774193548388e-05, | |
| "loss": 0.3068, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.7393617033958435, | |
| "eval_loss": 1.0832968950271606, | |
| "eval_runtime": 22.726, | |
| "eval_samples_per_second": 8.272, | |
| "eval_steps_per_second": 2.068, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 6.989247311827958e-05, | |
| "loss": 0.4347, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.6808510422706604, | |
| "eval_loss": 1.3584102392196655, | |
| "eval_runtime": 19.4201, | |
| "eval_samples_per_second": 9.681, | |
| "eval_steps_per_second": 2.42, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 6.881720430107527e-05, | |
| "loss": 0.2742, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.7765957713127136, | |
| "eval_loss": 0.9648256301879883, | |
| "eval_runtime": 20.0148, | |
| "eval_samples_per_second": 9.393, | |
| "eval_steps_per_second": 2.348, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 6.774193548387096e-05, | |
| "loss": 0.3018, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.7446808218955994, | |
| "eval_loss": 0.9408243298530579, | |
| "eval_runtime": 20.3383, | |
| "eval_samples_per_second": 9.244, | |
| "eval_steps_per_second": 2.311, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.4159, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "eval_accuracy": 0.7127659320831299, | |
| "eval_loss": 1.3147622346878052, | |
| "eval_runtime": 28.2012, | |
| "eval_samples_per_second": 6.666, | |
| "eval_steps_per_second": 1.667, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 6.559139784946236e-05, | |
| "loss": 0.35, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_accuracy": 0.7659574747085571, | |
| "eval_loss": 0.9583517909049988, | |
| "eval_runtime": 21.7271, | |
| "eval_samples_per_second": 8.653, | |
| "eval_steps_per_second": 2.163, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 6.451612903225807e-05, | |
| "loss": 0.1448, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_accuracy": 0.7819148898124695, | |
| "eval_loss": 0.7321730256080627, | |
| "eval_runtime": 21.3525, | |
| "eval_samples_per_second": 8.805, | |
| "eval_steps_per_second": 2.201, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 6.344086021505376e-05, | |
| "loss": 0.1969, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "eval_accuracy": 0.8297872543334961, | |
| "eval_loss": 0.7202281355857849, | |
| "eval_runtime": 19.4523, | |
| "eval_samples_per_second": 9.665, | |
| "eval_steps_per_second": 2.416, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 6.236559139784946e-05, | |
| "loss": 0.4023, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_accuracy": 0.7340425252914429, | |
| "eval_loss": 1.210102915763855, | |
| "eval_runtime": 19.9948, | |
| "eval_samples_per_second": 9.402, | |
| "eval_steps_per_second": 2.351, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 6.129032258064517e-05, | |
| "loss": 0.301, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.4878663122653961, | |
| "eval_runtime": 21.8287, | |
| "eval_samples_per_second": 8.613, | |
| "eval_steps_per_second": 2.153, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 6.021505376344086e-05, | |
| "loss": 0.1968, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_accuracy": 0.8191489577293396, | |
| "eval_loss": 0.721879243850708, | |
| "eval_runtime": 19.355, | |
| "eval_samples_per_second": 9.713, | |
| "eval_steps_per_second": 2.428, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 5.913978494623657e-05, | |
| "loss": 0.1026, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "eval_accuracy": 0.7234042286872864, | |
| "eval_loss": 1.1493706703186035, | |
| "eval_runtime": 19.4745, | |
| "eval_samples_per_second": 9.654, | |
| "eval_steps_per_second": 2.413, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 5.8064516129032266e-05, | |
| "loss": 0.2143, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.7446808218955994, | |
| "eval_loss": 1.1254618167877197, | |
| "eval_runtime": 19.5314, | |
| "eval_samples_per_second": 9.626, | |
| "eval_steps_per_second": 2.406, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 5.6989247311827965e-05, | |
| "loss": 0.2157, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "eval_accuracy": 0.7978723645210266, | |
| "eval_loss": 0.867353618144989, | |
| "eval_runtime": 19.5612, | |
| "eval_samples_per_second": 9.611, | |
| "eval_steps_per_second": 2.403, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 5.5913978494623656e-05, | |
| "loss": 0.1429, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "eval_accuracy": 0.8031914830207825, | |
| "eval_loss": 0.8620893359184265, | |
| "eval_runtime": 19.8011, | |
| "eval_samples_per_second": 9.494, | |
| "eval_steps_per_second": 2.374, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 5.4838709677419355e-05, | |
| "loss": 0.2338, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "eval_accuracy": 0.771276593208313, | |
| "eval_loss": 1.0524601936340332, | |
| "eval_runtime": 20.0193, | |
| "eval_samples_per_second": 9.391, | |
| "eval_steps_per_second": 2.348, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 5.3763440860215054e-05, | |
| "loss": 0.079, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "eval_accuracy": 0.7978723645210266, | |
| "eval_loss": 0.9354122877120972, | |
| "eval_runtime": 19.937, | |
| "eval_samples_per_second": 9.43, | |
| "eval_steps_per_second": 2.357, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 5.268817204301075e-05, | |
| "loss": 0.0652, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "eval_accuracy": 0.8297872543334961, | |
| "eval_loss": 0.8370540142059326, | |
| "eval_runtime": 20.2037, | |
| "eval_samples_per_second": 9.305, | |
| "eval_steps_per_second": 2.326, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 5.161290322580645e-05, | |
| "loss": 0.0464, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "eval_accuracy": 0.7978723645210266, | |
| "eval_loss": 0.9808104634284973, | |
| "eval_runtime": 21.6193, | |
| "eval_samples_per_second": 8.696, | |
| "eval_steps_per_second": 2.174, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 5.053763440860215e-05, | |
| "loss": 0.2738, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.7124775052070618, | |
| "eval_runtime": 20.7786, | |
| "eval_samples_per_second": 9.048, | |
| "eval_steps_per_second": 2.262, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 4.9462365591397855e-05, | |
| "loss": 0.0234, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_accuracy": 0.813829779624939, | |
| "eval_loss": 0.799345850944519, | |
| "eval_runtime": 22.0458, | |
| "eval_samples_per_second": 8.528, | |
| "eval_steps_per_second": 2.132, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.1504, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "eval_accuracy": 0.7978723645210266, | |
| "eval_loss": 0.9516879916191101, | |
| "eval_runtime": 20.0905, | |
| "eval_samples_per_second": 9.358, | |
| "eval_steps_per_second": 2.339, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.731182795698925e-05, | |
| "loss": 0.0289, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "eval_accuracy": 0.8404255509376526, | |
| "eval_loss": 0.7372261881828308, | |
| "eval_runtime": 19.4546, | |
| "eval_samples_per_second": 9.664, | |
| "eval_steps_per_second": 2.416, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 4.6236559139784944e-05, | |
| "loss": 0.2253, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.47867247462272644, | |
| "eval_runtime": 19.7513, | |
| "eval_samples_per_second": 9.518, | |
| "eval_steps_per_second": 2.38, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.516129032258064e-05, | |
| "loss": 0.0487, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "eval_accuracy": 0.8457446694374084, | |
| "eval_loss": 0.6794525384902954, | |
| "eval_runtime": 20.9065, | |
| "eval_samples_per_second": 8.992, | |
| "eval_steps_per_second": 2.248, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 4.408602150537635e-05, | |
| "loss": 0.0266, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_accuracy": 0.835106372833252, | |
| "eval_loss": 0.8275220990180969, | |
| "eval_runtime": 20.2058, | |
| "eval_samples_per_second": 9.304, | |
| "eval_steps_per_second": 2.326, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 4.301075268817205e-05, | |
| "loss": 0.07, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "eval_accuracy": 0.7978723645210266, | |
| "eval_loss": 1.0128568410873413, | |
| "eval_runtime": 20.839, | |
| "eval_samples_per_second": 9.022, | |
| "eval_steps_per_second": 2.255, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 4.1935483870967746e-05, | |
| "loss": 0.0439, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "eval_accuracy": 0.835106372833252, | |
| "eval_loss": 0.8743005394935608, | |
| "eval_runtime": 20.318, | |
| "eval_samples_per_second": 9.253, | |
| "eval_steps_per_second": 2.313, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 4.0860215053763444e-05, | |
| "loss": 0.0609, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_accuracy": 0.8457446694374084, | |
| "eval_loss": 0.7512069344520569, | |
| "eval_runtime": 19.5322, | |
| "eval_samples_per_second": 9.625, | |
| "eval_steps_per_second": 2.406, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 3.978494623655914e-05, | |
| "loss": 0.0902, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.702180802822113, | |
| "eval_runtime": 19.3513, | |
| "eval_samples_per_second": 9.715, | |
| "eval_steps_per_second": 2.429, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 3.870967741935484e-05, | |
| "loss": 0.0364, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.7819148898124695, | |
| "eval_loss": 1.0619040727615356, | |
| "eval_runtime": 21.233, | |
| "eval_samples_per_second": 8.854, | |
| "eval_steps_per_second": 2.214, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 3.763440860215054e-05, | |
| "loss": 0.033, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "eval_accuracy": 0.7765957713127136, | |
| "eval_loss": 1.243336796760559, | |
| "eval_runtime": 19.5002, | |
| "eval_samples_per_second": 9.641, | |
| "eval_steps_per_second": 2.41, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.1456, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "eval_accuracy": 0.8244680762290955, | |
| "eval_loss": 0.8834251165390015, | |
| "eval_runtime": 19.3312, | |
| "eval_samples_per_second": 9.725, | |
| "eval_steps_per_second": 2.431, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 3.5591397849462364e-05, | |
| "loss": 0.0049, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.8070632219314575, | |
| "eval_runtime": 20.1893, | |
| "eval_samples_per_second": 9.312, | |
| "eval_steps_per_second": 2.328, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 3.451612903225806e-05, | |
| "loss": 0.1007, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_accuracy": 0.8670212626457214, | |
| "eval_loss": 0.6686306595802307, | |
| "eval_runtime": 19.5009, | |
| "eval_samples_per_second": 9.641, | |
| "eval_steps_per_second": 2.41, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3.344086021505377e-05, | |
| "loss": 0.0054, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "eval_accuracy": 0.8776595592498779, | |
| "eval_loss": 0.6421190500259399, | |
| "eval_runtime": 21.8176, | |
| "eval_samples_per_second": 8.617, | |
| "eval_steps_per_second": 2.154, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 3.236559139784947e-05, | |
| "loss": 0.0027, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "eval_accuracy": 0.8563829660415649, | |
| "eval_loss": 0.7604753375053406, | |
| "eval_runtime": 19.6912, | |
| "eval_samples_per_second": 9.547, | |
| "eval_steps_per_second": 2.387, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 3.1290322580645166e-05, | |
| "loss": 0.1582, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "eval_accuracy": 0.8670212626457214, | |
| "eval_loss": 0.7657257914543152, | |
| "eval_runtime": 19.5763, | |
| "eval_samples_per_second": 9.603, | |
| "eval_steps_per_second": 2.401, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 3.0215053763440858e-05, | |
| "loss": 0.0106, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "eval_accuracy": 0.835106372833252, | |
| "eval_loss": 0.8570267558097839, | |
| "eval_runtime": 20.2786, | |
| "eval_samples_per_second": 9.271, | |
| "eval_steps_per_second": 2.318, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 2.913978494623656e-05, | |
| "loss": 0.0219, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "eval_accuracy": 0.8457446694374084, | |
| "eval_loss": 0.8243122696876526, | |
| "eval_runtime": 20.636, | |
| "eval_samples_per_second": 9.11, | |
| "eval_steps_per_second": 2.278, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 2.806451612903226e-05, | |
| "loss": 0.0032, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "eval_accuracy": 0.8085106611251831, | |
| "eval_loss": 0.9377107620239258, | |
| "eval_runtime": 19.4326, | |
| "eval_samples_per_second": 9.674, | |
| "eval_steps_per_second": 2.419, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 2.698924731182796e-05, | |
| "loss": 0.0037, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "eval_accuracy": 0.8031914830207825, | |
| "eval_loss": 1.0501412153244019, | |
| "eval_runtime": 19.4587, | |
| "eval_samples_per_second": 9.661, | |
| "eval_steps_per_second": 2.415, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 2.591397849462366e-05, | |
| "loss": 0.0072, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "eval_accuracy": 0.8297872543334961, | |
| "eval_loss": 0.8667866587638855, | |
| "eval_runtime": 19.4943, | |
| "eval_samples_per_second": 9.644, | |
| "eval_steps_per_second": 2.411, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 2.4838709677419354e-05, | |
| "loss": 0.0626, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "eval_accuracy": 0.8617021441459656, | |
| "eval_loss": 0.6626190543174744, | |
| "eval_runtime": 19.4854, | |
| "eval_samples_per_second": 9.648, | |
| "eval_steps_per_second": 2.412, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 2.3763440860215056e-05, | |
| "loss": 0.0237, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.6875038743019104, | |
| "eval_runtime": 21.2232, | |
| "eval_samples_per_second": 8.858, | |
| "eval_steps_per_second": 2.215, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 2.268817204301075e-05, | |
| "loss": 0.0046, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "eval_accuracy": 0.8297872543334961, | |
| "eval_loss": 0.8058395385742188, | |
| "eval_runtime": 21.9104, | |
| "eval_samples_per_second": 8.58, | |
| "eval_steps_per_second": 2.145, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 2.1612903225806454e-05, | |
| "loss": 0.0258, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "eval_accuracy": 0.8723404407501221, | |
| "eval_loss": 0.730888307094574, | |
| "eval_runtime": 19.7808, | |
| "eval_samples_per_second": 9.504, | |
| "eval_steps_per_second": 2.376, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 2.0537634408602152e-05, | |
| "loss": 0.0363, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_accuracy": 0.8510638475418091, | |
| "eval_loss": 0.7736497521400452, | |
| "eval_runtime": 20.7525, | |
| "eval_samples_per_second": 9.059, | |
| "eval_steps_per_second": 2.265, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1.9462365591397848e-05, | |
| "loss": 0.0012, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_accuracy": 0.8457446694374084, | |
| "eval_loss": 0.8431892991065979, | |
| "eval_runtime": 21.0633, | |
| "eval_samples_per_second": 8.925, | |
| "eval_steps_per_second": 2.231, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "learning_rate": 1.838709677419355e-05, | |
| "loss": 0.0112, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "eval_accuracy": 0.8617021441459656, | |
| "eval_loss": 0.7464305758476257, | |
| "eval_runtime": 19.2825, | |
| "eval_samples_per_second": 9.75, | |
| "eval_steps_per_second": 2.437, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 1.7311827956989248e-05, | |
| "loss": 0.0025, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "eval_accuracy": 0.8829787373542786, | |
| "eval_loss": 0.6915740370750427, | |
| "eval_runtime": 19.647, | |
| "eval_samples_per_second": 9.569, | |
| "eval_steps_per_second": 2.392, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 1.6236559139784947e-05, | |
| "loss": 0.003, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "eval_accuracy": 0.8882978558540344, | |
| "eval_loss": 0.6764039993286133, | |
| "eval_runtime": 21.6356, | |
| "eval_samples_per_second": 8.689, | |
| "eval_steps_per_second": 2.172, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 1.5161290322580646e-05, | |
| "loss": 0.0817, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "eval_accuracy": 0.8829787373542786, | |
| "eval_loss": 0.6824085116386414, | |
| "eval_runtime": 19.3614, | |
| "eval_samples_per_second": 9.71, | |
| "eval_steps_per_second": 2.428, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 1.4086021505376346e-05, | |
| "loss": 0.0059, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "eval_accuracy": 0.8882978558540344, | |
| "eval_loss": 0.6375851035118103, | |
| "eval_runtime": 19.4127, | |
| "eval_samples_per_second": 9.684, | |
| "eval_steps_per_second": 2.421, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 1.3010752688172043e-05, | |
| "loss": 0.0018, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "eval_accuracy": 0.8882978558540344, | |
| "eval_loss": 0.6130115389823914, | |
| "eval_runtime": 20.8292, | |
| "eval_samples_per_second": 9.026, | |
| "eval_steps_per_second": 2.256, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 1.1935483870967743e-05, | |
| "loss": 0.0012, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6152318716049194, | |
| "eval_runtime": 19.6293, | |
| "eval_samples_per_second": 9.578, | |
| "eval_steps_per_second": 2.394, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 1.086021505376344e-05, | |
| "loss": 0.0029, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6171016097068787, | |
| "eval_runtime": 19.602, | |
| "eval_samples_per_second": 9.591, | |
| "eval_steps_per_second": 2.398, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 9.78494623655914e-06, | |
| "loss": 0.0016, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.8882978558540344, | |
| "eval_loss": 0.6230288147926331, | |
| "eval_runtime": 19.6382, | |
| "eval_samples_per_second": 9.573, | |
| "eval_steps_per_second": 2.393, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 8.70967741935484e-06, | |
| "loss": 0.0016, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6247658729553223, | |
| "eval_runtime": 22.0158, | |
| "eval_samples_per_second": 8.539, | |
| "eval_steps_per_second": 2.135, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 7.634408602150538e-06, | |
| "loss": 0.0022, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6223304271697998, | |
| "eval_runtime": 22.0568, | |
| "eval_samples_per_second": 8.523, | |
| "eval_steps_per_second": 2.131, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 6.559139784946237e-06, | |
| "loss": 0.0011, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6233484745025635, | |
| "eval_runtime": 20.142, | |
| "eval_samples_per_second": 9.334, | |
| "eval_steps_per_second": 2.333, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 5.483870967741936e-06, | |
| "loss": 0.0008, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "eval_accuracy": 0.8936170339584351, | |
| "eval_loss": 0.6245375275611877, | |
| "eval_runtime": 20.6589, | |
| "eval_samples_per_second": 9.1, | |
| "eval_steps_per_second": 2.275, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 4.408602150537635e-06, | |
| "loss": 0.0109, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "eval_accuracy": 0.8829787373542786, | |
| "eval_loss": 0.6385069489479065, | |
| "eval_runtime": 19.883, | |
| "eval_samples_per_second": 9.455, | |
| "eval_steps_per_second": 2.364, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0011, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "eval_accuracy": 0.8776595592498779, | |
| "eval_loss": 0.65866619348526, | |
| "eval_runtime": 19.924, | |
| "eval_samples_per_second": 9.436, | |
| "eval_steps_per_second": 2.359, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 2.2580645161290324e-06, | |
| "loss": 0.0013, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "eval_accuracy": 0.8776595592498779, | |
| "eval_loss": 0.6651791930198669, | |
| "eval_runtime": 19.7346, | |
| "eval_samples_per_second": 9.526, | |
| "eval_steps_per_second": 2.382, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 1.1827956989247313e-06, | |
| "loss": 0.0548, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "eval_accuracy": 0.8776595592498779, | |
| "eval_loss": 0.6682336926460266, | |
| "eval_runtime": 19.8027, | |
| "eval_samples_per_second": 9.494, | |
| "eval_steps_per_second": 2.373, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 1.0752688172043011e-07, | |
| "loss": 0.0367, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_accuracy": 0.8776595592498779, | |
| "eval_loss": 0.6705843806266785, | |
| "eval_runtime": 19.6221, | |
| "eval_samples_per_second": 9.581, | |
| "eval_steps_per_second": 2.395, | |
| "step": 930 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 930, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 10, | |
| "total_flos": 1.1929088038243113e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |