| { | |
| "best_metric": 0.7391304347826086, | |
| "best_model_checkpoint": "swiftformer-xs-DMAE\\checkpoint-45", | |
| "epoch": 34.285714285714285, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.45652173913043476, | |
| "eval_loss": 1.3835715055465698, | |
| "eval_runtime": 0.849, | |
| "eval_samples_per_second": 54.181, | |
| "eval_steps_per_second": 3.534, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.3326613903045654, | |
| "eval_runtime": 0.6171, | |
| "eval_samples_per_second": 74.538, | |
| "eval_steps_per_second": 4.861, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.00125, | |
| "loss": 1.3567, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.168088436126709, | |
| "eval_runtime": 0.6072, | |
| "eval_samples_per_second": 75.764, | |
| "eval_steps_per_second": 4.941, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5652173913043478, | |
| "eval_loss": 1.044042944908142, | |
| "eval_runtime": 0.6302, | |
| "eval_samples_per_second": 72.998, | |
| "eval_steps_per_second": 4.761, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 1.0461686849594116, | |
| "eval_runtime": 0.6512, | |
| "eval_samples_per_second": 70.643, | |
| "eval_steps_per_second": 4.607, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.001388888888888889, | |
| "loss": 1.0903, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5869565217391305, | |
| "eval_loss": 0.9293990731239319, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 75.207, | |
| "eval_steps_per_second": 4.905, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.9571677446365356, | |
| "eval_runtime": 0.6031, | |
| "eval_samples_per_second": 76.268, | |
| "eval_steps_per_second": 4.974, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.9286340475082397, | |
| "eval_runtime": 0.6212, | |
| "eval_samples_per_second": 74.056, | |
| "eval_steps_per_second": 4.83, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 0.00125, | |
| "loss": 1.0969, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "eval_accuracy": 0.6304347826086957, | |
| "eval_loss": 0.9229152798652649, | |
| "eval_runtime": 0.6537, | |
| "eval_samples_per_second": 70.374, | |
| "eval_steps_per_second": 4.59, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 0.9061374664306641, | |
| "eval_runtime": 0.6311, | |
| "eval_samples_per_second": 72.884, | |
| "eval_steps_per_second": 4.753, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8341346383094788, | |
| "eval_runtime": 0.6232, | |
| "eval_samples_per_second": 73.818, | |
| "eval_steps_per_second": 4.814, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 0.0011111111111111111, | |
| "loss": 0.8923, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.878592848777771, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 75.766, | |
| "eval_steps_per_second": 4.941, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_loss": 0.8596428632736206, | |
| "eval_runtime": 0.6192, | |
| "eval_samples_per_second": 74.295, | |
| "eval_steps_per_second": 4.845, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.8902113437652588, | |
| "eval_runtime": 0.6051, | |
| "eval_samples_per_second": 76.016, | |
| "eval_steps_per_second": 4.958, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 0.0009722222222222222, | |
| "loss": 0.7289, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8024305105209351, | |
| "eval_runtime": 0.6026, | |
| "eval_samples_per_second": 76.33, | |
| "eval_steps_per_second": 4.978, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.9340682625770569, | |
| "eval_runtime": 0.6272, | |
| "eval_samples_per_second": 73.347, | |
| "eval_steps_per_second": 4.783, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 1.0463521480560303, | |
| "eval_runtime": 0.6191, | |
| "eval_samples_per_second": 74.297, | |
| "eval_steps_per_second": 4.845, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 0.6609, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6086956521739131, | |
| "eval_loss": 0.9922565817832947, | |
| "eval_runtime": 0.5991, | |
| "eval_samples_per_second": 76.776, | |
| "eval_steps_per_second": 5.007, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.86, | |
| "eval_accuracy": 0.717391304347826, | |
| "eval_loss": 0.8224782347679138, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 74.415, | |
| "eval_steps_per_second": 4.853, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0006944444444444445, | |
| "loss": 0.6527, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.874836802482605, | |
| "eval_runtime": 0.6096, | |
| "eval_samples_per_second": 75.454, | |
| "eval_steps_per_second": 4.921, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 0.8051551580429077, | |
| "eval_runtime": 0.5866, | |
| "eval_samples_per_second": 78.412, | |
| "eval_steps_per_second": 5.114, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.8861276507377625, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 74.417, | |
| "eval_steps_per_second": 4.853, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "learning_rate": 0.0005555555555555556, | |
| "loss": 0.493, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.86, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.9554585218429565, | |
| "eval_runtime": 0.6196, | |
| "eval_samples_per_second": 74.236, | |
| "eval_steps_per_second": 4.841, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.0335818529129028, | |
| "eval_runtime": 0.6146, | |
| "eval_samples_per_second": 74.84, | |
| "eval_steps_per_second": 4.881, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.86, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 0.9960870146751404, | |
| "eval_runtime": 0.6046, | |
| "eval_samples_per_second": 76.078, | |
| "eval_steps_per_second": 4.962, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 0.4088, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.0399607419967651, | |
| "eval_runtime": 0.6222, | |
| "eval_samples_per_second": 73.937, | |
| "eval_steps_per_second": 4.822, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.86, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.053550362586975, | |
| "eval_runtime": 0.6206, | |
| "eval_samples_per_second": 74.116, | |
| "eval_steps_per_second": 4.834, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.1388055086135864, | |
| "eval_runtime": 0.6152, | |
| "eval_samples_per_second": 74.778, | |
| "eval_steps_per_second": 4.877, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 0.4047, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.2294652462005615, | |
| "eval_runtime": 0.6572, | |
| "eval_samples_per_second": 69.998, | |
| "eval_steps_per_second": 4.565, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.2627309560775757, | |
| "eval_runtime": 0.6271, | |
| "eval_samples_per_second": 73.349, | |
| "eval_steps_per_second": 4.784, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.86, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.2371925115585327, | |
| "eval_runtime": 0.6617, | |
| "eval_samples_per_second": 69.522, | |
| "eval_steps_per_second": 4.534, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.43, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 0.3681, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.2918990850448608, | |
| "eval_runtime": 0.6482, | |
| "eval_samples_per_second": 70.97, | |
| "eval_steps_per_second": 4.628, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.86, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_loss": 1.2453422546386719, | |
| "eval_runtime": 0.6331, | |
| "eval_samples_per_second": 72.653, | |
| "eval_steps_per_second": 4.738, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.6739130434782609, | |
| "eval_loss": 1.2612279653549194, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 74.417, | |
| "eval_steps_per_second": 4.853, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "learning_rate": 0.0, | |
| "loss": 0.353, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_loss": 1.2610585689544678, | |
| "eval_runtime": 0.6256, | |
| "eval_samples_per_second": 73.524, | |
| "eval_steps_per_second": 4.795, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.29, | |
| "step": 120, | |
| "total_flos": 2.0027429927092224e+16, | |
| "train_loss": 0.7088598291079203, | |
| "train_runtime": 131.3917, | |
| "train_samples_per_second": 64.844, | |
| "train_steps_per_second": 0.913 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 2.0027429927092224e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |