| { | |
| "best_metric": 0.6082, | |
| "best_model_checkpoint": "save_model/deberta-v3-base-amazon-reviews-multi-fp16/checkpoint-12500", | |
| "epoch": 5.0, | |
| "global_step": 31250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.92016e-05, | |
| "loss": 1.0654, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.84016e-05, | |
| "loss": 0.9825, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.76048e-05, | |
| "loss": 0.9506, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.68048e-05, | |
| "loss": 0.943, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.6004800000000004e-05, | |
| "loss": 0.9473, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.5204800000000005e-05, | |
| "loss": 0.9369, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.44048e-05, | |
| "loss": 0.9264, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.36064e-05, | |
| "loss": 0.9267, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.2806400000000004e-05, | |
| "loss": 0.9247, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.20064e-05, | |
| "loss": 0.9149, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.12064e-05, | |
| "loss": 0.9236, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.04064e-05, | |
| "loss": 0.9134, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9276404976844788, | |
| "eval_runtime": 5.8479, | |
| "eval_samples_per_second": 855.013, | |
| "eval_steps_per_second": 26.847, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.960640000000001e-05, | |
| "loss": 0.8749, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 3.88064e-05, | |
| "loss": 0.8431, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.80064e-05, | |
| "loss": 0.8411, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.7206400000000003e-05, | |
| "loss": 0.8368, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 3.6406400000000004e-05, | |
| "loss": 0.8489, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 3.56064e-05, | |
| "loss": 0.8415, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 3.48064e-05, | |
| "loss": 0.8473, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.40064e-05, | |
| "loss": 0.8411, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.3208e-05, | |
| "loss": 0.8517, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.2408e-05, | |
| "loss": 0.8438, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.1608000000000006e-05, | |
| "loss": 0.8443, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.0808e-05, | |
| "loss": 0.847, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.0009599999999998e-05, | |
| "loss": 0.8412, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6082, | |
| "eval_loss": 0.928673505783081, | |
| "eval_runtime": 5.8358, | |
| "eval_samples_per_second": 856.786, | |
| "eval_steps_per_second": 26.903, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.92112e-05, | |
| "loss": 0.7459, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.84112e-05, | |
| "loss": 0.7468, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.7611200000000004e-05, | |
| "loss": 0.7531, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.68128e-05, | |
| "loss": 0.7518, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.6012800000000003e-05, | |
| "loss": 0.7548, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 2.52144e-05, | |
| "loss": 0.7458, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.44144e-05, | |
| "loss": 0.7474, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2.36144e-05, | |
| "loss": 0.7498, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.2814400000000003e-05, | |
| "loss": 0.7463, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.20144e-05, | |
| "loss": 0.7495, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.12144e-05, | |
| "loss": 0.7463, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.04144e-05, | |
| "loss": 0.7523, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6078, | |
| "eval_loss": 0.9500126242637634, | |
| "eval_runtime": 5.8734, | |
| "eval_samples_per_second": 851.298, | |
| "eval_steps_per_second": 26.731, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 1.9614400000000003e-05, | |
| "loss": 0.6805, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.88144e-05, | |
| "loss": 0.6338, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.80144e-05, | |
| "loss": 0.633, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 1.72144e-05, | |
| "loss": 0.6267, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 1.6416e-05, | |
| "loss": 0.6349, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.5616e-05, | |
| "loss": 0.6409, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 1.4816e-05, | |
| "loss": 0.6331, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.4016000000000001e-05, | |
| "loss": 0.6194, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.3217599999999999e-05, | |
| "loss": 0.6376, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 1.24176e-05, | |
| "loss": 0.6279, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.16176e-05, | |
| "loss": 0.6285, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.08176e-05, | |
| "loss": 0.6347, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.00176e-05, | |
| "loss": 0.6247, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5984, | |
| "eval_loss": 1.0481306314468384, | |
| "eval_runtime": 5.8354, | |
| "eval_samples_per_second": 856.841, | |
| "eval_steps_per_second": 26.905, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 9.2176e-06, | |
| "loss": 0.5317, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 8.4176e-06, | |
| "loss": 0.5172, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 7.6176e-06, | |
| "loss": 0.5277, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 6.817600000000001e-06, | |
| "loss": 0.5288, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 6.017600000000001e-06, | |
| "loss": 0.5189, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 5.219200000000001e-06, | |
| "loss": 0.5219, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 4.419200000000001e-06, | |
| "loss": 0.5229, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 3.6192000000000004e-06, | |
| "loss": 0.5219, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 2.8192e-06, | |
| "loss": 0.5081, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 2.0192e-06, | |
| "loss": 0.5169, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 1.2192000000000001e-06, | |
| "loss": 0.5051, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 4.1920000000000005e-07, | |
| "loss": 0.51, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6016, | |
| "eval_loss": 1.1977229118347168, | |
| "eval_runtime": 5.8354, | |
| "eval_samples_per_second": 856.846, | |
| "eval_steps_per_second": 26.905, | |
| "step": 31250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 31250, | |
| "total_flos": 6.5780715264e+16, | |
| "train_loss": 0.7373493041992187, | |
| "train_runtime": 4266.3791, | |
| "train_samples_per_second": 234.391, | |
| "train_steps_per_second": 7.325 | |
| } | |
| ], | |
| "max_steps": 31250, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.5780715264e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |