| { | |
| "best_metric": 60.099999999999994, | |
| "best_model_checkpoint": "outputs/soft_prompt/t5-base/sst2/checkpoint-5800", | |
| "epoch": 3.0, | |
| "global_step": 6222, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.0, | |
| "eval_average_metrics": 0.0, | |
| "eval_loss": 1.9322336912155151, | |
| "eval_runtime": 9.7016, | |
| "eval_samples_per_second": 103.076, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 51.0, | |
| "eval_average_metrics": 51.0, | |
| "eval_loss": 0.7396745681762695, | |
| "eval_runtime": 3.3022, | |
| "eval_samples_per_second": 302.832, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00027589199614271937, | |
| "loss": 2.5059, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 55.2, | |
| "eval_average_metrics": 55.2, | |
| "eval_loss": 0.5560445785522461, | |
| "eval_runtime": 3.2155, | |
| "eval_samples_per_second": 310.993, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.463209867477417, | |
| "eval_runtime": 3.1314, | |
| "eval_samples_per_second": 319.345, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002517839922854387, | |
| "loss": 0.7329, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.4116092920303345, | |
| "eval_runtime": 3.3786, | |
| "eval_samples_per_second": 295.977, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.38179224729537964, | |
| "eval_runtime": 3.2857, | |
| "eval_samples_per_second": 304.351, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.35022690892219543, | |
| "eval_runtime": 2.8293, | |
| "eval_samples_per_second": 353.443, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00022767598842815813, | |
| "loss": 0.4828, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.3595990538597107, | |
| "eval_runtime": 3.2567, | |
| "eval_samples_per_second": 307.055, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.3511555790901184, | |
| "eval_runtime": 3.0816, | |
| "eval_samples_per_second": 324.506, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00020356798457087753, | |
| "loss": 0.421, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.3314475417137146, | |
| "eval_runtime": 2.522, | |
| "eval_samples_per_second": 396.513, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_accuracy": 55.300000000000004, | |
| "eval_average_metrics": 55.300000000000004, | |
| "eval_loss": 0.3108561336994171, | |
| "eval_runtime": 2.7063, | |
| "eval_samples_per_second": 369.512, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_accuracy": 55.7, | |
| "eval_average_metrics": 55.7, | |
| "eval_loss": 0.2974016070365906, | |
| "eval_runtime": 3.5599, | |
| "eval_samples_per_second": 280.91, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.0001794599807135969, | |
| "loss": 0.3931, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 55.7, | |
| "eval_average_metrics": 55.7, | |
| "eval_loss": 0.3047039210796356, | |
| "eval_runtime": 3.0837, | |
| "eval_samples_per_second": 324.282, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_accuracy": 55.900000000000006, | |
| "eval_average_metrics": 55.900000000000006, | |
| "eval_loss": 0.2895060181617737, | |
| "eval_runtime": 3.9994, | |
| "eval_samples_per_second": 250.039, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00015535197685631627, | |
| "loss": 0.3777, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 56.2, | |
| "eval_average_metrics": 56.2, | |
| "eval_loss": 0.28209826350212097, | |
| "eval_runtime": 3.5924, | |
| "eval_samples_per_second": 278.366, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 57.4, | |
| "eval_average_metrics": 57.4, | |
| "eval_loss": 0.2806238532066345, | |
| "eval_runtime": 3.1947, | |
| "eval_samples_per_second": 313.019, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 56.699999999999996, | |
| "eval_average_metrics": 56.699999999999996, | |
| "eval_loss": 0.2809857428073883, | |
| "eval_runtime": 3.3046, | |
| "eval_samples_per_second": 302.612, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00013124397299903566, | |
| "loss": 0.372, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 56.89999999999999, | |
| "eval_average_metrics": 56.89999999999999, | |
| "eval_loss": 0.2799268066883087, | |
| "eval_runtime": 3.5652, | |
| "eval_samples_per_second": 280.491, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 56.599999999999994, | |
| "eval_average_metrics": 56.599999999999994, | |
| "eval_loss": 0.2795256972312927, | |
| "eval_runtime": 3.5541, | |
| "eval_samples_per_second": 281.363, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00010713596914175504, | |
| "loss": 0.3629, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_accuracy": 57.099999999999994, | |
| "eval_average_metrics": 57.099999999999994, | |
| "eval_loss": 0.27900928258895874, | |
| "eval_runtime": 3.42, | |
| "eval_samples_per_second": 292.396, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 57.699999999999996, | |
| "eval_average_metrics": 57.699999999999996, | |
| "eval_loss": 0.27878034114837646, | |
| "eval_runtime": 3.38, | |
| "eval_samples_per_second": 295.859, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 57.699999999999996, | |
| "eval_average_metrics": 57.699999999999996, | |
| "eval_loss": 0.27947279810905457, | |
| "eval_runtime": 3.1427, | |
| "eval_samples_per_second": 318.196, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 8.302796528447444e-05, | |
| "loss": 0.3597, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 57.99999999999999, | |
| "eval_average_metrics": 57.99999999999999, | |
| "eval_loss": 0.2773045599460602, | |
| "eval_runtime": 3.1388, | |
| "eval_samples_per_second": 318.593, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_accuracy": 59.8, | |
| "eval_average_metrics": 59.8, | |
| "eval_loss": 0.2791491448879242, | |
| "eval_runtime": 3.8832, | |
| "eval_samples_per_second": 257.519, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 5.891996142719383e-05, | |
| "loss": 0.3555, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_accuracy": 59.4, | |
| "eval_average_metrics": 59.4, | |
| "eval_loss": 0.2792259454727173, | |
| "eval_runtime": 3.3994, | |
| "eval_samples_per_second": 294.169, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_accuracy": 59.3, | |
| "eval_average_metrics": 59.3, | |
| "eval_loss": 0.27831217646598816, | |
| "eval_runtime": 3.2184, | |
| "eval_samples_per_second": 310.718, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 59.199999999999996, | |
| "eval_average_metrics": 59.199999999999996, | |
| "eval_loss": 0.2772988975048065, | |
| "eval_runtime": 3.2951, | |
| "eval_samples_per_second": 303.485, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3.481195756991321e-05, | |
| "loss": 0.3528, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_accuracy": 59.9, | |
| "eval_average_metrics": 59.9, | |
| "eval_loss": 0.2777373790740967, | |
| "eval_runtime": 3.3995, | |
| "eval_samples_per_second": 294.161, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 60.099999999999994, | |
| "eval_average_metrics": 60.099999999999994, | |
| "eval_loss": 0.2783910036087036, | |
| "eval_runtime": 3.8322, | |
| "eval_samples_per_second": 260.948, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.0703953712632592e-05, | |
| "loss": 0.3503, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 59.699999999999996, | |
| "eval_average_metrics": 59.699999999999996, | |
| "eval_loss": 0.27744776010513306, | |
| "eval_runtime": 3.3726, | |
| "eval_samples_per_second": 296.511, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 59.9, | |
| "eval_average_metrics": 59.9, | |
| "eval_loss": 0.27743667364120483, | |
| "eval_runtime": 3.6434, | |
| "eval_samples_per_second": 274.47, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 6222, | |
| "total_flos": 1.316605032958464e+16, | |
| "train_loss": 0.5804944922785328, | |
| "train_runtime": 1254.7674, | |
| "train_samples_per_second": 158.633, | |
| "train_steps_per_second": 4.959 | |
| } | |
| ], | |
| "max_steps": 6222, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.316605032958464e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |