| { | |
| "best_metric": 72.47706422018348, | |
| "best_model_checkpoint": "outputs/adapter/superglue-boolq/checkpoint-3835", | |
| "epoch": 20.0, | |
| "global_step": 5900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 62.62996941896024, | |
| "eval_average_metrics": 62.62996941896024, | |
| "eval_loss": 0.2848173677921295, | |
| "eval_runtime": 4.5081, | |
| "eval_samples_per_second": 362.682, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00027457627118644066, | |
| "loss": 0.3238, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 62.62996941896024, | |
| "eval_average_metrics": 62.62996941896024, | |
| "eval_loss": 0.27746617794036865, | |
| "eval_runtime": 4.4839, | |
| "eval_samples_per_second": 364.64, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 62.62996941896024, | |
| "eval_average_metrics": 62.62996941896024, | |
| "eval_loss": 0.2767568826675415, | |
| "eval_runtime": 4.4969, | |
| "eval_samples_per_second": 363.585, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00024915254237288135, | |
| "loss": 0.2896, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 62.75229357798165, | |
| "eval_average_metrics": 62.75229357798165, | |
| "eval_loss": 0.2767893373966217, | |
| "eval_runtime": 4.7449, | |
| "eval_samples_per_second": 344.579, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 63.36391437308868, | |
| "eval_average_metrics": 63.36391437308868, | |
| "eval_loss": 0.26716169714927673, | |
| "eval_runtime": 4.7487, | |
| "eval_samples_per_second": 344.302, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.000223728813559322, | |
| "loss": 0.2823, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 66.11620795107034, | |
| "eval_average_metrics": 66.11620795107034, | |
| "eval_loss": 0.2606015205383301, | |
| "eval_runtime": 4.7538, | |
| "eval_samples_per_second": 343.936, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0001983050847457627, | |
| "loss": 0.2715, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 69.0519877675841, | |
| "eval_average_metrics": 69.0519877675841, | |
| "eval_loss": 0.25209498405456543, | |
| "eval_runtime": 4.7471, | |
| "eval_samples_per_second": 344.422, | |
| "step": 2065 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 69.2354740061162, | |
| "eval_average_metrics": 69.2354740061162, | |
| "eval_loss": 0.25652435421943665, | |
| "eval_runtime": 4.7461, | |
| "eval_samples_per_second": 344.493, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.0001728813559322034, | |
| "loss": 0.236, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 71.0091743119266, | |
| "eval_average_metrics": 71.0091743119266, | |
| "eval_loss": 0.24900275468826294, | |
| "eval_runtime": 4.7529, | |
| "eval_samples_per_second": 344.002, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 71.80428134556574, | |
| "eval_average_metrics": 71.80428134556574, | |
| "eval_loss": 0.2635628283023834, | |
| "eval_runtime": 4.7533, | |
| "eval_samples_per_second": 343.974, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00014745762711864405, | |
| "loss": 0.2038, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 70.70336391437309, | |
| "eval_average_metrics": 70.70336391437309, | |
| "eval_loss": 0.26442670822143555, | |
| "eval_runtime": 4.7661, | |
| "eval_samples_per_second": 343.05, | |
| "step": 3245 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00012203389830508474, | |
| "loss": 0.1743, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 71.49847094801223, | |
| "eval_average_metrics": 71.49847094801223, | |
| "eval_loss": 0.2672346830368042, | |
| "eval_runtime": 4.761, | |
| "eval_samples_per_second": 343.414, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 72.47706422018348, | |
| "eval_average_metrics": 72.47706422018348, | |
| "eval_loss": 0.27757981419563293, | |
| "eval_runtime": 4.7532, | |
| "eval_samples_per_second": 343.982, | |
| "step": 3835 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 9.661016949152541e-05, | |
| "loss": 0.1493, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 70.8868501529052, | |
| "eval_average_metrics": 70.8868501529052, | |
| "eval_loss": 0.29447805881500244, | |
| "eval_runtime": 4.7492, | |
| "eval_samples_per_second": 344.268, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 70.21406727828746, | |
| "eval_average_metrics": 70.21406727828746, | |
| "eval_loss": 0.2965507209300995, | |
| "eval_runtime": 4.7633, | |
| "eval_samples_per_second": 343.251, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 7.11864406779661e-05, | |
| "loss": 0.1336, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 72.29357798165138, | |
| "eval_average_metrics": 72.29357798165138, | |
| "eval_loss": 0.3127536177635193, | |
| "eval_runtime": 4.7675, | |
| "eval_samples_per_second": 342.95, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 4.576271186440678e-05, | |
| "loss": 0.1166, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 70.9480122324159, | |
| "eval_average_metrics": 70.9480122324159, | |
| "eval_loss": 0.3478758931159973, | |
| "eval_runtime": 4.7792, | |
| "eval_samples_per_second": 342.105, | |
| "step": 5015 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 72.17125382262996, | |
| "eval_average_metrics": 72.17125382262996, | |
| "eval_loss": 0.33715757727622986, | |
| "eval_runtime": 4.7549, | |
| "eval_samples_per_second": 343.858, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 2.0338983050847455e-05, | |
| "loss": 0.1059, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 71.92660550458716, | |
| "eval_average_metrics": 71.92660550458716, | |
| "eval_loss": 0.34181272983551025, | |
| "eval_runtime": 4.753, | |
| "eval_samples_per_second": 343.993, | |
| "step": 5605 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 71.80428134556574, | |
| "eval_average_metrics": 71.80428134556574, | |
| "eval_loss": 0.34913983941078186, | |
| "eval_runtime": 4.7538, | |
| "eval_samples_per_second": 343.937, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5900, | |
| "total_flos": 5.78564930408256e+16, | |
| "train_loss": 0.2005054286374884, | |
| "train_runtime": 1293.5612, | |
| "train_samples_per_second": 145.753, | |
| "train_steps_per_second": 4.561 | |
| } | |
| ], | |
| "max_steps": 5900, | |
| "num_train_epochs": 20, | |
| "total_flos": 5.78564930408256e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |