| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 2000, | |
| "global_step": 12575, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.801192842942346e-05, | |
| "loss": 0.2793, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.602385685884692e-05, | |
| "loss": 0.167, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.403578528827038e-05, | |
| "loss": 0.1142, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.204771371769384e-05, | |
| "loss": 0.1172, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_f1": 0.8002317832826308, | |
| "eval_loss": 0.6091572739421863, | |
| "eval_runtime": 11.3587, | |
| "eval_samples_per_second": 7.571, | |
| "eval_steps_per_second": 7.571, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.00596421471173e-05, | |
| "loss": 0.1046, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.8071570576540756e-05, | |
| "loss": 0.1103, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 3.6083499005964215e-05, | |
| "loss": 0.1216, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 3.409542743538768e-05, | |
| "loss": 0.1201, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_f1": 0.797415004214667, | |
| "eval_loss": 0.570012144717545, | |
| "eval_runtime": 11.2237, | |
| "eval_samples_per_second": 7.662, | |
| "eval_steps_per_second": 7.662, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.210735586481113e-05, | |
| "loss": 0.1239, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 3.0119284294234595e-05, | |
| "loss": 0.1259, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.813121272365805e-05, | |
| "loss": 0.1254, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 2.6143141153081513e-05, | |
| "loss": 0.1738, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_f1": 0.785323775968898, | |
| "eval_loss": 1.3082878860688905, | |
| "eval_runtime": 11.1911, | |
| "eval_samples_per_second": 7.685, | |
| "eval_steps_per_second": 7.685, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.415506958250497e-05, | |
| "loss": 0.1656, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2.216699801192843e-05, | |
| "loss": 0.1316, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.017892644135189e-05, | |
| "loss": 0.1284, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 1.8190854870775348e-05, | |
| "loss": 0.1431, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_f1": 0.7811158798283263, | |
| "eval_loss": 0.6317664683518408, | |
| "eval_runtime": 11.2067, | |
| "eval_samples_per_second": 7.674, | |
| "eval_steps_per_second": 7.674, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 1.6202783300198807e-05, | |
| "loss": 0.1353, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 1.4214711729622268e-05, | |
| "loss": 0.1013, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.2226640159045727e-05, | |
| "loss": 0.1174, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.0238568588469186e-05, | |
| "loss": 0.1213, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "eval_f1": 0.8029407524866656, | |
| "eval_loss": 0.4704339896264236, | |
| "eval_runtime": 11.4371, | |
| "eval_samples_per_second": 7.519, | |
| "eval_steps_per_second": 7.519, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 8.250497017892645e-06, | |
| "loss": 0.0851, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.2624254473161034e-06, | |
| "loss": 0.096, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 4.274353876739562e-06, | |
| "loss": 0.0727, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 2.286282306163022e-06, | |
| "loss": 0.0764, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "eval_f1": 0.8055235903337169, | |
| "eval_loss": 0.4711939039551899, | |
| "eval_runtime": 11.2703, | |
| "eval_samples_per_second": 7.631, | |
| "eval_steps_per_second": 7.631, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 2.982107355864811e-07, | |
| "loss": 0.09, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 12575, | |
| "total_flos": 1.933020837593007e+16, | |
| "train_loss": 0.12557823169776508, | |
| "train_runtime": 2775.7723, | |
| "train_samples_per_second": 4.53, | |
| "train_steps_per_second": 4.53 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12575, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.933020837593007e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |