| { | |
| "best_metric": 2.396937131881714, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/jav-Latn/checkpoint-6000", | |
| "epoch": 20.338983050847457, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.694915254237288, | |
| "grad_norm": 4.60462760925293, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.8563, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.694915254237288, | |
| "eval_accuracy": 0.7129037313539158, | |
| "eval_loss": 2.615060329437256, | |
| "eval_runtime": 16.1415, | |
| "eval_samples_per_second": 261.934, | |
| "eval_steps_per_second": 8.24, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "grad_norm": 4.434678077697754, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.6006, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "eval_accuracy": 0.7279286229632328, | |
| "eval_loss": 2.511897563934326, | |
| "eval_runtime": 16.7992, | |
| "eval_samples_per_second": 251.678, | |
| "eval_steps_per_second": 7.917, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.084745762711864, | |
| "grad_norm": 3.77703595161438, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.4826, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.084745762711864, | |
| "eval_accuracy": 0.739675804970421, | |
| "eval_loss": 2.5077760219573975, | |
| "eval_runtime": 17.3002, | |
| "eval_samples_per_second": 244.39, | |
| "eval_steps_per_second": 7.688, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.779661016949152, | |
| "grad_norm": 3.7486958503723145, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.4095, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.779661016949152, | |
| "eval_accuracy": 0.742510144555458, | |
| "eval_loss": 2.4969942569732666, | |
| "eval_runtime": 17.6233, | |
| "eval_samples_per_second": 239.91, | |
| "eval_steps_per_second": 7.547, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.474576271186441, | |
| "grad_norm": 3.9276657104492188, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.3522, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.474576271186441, | |
| "eval_accuracy": 0.7483269445491463, | |
| "eval_loss": 2.483748197555542, | |
| "eval_runtime": 18.2776, | |
| "eval_samples_per_second": 231.322, | |
| "eval_steps_per_second": 7.277, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.169491525423728, | |
| "grad_norm": 3.0509426593780518, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.3055, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.169491525423728, | |
| "eval_accuracy": 0.7534293503861141, | |
| "eval_loss": 2.4522793292999268, | |
| "eval_runtime": 17.8658, | |
| "eval_samples_per_second": 236.654, | |
| "eval_steps_per_second": 7.444, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.864406779661017, | |
| "grad_norm": 3.749621629714966, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.2637, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.864406779661017, | |
| "eval_accuracy": 0.7564406320026262, | |
| "eval_loss": 2.5035338401794434, | |
| "eval_runtime": 16.226, | |
| "eval_samples_per_second": 260.569, | |
| "eval_steps_per_second": 8.197, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.559322033898304, | |
| "grad_norm": 3.407952308654785, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2289, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.559322033898304, | |
| "eval_accuracy": 0.7572301628520813, | |
| "eval_loss": 2.5971286296844482, | |
| "eval_runtime": 18.6487, | |
| "eval_samples_per_second": 226.718, | |
| "eval_steps_per_second": 7.132, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.254237288135593, | |
| "grad_norm": 3.5328588485717773, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.2005, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.254237288135593, | |
| "eval_accuracy": 0.760487028966383, | |
| "eval_loss": 2.424844980239868, | |
| "eval_runtime": 18.0059, | |
| "eval_samples_per_second": 234.812, | |
| "eval_steps_per_second": 7.386, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.949152542372882, | |
| "grad_norm": 3.54754638671875, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.1669, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.949152542372882, | |
| "eval_accuracy": 0.7611065918853734, | |
| "eval_loss": 2.494778871536255, | |
| "eval_runtime": 18.045, | |
| "eval_samples_per_second": 234.304, | |
| "eval_steps_per_second": 7.37, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 18.64406779661017, | |
| "grad_norm": 3.227668046951294, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.1376, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.64406779661017, | |
| "eval_accuracy": 0.7634028732398379, | |
| "eval_loss": 2.465242862701416, | |
| "eval_runtime": 17.9581, | |
| "eval_samples_per_second": 235.437, | |
| "eval_steps_per_second": 7.406, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 20.338983050847457, | |
| "grad_norm": 3.670949697494507, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.1196, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.338983050847457, | |
| "eval_accuracy": 0.7678996847315939, | |
| "eval_loss": 2.396937131881714, | |
| "eval_runtime": 17.9579, | |
| "eval_samples_per_second": 235.439, | |
| "eval_steps_per_second": 7.406, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 339, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.066477747463782e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |