| { | |
| "best_global_step": 885, | |
| "best_metric": 1.009942650794983, | |
| "best_model_checkpoint": "./fine-tuned-model\\checkpoint-885", | |
| "epoch": 19.0, | |
| "eval_steps": 500, | |
| "global_step": 1121, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.847457627118644, | |
| "grad_norm": 0.939425528049469, | |
| "learning_rate": 5.8983050847457634e-05, | |
| "loss": 7.3825, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.3358267545700073, | |
| "eval_runtime": 7.2445, | |
| "eval_samples_per_second": 14.494, | |
| "eval_steps_per_second": 0.966, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.694915254237288, | |
| "grad_norm": 2.5237913131713867, | |
| "learning_rate": 5.796610169491525e-05, | |
| "loss": 1.6245, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1412484645843506, | |
| "eval_runtime": 7.1915, | |
| "eval_samples_per_second": 14.601, | |
| "eval_steps_per_second": 0.973, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.542372881355932, | |
| "grad_norm": 7.8459625244140625, | |
| "learning_rate": 5.6949152542372884e-05, | |
| "loss": 1.4469, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.0957719087600708, | |
| "eval_runtime": 7.3705, | |
| "eval_samples_per_second": 14.246, | |
| "eval_steps_per_second": 0.95, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "grad_norm": 1.5296450853347778, | |
| "learning_rate": 5.593220338983051e-05, | |
| "loss": 1.3912, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.072253704071045, | |
| "eval_runtime": 7.3695, | |
| "eval_samples_per_second": 14.248, | |
| "eval_steps_per_second": 0.95, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 4.237288135593221, | |
| "grad_norm": 1.9591976404190063, | |
| "learning_rate": 5.4915254237288135e-05, | |
| "loss": 1.2829, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.1338605880737305, | |
| "eval_runtime": 7.1975, | |
| "eval_samples_per_second": 14.588, | |
| "eval_steps_per_second": 0.973, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 5.084745762711864, | |
| "grad_norm": 1.4545025825500488, | |
| "learning_rate": 5.389830508474577e-05, | |
| "loss": 1.3583, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.932203389830509, | |
| "grad_norm": 1.4759844541549683, | |
| "learning_rate": 5.288135593220339e-05, | |
| "loss": 1.295, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.043823480606079, | |
| "eval_runtime": 7.226, | |
| "eval_samples_per_second": 14.531, | |
| "eval_steps_per_second": 0.969, | |
| "step": 354 | |
| }, | |
| { | |
| "epoch": 6.779661016949152, | |
| "grad_norm": 1.5640958547592163, | |
| "learning_rate": 5.186440677966102e-05, | |
| "loss": 1.2581, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.0363339185714722, | |
| "eval_runtime": 7.3055, | |
| "eval_samples_per_second": 14.373, | |
| "eval_steps_per_second": 0.958, | |
| "step": 413 | |
| }, | |
| { | |
| "epoch": 7.627118644067797, | |
| "grad_norm": 2.0497965812683105, | |
| "learning_rate": 5.0847457627118643e-05, | |
| "loss": 1.2544, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.067766785621643, | |
| "eval_runtime": 7.3717, | |
| "eval_samples_per_second": 14.244, | |
| "eval_steps_per_second": 0.95, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 8.474576271186441, | |
| "grad_norm": 1.2606173753738403, | |
| "learning_rate": 4.9830508474576276e-05, | |
| "loss": 1.3097, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.02413809299469, | |
| "eval_runtime": 7.1818, | |
| "eval_samples_per_second": 14.62, | |
| "eval_steps_per_second": 0.975, | |
| "step": 531 | |
| }, | |
| { | |
| "epoch": 9.322033898305085, | |
| "grad_norm": 2.3118815422058105, | |
| "learning_rate": 4.88135593220339e-05, | |
| "loss": 1.2787, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.013655424118042, | |
| "eval_runtime": 7.1577, | |
| "eval_samples_per_second": 14.67, | |
| "eval_steps_per_second": 0.978, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 10.169491525423728, | |
| "grad_norm": 1.2859658002853394, | |
| "learning_rate": 4.7796610169491526e-05, | |
| "loss": 1.2354, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.0547661781311035, | |
| "eval_runtime": 7.1926, | |
| "eval_samples_per_second": 14.598, | |
| "eval_steps_per_second": 0.973, | |
| "step": 649 | |
| }, | |
| { | |
| "epoch": 11.016949152542374, | |
| "grad_norm": 2.121445417404175, | |
| "learning_rate": 4.677966101694916e-05, | |
| "loss": 1.2596, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 11.864406779661017, | |
| "grad_norm": 3.0464370250701904, | |
| "learning_rate": 4.576271186440678e-05, | |
| "loss": 1.2646, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0133599042892456, | |
| "eval_runtime": 10.6666, | |
| "eval_samples_per_second": 9.844, | |
| "eval_steps_per_second": 0.656, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 12.711864406779661, | |
| "grad_norm": 1.1342540979385376, | |
| "learning_rate": 4.474576271186441e-05, | |
| "loss": 1.2068, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 1.0467838048934937, | |
| "eval_runtime": 11.4351, | |
| "eval_samples_per_second": 9.182, | |
| "eval_steps_per_second": 0.612, | |
| "step": 767 | |
| }, | |
| { | |
| "epoch": 13.559322033898304, | |
| "grad_norm": 2.094381093978882, | |
| "learning_rate": 4.3728813559322035e-05, | |
| "loss": 1.2955, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.0249124765396118, | |
| "eval_runtime": 13.2701, | |
| "eval_samples_per_second": 7.913, | |
| "eval_steps_per_second": 0.528, | |
| "step": 826 | |
| }, | |
| { | |
| "epoch": 14.40677966101695, | |
| "grad_norm": 1.0174381732940674, | |
| "learning_rate": 4.271186440677966e-05, | |
| "loss": 1.2215, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.009942650794983, | |
| "eval_runtime": 10.124, | |
| "eval_samples_per_second": 10.371, | |
| "eval_steps_per_second": 0.691, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 15.254237288135593, | |
| "grad_norm": 1.1202493906021118, | |
| "learning_rate": 4.169491525423729e-05, | |
| "loss": 1.2365, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.0121246576309204, | |
| "eval_runtime": 9.974, | |
| "eval_samples_per_second": 10.527, | |
| "eval_steps_per_second": 0.702, | |
| "step": 944 | |
| }, | |
| { | |
| "epoch": 16.10169491525424, | |
| "grad_norm": 1.1021959781646729, | |
| "learning_rate": 4.067796610169492e-05, | |
| "loss": 1.2412, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 16.949152542372882, | |
| "grad_norm": 0.9624550938606262, | |
| "learning_rate": 3.966101694915254e-05, | |
| "loss": 1.2348, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 1.0155479907989502, | |
| "eval_runtime": 8.9635, | |
| "eval_samples_per_second": 11.714, | |
| "eval_steps_per_second": 0.781, | |
| "step": 1003 | |
| }, | |
| { | |
| "epoch": 17.796610169491526, | |
| "grad_norm": 0.9586867094039917, | |
| "learning_rate": 3.864406779661017e-05, | |
| "loss": 1.2455, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.0335369110107422, | |
| "eval_runtime": 9.0555, | |
| "eval_samples_per_second": 11.595, | |
| "eval_steps_per_second": 0.773, | |
| "step": 1062 | |
| }, | |
| { | |
| "epoch": 18.64406779661017, | |
| "grad_norm": 1.7303390502929688, | |
| "learning_rate": 3.76271186440678e-05, | |
| "loss": 1.2238, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 1.020735263824463, | |
| "eval_runtime": 9.6479, | |
| "eval_samples_per_second": 10.883, | |
| "eval_steps_per_second": 0.726, | |
| "step": 1121 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2950, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 4, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 4 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.5260523640520704e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |