| { | |
| "best_metric": 0.21938171982765198, | |
| "best_model_checkpoint": "plantuml-code-generation/checkpoint-300", | |
| "epoch": 9.30232558139535, | |
| "eval_steps": 50, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31007751937984496, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 5.2132, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "grad_norm": 8.985054969787598, | |
| "learning_rate": 7.2e-05, | |
| "loss": 3.2089, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 0.10187079757452011, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 0.7806, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2403100775193798, | |
| "grad_norm": 0.10785023868083954, | |
| "learning_rate": 0.000152, | |
| "loss": 0.6766, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.550387596899225, | |
| "grad_norm": 0.16855500638484955, | |
| "learning_rate": 0.000192, | |
| "loss": 0.5977, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.550387596899225, | |
| "eval_bleu_score": 58.6485, | |
| "eval_levenshtein_similarity": 0.6472, | |
| "eval_loss": 0.5428789854049683, | |
| "eval_runtime": 21.6968, | |
| "eval_samples_per_second": 11.891, | |
| "eval_steps_per_second": 1.521, | |
| "eval_structural_distance": 0.6925, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.8604651162790697, | |
| "grad_norm": 0.09580887109041214, | |
| "learning_rate": 0.00019360000000000002, | |
| "loss": 0.5204, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.1705426356589146, | |
| "grad_norm": 0.09782849997282028, | |
| "learning_rate": 0.0001856, | |
| "loss": 0.4559, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.4806201550387597, | |
| "grad_norm": 0.16383880376815796, | |
| "learning_rate": 0.0001776, | |
| "loss": 0.4082, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.7906976744186047, | |
| "grad_norm": 0.09721529483795166, | |
| "learning_rate": 0.0001696, | |
| "loss": 0.3532, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.10077519379845, | |
| "grad_norm": 0.12678824365139008, | |
| "learning_rate": 0.00016160000000000002, | |
| "loss": 0.3229, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.10077519379845, | |
| "eval_bleu_score": 67.9654, | |
| "eval_levenshtein_similarity": 0.7437, | |
| "eval_loss": 0.3111547529697418, | |
| "eval_runtime": 21.7177, | |
| "eval_samples_per_second": 11.88, | |
| "eval_steps_per_second": 1.519, | |
| "eval_structural_distance": 0.5332, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.4108527131782944, | |
| "grad_norm": 0.09589164704084396, | |
| "learning_rate": 0.00015360000000000002, | |
| "loss": 0.3155, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.7209302325581395, | |
| "grad_norm": 0.13837170600891113, | |
| "learning_rate": 0.00014560000000000002, | |
| "loss": 0.2995, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.0310077519379846, | |
| "grad_norm": 0.10599964112043381, | |
| "learning_rate": 0.00013759999999999998, | |
| "loss": 0.2918, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.341085271317829, | |
| "grad_norm": 0.09682385623455048, | |
| "learning_rate": 0.0001296, | |
| "loss": 0.26, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.651162790697675, | |
| "grad_norm": 0.12630213797092438, | |
| "learning_rate": 0.0001216, | |
| "loss": 0.2706, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.651162790697675, | |
| "eval_bleu_score": 71.5314, | |
| "eval_levenshtein_similarity": 0.7707, | |
| "eval_loss": 0.26290762424468994, | |
| "eval_runtime": 21.6797, | |
| "eval_samples_per_second": 11.901, | |
| "eval_steps_per_second": 1.522, | |
| "eval_structural_distance": 0.4633, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.961240310077519, | |
| "grad_norm": 0.12997351586818695, | |
| "learning_rate": 0.0001136, | |
| "loss": 0.2801, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.271317829457364, | |
| "grad_norm": 0.12334702163934708, | |
| "learning_rate": 0.0001056, | |
| "loss": 0.2412, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.5813953488372094, | |
| "grad_norm": 0.14276793599128723, | |
| "learning_rate": 9.76e-05, | |
| "loss": 0.229, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.891472868217054, | |
| "grad_norm": 0.14187170565128326, | |
| "learning_rate": 8.960000000000001e-05, | |
| "loss": 0.2426, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.2015503875969, | |
| "grad_norm": 0.20206551253795624, | |
| "learning_rate": 8.16e-05, | |
| "loss": 0.2548, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.2015503875969, | |
| "eval_bleu_score": 73.6148, | |
| "eval_levenshtein_similarity": 0.7889, | |
| "eval_loss": 0.23503056168556213, | |
| "eval_runtime": 21.6369, | |
| "eval_samples_per_second": 11.924, | |
| "eval_steps_per_second": 1.525, | |
| "eval_structural_distance": 0.4277, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.511627906976744, | |
| "grad_norm": 0.14615343511104584, | |
| "learning_rate": 7.36e-05, | |
| "loss": 0.226, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.821705426356589, | |
| "grad_norm": 0.14554168283939362, | |
| "learning_rate": 6.560000000000001e-05, | |
| "loss": 0.2152, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.131782945736434, | |
| "grad_norm": 0.1501888930797577, | |
| "learning_rate": 5.76e-05, | |
| "loss": 0.2175, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.441860465116279, | |
| "grad_norm": 0.2138899862766266, | |
| "learning_rate": 4.96e-05, | |
| "loss": 0.2148, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.751937984496124, | |
| "grad_norm": 0.16335947811603546, | |
| "learning_rate": 4.16e-05, | |
| "loss": 0.2159, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.751937984496124, | |
| "eval_bleu_score": 74.3094, | |
| "eval_levenshtein_similarity": 0.799, | |
| "eval_loss": 0.22344815731048584, | |
| "eval_runtime": 21.6149, | |
| "eval_samples_per_second": 11.936, | |
| "eval_steps_per_second": 1.527, | |
| "eval_structural_distance": 0.411, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.062015503875969, | |
| "grad_norm": 0.23698154091835022, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 0.2005, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.372093023255815, | |
| "grad_norm": 0.14778129756450653, | |
| "learning_rate": 2.5600000000000002e-05, | |
| "loss": 0.1981, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.682170542635658, | |
| "grad_norm": 0.15810427069664001, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.2173, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.992248062015504, | |
| "grad_norm": 0.1843605935573578, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.1969, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.30232558139535, | |
| "grad_norm": 0.14364413917064667, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.1963, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.30232558139535, | |
| "eval_bleu_score": 74.7205, | |
| "eval_levenshtein_similarity": 0.7988, | |
| "eval_loss": 0.21938171982765198, | |
| "eval_runtime": 21.6361, | |
| "eval_samples_per_second": 11.924, | |
| "eval_steps_per_second": 1.525, | |
| "eval_structural_distance": 0.4108, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.948057685018542e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |