| { | |
| "best_metric": 5.346867561340332, | |
| "best_model_checkpoint": "./results/models/mistral-dna/checkpoint-22675", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 22675, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11025358324145534, | |
| "grad_norm": 0.142578125, | |
| "learning_rate": 0.0003991179713340684, | |
| "loss": 6.6131, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2205071664829107, | |
| "grad_norm": 0.1298828125, | |
| "learning_rate": 0.00039823594266813673, | |
| "loss": 6.0212, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33076074972436603, | |
| "grad_norm": 0.12255859375, | |
| "learning_rate": 0.00039735391400220506, | |
| "loss": 5.8931, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4410143329658214, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.00039647188533627344, | |
| "loss": 5.822, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5512679162072768, | |
| "grad_norm": 0.125, | |
| "learning_rate": 0.0003955898566703418, | |
| "loss": 5.7678, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6615214994487321, | |
| "grad_norm": 0.119140625, | |
| "learning_rate": 0.00039470782800441015, | |
| "loss": 5.7204, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7717750826901875, | |
| "grad_norm": 0.12255859375, | |
| "learning_rate": 0.00039382579933847854, | |
| "loss": 5.7025, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8820286659316428, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 0.00039294377067254687, | |
| "loss": 5.6733, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9922822491730982, | |
| "grad_norm": 0.1328125, | |
| "learning_rate": 0.00039206174200661525, | |
| "loss": 5.6491, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 5.551197528839111, | |
| "eval_runtime": 3.3539, | |
| "eval_samples_per_second": 86.765, | |
| "eval_steps_per_second": 1.491, | |
| "step": 4535 | |
| }, | |
| { | |
| "epoch": 1.1025358324145536, | |
| "grad_norm": 0.126953125, | |
| "learning_rate": 0.0003911797133406836, | |
| "loss": 5.6279, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.2127894156560088, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 0.00039029768467475196, | |
| "loss": 5.6068, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.3230429988974641, | |
| "grad_norm": 0.1455078125, | |
| "learning_rate": 0.0003894156560088203, | |
| "loss": 5.6068, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.4332965821389196, | |
| "grad_norm": 0.130859375, | |
| "learning_rate": 0.00038853362734288867, | |
| "loss": 5.5658, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.543550165380375, | |
| "grad_norm": 0.1318359375, | |
| "learning_rate": 0.000387651598676957, | |
| "loss": 5.5665, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.6538037486218302, | |
| "grad_norm": 0.138671875, | |
| "learning_rate": 0.0003867695700110254, | |
| "loss": 5.5565, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.7640573318632855, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 0.00038588754134509377, | |
| "loss": 5.5477, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.8743109151047408, | |
| "grad_norm": 0.1298828125, | |
| "learning_rate": 0.0003850055126791621, | |
| "loss": 5.5295, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.9845644983461963, | |
| "grad_norm": 0.140625, | |
| "learning_rate": 0.0003841234840132304, | |
| "loss": 5.5208, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 5.451062202453613, | |
| "eval_runtime": 3.2969, | |
| "eval_samples_per_second": 88.264, | |
| "eval_steps_per_second": 1.517, | |
| "step": 9070 | |
| }, | |
| { | |
| "epoch": 2.0948180815876514, | |
| "grad_norm": 0.138671875, | |
| "learning_rate": 0.0003832414553472988, | |
| "loss": 5.5017, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.205071664829107, | |
| "grad_norm": 0.142578125, | |
| "learning_rate": 0.0003823594266813672, | |
| "loss": 5.4918, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.3153252480705624, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.0003814773980154355, | |
| "loss": 5.4806, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.4255788313120177, | |
| "grad_norm": 0.13671875, | |
| "learning_rate": 0.00038059536934950385, | |
| "loss": 5.4877, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.535832414553473, | |
| "grad_norm": 0.138671875, | |
| "learning_rate": 0.00037971334068357223, | |
| "loss": 5.4696, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.6460859977949283, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 0.0003788313120176406, | |
| "loss": 5.4781, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.7563395810363835, | |
| "grad_norm": 0.140625, | |
| "learning_rate": 0.00037794928335170894, | |
| "loss": 5.4748, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.8665931642778393, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 0.0003770672546857773, | |
| "loss": 5.4608, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.9768467475192946, | |
| "grad_norm": 0.1923828125, | |
| "learning_rate": 0.00037618522601984565, | |
| "loss": 5.453, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 5.399576663970947, | |
| "eval_runtime": 3.4053, | |
| "eval_samples_per_second": 85.455, | |
| "eval_steps_per_second": 1.468, | |
| "step": 13605 | |
| }, | |
| { | |
| "epoch": 3.08710033076075, | |
| "grad_norm": 0.1865234375, | |
| "learning_rate": 0.00037530319735391404, | |
| "loss": 5.4278, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.197353914002205, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 0.00037442116868798236, | |
| "loss": 5.4226, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.3076074972436604, | |
| "grad_norm": 0.1484375, | |
| "learning_rate": 0.00037353914002205075, | |
| "loss": 5.4334, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.4178610804851157, | |
| "grad_norm": 0.162109375, | |
| "learning_rate": 0.0003726571113561191, | |
| "loss": 5.4321, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.528114663726571, | |
| "grad_norm": 0.1572265625, | |
| "learning_rate": 0.00037177508269018746, | |
| "loss": 5.4277, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.6383682469680263, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 0.0003708930540242558, | |
| "loss": 5.4193, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.7486218302094816, | |
| "grad_norm": 0.1611328125, | |
| "learning_rate": 0.00037001102535832417, | |
| "loss": 5.4226, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.8588754134509373, | |
| "grad_norm": 0.154296875, | |
| "learning_rate": 0.00036912899669239255, | |
| "loss": 5.4214, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.9691289966923926, | |
| "grad_norm": 0.1533203125, | |
| "learning_rate": 0.0003682469680264609, | |
| "loss": 5.4218, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 5.367885112762451, | |
| "eval_runtime": 3.6662, | |
| "eval_samples_per_second": 79.373, | |
| "eval_steps_per_second": 1.364, | |
| "step": 18140 | |
| }, | |
| { | |
| "epoch": 4.0793825799338475, | |
| "grad_norm": 0.1748046875, | |
| "learning_rate": 0.0003673649393605292, | |
| "loss": 5.4067, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.189636163175303, | |
| "grad_norm": 0.1572265625, | |
| "learning_rate": 0.0003664829106945976, | |
| "loss": 5.3963, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.299889746416759, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 0.000365600882028666, | |
| "loss": 5.3929, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.410143329658214, | |
| "grad_norm": 0.1630859375, | |
| "learning_rate": 0.0003647188533627343, | |
| "loss": 5.3888, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.5203969128996695, | |
| "grad_norm": 0.16015625, | |
| "learning_rate": 0.00036383682469680263, | |
| "loss": 5.3982, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.630650496141125, | |
| "grad_norm": 0.173828125, | |
| "learning_rate": 0.000362954796030871, | |
| "loss": 5.3754, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.74090407938258, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 0.0003620727673649394, | |
| "loss": 5.3854, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.851157662624035, | |
| "grad_norm": 0.1748046875, | |
| "learning_rate": 0.00036119073869900773, | |
| "loss": 5.3876, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.961411245865491, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.0003603087100330761, | |
| "loss": 5.3899, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 5.346867561340332, | |
| "eval_runtime": 3.7974, | |
| "eval_samples_per_second": 76.63, | |
| "eval_steps_per_second": 1.317, | |
| "step": 22675 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 226750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.976338069294799e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |