| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.15687043482523652, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0015687043482523653, | |
| "eval_loss": 1.8242242336273193, | |
| "eval_runtime": 142.2157, | |
| "eval_samples_per_second": 15.104, | |
| "eval_steps_per_second": 7.552, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004706113044757096, | |
| "grad_norm": 0.7702743411064148, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8967, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009412226089514192, | |
| "grad_norm": 0.7187179327011108, | |
| "learning_rate": 6e-05, | |
| "loss": 1.7129, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.014118339134271287, | |
| "grad_norm": 0.5630563497543335, | |
| "learning_rate": 9e-05, | |
| "loss": 1.6705, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014118339134271287, | |
| "eval_loss": 1.6580013036727905, | |
| "eval_runtime": 137.3695, | |
| "eval_samples_per_second": 15.637, | |
| "eval_steps_per_second": 7.818, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.018824452179028384, | |
| "grad_norm": 0.6352736353874207, | |
| "learning_rate": 0.00012, | |
| "loss": 1.6757, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02353056522378548, | |
| "grad_norm": 0.5756219029426575, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.6138, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.028236678268542575, | |
| "grad_norm": 0.5083643794059753, | |
| "learning_rate": 0.00018, | |
| "loss": 1.6546, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028236678268542575, | |
| "eval_loss": 1.5876401662826538, | |
| "eval_runtime": 145.4617, | |
| "eval_samples_per_second": 14.767, | |
| "eval_steps_per_second": 7.383, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03294279131329967, | |
| "grad_norm": 0.4145301878452301, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 1.5349, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03764890435805677, | |
| "grad_norm": 0.42459216713905334, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.5664, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04235501740281386, | |
| "grad_norm": 0.4329008460044861, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 1.5261, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04235501740281386, | |
| "eval_loss": 1.5502554178237915, | |
| "eval_runtime": 148.1658, | |
| "eval_samples_per_second": 14.497, | |
| "eval_steps_per_second": 7.249, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04706113044757096, | |
| "grad_norm": 0.4267719089984894, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.6148, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05176724349232806, | |
| "grad_norm": 0.4181731641292572, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 1.5572, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05647335653708515, | |
| "grad_norm": 0.4355478584766388, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.4369, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05647335653708515, | |
| "eval_loss": 1.531390905380249, | |
| "eval_runtime": 137.4641, | |
| "eval_samples_per_second": 15.626, | |
| "eval_steps_per_second": 7.813, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.06117946958184225, | |
| "grad_norm": 0.4404115080833435, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 1.515, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06588558262659934, | |
| "grad_norm": 0.39275091886520386, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.5592, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.07059169567135644, | |
| "grad_norm": 0.37436187267303467, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.5643, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07059169567135644, | |
| "eval_loss": 1.515994906425476, | |
| "eval_runtime": 137.4794, | |
| "eval_samples_per_second": 15.624, | |
| "eval_steps_per_second": 7.812, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07529780871611354, | |
| "grad_norm": 0.39492252469062805, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 1.5264, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.08000392176087064, | |
| "grad_norm": 0.39957186579704285, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 1.5242, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.08471003480562772, | |
| "grad_norm": 0.3886667490005493, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.4754, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.08471003480562772, | |
| "eval_loss": 1.5080097913742065, | |
| "eval_runtime": 148.2053, | |
| "eval_samples_per_second": 14.493, | |
| "eval_steps_per_second": 7.247, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.08941614785038482, | |
| "grad_norm": 0.3844744861125946, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 1.4846, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.09412226089514192, | |
| "grad_norm": 0.37326759099960327, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4674, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09882837393989902, | |
| "grad_norm": 0.3983359634876251, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.4584, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09882837393989902, | |
| "eval_loss": 1.5008231401443481, | |
| "eval_runtime": 148.4011, | |
| "eval_samples_per_second": 14.474, | |
| "eval_steps_per_second": 7.237, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.10353448698465612, | |
| "grad_norm": 0.3726574778556824, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.4221, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.1082406000294132, | |
| "grad_norm": 0.3979525566101074, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 1.5156, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.1129467130741703, | |
| "grad_norm": 0.3521616756916046, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.4337, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1129467130741703, | |
| "eval_loss": 1.4966624975204468, | |
| "eval_runtime": 137.8159, | |
| "eval_samples_per_second": 15.586, | |
| "eval_steps_per_second": 7.793, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1176528261189274, | |
| "grad_norm": 0.3657877743244171, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.5035, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1223589391636845, | |
| "grad_norm": 0.3593926727771759, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.4785, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.12706505220844158, | |
| "grad_norm": 0.37910911440849304, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.5454, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.12706505220844158, | |
| "eval_loss": 1.4934673309326172, | |
| "eval_runtime": 137.7475, | |
| "eval_samples_per_second": 15.594, | |
| "eval_steps_per_second": 7.797, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.13177116525319868, | |
| "grad_norm": 0.38638541102409363, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.4935, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.13647727829795578, | |
| "grad_norm": 0.3478032350540161, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.4452, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.14118339134271288, | |
| "grad_norm": 0.34858161211013794, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.4752, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14118339134271288, | |
| "eval_loss": 1.492301106452942, | |
| "eval_runtime": 148.3193, | |
| "eval_samples_per_second": 14.482, | |
| "eval_steps_per_second": 7.241, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14588950438746998, | |
| "grad_norm": 0.37905701994895935, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.5333, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.15059561743222707, | |
| "grad_norm": 0.37757179141044617, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.4955, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.15530173047698417, | |
| "grad_norm": 0.3686583936214447, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.4875, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.15530173047698417, | |
| "eval_loss": 1.4919251203536987, | |
| "eval_runtime": 145.2178, | |
| "eval_samples_per_second": 14.792, | |
| "eval_steps_per_second": 7.396, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.589828125877862e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |