| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8298755186721992, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02074688796680498, | |
| "grad_norm": 0.4129277765750885, | |
| "learning_rate": 8.264462809917356e-05, | |
| "loss": 1.7066, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04149377593360996, | |
| "grad_norm": 0.2623195946216583, | |
| "learning_rate": 0.00016528925619834712, | |
| "loss": 0.7931, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06224066390041494, | |
| "grad_norm": 0.29180774092674255, | |
| "learning_rate": 0.00019746614242027087, | |
| "loss": 0.6638, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08298755186721991, | |
| "grad_norm": 0.2945658564567566, | |
| "learning_rate": 0.0001930974224552206, | |
| "loss": 0.6448, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1037344398340249, | |
| "grad_norm": 0.28438836336135864, | |
| "learning_rate": 0.00018872870249017038, | |
| "loss": 0.6229, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.12448132780082988, | |
| "grad_norm": 0.27639704942703247, | |
| "learning_rate": 0.00018435998252512014, | |
| "loss": 0.6064, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.14522821576763487, | |
| "grad_norm": 0.2640627920627594, | |
| "learning_rate": 0.0001799912625600699, | |
| "loss": 0.591, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16597510373443983, | |
| "grad_norm": 0.258998304605484, | |
| "learning_rate": 0.00017562254259501968, | |
| "loss": 0.5753, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18672199170124482, | |
| "grad_norm": 0.24890367686748505, | |
| "learning_rate": 0.00017125382262996944, | |
| "loss": 0.5734, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.2074688796680498, | |
| "grad_norm": 0.2385823130607605, | |
| "learning_rate": 0.00016688510266491918, | |
| "loss": 0.5721, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2074688796680498, | |
| "eval_loss": 0.573611319065094, | |
| "eval_runtime": 141.3666, | |
| "eval_samples_per_second": 7.166, | |
| "eval_steps_per_second": 0.898, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22821576763485477, | |
| "grad_norm": 0.21996833384037018, | |
| "learning_rate": 0.00016251638269986895, | |
| "loss": 0.5688, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24896265560165975, | |
| "grad_norm": 0.23547005653381348, | |
| "learning_rate": 0.0001581476627348187, | |
| "loss": 0.554, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2697095435684647, | |
| "grad_norm": 0.23831422626972198, | |
| "learning_rate": 0.00015377894276976848, | |
| "loss": 0.5488, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.29045643153526973, | |
| "grad_norm": 0.24010932445526123, | |
| "learning_rate": 0.00014941022280471822, | |
| "loss": 0.5445, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3112033195020747, | |
| "grad_norm": 0.22254310548305511, | |
| "learning_rate": 0.00014504150283966798, | |
| "loss": 0.5491, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.33195020746887965, | |
| "grad_norm": 0.23563440144062042, | |
| "learning_rate": 0.00014067278287461772, | |
| "loss": 0.5428, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.35269709543568467, | |
| "grad_norm": 0.23261123895645142, | |
| "learning_rate": 0.0001363040629095675, | |
| "loss": 0.5444, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.37344398340248963, | |
| "grad_norm": 0.2282126247882843, | |
| "learning_rate": 0.00013193534294451726, | |
| "loss": 0.5298, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3941908713692946, | |
| "grad_norm": 0.28895512223243713, | |
| "learning_rate": 0.00012756662297946702, | |
| "loss": 0.5368, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.4149377593360996, | |
| "grad_norm": 0.2174956202507019, | |
| "learning_rate": 0.0001231979030144168, | |
| "loss": 0.5333, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4149377593360996, | |
| "eval_loss": 0.5404650568962097, | |
| "eval_runtime": 140.5485, | |
| "eval_samples_per_second": 7.207, | |
| "eval_steps_per_second": 0.904, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.43568464730290457, | |
| "grad_norm": 0.22959038615226746, | |
| "learning_rate": 0.00011882918304936654, | |
| "loss": 0.5254, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.45643153526970953, | |
| "grad_norm": 0.20930084586143494, | |
| "learning_rate": 0.00011446046308431631, | |
| "loss": 0.53, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.47717842323651455, | |
| "grad_norm": 0.2505868077278137, | |
| "learning_rate": 0.00011009174311926606, | |
| "loss": 0.5344, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.4979253112033195, | |
| "grad_norm": 0.24351277947425842, | |
| "learning_rate": 0.00010572302315421583, | |
| "loss": 0.5317, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5186721991701245, | |
| "grad_norm": 0.24290521442890167, | |
| "learning_rate": 0.00010135430318916559, | |
| "loss": 0.5265, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.5394190871369294, | |
| "grad_norm": 0.2370923012495041, | |
| "learning_rate": 9.698558322411535e-05, | |
| "loss": 0.5255, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.5601659751037344, | |
| "grad_norm": 0.2208719551563263, | |
| "learning_rate": 9.26168632590651e-05, | |
| "loss": 0.5182, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.5809128630705395, | |
| "grad_norm": 0.24646012485027313, | |
| "learning_rate": 8.824814329401486e-05, | |
| "loss": 0.5117, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.6016597510373444, | |
| "grad_norm": 0.21846014261245728, | |
| "learning_rate": 8.387942332896462e-05, | |
| "loss": 0.5275, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.6224066390041494, | |
| "grad_norm": 0.23680146038532257, | |
| "learning_rate": 7.951070336391437e-05, | |
| "loss": 0.5252, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6224066390041494, | |
| "eval_loss": 0.5271425247192383, | |
| "eval_runtime": 232.8953, | |
| "eval_samples_per_second": 4.35, | |
| "eval_steps_per_second": 0.545, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6431535269709544, | |
| "grad_norm": 0.24285145103931427, | |
| "learning_rate": 7.514198339886414e-05, | |
| "loss": 0.5186, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.6639004149377593, | |
| "grad_norm": 0.25104212760925293, | |
| "learning_rate": 7.07732634338139e-05, | |
| "loss": 0.5154, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.6846473029045643, | |
| "grad_norm": 0.23628434538841248, | |
| "learning_rate": 6.640454346876365e-05, | |
| "loss": 0.5127, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.7053941908713693, | |
| "grad_norm": 0.25255104899406433, | |
| "learning_rate": 6.203582350371342e-05, | |
| "loss": 0.5174, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7261410788381742, | |
| "grad_norm": 0.23285475373268127, | |
| "learning_rate": 5.766710353866317e-05, | |
| "loss": 0.5059, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.7468879668049793, | |
| "grad_norm": 0.24090896546840668, | |
| "learning_rate": 5.329838357361293e-05, | |
| "loss": 0.5179, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7676348547717843, | |
| "grad_norm": 0.23537760972976685, | |
| "learning_rate": 4.892966360856269e-05, | |
| "loss": 0.5164, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.7883817427385892, | |
| "grad_norm": 0.23769009113311768, | |
| "learning_rate": 4.456094364351245e-05, | |
| "loss": 0.5103, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.8091286307053942, | |
| "grad_norm": 0.2349609136581421, | |
| "learning_rate": 4.019222367846222e-05, | |
| "loss": 0.5108, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.8298755186721992, | |
| "grad_norm": 0.232425257563591, | |
| "learning_rate": 3.582350371341197e-05, | |
| "loss": 0.5074, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8298755186721992, | |
| "eval_loss": 0.5192467570304871, | |
| "eval_runtime": 206.8865, | |
| "eval_samples_per_second": 4.896, | |
| "eval_steps_per_second": 0.614, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2410, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.8341028691968e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |