| { | |
| "best_global_step": 1000, | |
| "best_metric": 2.0130758995663705, | |
| "best_model_checkpoint": "./SALAMA_C5/checkpoint-1000", | |
| "epoch": 0.6807351940095303, | |
| "eval_steps": 1000, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013614703880190605, | |
| "grad_norm": 5.017816066741943, | |
| "learning_rate": 3.8e-07, | |
| "loss": 0.0279, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02722940776038121, | |
| "grad_norm": 3.6539018154144287, | |
| "learning_rate": 7.8e-07, | |
| "loss": 0.0334, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04084411164057182, | |
| "grad_norm": 2.4141480922698975, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.0266, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05445881552076242, | |
| "grad_norm": 2.1062567234039307, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 0.0197, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06807351940095303, | |
| "grad_norm": 4.6255106925964355, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.0279, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08168822328114364, | |
| "grad_norm": 1.933724284172058, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.0261, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09530292716133425, | |
| "grad_norm": 2.3127458095550537, | |
| "learning_rate": 2.7800000000000005e-06, | |
| "loss": 0.0261, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.10891763104152484, | |
| "grad_norm": 2.6971325874328613, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.0186, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.12253233492171545, | |
| "grad_norm": 3.273725986480713, | |
| "learning_rate": 3.58e-06, | |
| "loss": 0.0278, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.13614703880190607, | |
| "grad_norm": 2.669668197631836, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 0.0308, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14976174268209666, | |
| "grad_norm": 0.6969190239906311, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.0271, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.16337644656228728, | |
| "grad_norm": 2.6846354007720947, | |
| "learning_rate": 4.78e-06, | |
| "loss": 0.0298, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.17699115044247787, | |
| "grad_norm": 3.2715647220611572, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.0223, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1906058543226685, | |
| "grad_norm": 3.50703763961792, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 0.0273, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.2042205582028591, | |
| "grad_norm": 1.7984274625778198, | |
| "learning_rate": 5.98e-06, | |
| "loss": 0.0305, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.21783526208304968, | |
| "grad_norm": 2.7656664848327637, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 0.0484, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.2314499659632403, | |
| "grad_norm": 0.7585025429725647, | |
| "learning_rate": 6.780000000000001e-06, | |
| "loss": 0.0213, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.2450646698434309, | |
| "grad_norm": 4.038341522216797, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 0.0333, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.2586793737236215, | |
| "grad_norm": 2.8464272022247314, | |
| "learning_rate": 7.58e-06, | |
| "loss": 0.0269, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.27229407760381213, | |
| "grad_norm": 3.064093589782715, | |
| "learning_rate": 7.980000000000002e-06, | |
| "loss": 0.0169, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2859087814840027, | |
| "grad_norm": 5.7664971351623535, | |
| "learning_rate": 8.380000000000001e-06, | |
| "loss": 0.0275, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.2995234853641933, | |
| "grad_norm": 1.4587912559509277, | |
| "learning_rate": 8.78e-06, | |
| "loss": 0.0258, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.3131381892443839, | |
| "grad_norm": 2.867338180541992, | |
| "learning_rate": 9.180000000000002e-06, | |
| "loss": 0.0279, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.32675289312457456, | |
| "grad_norm": 4.9567484855651855, | |
| "learning_rate": 9.58e-06, | |
| "loss": 0.0301, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.34036759700476515, | |
| "grad_norm": 4.536691665649414, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 0.0329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 2.4109556674957275, | |
| "learning_rate": 9.95136933708728e-06, | |
| "loss": 0.0351, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.36759700476514634, | |
| "grad_norm": 3.561450958251953, | |
| "learning_rate": 9.900179165600206e-06, | |
| "loss": 0.0368, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.381211708645337, | |
| "grad_norm": 1.7630550861358643, | |
| "learning_rate": 9.848988994113131e-06, | |
| "loss": 0.0388, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.3948264125255276, | |
| "grad_norm": 1.5144799947738647, | |
| "learning_rate": 9.797798822626056e-06, | |
| "loss": 0.0304, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.4084411164057182, | |
| "grad_norm": 4.403763771057129, | |
| "learning_rate": 9.746608651138983e-06, | |
| "loss": 0.033, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.42205582028590877, | |
| "grad_norm": 3.6199100017547607, | |
| "learning_rate": 9.695418479651908e-06, | |
| "loss": 0.0377, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.43567052416609936, | |
| "grad_norm": 3.1659555435180664, | |
| "learning_rate": 9.644228308164833e-06, | |
| "loss": 0.0381, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.44928522804629, | |
| "grad_norm": 4.508129596710205, | |
| "learning_rate": 9.59303813667776e-06, | |
| "loss": 0.0277, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.4628999319264806, | |
| "grad_norm": 3.981079578399658, | |
| "learning_rate": 9.541847965190683e-06, | |
| "loss": 0.0297, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.4765146358066712, | |
| "grad_norm": 2.7824220657348633, | |
| "learning_rate": 9.49065779370361e-06, | |
| "loss": 0.0398, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4901293396868618, | |
| "grad_norm": 5.651052951812744, | |
| "learning_rate": 9.439467622216535e-06, | |
| "loss": 0.0448, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.5037440435670524, | |
| "grad_norm": 4.3824920654296875, | |
| "learning_rate": 9.38827745072946e-06, | |
| "loss": 0.0314, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.517358747447243, | |
| "grad_norm": 4.516641139984131, | |
| "learning_rate": 9.337087279242385e-06, | |
| "loss": 0.0437, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.5309734513274337, | |
| "grad_norm": 3.0896172523498535, | |
| "learning_rate": 9.285897107755312e-06, | |
| "loss": 0.0389, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.5445881552076243, | |
| "grad_norm": 2.9489505290985107, | |
| "learning_rate": 9.234706936268237e-06, | |
| "loss": 0.036, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5582028590878149, | |
| "grad_norm": 1.6065051555633545, | |
| "learning_rate": 9.183516764781162e-06, | |
| "loss": 0.0431, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.5718175629680055, | |
| "grad_norm": 2.8498787879943848, | |
| "learning_rate": 9.132326593294089e-06, | |
| "loss": 0.035, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.585432266848196, | |
| "grad_norm": 1.6527795791625977, | |
| "learning_rate": 9.081136421807014e-06, | |
| "loss": 0.0351, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.5990469707283866, | |
| "grad_norm": 3.358445405960083, | |
| "learning_rate": 9.02994625031994e-06, | |
| "loss": 0.0287, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.6126616746085772, | |
| "grad_norm": 3.706794500350952, | |
| "learning_rate": 8.978756078832866e-06, | |
| "loss": 0.0413, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6262763784887678, | |
| "grad_norm": 3.2217488288879395, | |
| "learning_rate": 8.92756590734579e-06, | |
| "loss": 0.0335, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.6398910823689585, | |
| "grad_norm": 4.742300987243652, | |
| "learning_rate": 8.876375735858716e-06, | |
| "loss": 0.0365, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.6535057862491491, | |
| "grad_norm": 2.290529727935791, | |
| "learning_rate": 8.825185564371641e-06, | |
| "loss": 0.0288, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.6671204901293397, | |
| "grad_norm": 1.5146220922470093, | |
| "learning_rate": 8.773995392884566e-06, | |
| "loss": 0.0379, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.6807351940095303, | |
| "grad_norm": 5.127984046936035, | |
| "learning_rate": 8.722805221397493e-06, | |
| "loss": 0.0256, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6807351940095303, | |
| "eval_loss": 0.02619771473109722, | |
| "eval_runtime": 4340.6262, | |
| "eval_samples_per_second": 2.707, | |
| "eval_steps_per_second": 0.338, | |
| "eval_wer": 2.0130758995663705, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 4407, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.23473281024e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |