| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.0332210931664656, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1785, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.03271791537281304, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1808, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.03167338979293765, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.178, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.03663663541401911, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1887, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.0371641909182427, | |
| "learning_rate": 4.990486745229364e-05, | |
| "loss": 0.1772, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.04007854069569051, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.1683, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.04347491830135535, | |
| "learning_rate": 4.914814565722671e-05, | |
| "loss": 0.1589, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.032813280601834985, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 0.1545, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.0320349651596436, | |
| "learning_rate": 4.765769467591625e-05, | |
| "loss": 0.1431, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.03550685471931854, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 0.142, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.036775495317884224, | |
| "learning_rate": 4.54788011072248e-05, | |
| "loss": 0.1404, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.037308092127002694, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.1304, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.03593110769573973, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.1249, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.03483657463560176, | |
| "learning_rate": 4.1069690242163484e-05, | |
| "loss": 0.1229, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.03368572193911597, | |
| "learning_rate": 3.933941090877615e-05, | |
| "loss": 0.1138, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.030381902117838585, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1125, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.03015218446589334, | |
| "learning_rate": 3.556545654351749e-05, | |
| "loss": 0.1058, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.03365922430315208, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.1013, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.032113749878833836, | |
| "learning_rate": 3.147047612756302e-05, | |
| "loss": 0.0998, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.031038779563112767, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 0.0907, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.02910687996492009, | |
| "learning_rate": 2.717889356869146e-05, | |
| "loss": 0.0891, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.02947620971365539, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0836, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.029528307620463545, | |
| "learning_rate": 2.2821106431308544e-05, | |
| "loss": 0.0785, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.028355800917651552, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.0753, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.030374125439615877, | |
| "learning_rate": 1.852952387243698e-05, | |
| "loss": 0.0775, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.02812084274206112, | |
| "learning_rate": 1.6449496416858284e-05, | |
| "loss": 0.0782, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.025829020482598886, | |
| "learning_rate": 1.443454345648252e-05, | |
| "loss": 0.0707, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.0280188739734266, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 0.0637, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.026624362293195123, | |
| "learning_rate": 1.0660589091223855e-05, | |
| "loss": 0.0631, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.025730151118709174, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.0621, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.027901697536098044, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.0678, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.030664427083571776, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 0.0578, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.02675877087116603, | |
| "learning_rate": 4.521198892775203e-06, | |
| "loss": 0.0576, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.029533712143689446, | |
| "learning_rate": 3.3493649053890326e-06, | |
| "loss": 0.0606, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.0274840722382233, | |
| "learning_rate": 2.3423053240837515e-06, | |
| "loss": 0.0588, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.026833715799204856, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.0606, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.02756778754172055, | |
| "learning_rate": 8.51854342773295e-07, | |
| "loss": 0.0621, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.02573674474074463, | |
| "learning_rate": 3.7980617469479953e-07, | |
| "loss": 0.0575, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.02413150139785941, | |
| "learning_rate": 9.513254770636137e-08, | |
| "loss": 0.0635, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.028153696733032887, | |
| "learning_rate": 0.0, | |
| "loss": 0.0553, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 40, | |
| "total_flos": 218478981677056.0, | |
| "train_loss": 0.1039005771279335, | |
| "train_runtime": 405.7029, | |
| "train_samples_per_second": 0.774, | |
| "train_steps_per_second": 0.099 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 40, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 218478981677056.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |