| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9615384615384617, | |
| "eval_steps": 7, | |
| "global_step": 77, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.038461538461538464, | |
| "grad_norm": 1.3442820310592651, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8401, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.038461538461538464, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.3672, | |
| "eval_samples_per_second": 4.099, | |
| "eval_steps_per_second": 0.559, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 0.9724370837211609, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8328, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11538461538461539, | |
| "grad_norm": 1.101107120513916, | |
| "learning_rate": 3e-05, | |
| "loss": 1.678, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.3484959602355957, | |
| "learning_rate": 4e-05, | |
| "loss": 1.912, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9033, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2692307692307692, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2692307692307692, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8644, | |
| "eval_samples_per_second": 4.523, | |
| "eval_steps_per_second": 0.617, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.34615384615384615, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.994504457428558e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.978029910109491e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.950612572673255e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.91231271437788e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8639, | |
| "eval_samples_per_second": 4.523, | |
| "eval_steps_per_second": 0.617, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.863214526624065e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.8034259378842e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6538461538461539, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.733078376452171e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.652326481535435e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7307692307692307, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.561347763324484e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.460342212786932e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8076923076923077, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.349531862043952e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8076923076923077, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.865, | |
| "eval_samples_per_second": 4.522, | |
| "eval_steps_per_second": 0.617, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.229160296295488e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.8846153846153846, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.099492118367123e-05, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.960812367055646e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.81342589055191e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.657656676318346e-05, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0384615384615385, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.493847138894209e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.322357367194109e-05, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8674, | |
| "eval_samples_per_second": 4.52, | |
| "eval_steps_per_second": 0.616, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.1153846153846154, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.143564332954425e-05, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.957861062067614e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1923076923076923, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.765655770625997e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.56737096757421e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.2692307692307692, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.363442525942826e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.154318724704853e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.940459263361249e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8653, | |
| "eval_samples_per_second": 4.522, | |
| "eval_steps_per_second": 0.617, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.722334251421665e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.500423175001705e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.275213842808383e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.0472013138307235e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.816886809092651e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5769230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.584776609860414e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.351380944726465e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8664, | |
| "eval_samples_per_second": 4.521, | |
| "eval_steps_per_second": 0.616, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6538461538461537, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.117212868016303e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.882787131983698e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.648619055273537e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.415223390139588e-05, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8076923076923077, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.183113190907349e-05, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.952798686169279e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.8846153846153846, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.7247861571916185e-05, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.8846153846153846, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8653, | |
| "eval_samples_per_second": 4.522, | |
| "eval_steps_per_second": 0.617, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.499576824998298e-05, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9615384615384617, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.277665748578336e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0595407366387504e-05, | |
| "loss": 0.0, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.0384615384615383, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.8456812752951485e-05, | |
| "loss": 0.0, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.636557474057173e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4326290324257894e-05, | |
| "loss": 0.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.234344229374003e-05, | |
| "loss": 0.0, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8673, | |
| "eval_samples_per_second": 4.52, | |
| "eval_steps_per_second": 0.616, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.1923076923076925, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.042138937932388e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.8564356670455767e-05, | |
| "loss": 0.0, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.269230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.677642632805892e-05, | |
| "loss": 0.0, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.5061528611057918e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3461538461538463, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3423433236816563e-05, | |
| "loss": 0.0, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1865741094480909e-05, | |
| "loss": 0.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.423076923076923, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0391876329443533e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.423076923076923, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8685, | |
| "eval_samples_per_second": 4.519, | |
| "eval_steps_per_second": 0.616, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.005078816328771e-06, | |
| "loss": 0.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.708397037045129e-06, | |
| "loss": 0.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.50468137956049e-06, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.5769230769230766, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.3965778721306755e-06, | |
| "loss": 0.0, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.386522366755169e-06, | |
| "loss": 0.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.6538461538461537, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.476735184645674e-06, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6692162354782944e-06, | |
| "loss": 0.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.868, | |
| "eval_samples_per_second": 4.519, | |
| "eval_steps_per_second": 0.616, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.730769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9657406211579966e-06, | |
| "loss": 0.0, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3678547337593494e-06, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8076923076923075, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.768728562211947e-07, | |
| "loss": 0.0, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.938742732674529e-07, | |
| "loss": 0.0, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.1970089890509527e-07, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4955425714431353e-08, | |
| "loss": 0.0, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.9615384615384617, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.9615384615384617, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.8662, | |
| "eval_samples_per_second": 4.521, | |
| "eval_steps_per_second": 0.616, | |
| "step": 77 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 77, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1313781074296832e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |