| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03017832647462277, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00030483158055174517, | |
| "eval_loss": 1.6358747482299805, | |
| "eval_runtime": 393.362, | |
| "eval_samples_per_second": 14.046, | |
| "eval_steps_per_second": 1.757, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009144947416552355, | |
| "grad_norm": 17.194887161254883, | |
| "learning_rate": 1.5e-05, | |
| "loss": 6.7621, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001828989483310471, | |
| "grad_norm": 10.281621932983398, | |
| "learning_rate": 3e-05, | |
| "loss": 5.8586, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0027434842249657062, | |
| "grad_norm": 6.017336845397949, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.2144, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0027434842249657062, | |
| "eval_loss": 1.4405845403671265, | |
| "eval_runtime": 397.296, | |
| "eval_samples_per_second": 13.907, | |
| "eval_steps_per_second": 1.739, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003657978966620942, | |
| "grad_norm": 5.029771327972412, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 5.5157, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004572473708276178, | |
| "grad_norm": 4.417510032653809, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 5.1545, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0054869684499314125, | |
| "grad_norm": 4.255384922027588, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 5.7058, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0054869684499314125, | |
| "eval_loss": 1.3602502346038818, | |
| "eval_runtime": 397.5904, | |
| "eval_samples_per_second": 13.896, | |
| "eval_steps_per_second": 1.738, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006401463191586648, | |
| "grad_norm": 4.237952709197998, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 5.1144, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.007315957933241884, | |
| "grad_norm": 4.633209228515625, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 5.8178, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00823045267489712, | |
| "grad_norm": 3.6630399227142334, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 5.1011, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00823045267489712, | |
| "eval_loss": 1.3356471061706543, | |
| "eval_runtime": 397.6104, | |
| "eval_samples_per_second": 13.896, | |
| "eval_steps_per_second": 1.738, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.009144947416552356, | |
| "grad_norm": 3.946521043777466, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 5.4655, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010059442158207591, | |
| "grad_norm": 3.9313600063323975, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 5.3857, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.010973936899862825, | |
| "grad_norm": 3.9708104133605957, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 5.1117, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010973936899862825, | |
| "eval_loss": 1.3252490758895874, | |
| "eval_runtime": 397.7349, | |
| "eval_samples_per_second": 13.891, | |
| "eval_steps_per_second": 1.737, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01188843164151806, | |
| "grad_norm": 4.425464630126953, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 5.1745, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.012802926383173296, | |
| "grad_norm": 3.689208745956421, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 4.9595, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.013717421124828532, | |
| "grad_norm": 4.085570812225342, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 5.382, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013717421124828532, | |
| "eval_loss": 1.3172868490219116, | |
| "eval_runtime": 397.5764, | |
| "eval_samples_per_second": 13.897, | |
| "eval_steps_per_second": 1.738, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014631915866483767, | |
| "grad_norm": 3.725468635559082, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 5.7061, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015546410608139003, | |
| "grad_norm": 3.879150867462158, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 5.1629, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01646090534979424, | |
| "grad_norm": 3.862717390060425, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 5.5037, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01646090534979424, | |
| "eval_loss": 1.312578797340393, | |
| "eval_runtime": 397.5795, | |
| "eval_samples_per_second": 13.897, | |
| "eval_steps_per_second": 1.738, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.017375400091449476, | |
| "grad_norm": 4.1746826171875, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 5.1687, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01828989483310471, | |
| "grad_norm": 3.7560031414031982, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 5.3723, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.019204389574759947, | |
| "grad_norm": 3.2203760147094727, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.927, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.019204389574759947, | |
| "eval_loss": 1.309173583984375, | |
| "eval_runtime": 397.7424, | |
| "eval_samples_per_second": 13.891, | |
| "eval_steps_per_second": 1.737, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.020118884316415182, | |
| "grad_norm": 3.7798800468444824, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.891, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.021033379058070414, | |
| "grad_norm": 3.075923442840576, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 5.1185, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.02194787379972565, | |
| "grad_norm": 3.5462028980255127, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 5.1134, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02194787379972565, | |
| "eval_loss": 1.306206226348877, | |
| "eval_runtime": 397.4719, | |
| "eval_samples_per_second": 13.9, | |
| "eval_steps_per_second": 1.738, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.022862368541380886, | |
| "grad_norm": 3.332122564315796, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 5.6472, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02377686328303612, | |
| "grad_norm": 3.3746860027313232, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 5.0507, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "grad_norm": 3.771028757095337, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.741, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "eval_loss": 1.3051848411560059, | |
| "eval_runtime": 397.5922, | |
| "eval_samples_per_second": 13.896, | |
| "eval_steps_per_second": 1.738, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.025605852766346592, | |
| "grad_norm": 4.040452003479004, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 5.5279, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.026520347508001828, | |
| "grad_norm": 3.8290328979492188, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 5.4467, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.027434842249657063, | |
| "grad_norm": 4.0576395988464355, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 5.546, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.027434842249657063, | |
| "eval_loss": 1.3046977519989014, | |
| "eval_runtime": 397.5869, | |
| "eval_samples_per_second": 13.896, | |
| "eval_steps_per_second": 1.738, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0283493369913123, | |
| "grad_norm": 3.4416396617889404, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 5.3352, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.029263831732967534, | |
| "grad_norm": 3.388613700866699, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 5.5734, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.03017832647462277, | |
| "grad_norm": 3.199336290359497, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 5.2268, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.03017832647462277, | |
| "eval_loss": 1.3045598268508911, | |
| "eval_runtime": 397.9063, | |
| "eval_samples_per_second": 13.885, | |
| "eval_steps_per_second": 1.737, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3921980018956698e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |