| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.33755274261603374, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0033755274261603376, | |
| "eval_loss": 1.6038267612457275, | |
| "eval_runtime": 18.5194, | |
| "eval_samples_per_second": 26.945, | |
| "eval_steps_per_second": 3.402, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010126582278481013, | |
| "grad_norm": 2.1169235706329346, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.6178, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020253164556962026, | |
| "grad_norm": 2.1117138862609863, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5978, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030379746835443037, | |
| "grad_norm": 1.5161991119384766, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4374, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030379746835443037, | |
| "eval_loss": 1.2946577072143555, | |
| "eval_runtime": 18.883, | |
| "eval_samples_per_second": 26.426, | |
| "eval_steps_per_second": 3.336, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04050632911392405, | |
| "grad_norm": 0.7761142253875732, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.2455, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05063291139240506, | |
| "grad_norm": 0.92899090051651, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.1969, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.060759493670886074, | |
| "grad_norm": 0.8056496977806091, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.1269, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.060759493670886074, | |
| "eval_loss": 1.0773837566375732, | |
| "eval_runtime": 18.9883, | |
| "eval_samples_per_second": 26.279, | |
| "eval_steps_per_second": 3.318, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07088607594936709, | |
| "grad_norm": 0.5774959921836853, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.09, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0810126582278481, | |
| "grad_norm": 0.491992712020874, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.0201, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09113924050632911, | |
| "grad_norm": 0.5166187882423401, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.0201, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09113924050632911, | |
| "eval_loss": 0.9979392290115356, | |
| "eval_runtime": 18.9935, | |
| "eval_samples_per_second": 26.272, | |
| "eval_steps_per_second": 3.317, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10126582278481013, | |
| "grad_norm": 0.548265278339386, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.0072, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11139240506329114, | |
| "grad_norm": 0.5094746947288513, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.0009, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12151898734177215, | |
| "grad_norm": 0.5233523845672607, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.9853, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12151898734177215, | |
| "eval_loss": 0.9588655829429626, | |
| "eval_runtime": 19.0068, | |
| "eval_samples_per_second": 26.254, | |
| "eval_steps_per_second": 3.315, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13164556962025317, | |
| "grad_norm": 0.4625486433506012, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.9428, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14177215189873418, | |
| "grad_norm": 0.44984695315361023, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.9767, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1518987341772152, | |
| "grad_norm": 0.4471226632595062, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.9393, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1518987341772152, | |
| "eval_loss": 0.9369592666625977, | |
| "eval_runtime": 19.0325, | |
| "eval_samples_per_second": 26.218, | |
| "eval_steps_per_second": 3.31, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1620253164556962, | |
| "grad_norm": 0.49932029843330383, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.9306, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.17215189873417722, | |
| "grad_norm": 0.4568023979663849, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.9029, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.18227848101265823, | |
| "grad_norm": 0.482744425535202, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.9557, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18227848101265823, | |
| "eval_loss": 0.9249128103256226, | |
| "eval_runtime": 19.0163, | |
| "eval_samples_per_second": 26.241, | |
| "eval_steps_per_second": 3.313, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.19240506329113924, | |
| "grad_norm": 0.4680204689502716, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.9541, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.20253164556962025, | |
| "grad_norm": 0.4750822186470032, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.9059, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21265822784810126, | |
| "grad_norm": 0.5030151009559631, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.9317, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21265822784810126, | |
| "eval_loss": 0.9165537357330322, | |
| "eval_runtime": 19.019, | |
| "eval_samples_per_second": 26.237, | |
| "eval_steps_per_second": 3.312, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.22278481012658227, | |
| "grad_norm": 0.4766693711280823, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.924, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.23291139240506328, | |
| "grad_norm": 0.5020814538002014, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.9456, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2430379746835443, | |
| "grad_norm": 0.48271435499191284, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.947, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2430379746835443, | |
| "eval_loss": 0.9115452766418457, | |
| "eval_runtime": 19.0108, | |
| "eval_samples_per_second": 26.248, | |
| "eval_steps_per_second": 3.314, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.25316455696202533, | |
| "grad_norm": 0.49486243724823, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.9081, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.26329113924050634, | |
| "grad_norm": 0.47872477769851685, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.9158, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.27341772151898736, | |
| "grad_norm": 0.45988014340400696, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.9381, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.27341772151898736, | |
| "eval_loss": 0.9089102149009705, | |
| "eval_runtime": 19.0042, | |
| "eval_samples_per_second": 26.257, | |
| "eval_steps_per_second": 3.315, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.28354430379746837, | |
| "grad_norm": 0.4782557189464569, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.9225, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2936708860759494, | |
| "grad_norm": 0.4685029685497284, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.8903, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.3037974683544304, | |
| "grad_norm": 0.48110467195510864, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.9264, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3037974683544304, | |
| "eval_loss": 0.9077513217926025, | |
| "eval_runtime": 19.0064, | |
| "eval_samples_per_second": 26.254, | |
| "eval_steps_per_second": 3.315, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3139240506329114, | |
| "grad_norm": 0.5063574910163879, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.9341, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3240506329113924, | |
| "grad_norm": 0.5072469711303711, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.9095, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3341772151898734, | |
| "grad_norm": 0.48717790842056274, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.9301, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3341772151898734, | |
| "eval_loss": 0.9075644016265869, | |
| "eval_runtime": 18.987, | |
| "eval_samples_per_second": 26.281, | |
| "eval_steps_per_second": 3.318, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.308936012752486e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |