| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.999939507591797, | |
| "global_step": 82650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5e-05, | |
| "loss": 9.0179, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 8.008028984069824, | |
| "eval_runtime": 354.8383, | |
| "eval_samples_per_second": 41.117, | |
| "eval_steps_per_second": 5.14, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0001, | |
| "loss": 7.589, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 7.217428684234619, | |
| "eval_runtime": 357.6188, | |
| "eval_samples_per_second": 40.798, | |
| "eval_steps_per_second": 5.1, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.876007439553628e-05, | |
| "loss": 7.0805, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 6.8754563331604, | |
| "eval_runtime": 339.2212, | |
| "eval_samples_per_second": 43.01, | |
| "eval_steps_per_second": 5.377, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.752014879107254e-05, | |
| "loss": 6.8341, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 6.731923580169678, | |
| "eval_runtime": 339.5448, | |
| "eval_samples_per_second": 42.969, | |
| "eval_steps_per_second": 5.372, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 9.628022318660881e-05, | |
| "loss": 6.5597, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 6.266661643981934, | |
| "eval_runtime": 339.3392, | |
| "eval_samples_per_second": 42.995, | |
| "eval_steps_per_second": 5.375, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 9.504029758214508e-05, | |
| "loss": 6.1904, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 6.067315101623535, | |
| "eval_runtime": 354.6519, | |
| "eval_samples_per_second": 41.139, | |
| "eval_steps_per_second": 5.143, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 9.380037197768134e-05, | |
| "loss": 6.05, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 5.955410957336426, | |
| "eval_runtime": 357.6298, | |
| "eval_samples_per_second": 40.796, | |
| "eval_steps_per_second": 5.1, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.132176069435834e-05, | |
| "loss": 5.8802, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 5.814595699310303, | |
| "eval_runtime": 333.2969, | |
| "eval_samples_per_second": 43.775, | |
| "eval_steps_per_second": 5.473, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 8.76032238065716e-05, | |
| "loss": 5.7503, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 5.659432888031006, | |
| "eval_runtime": 337.2485, | |
| "eval_samples_per_second": 43.262, | |
| "eval_steps_per_second": 5.408, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 8.388468691878487e-05, | |
| "loss": 5.6093, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 5.4909515380859375, | |
| "eval_runtime": 333.385, | |
| "eval_samples_per_second": 43.763, | |
| "eval_steps_per_second": 5.471, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.016615003099814e-05, | |
| "loss": 5.3841, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 5.236297130584717, | |
| "eval_runtime": 337.7351, | |
| "eval_samples_per_second": 43.2, | |
| "eval_steps_per_second": 5.401, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 7.644761314321141e-05, | |
| "loss": 5.098, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 4.817739486694336, | |
| "eval_runtime": 337.6417, | |
| "eval_samples_per_second": 43.211, | |
| "eval_steps_per_second": 5.402, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 7.272907625542468e-05, | |
| "loss": 4.71, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 4.514278411865234, | |
| "eval_runtime": 333.4053, | |
| "eval_samples_per_second": 43.761, | |
| "eval_steps_per_second": 5.471, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 6.901177929324241e-05, | |
| "loss": 4.4553, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "eval_loss": 4.293982028961182, | |
| "eval_runtime": 337.4791, | |
| "eval_samples_per_second": 43.232, | |
| "eval_steps_per_second": 5.405, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 6.529324240545568e-05, | |
| "loss": 4.2619, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "eval_loss": 4.119041442871094, | |
| "eval_runtime": 333.3435, | |
| "eval_samples_per_second": 43.769, | |
| "eval_steps_per_second": 5.472, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 6.157470551766893e-05, | |
| "loss": 4.0868, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_loss": 3.902297258377075, | |
| "eval_runtime": 337.8221, | |
| "eval_samples_per_second": 43.188, | |
| "eval_steps_per_second": 5.399, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 5.785740855548667e-05, | |
| "loss": 3.8931, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_loss": 3.7250819206237793, | |
| "eval_runtime": 337.5683, | |
| "eval_samples_per_second": 43.221, | |
| "eval_steps_per_second": 5.403, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 5.413887166769994e-05, | |
| "loss": 3.7433, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 3.6013708114624023, | |
| "eval_runtime": 333.4516, | |
| "eval_samples_per_second": 43.754, | |
| "eval_steps_per_second": 5.47, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 5.0420334779913205e-05, | |
| "loss": 3.6398, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "eval_loss": 3.514371633529663, | |
| "eval_runtime": 337.9252, | |
| "eval_samples_per_second": 43.175, | |
| "eval_steps_per_second": 5.398, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 4.670179789212647e-05, | |
| "loss": 3.5446, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_loss": 3.439208984375, | |
| "eval_runtime": 333.2805, | |
| "eval_samples_per_second": 43.777, | |
| "eval_steps_per_second": 5.473, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 4.298326100433974e-05, | |
| "loss": 3.4674, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "eval_loss": 3.360422372817993, | |
| "eval_runtime": 337.4986, | |
| "eval_samples_per_second": 43.23, | |
| "eval_steps_per_second": 5.404, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 3.926472411655301e-05, | |
| "loss": 3.3896, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "eval_loss": 3.3044984340667725, | |
| "eval_runtime": 337.3599, | |
| "eval_samples_per_second": 43.248, | |
| "eval_steps_per_second": 5.407, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 3.554618722876628e-05, | |
| "loss": 3.3327, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "eval_loss": 3.2497129440307617, | |
| "eval_runtime": 333.3738, | |
| "eval_samples_per_second": 43.765, | |
| "eval_steps_per_second": 5.471, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 3.1828890266584005e-05, | |
| "loss": 3.2596, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 3.208232879638672, | |
| "eval_runtime": 333.2588, | |
| "eval_samples_per_second": 43.78, | |
| "eval_steps_per_second": 5.473, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 2.811035337879727e-05, | |
| "loss": 3.2211, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "eval_loss": 3.1547913551330566, | |
| "eval_runtime": 337.1177, | |
| "eval_samples_per_second": 43.279, | |
| "eval_steps_per_second": 5.411, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 2.4391816491010542e-05, | |
| "loss": 3.1913, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "eval_loss": 3.1304616928100586, | |
| "eval_runtime": 337.6326, | |
| "eval_samples_per_second": 43.213, | |
| "eval_steps_per_second": 5.402, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 2.0673279603223807e-05, | |
| "loss": 3.1425, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_loss": 3.0831258296966553, | |
| "eval_runtime": 333.274, | |
| "eval_samples_per_second": 43.778, | |
| "eval_steps_per_second": 5.473, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 1.6955982641041537e-05, | |
| "loss": 3.1002, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "eval_loss": 3.049844741821289, | |
| "eval_runtime": 337.2388, | |
| "eval_samples_per_second": 43.263, | |
| "eval_steps_per_second": 5.409, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 1.3237445753254804e-05, | |
| "loss": 3.0675, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "eval_loss": 3.028550624847412, | |
| "eval_runtime": 333.2341, | |
| "eval_samples_per_second": 43.783, | |
| "eval_steps_per_second": 5.474, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 9.517668939863609e-06, | |
| "loss": 3.0515, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_loss": 2.9997730255126953, | |
| "eval_runtime": 337.3815, | |
| "eval_samples_per_second": 43.245, | |
| "eval_steps_per_second": 5.406, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 5.80037197768134e-06, | |
| "loss": 3.021, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "eval_loss": 2.982238292694092, | |
| "eval_runtime": 337.6502, | |
| "eval_samples_per_second": 43.21, | |
| "eval_steps_per_second": 5.402, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 2.0805951642901425e-06, | |
| "loss": 2.996, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_loss": 2.9674391746520996, | |
| "eval_runtime": 333.3185, | |
| "eval_samples_per_second": 43.772, | |
| "eval_steps_per_second": 5.472, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 82650, | |
| "total_flos": 2.2475936453492736e+17, | |
| "train_loss": 0.11406417782062916, | |
| "train_runtime": 1932.5263, | |
| "train_samples_per_second": 684.301, | |
| "train_steps_per_second": 42.768 | |
| } | |
| ], | |
| "max_steps": 82650, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.2475936453492736e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |