| { | |
| "best_metric": 0.049842845648527145, | |
| "best_model_checkpoint": "./trained_models/Llama-3.1-8B-Instruct-CNADFTD-ADNI2NIFD-AN-fold-0-gathered-equally-represented-0234-v2/checkpoint-65", | |
| "epoch": 9.405063291139241, | |
| "eval_steps": 5, | |
| "global_step": 85, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20253164556962025, | |
| "grad_norm": 20.267499923706055, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 1.7395, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.4050632911392405, | |
| "grad_norm": 19.267681121826172, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 1.5413, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5063291139240507, | |
| "eval_loss": 0.1354750096797943, | |
| "eval_runtime": 10.1119, | |
| "eval_samples_per_second": 8.9, | |
| "eval_steps_per_second": 2.275, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.6075949367088608, | |
| "grad_norm": 9.370927810668945, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 0.928, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.810126582278481, | |
| "grad_norm": 14.292084693908691, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.0332, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1012658227848102, | |
| "grad_norm": 16.358558654785156, | |
| "learning_rate": 0.00015, | |
| "loss": 0.7069, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1012658227848102, | |
| "eval_loss": 0.09530380368232727, | |
| "eval_runtime": 10.229, | |
| "eval_samples_per_second": 8.799, | |
| "eval_steps_per_second": 2.249, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.3037974683544304, | |
| "grad_norm": 4.635617733001709, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.7555, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.5063291139240507, | |
| "grad_norm": 5.011940002441406, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.4148, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.6075949367088609, | |
| "eval_loss": 0.07413829118013382, | |
| "eval_runtime": 10.2852, | |
| "eval_samples_per_second": 8.75, | |
| "eval_steps_per_second": 2.236, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.7088607594936709, | |
| "grad_norm": 3.0944323539733887, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.4899, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.9113924050632911, | |
| "grad_norm": 9.65760612487793, | |
| "learning_rate": 0.00027, | |
| "loss": 0.7349, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.2025316455696204, | |
| "grad_norm": 6.6734418869018555, | |
| "learning_rate": 0.0003, | |
| "loss": 0.7657, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.2025316455696204, | |
| "eval_loss": 0.06096666678786278, | |
| "eval_runtime": 10.2992, | |
| "eval_samples_per_second": 8.739, | |
| "eval_steps_per_second": 2.233, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.4050632911392404, | |
| "grad_norm": 4.764143943786621, | |
| "learning_rate": 0.00029939614409928584, | |
| "loss": 0.5037, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.607594936708861, | |
| "grad_norm": 2.2680883407592773, | |
| "learning_rate": 0.00029758943828979444, | |
| "loss": 0.4673, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.708860759493671, | |
| "eval_loss": 0.05865984410047531, | |
| "eval_runtime": 10.3046, | |
| "eval_samples_per_second": 8.734, | |
| "eval_steps_per_second": 2.232, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.810126582278481, | |
| "grad_norm": 4.016191005706787, | |
| "learning_rate": 0.00029459442910437797, | |
| "loss": 0.4189, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.1012658227848102, | |
| "grad_norm": 6.397407531738281, | |
| "learning_rate": 0.00029043523059596053, | |
| "loss": 0.6435, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.3037974683544302, | |
| "grad_norm": 5.727280139923096, | |
| "learning_rate": 0.0002851453301853628, | |
| "loss": 0.3936, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.3037974683544302, | |
| "eval_loss": 0.06180719658732414, | |
| "eval_runtime": 10.3031, | |
| "eval_samples_per_second": 8.735, | |
| "eval_steps_per_second": 2.232, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.5063291139240507, | |
| "grad_norm": 4.997829437255859, | |
| "learning_rate": 0.0002787673190402799, | |
| "loss": 0.3523, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.708860759493671, | |
| "grad_norm": 4.903127670288086, | |
| "learning_rate": 0.0002713525491562421, | |
| "loss": 0.5887, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.810126582278481, | |
| "eval_loss": 0.05977346748113632, | |
| "eval_runtime": 10.2946, | |
| "eval_samples_per_second": 8.742, | |
| "eval_steps_per_second": 2.234, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.911392405063291, | |
| "grad_norm": 6.872932434082031, | |
| "learning_rate": 0.00026296071990054165, | |
| "loss": 0.4239, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.2025316455696204, | |
| "grad_norm": 3.7034521102905273, | |
| "learning_rate": 0.0002536593973480297, | |
| "loss": 0.437, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.405063291139241, | |
| "grad_norm": 4.1444244384765625, | |
| "learning_rate": 0.00024352347027881003, | |
| "loss": 0.2975, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.405063291139241, | |
| "eval_loss": 0.06828329712152481, | |
| "eval_runtime": 10.2997, | |
| "eval_samples_per_second": 8.738, | |
| "eval_steps_per_second": 2.233, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.6075949367088604, | |
| "grad_norm": 4.268646717071533, | |
| "learning_rate": 0.00023263454721781537, | |
| "loss": 0.2971, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.810126582278481, | |
| "grad_norm": 9.77409839630127, | |
| "learning_rate": 0.0002210802993709498, | |
| "loss": 0.3884, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.911392405063291, | |
| "eval_loss": 0.0632724016904831, | |
| "eval_runtime": 10.2951, | |
| "eval_samples_per_second": 8.742, | |
| "eval_steps_per_second": 2.234, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.10126582278481, | |
| "grad_norm": 12.521153450012207, | |
| "learning_rate": 0.00020895375474808852, | |
| "loss": 0.634, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.30379746835443, | |
| "grad_norm": 7.4212541580200195, | |
| "learning_rate": 0.0001963525491562421, | |
| "loss": 0.3246, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.506329113924051, | |
| "grad_norm": 10.269664764404297, | |
| "learning_rate": 0.00018337814009344714, | |
| "loss": 0.4664, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.506329113924051, | |
| "eval_loss": 0.09271334856748581, | |
| "eval_runtime": 10.2899, | |
| "eval_samples_per_second": 8.746, | |
| "eval_steps_per_second": 2.235, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.708860759493671, | |
| "grad_norm": 6.81392765045166, | |
| "learning_rate": 0.00017013498987264832, | |
| "loss": 0.3999, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.911392405063291, | |
| "grad_norm": 3.8285505771636963, | |
| "learning_rate": 0.00015672972455257723, | |
| "loss": 0.2666, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.10126582278481, | |
| "eval_loss": 0.05789216235280037, | |
| "eval_runtime": 10.2896, | |
| "eval_samples_per_second": 8.747, | |
| "eval_steps_per_second": 2.235, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.2025316455696204, | |
| "grad_norm": 3.7714805603027344, | |
| "learning_rate": 0.0001432702754474228, | |
| "loss": 0.3953, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.405063291139241, | |
| "grad_norm": 2.9867966175079346, | |
| "learning_rate": 0.0001298650101273517, | |
| "loss": 0.227, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.6075949367088604, | |
| "grad_norm": 4.1205010414123535, | |
| "learning_rate": 0.00011662185990655284, | |
| "loss": 0.2554, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.6075949367088604, | |
| "eval_loss": 0.061419058591127396, | |
| "eval_runtime": 10.27, | |
| "eval_samples_per_second": 8.763, | |
| "eval_steps_per_second": 2.24, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.810126582278481, | |
| "grad_norm": 2.6789913177490234, | |
| "learning_rate": 0.0001036474508437579, | |
| "loss": 0.2873, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.10126582278481, | |
| "grad_norm": 4.4857683181762695, | |
| "learning_rate": 9.104624525191145e-05, | |
| "loss": 0.2828, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.2025316455696204, | |
| "eval_loss": 0.049842845648527145, | |
| "eval_runtime": 10.2776, | |
| "eval_samples_per_second": 8.757, | |
| "eval_steps_per_second": 2.238, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.30379746835443, | |
| "grad_norm": 1.8388667106628418, | |
| "learning_rate": 7.89197006290502e-05, | |
| "loss": 0.1712, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.506329113924051, | |
| "grad_norm": 3.90205454826355, | |
| "learning_rate": 6.736545278218463e-05, | |
| "loss": 0.1491, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 7.708860759493671, | |
| "grad_norm": 4.954835891723633, | |
| "learning_rate": 5.6476529721189974e-05, | |
| "loss": 0.233, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.708860759493671, | |
| "eval_loss": 0.05690564960241318, | |
| "eval_runtime": 10.2862, | |
| "eval_samples_per_second": 8.75, | |
| "eval_steps_per_second": 2.236, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.911392405063291, | |
| "grad_norm": 4.313951015472412, | |
| "learning_rate": 4.63406026519703e-05, | |
| "loss": 0.182, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.20253164556962, | |
| "grad_norm": 3.9724860191345215, | |
| "learning_rate": 3.7039280099458366e-05, | |
| "loss": 0.1574, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.30379746835443, | |
| "eval_loss": 0.06441806256771088, | |
| "eval_runtime": 10.2783, | |
| "eval_samples_per_second": 8.756, | |
| "eval_steps_per_second": 2.238, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.405063291139241, | |
| "grad_norm": 7.101994037628174, | |
| "learning_rate": 2.8647450843757897e-05, | |
| "loss": 0.2221, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.60759493670886, | |
| "grad_norm": 4.0686845779418945, | |
| "learning_rate": 2.1232680959720082e-05, | |
| "loss": 0.1784, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 8.810126582278482, | |
| "grad_norm": 3.0517356395721436, | |
| "learning_rate": 1.4854669814637143e-05, | |
| "loss": 0.1949, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.810126582278482, | |
| "eval_loss": 0.05705536901950836, | |
| "eval_runtime": 10.2777, | |
| "eval_samples_per_second": 8.757, | |
| "eval_steps_per_second": 2.238, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.10126582278481, | |
| "grad_norm": 5.039515018463135, | |
| "learning_rate": 9.564769404039419e-06, | |
| "loss": 0.1881, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.30379746835443, | |
| "grad_norm": 1.9921464920043945, | |
| "learning_rate": 5.405570895622013e-06, | |
| "loss": 0.0491, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 9.405063291139241, | |
| "eval_loss": 0.057273320853710175, | |
| "eval_runtime": 10.276, | |
| "eval_samples_per_second": 8.758, | |
| "eval_steps_per_second": 2.238, | |
| "step": 85 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.967898680180736e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |