| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.917525773195877, | |
| "eval_steps": 30, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 1.6045022010803223, | |
| "learning_rate": 2.9375e-05, | |
| "loss": 10.1016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 1.7116177082061768, | |
| "learning_rate": 2.875e-05, | |
| "loss": 10.2357, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.4948453608247423, | |
| "grad_norm": 1.7441893815994263, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 10.0531, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.6597938144329897, | |
| "grad_norm": 2.1694791316986084, | |
| "learning_rate": 2.75e-05, | |
| "loss": 10.2016, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8247422680412371, | |
| "grad_norm": 2.1092519760131836, | |
| "learning_rate": 2.6875000000000003e-05, | |
| "loss": 10.0665, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "grad_norm": 2.16532301902771, | |
| "learning_rate": 2.625e-05, | |
| "loss": 9.8524, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.1546391752577319, | |
| "grad_norm": 2.3102128505706787, | |
| "learning_rate": 2.5625e-05, | |
| "loss": 9.9192, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.3195876288659794, | |
| "grad_norm": 2.627959966659546, | |
| "learning_rate": 2.5e-05, | |
| "loss": 9.8474, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.4845360824742269, | |
| "grad_norm": 3.065120220184326, | |
| "learning_rate": 2.4375e-05, | |
| "loss": 9.8735, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.6494845360824741, | |
| "grad_norm": 2.6623458862304688, | |
| "learning_rate": 2.3749999999999998e-05, | |
| "loss": 9.6075, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.8144329896907216, | |
| "grad_norm": 2.8638663291931152, | |
| "learning_rate": 2.3125000000000003e-05, | |
| "loss": 9.6006, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.9793814432989691, | |
| "grad_norm": 3.157634735107422, | |
| "learning_rate": 2.25e-05, | |
| "loss": 9.511, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.1443298969072164, | |
| "grad_norm": 3.286606788635254, | |
| "learning_rate": 2.1875e-05, | |
| "loss": 9.4582, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.3092783505154637, | |
| "grad_norm": 3.352224588394165, | |
| "learning_rate": 2.125e-05, | |
| "loss": 9.3268, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.4742268041237114, | |
| "grad_norm": 3.678314208984375, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 9.3043, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.6391752577319587, | |
| "grad_norm": 3.6437971591949463, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 9.1926, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.804123711340206, | |
| "grad_norm": 3.841749429702759, | |
| "learning_rate": 1.9375e-05, | |
| "loss": 9.0797, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.9690721649484537, | |
| "grad_norm": 3.7387771606445312, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 9.0155, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.134020618556701, | |
| "grad_norm": 3.5350756645202637, | |
| "learning_rate": 1.8125e-05, | |
| "loss": 8.8982, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.2989690721649483, | |
| "grad_norm": 3.9751718044281006, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 8.8391, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.463917525773196, | |
| "grad_norm": 3.5453691482543945, | |
| "learning_rate": 1.6875e-05, | |
| "loss": 8.8007, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.6288659793814433, | |
| "grad_norm": 3.744359254837036, | |
| "learning_rate": 1.625e-05, | |
| "loss": 8.6382, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.7938144329896906, | |
| "grad_norm": 4.300820350646973, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 8.5496, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.9587628865979383, | |
| "grad_norm": 3.571364164352417, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.4785, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.123711340206185, | |
| "grad_norm": 3.5201902389526367, | |
| "learning_rate": 1.4375e-05, | |
| "loss": 8.4425, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.288659793814433, | |
| "grad_norm": 3.3802483081817627, | |
| "learning_rate": 1.375e-05, | |
| "loss": 8.4132, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.453608247422681, | |
| "grad_norm": 3.3173177242279053, | |
| "learning_rate": 1.3125e-05, | |
| "loss": 8.3052, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.618556701030927, | |
| "grad_norm": 3.4614617824554443, | |
| "learning_rate": 1.25e-05, | |
| "loss": 8.1881, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.783505154639175, | |
| "grad_norm": 3.216718912124634, | |
| "learning_rate": 1.1874999999999999e-05, | |
| "loss": 8.2378, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "grad_norm": 3.36505126953125, | |
| "learning_rate": 1.125e-05, | |
| "loss": 8.1809, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "eval_audio_cosine_sim": 0.557277500629425, | |
| "eval_loss": 3.095508575439453, | |
| "eval_runtime": 2121.8578, | |
| "eval_samples_per_second": 0.006, | |
| "eval_steps_per_second": 0.006, | |
| "eval_text_cosine_sim": 0.3839966356754303, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.11340206185567, | |
| "grad_norm": 3.3731958866119385, | |
| "learning_rate": 1.0625e-05, | |
| "loss": 8.1047, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.278350515463917, | |
| "grad_norm": 3.572460651397705, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 7.9906, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.443298969072165, | |
| "grad_norm": 3.343137502670288, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 8.1973, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.608247422680412, | |
| "grad_norm": 3.5458569526672363, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 8.0155, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.77319587628866, | |
| "grad_norm": 3.403402328491211, | |
| "learning_rate": 8.125e-06, | |
| "loss": 8.0047, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.938144329896907, | |
| "grad_norm": 3.5528311729431152, | |
| "learning_rate": 7.5e-06, | |
| "loss": 7.8875, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.103092783505154, | |
| "grad_norm": 3.659574270248413, | |
| "learning_rate": 6.875e-06, | |
| "loss": 7.82, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.268041237113402, | |
| "grad_norm": 3.419759511947632, | |
| "learning_rate": 6.25e-06, | |
| "loss": 7.8027, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.43298969072165, | |
| "grad_norm": 3.7508034706115723, | |
| "learning_rate": 5.625e-06, | |
| "loss": 7.8398, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.597938144329897, | |
| "grad_norm": 3.735914945602417, | |
| "learning_rate": 4.9999999999999996e-06, | |
| "loss": 7.8924, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.762886597938144, | |
| "grad_norm": 3.593177318572998, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 7.9886, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 6.927835051546392, | |
| "grad_norm": 3.5156137943267822, | |
| "learning_rate": 3.75e-06, | |
| "loss": 7.8703, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.092783505154639, | |
| "grad_norm": 3.6164469718933105, | |
| "learning_rate": 3.125e-06, | |
| "loss": 7.7555, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 7.257731958762887, | |
| "grad_norm": 3.5708608627319336, | |
| "learning_rate": 2.4999999999999998e-06, | |
| "loss": 7.7007, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 7.422680412371134, | |
| "grad_norm": 3.396042585372925, | |
| "learning_rate": 1.875e-06, | |
| "loss": 7.9848, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.587628865979381, | |
| "grad_norm": 3.5203356742858887, | |
| "learning_rate": 1.2499999999999999e-06, | |
| "loss": 7.85, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 7.752577319587629, | |
| "grad_norm": 3.814443349838257, | |
| "learning_rate": 6.249999999999999e-07, | |
| "loss": 7.566, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.917525773195877, | |
| "grad_norm": 3.4914655685424805, | |
| "learning_rate": 0.0, | |
| "loss": 7.8495, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 7.917525773195877, | |
| "step": 48, | |
| "total_flos": 962957869218000.0, | |
| "train_loss": 8.715420136849085, | |
| "train_runtime": 7523.7142, | |
| "train_samples_per_second": 0.103, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 962957869218000.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |