| { | |
| "best_metric": 0.20606324076652527, | |
| "best_model_checkpoint": "/scratch/skscla001/speech/results/mms-1b-swagen-female-model/checkpoint-2100", | |
| "epoch": 2.9411764705882355, | |
| "eval_steps": 100, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 42.9699821472168, | |
| "learning_rate": 0.00029099999999999997, | |
| "loss": 8.5907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11764705882352941, | |
| "eval_loss": 4.55902099609375, | |
| "eval_runtime": 32.5172, | |
| "eval_samples_per_second": 17.744, | |
| "eval_steps_per_second": 4.459, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 23.00685691833496, | |
| "learning_rate": 0.00028812499999999997, | |
| "loss": 4.3006, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "eval_loss": 4.2573113441467285, | |
| "eval_runtime": 32.1998, | |
| "eval_samples_per_second": 17.919, | |
| "eval_steps_per_second": 4.503, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 7.895191192626953, | |
| "learning_rate": 0.00027575, | |
| "loss": 4.1906, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "eval_loss": 3.843606948852539, | |
| "eval_runtime": 32.1765, | |
| "eval_samples_per_second": 17.932, | |
| "eval_steps_per_second": 4.506, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 1.9934626817703247, | |
| "learning_rate": 0.00026325, | |
| "loss": 1.1198, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "eval_loss": 0.25689366459846497, | |
| "eval_runtime": 32.1, | |
| "eval_samples_per_second": 17.975, | |
| "eval_steps_per_second": 4.517, | |
| "eval_wer": 0.19961427193828352, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 1.5836255550384521, | |
| "learning_rate": 0.00025075, | |
| "loss": 0.3012, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "eval_loss": 0.2465960681438446, | |
| "eval_runtime": 32.4617, | |
| "eval_samples_per_second": 17.775, | |
| "eval_steps_per_second": 4.467, | |
| "eval_wer": 0.19421407907425264, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 1.9752272367477417, | |
| "learning_rate": 0.00023825, | |
| "loss": 0.2919, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "eval_loss": 0.25010332465171814, | |
| "eval_runtime": 32.532, | |
| "eval_samples_per_second": 17.736, | |
| "eval_steps_per_second": 4.457, | |
| "eval_wer": 0.19903567984570877, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 4.631561279296875, | |
| "learning_rate": 0.00022574999999999996, | |
| "loss": 0.267, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "eval_loss": 0.2353355437517166, | |
| "eval_runtime": 32.2905, | |
| "eval_samples_per_second": 17.869, | |
| "eval_steps_per_second": 4.49, | |
| "eval_wer": 0.18939247830279654, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 3.183828830718994, | |
| "learning_rate": 0.00021324999999999998, | |
| "loss": 0.2666, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 0.22800631821155548, | |
| "eval_runtime": 32.2537, | |
| "eval_samples_per_second": 17.889, | |
| "eval_steps_per_second": 4.496, | |
| "eval_wer": 0.19035679845708775, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 1.4982892274856567, | |
| "learning_rate": 0.00020075, | |
| "loss": 0.2396, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "eval_loss": 0.22375665605068207, | |
| "eval_runtime": 32.5796, | |
| "eval_samples_per_second": 17.71, | |
| "eval_steps_per_second": 4.451, | |
| "eval_wer": 0.1891996142719383, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.2470210790634155, | |
| "learning_rate": 0.00018824999999999997, | |
| "loss": 0.258, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "eval_loss": 0.22019432485103607, | |
| "eval_runtime": 32.5522, | |
| "eval_samples_per_second": 17.725, | |
| "eval_steps_per_second": 4.454, | |
| "eval_wer": 0.189778206364513, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 2.3243045806884766, | |
| "learning_rate": 0.00017575, | |
| "loss": 0.2192, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "eval_loss": 0.21961148083209991, | |
| "eval_runtime": 32.5996, | |
| "eval_samples_per_second": 17.7, | |
| "eval_steps_per_second": 4.448, | |
| "eval_wer": 0.19382835101253615, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 2.2011804580688477, | |
| "learning_rate": 0.00016324999999999998, | |
| "loss": 0.2353, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "eval_loss": 0.2168532758951187, | |
| "eval_runtime": 32.3432, | |
| "eval_samples_per_second": 17.84, | |
| "eval_steps_per_second": 4.483, | |
| "eval_wer": 0.18746383799421407, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 2.5243468284606934, | |
| "learning_rate": 0.00015074999999999998, | |
| "loss": 0.2398, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "eval_loss": 0.2163960486650467, | |
| "eval_runtime": 32.6916, | |
| "eval_samples_per_second": 17.65, | |
| "eval_steps_per_second": 4.435, | |
| "eval_wer": 0.18958534233365476, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 2.1168859004974365, | |
| "learning_rate": 0.00013824999999999997, | |
| "loss": 0.2419, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "eval_loss": 0.21455629169940948, | |
| "eval_runtime": 32.7536, | |
| "eval_samples_per_second": 17.616, | |
| "eval_steps_per_second": 4.427, | |
| "eval_wer": 0.191321118611379, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 2.818471670150757, | |
| "learning_rate": 0.00012575, | |
| "loss": 0.2582, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "eval_loss": 0.21221759915351868, | |
| "eval_runtime": 32.4823, | |
| "eval_samples_per_second": 17.764, | |
| "eval_steps_per_second": 4.464, | |
| "eval_wer": 0.190549662487946, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.6471282243728638, | |
| "learning_rate": 0.00011324999999999999, | |
| "loss": 0.2417, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "eval_loss": 0.2105189561843872, | |
| "eval_runtime": 32.2537, | |
| "eval_samples_per_second": 17.889, | |
| "eval_steps_per_second": 4.496, | |
| "eval_wer": 0.18611378977820636, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.615028977394104, | |
| "learning_rate": 0.00010074999999999998, | |
| "loss": 0.2375, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.20913735032081604, | |
| "eval_runtime": 32.7492, | |
| "eval_samples_per_second": 17.619, | |
| "eval_steps_per_second": 4.428, | |
| "eval_wer": 0.1891996142719383, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 2.387188673019409, | |
| "learning_rate": 8.825e-05, | |
| "loss": 0.2353, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "eval_loss": 0.20865142345428467, | |
| "eval_runtime": 32.6893, | |
| "eval_samples_per_second": 17.651, | |
| "eval_steps_per_second": 4.436, | |
| "eval_wer": 0.18823529411764706, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 7.139828681945801, | |
| "learning_rate": 7.575e-05, | |
| "loss": 0.23, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "eval_loss": 0.20906075835227966, | |
| "eval_runtime": 32.4818, | |
| "eval_samples_per_second": 17.764, | |
| "eval_steps_per_second": 4.464, | |
| "eval_wer": 0.19035679845708775, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.5875645875930786, | |
| "learning_rate": 6.324999999999999e-05, | |
| "loss": 0.2378, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "eval_loss": 0.20674680173397064, | |
| "eval_runtime": 32.5201, | |
| "eval_samples_per_second": 17.743, | |
| "eval_steps_per_second": 4.459, | |
| "eval_wer": 0.189778206364513, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 3.247025966644287, | |
| "learning_rate": 5.0749999999999994e-05, | |
| "loss": 0.2343, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "eval_loss": 0.20606324076652527, | |
| "eval_runtime": 32.7292, | |
| "eval_samples_per_second": 17.63, | |
| "eval_steps_per_second": 4.43, | |
| "eval_wer": 0.18900675024108005, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 15.340932846069336, | |
| "learning_rate": 3.8249999999999995e-05, | |
| "loss": 0.2083, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "eval_loss": 0.2084166258573532, | |
| "eval_runtime": 32.8037, | |
| "eval_samples_per_second": 17.59, | |
| "eval_steps_per_second": 4.42, | |
| "eval_wer": 0.1884281581485053, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 1.6703847646713257, | |
| "learning_rate": 2.5749999999999996e-05, | |
| "loss": 0.2058, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "eval_loss": 0.20847396552562714, | |
| "eval_runtime": 32.6507, | |
| "eval_samples_per_second": 17.672, | |
| "eval_steps_per_second": 4.441, | |
| "eval_wer": 0.18823529411764706, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.9312336444854736, | |
| "learning_rate": 1.3249999999999999e-05, | |
| "loss": 0.2197, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "eval_loss": 0.2061544954776764, | |
| "eval_runtime": 32.4321, | |
| "eval_samples_per_second": 17.791, | |
| "eval_steps_per_second": 4.471, | |
| "eval_wer": 0.19035679845708775, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 1.8859847784042358, | |
| "learning_rate": 7.499999999999999e-07, | |
| "loss": 0.2317, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 0.2064579278230667, | |
| "eval_runtime": 33.0285, | |
| "eval_samples_per_second": 17.47, | |
| "eval_steps_per_second": 4.39, | |
| "eval_wer": 0.18958534233365476, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "step": 2500, | |
| "total_flos": 5.091528497919981e+18, | |
| "train_loss": 0.9320957969665528, | |
| "train_runtime": 2465.2512, | |
| "train_samples_per_second": 4.056, | |
| "train_steps_per_second": 1.014 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 4, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 4 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.091528497919981e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |