| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.9696, | |
| "eval_steps": 500, | |
| "global_step": 3120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.851910588425601, | |
| "learning_rate": 1.967948717948718e-05, | |
| "loss": 0.5612, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.2315880528966252, | |
| "learning_rate": 1.935897435897436e-05, | |
| "loss": 0.3709, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5704120750666282, | |
| "learning_rate": 1.903846153846154e-05, | |
| "loss": 0.3587, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5819247571936106, | |
| "learning_rate": 1.8717948717948718e-05, | |
| "loss": 0.35, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5367665073602743, | |
| "learning_rate": 1.8397435897435897e-05, | |
| "loss": 0.3477, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.49331381086425624, | |
| "learning_rate": 1.807692307692308e-05, | |
| "loss": 0.3454, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1184, | |
| "grad_norm": 0.40621244969105647, | |
| "learning_rate": 1.775641025641026e-05, | |
| "loss": 0.3286, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2784, | |
| "grad_norm": 0.5399875305937165, | |
| "learning_rate": 1.7435897435897438e-05, | |
| "loss": 0.3302, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4384000000000001, | |
| "grad_norm": 0.4867667504051181, | |
| "learning_rate": 1.7115384615384617e-05, | |
| "loss": 0.329, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5984, | |
| "grad_norm": 0.39607945861138233, | |
| "learning_rate": 1.6794871794871796e-05, | |
| "loss": 0.3285, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.7584, | |
| "grad_norm": 0.5226369083622061, | |
| "learning_rate": 1.6474358974358975e-05, | |
| "loss": 0.3287, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.9184, | |
| "grad_norm": 0.45447653237240976, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 0.3294, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0768, | |
| "grad_norm": 0.47874441959819114, | |
| "learning_rate": 1.5833333333333333e-05, | |
| "loss": 0.3139, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.2368, | |
| "grad_norm": 0.41151764069492225, | |
| "learning_rate": 1.5512820512820516e-05, | |
| "loss": 0.3062, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3968, | |
| "grad_norm": 0.4463050953450543, | |
| "learning_rate": 1.5192307692307693e-05, | |
| "loss": 0.3092, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.5568, | |
| "grad_norm": 0.3643203177201396, | |
| "learning_rate": 1.4871794871794874e-05, | |
| "loss": 0.3111, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.7168, | |
| "grad_norm": 0.44551792133446877, | |
| "learning_rate": 1.4551282051282051e-05, | |
| "loss": 0.3082, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.8768000000000002, | |
| "grad_norm": 0.3543322141266429, | |
| "learning_rate": 1.4230769230769232e-05, | |
| "loss": 0.3093, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.0352, | |
| "grad_norm": 0.4113939169966545, | |
| "learning_rate": 1.3910256410256411e-05, | |
| "loss": 0.2976, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.1952, | |
| "grad_norm": 0.4409058514312091, | |
| "learning_rate": 1.3589743589743592e-05, | |
| "loss": 0.2713, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.3552, | |
| "grad_norm": 0.42568805167898643, | |
| "learning_rate": 1.3269230769230769e-05, | |
| "loss": 0.2747, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.5152, | |
| "grad_norm": 0.3827205184015781, | |
| "learning_rate": 1.294871794871795e-05, | |
| "loss": 0.278, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.6752000000000002, | |
| "grad_norm": 0.3979662194624538, | |
| "learning_rate": 1.2628205128205129e-05, | |
| "loss": 0.2777, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.8352, | |
| "grad_norm": 0.39243876049749526, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.2763, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.9952, | |
| "grad_norm": 0.361777833579101, | |
| "learning_rate": 1.1987179487179487e-05, | |
| "loss": 0.2776, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.1536, | |
| "grad_norm": 0.43490004305200247, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 0.2168, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.3136, | |
| "grad_norm": 0.42922627875160785, | |
| "learning_rate": 1.1346153846153847e-05, | |
| "loss": 0.2193, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.4736, | |
| "grad_norm": 0.40832706764608834, | |
| "learning_rate": 1.1025641025641028e-05, | |
| "loss": 0.2211, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.6336, | |
| "grad_norm": 0.4157572186874582, | |
| "learning_rate": 1.0705128205128205e-05, | |
| "loss": 0.2219, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.7936, | |
| "grad_norm": 0.4061532656958216, | |
| "learning_rate": 1.0384615384615386e-05, | |
| "loss": 0.2224, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.9536, | |
| "grad_norm": 0.39607581683245563, | |
| "learning_rate": 1.0064102564102565e-05, | |
| "loss": 0.2236, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 5.112, | |
| "grad_norm": 0.45204634084477424, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.1718, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.272, | |
| "grad_norm": 0.45038388201781476, | |
| "learning_rate": 9.423076923076923e-06, | |
| "loss": 0.1521, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.432, | |
| "grad_norm": 0.432268289063817, | |
| "learning_rate": 9.102564102564104e-06, | |
| "loss": 0.1545, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.592, | |
| "grad_norm": 0.45687320006605464, | |
| "learning_rate": 8.782051282051283e-06, | |
| "loss": 0.155, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.752, | |
| "grad_norm": 0.482007411435189, | |
| "learning_rate": 8.461538461538462e-06, | |
| "loss": 0.1555, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.912, | |
| "grad_norm": 0.43993192221405386, | |
| "learning_rate": 8.141025641025641e-06, | |
| "loss": 0.155, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 6.0704, | |
| "grad_norm": 0.4300425670111472, | |
| "learning_rate": 7.820512820512822e-06, | |
| "loss": 0.1285, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.2304, | |
| "grad_norm": 0.43977728574911473, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0959, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.3904, | |
| "grad_norm": 0.4306659398586491, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.0962, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.5504, | |
| "grad_norm": 0.4256013017505169, | |
| "learning_rate": 6.858974358974359e-06, | |
| "loss": 0.0971, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.7104, | |
| "grad_norm": 0.433814440985733, | |
| "learning_rate": 6.538461538461539e-06, | |
| "loss": 0.0972, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.8704, | |
| "grad_norm": 0.43711207631691046, | |
| "learning_rate": 6.217948717948718e-06, | |
| "loss": 0.0979, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 7.0288, | |
| "grad_norm": 0.4409772510395749, | |
| "learning_rate": 5.897435897435898e-06, | |
| "loss": 0.0901, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.1888, | |
| "grad_norm": 0.39891255071766135, | |
| "learning_rate": 5.576923076923077e-06, | |
| "loss": 0.0586, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.3488, | |
| "grad_norm": 0.3823770127267414, | |
| "learning_rate": 5.256410256410257e-06, | |
| "loss": 0.0591, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.5088, | |
| "grad_norm": 0.37627964770669303, | |
| "learning_rate": 4.935897435897436e-06, | |
| "loss": 0.0597, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.6688, | |
| "grad_norm": 0.38903226358175974, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.0596, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.8288, | |
| "grad_norm": 0.3718579723640309, | |
| "learning_rate": 4.294871794871795e-06, | |
| "loss": 0.0601, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.9888, | |
| "grad_norm": 0.3853191872797418, | |
| "learning_rate": 3.974358974358974e-06, | |
| "loss": 0.0591, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.1472, | |
| "grad_norm": 0.28699527941653097, | |
| "learning_rate": 3.653846153846154e-06, | |
| "loss": 0.0393, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.3072, | |
| "grad_norm": 0.2913275213251273, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0378, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.4672, | |
| "grad_norm": 0.29222648208895774, | |
| "learning_rate": 3.012820512820513e-06, | |
| "loss": 0.0378, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 8.6272, | |
| "grad_norm": 0.34018938277280725, | |
| "learning_rate": 2.6923076923076923e-06, | |
| "loss": 0.0377, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.7872, | |
| "grad_norm": 0.3093825001560387, | |
| "learning_rate": 2.371794871794872e-06, | |
| "loss": 0.0378, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 8.9472, | |
| "grad_norm": 0.2970931747170932, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 0.0374, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.1056, | |
| "grad_norm": 0.2213245615323131, | |
| "learning_rate": 1.7307692307692308e-06, | |
| "loss": 0.0301, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 9.2656, | |
| "grad_norm": 0.2390672228760369, | |
| "learning_rate": 1.4102564102564104e-06, | |
| "loss": 0.0267, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.4256, | |
| "grad_norm": 0.2243952551600555, | |
| "learning_rate": 1.0897435897435899e-06, | |
| "loss": 0.0267, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 9.5856, | |
| "grad_norm": 0.22046009291111926, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.0266, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.7456, | |
| "grad_norm": 0.224521914783935, | |
| "learning_rate": 4.4871794871794876e-07, | |
| "loss": 0.0265, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 9.9056, | |
| "grad_norm": 0.23483782299927702, | |
| "learning_rate": 1.282051282051282e-07, | |
| "loss": 0.0264, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.9696, | |
| "step": 3120, | |
| "total_flos": 2.62926380302336e+16, | |
| "train_loss": 0.18988347003857295, | |
| "train_runtime": 234399.9158, | |
| "train_samples_per_second": 3.412, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.62926380302336e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |