| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 35, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 20.565131124519954, | |
| "learning_rate": 0.0, | |
| "loss": 1.6904, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 20.126905785368916, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6039, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 11.090236436649034, | |
| "learning_rate": 9.978670881475173e-06, | |
| "loss": 1.5826, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 5.614097530062799, | |
| "learning_rate": 9.91486549841951e-06, | |
| "loss": 1.4068, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 5.234708126178992, | |
| "learning_rate": 9.809128215864096e-06, | |
| "loss": 1.4604, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 4.306326905631283, | |
| "learning_rate": 9.66236114702178e-06, | |
| "loss": 1.3071, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.8203719300545993, | |
| "learning_rate": 9.475816456775313e-06, | |
| "loss": 1.2656, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 3.336385963547391, | |
| "learning_rate": 9.251085678648072e-06, | |
| "loss": 1.1997, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 3.0415666992391386, | |
| "learning_rate": 8.990086136401199e-06, | |
| "loss": 1.1369, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 3.0682485455683, | |
| "learning_rate": 8.695044586103297e-06, | |
| "loss": 1.153, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 2.660915607465715, | |
| "learning_rate": 8.368478218232787e-06, | |
| "loss": 1.0615, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 2.3104573446692664, | |
| "learning_rate": 8.013173181896283e-06, | |
| "loss": 1.015, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 2.1383910273570086, | |
| "learning_rate": 7.63216081438678e-06, | |
| "loss": 1.06, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.9283231920822732, | |
| "learning_rate": 7.2286917788826926e-06, | |
| "loss": 0.9481, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 1.8886170821861419, | |
| "learning_rate": 6.806208330935766e-06, | |
| "loss": 0.9303, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 1.8907890024162215, | |
| "learning_rate": 6.368314950360416e-06, | |
| "loss": 0.9709, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.4285714285714284, | |
| "grad_norm": 1.9093053411125347, | |
| "learning_rate": 5.918747589082853e-06, | |
| "loss": 0.9213, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 1.866649612150217, | |
| "learning_rate": 5.46134179731651e-06, | |
| "loss": 0.8741, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.7142857142857144, | |
| "grad_norm": 1.7099290243677971, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8563, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.7446978143600087, | |
| "learning_rate": 4.53865820268349e-06, | |
| "loss": 0.8587, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.6658354125290393, | |
| "learning_rate": 4.081252410917148e-06, | |
| "loss": 0.7558, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 1.6588731170974642, | |
| "learning_rate": 3.6316850496395863e-06, | |
| "loss": 0.7808, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.2857142857142856, | |
| "grad_norm": 1.7054713909510084, | |
| "learning_rate": 3.1937916690642356e-06, | |
| "loss": 0.8018, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 1.5044309582959277, | |
| "learning_rate": 2.771308221117309e-06, | |
| "loss": 0.785, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 1.6109753709228194, | |
| "learning_rate": 2.3678391856132203e-06, | |
| "loss": 0.7427, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 1.6141990688430101, | |
| "learning_rate": 1.9868268181037186e-06, | |
| "loss": 0.7829, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.857142857142857, | |
| "grad_norm": 1.6811314196010585, | |
| "learning_rate": 1.6315217817672142e-06, | |
| "loss": 0.785, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.6017235236105927, | |
| "learning_rate": 1.3049554138967052e-06, | |
| "loss": 0.7273, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.142857142857143, | |
| "grad_norm": 1.5419350535809524, | |
| "learning_rate": 1.0099138635988026e-06, | |
| "loss": 0.7737, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 1.5992532932151025, | |
| "learning_rate": 7.489143213519301e-07, | |
| "loss": 0.6784, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.428571428571429, | |
| "grad_norm": 1.6017284844116721, | |
| "learning_rate": 5.241835432246888e-07, | |
| "loss": 0.7267, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 1.5403007797347406, | |
| "learning_rate": 3.3763885297822153e-07, | |
| "loss": 0.767, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.714285714285714, | |
| "grad_norm": 1.515361106622691, | |
| "learning_rate": 1.908717841359048e-07, | |
| "loss": 0.6971, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 1.5420726170851473, | |
| "learning_rate": 8.513450158049109e-08, | |
| "loss": 0.7647, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.5561245777833046, | |
| "learning_rate": 2.1329118524827662e-08, | |
| "loss": 0.7352, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 35, | |
| "total_flos": 852618117120.0, | |
| "train_loss": 0.9887651102883476, | |
| "train_runtime": 133.1517, | |
| "train_samples_per_second": 3.793, | |
| "train_steps_per_second": 0.263 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 35, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 42, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 852618117120.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |