| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02008637139700713, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008034548558802853, | |
| "grad_norm": 2.1960723400115967, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0697, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008034548558802853, | |
| "eval_loss": 1.163741111755371, | |
| "eval_runtime": 27.8219, | |
| "eval_samples_per_second": 37.704, | |
| "eval_steps_per_second": 18.87, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016069097117605705, | |
| "grad_norm": 2.7917819023132324, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2096, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0024103645676408557, | |
| "grad_norm": 2.0014545917510986, | |
| "learning_rate": 6e-05, | |
| "loss": 1.0414, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0024103645676408557, | |
| "eval_loss": 1.1070276498794556, | |
| "eval_runtime": 27.3924, | |
| "eval_samples_per_second": 38.295, | |
| "eval_steps_per_second": 19.166, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003213819423521141, | |
| "grad_norm": 2.172611951828003, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1045, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004017274279401426, | |
| "grad_norm": 1.4744445085525513, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8924, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004820729135281711, | |
| "grad_norm": 1.8950849771499634, | |
| "learning_rate": 0.00012, | |
| "loss": 0.9391, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004820729135281711, | |
| "eval_loss": 0.8586330413818359, | |
| "eval_runtime": 27.5044, | |
| "eval_samples_per_second": 38.139, | |
| "eval_steps_per_second": 19.088, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005624183991161997, | |
| "grad_norm": 2.3339345455169678, | |
| "learning_rate": 0.00014, | |
| "loss": 0.8422, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006427638847042282, | |
| "grad_norm": 1.636521816253662, | |
| "learning_rate": 0.00016, | |
| "loss": 0.6522, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0072310937029225674, | |
| "grad_norm": 2.06076717376709, | |
| "learning_rate": 0.00018, | |
| "loss": 0.6197, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0072310937029225674, | |
| "eval_loss": 0.521320641040802, | |
| "eval_runtime": 28.7745, | |
| "eval_samples_per_second": 36.456, | |
| "eval_steps_per_second": 18.245, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008034548558802852, | |
| "grad_norm": 1.8209742307662964, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4981, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008838003414683137, | |
| "grad_norm": 2.4247426986694336, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.4462, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009641458270563423, | |
| "grad_norm": 1.2692722082138062, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.3823, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.009641458270563423, | |
| "eval_loss": 0.3399949073791504, | |
| "eval_runtime": 27.4805, | |
| "eval_samples_per_second": 38.173, | |
| "eval_steps_per_second": 19.104, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010444913126443708, | |
| "grad_norm": 0.9483970999717712, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.3157, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011248367982323993, | |
| "grad_norm": 1.1190989017486572, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.3227, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.012051822838204279, | |
| "grad_norm": 1.410027265548706, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.3077, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012051822838204279, | |
| "eval_loss": 0.2781650722026825, | |
| "eval_runtime": 27.5144, | |
| "eval_samples_per_second": 38.126, | |
| "eval_steps_per_second": 19.081, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012855277694084564, | |
| "grad_norm": 1.0272347927093506, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.2837, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01365873254996485, | |
| "grad_norm": 1.063129186630249, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.2678, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.014462187405845135, | |
| "grad_norm": 1.1358978748321533, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.265, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014462187405845135, | |
| "eval_loss": 0.2543899118900299, | |
| "eval_runtime": 27.4127, | |
| "eval_samples_per_second": 38.267, | |
| "eval_steps_per_second": 19.152, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015265642261725419, | |
| "grad_norm": 1.0389295816421509, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2804, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.016069097117605704, | |
| "grad_norm": 1.001739263534546, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.2773, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01687255197348599, | |
| "grad_norm": 0.8247617483139038, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.2343, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01687255197348599, | |
| "eval_loss": 0.2293698489665985, | |
| "eval_runtime": 27.5151, | |
| "eval_samples_per_second": 38.124, | |
| "eval_steps_per_second": 19.08, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017676006829366275, | |
| "grad_norm": 0.8856362700462341, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.236, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01847946168524656, | |
| "grad_norm": 0.6518480777740479, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.2408, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.019282916541126845, | |
| "grad_norm": 0.777733325958252, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.2459, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.019282916541126845, | |
| "eval_loss": 0.2238992303609848, | |
| "eval_runtime": 27.5262, | |
| "eval_samples_per_second": 38.109, | |
| "eval_steps_per_second": 19.073, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02008637139700713, | |
| "grad_norm": 0.6574348211288452, | |
| "learning_rate": 0.0, | |
| "loss": 0.2059, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1802385752064000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |