| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07028641714988579, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007028641714988579, | |
| "eval_loss": 1.1621719598770142, | |
| "eval_runtime": 333.2956, | |
| "eval_samples_per_second": 7.192, | |
| "eval_steps_per_second": 0.9, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0021085925144965737, | |
| "grad_norm": 0.7200496196746826, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.1481, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004217185028993147, | |
| "grad_norm": 0.4291810393333435, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1496, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0063257775434897206, | |
| "grad_norm": 0.6174260377883911, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1159, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0063257775434897206, | |
| "eval_loss": 1.077151894569397, | |
| "eval_runtime": 336.606, | |
| "eval_samples_per_second": 7.121, | |
| "eval_steps_per_second": 0.891, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008434370057986295, | |
| "grad_norm": 0.5994438529014587, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.0264, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010542962572482868, | |
| "grad_norm": 0.25993818044662476, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.8913, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012651555086979441, | |
| "grad_norm": 0.2148323357105255, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.9068, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012651555086979441, | |
| "eval_loss": 0.8854690194129944, | |
| "eval_runtime": 336.5338, | |
| "eval_samples_per_second": 7.123, | |
| "eval_steps_per_second": 0.891, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014760147601476014, | |
| "grad_norm": 0.2650648057460785, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.8897, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01686874011597259, | |
| "grad_norm": 0.2767353355884552, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.805, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.018977332630469163, | |
| "grad_norm": 0.21788977086544037, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.8363, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.018977332630469163, | |
| "eval_loss": 0.8280394673347473, | |
| "eval_runtime": 336.3295, | |
| "eval_samples_per_second": 7.127, | |
| "eval_steps_per_second": 0.892, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.021085925144965736, | |
| "grad_norm": 0.19669905304908752, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.8562, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02319451765946231, | |
| "grad_norm": 0.16489790380001068, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.7713, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.025303110173958882, | |
| "grad_norm": 0.18417863547801971, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.7589, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.025303110173958882, | |
| "eval_loss": 0.7888206243515015, | |
| "eval_runtime": 336.3758, | |
| "eval_samples_per_second": 7.126, | |
| "eval_steps_per_second": 0.892, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.027411702688455455, | |
| "grad_norm": 0.1926659792661667, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.7353, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.02952029520295203, | |
| "grad_norm": 0.20542505383491516, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.8628, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0316288877174486, | |
| "grad_norm": 0.1803169548511505, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.8016, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0316288877174486, | |
| "eval_loss": 0.7604277729988098, | |
| "eval_runtime": 336.9862, | |
| "eval_samples_per_second": 7.113, | |
| "eval_steps_per_second": 0.89, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03373748023194518, | |
| "grad_norm": 0.19310009479522705, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.7162, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03584607274644175, | |
| "grad_norm": 0.18422365188598633, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.7279, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.037954665260938325, | |
| "grad_norm": 0.2260413020849228, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.7397, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.037954665260938325, | |
| "eval_loss": 0.7391679286956787, | |
| "eval_runtime": 337.1783, | |
| "eval_samples_per_second": 7.109, | |
| "eval_steps_per_second": 0.89, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.040063257775434895, | |
| "grad_norm": 0.22594496607780457, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.7247, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.04217185028993147, | |
| "grad_norm": 0.17635485529899597, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.6246, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "grad_norm": 0.1977640837430954, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.7231, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "eval_loss": 0.7249730825424194, | |
| "eval_runtime": 337.0353, | |
| "eval_samples_per_second": 7.112, | |
| "eval_steps_per_second": 0.89, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04638903531892462, | |
| "grad_norm": 0.22084610164165497, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.7496, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.048497627833421195, | |
| "grad_norm": 0.18980462849140167, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.7441, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.050606220347917764, | |
| "grad_norm": 0.18127837777137756, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.6734, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.050606220347917764, | |
| "eval_loss": 0.7157524824142456, | |
| "eval_runtime": 337.0656, | |
| "eval_samples_per_second": 7.111, | |
| "eval_steps_per_second": 0.89, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05271481286241434, | |
| "grad_norm": 0.17254553735256195, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.6744, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05482340537691091, | |
| "grad_norm": 0.20635554194450378, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.6527, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.05693199789140749, | |
| "grad_norm": 0.24892695248126984, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.7053, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05693199789140749, | |
| "eval_loss": 0.710917055606842, | |
| "eval_runtime": 337.0662, | |
| "eval_samples_per_second": 7.111, | |
| "eval_steps_per_second": 0.89, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05904059040590406, | |
| "grad_norm": 0.19214046001434326, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.7157, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.061149182920400634, | |
| "grad_norm": 0.20953665673732758, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.7124, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0632577754348972, | |
| "grad_norm": 0.18339523673057556, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.7001, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0632577754348972, | |
| "eval_loss": 0.708888828754425, | |
| "eval_runtime": 336.9158, | |
| "eval_samples_per_second": 7.115, | |
| "eval_steps_per_second": 0.89, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06536636794939378, | |
| "grad_norm": 0.21016551554203033, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.6601, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.06747496046389036, | |
| "grad_norm": 0.18210528790950775, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.7114, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.06958355297838692, | |
| "grad_norm": 0.20607209205627441, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.684, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.06958355297838692, | |
| "eval_loss": 0.7085462808609009, | |
| "eval_runtime": 336.9563, | |
| "eval_samples_per_second": 7.114, | |
| "eval_steps_per_second": 0.89, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.64437646753792e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |