|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 27582, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 4.909361177579581e-05, |
|
"loss": 1.1223, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 4.818722355159162e-05, |
|
"loss": 1.035, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 4.728083532738743e-05, |
|
"loss": 1.0383, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 4.637444710318324e-05, |
|
"loss": 0.9976, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 4.546805887897905e-05, |
|
"loss": 0.9862, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 4.4561670654774854e-05, |
|
"loss": 0.9828, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 4.3655282430570665e-05, |
|
"loss": 0.991, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 4.2748894206366476e-05, |
|
"loss": 0.9492, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 4.184250598216228e-05, |
|
"loss": 0.9492, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 4.093611775795809e-05, |
|
"loss": 0.9483, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 4.00297295337539e-05, |
|
"loss": 0.9157, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 3.912334130954971e-05, |
|
"loss": 0.9092, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 3.821695308534552e-05, |
|
"loss": 0.8941, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 3.731056486114133e-05, |
|
"loss": 0.9126, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 3.640417663693714e-05, |
|
"loss": 0.8869, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 3.549778841273294e-05, |
|
"loss": 0.8928, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 3.459140018852875e-05, |
|
"loss": 0.8764, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 3.368501196432456e-05, |
|
"loss": 0.8788, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 3.277862374012037e-05, |
|
"loss": 0.7333, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 3.187223551591618e-05, |
|
"loss": 0.6604, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 3.0965847291711983e-05, |
|
"loss": 0.6689, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 3.0059459067507794e-05, |
|
"loss": 0.6555, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 2.9153070843303605e-05, |
|
"loss": 0.658, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 2.8246682619099413e-05, |
|
"loss": 0.6627, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"learning_rate": 2.734029439489522e-05, |
|
"loss": 0.6626, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"learning_rate": 2.643390617069103e-05, |
|
"loss": 0.6381, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 2.552751794648684e-05, |
|
"loss": 0.6319, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 2.462112972228265e-05, |
|
"loss": 0.6766, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 2.3714741498078457e-05, |
|
"loss": 0.6428, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 2.2808353273874268e-05, |
|
"loss": 0.622, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"learning_rate": 2.1901965049670075e-05, |
|
"loss": 0.6313, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 2.0995576825465886e-05, |
|
"loss": 0.6466, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 2.0089188601261694e-05, |
|
"loss": 0.6203, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"learning_rate": 1.91828003770575e-05, |
|
"loss": 0.6103, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"learning_rate": 1.8276412152853312e-05, |
|
"loss": 0.6085, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 1.737002392864912e-05, |
|
"loss": 0.6039, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 1.646363570444493e-05, |
|
"loss": 0.5586, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 2.07, |
|
"learning_rate": 1.5557247480240738e-05, |
|
"loss": 0.4023, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"learning_rate": 1.4650859256036547e-05, |
|
"loss": 0.39, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 2.18, |
|
"learning_rate": 1.3744471031832355e-05, |
|
"loss": 0.3777, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"learning_rate": 1.2838082807628162e-05, |
|
"loss": 0.3958, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 2.28, |
|
"learning_rate": 1.1931694583423973e-05, |
|
"loss": 0.4169, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 2.34, |
|
"learning_rate": 1.1025306359219782e-05, |
|
"loss": 0.3802, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"learning_rate": 1.011891813501559e-05, |
|
"loss": 0.3793, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 9.212529910811399e-06, |
|
"loss": 0.4118, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"learning_rate": 8.306141686607208e-06, |
|
"loss": 0.3776, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 7.399753462403016e-06, |
|
"loss": 0.3892, |
|
"step": 23500 |
|
}, |
|
{ |
|
"epoch": 2.61, |
|
"learning_rate": 6.493365238198826e-06, |
|
"loss": 0.3759, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 5.586977013994634e-06, |
|
"loss": 0.3663, |
|
"step": 24500 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"learning_rate": 4.680588789790443e-06, |
|
"loss": 0.3836, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 2.77, |
|
"learning_rate": 3.7742005655862523e-06, |
|
"loss": 0.3731, |
|
"step": 25500 |
|
}, |
|
{ |
|
"epoch": 2.83, |
|
"learning_rate": 2.8678123413820606e-06, |
|
"loss": 0.3657, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 1.96142411717787e-06, |
|
"loss": 0.3635, |
|
"step": 26500 |
|
}, |
|
{ |
|
"epoch": 2.94, |
|
"learning_rate": 1.0550358929736786e-06, |
|
"loss": 0.3885, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"learning_rate": 1.4864766876948735e-07, |
|
"loss": 0.3573, |
|
"step": 27500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 27582, |
|
"total_flos": 2.836008195999293e+16, |
|
"train_loss": 0.6582008492790169, |
|
"train_runtime": 15084.1881, |
|
"train_samples_per_second": 14.627, |
|
"train_steps_per_second": 1.829 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 27582, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 2.836008195999293e+16, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|