|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.9985185185185185, |
|
"eval_steps": 500, |
|
"global_step": 337, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 0.1301183917120682, |
|
"learning_rate": 5.882352941176471e-06, |
|
"loss": 1.2653, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 0.13423529665439626, |
|
"learning_rate": 2.9411764705882354e-05, |
|
"loss": 1.2371, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 0.23748667333650045, |
|
"learning_rate": 5.882352941176471e-05, |
|
"loss": 1.1776, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 0.31960744166968214, |
|
"learning_rate": 8.823529411764706e-05, |
|
"loss": 1.1745, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 0.30058599861267066, |
|
"learning_rate": 0.00011764705882352942, |
|
"loss": 1.0782, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"grad_norm": 0.2768843491438946, |
|
"learning_rate": 0.00014705882352941178, |
|
"loss": 0.9767, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 0.3704908664562922, |
|
"learning_rate": 0.00017647058823529413, |
|
"loss": 0.9157, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 0.10417281565059555, |
|
"learning_rate": 0.00019999462497359466, |
|
"loss": 0.8196, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 0.08961437607599079, |
|
"learning_rate": 0.00019980655971335945, |
|
"loss": 0.7921, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"grad_norm": 0.09488226487438857, |
|
"learning_rate": 0.0001993503206718859, |
|
"loss": 0.7887, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.08675837478301882, |
|
"learning_rate": 0.0001986271337340182, |
|
"loss": 0.7985, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 0.08072222689651035, |
|
"learning_rate": 0.00019763894205636072, |
|
"loss": 0.8127, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 0.08501281247826332, |
|
"learning_rate": 0.00019638840084614182, |
|
"loss": 0.8066, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"grad_norm": 0.06341225515947009, |
|
"learning_rate": 0.00019487887022684336, |
|
"loss": 0.7701, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"grad_norm": 0.06812259862629667, |
|
"learning_rate": 0.00019311440620976597, |
|
"loss": 0.7974, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 0.08010377161115362, |
|
"learning_rate": 0.0001910997497957885, |
|
"loss": 0.8191, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"grad_norm": 0.06550467078105318, |
|
"learning_rate": 0.0001888403142366049, |
|
"loss": 0.7812, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 0.07253067486847224, |
|
"learning_rate": 0.00018634217048966637, |
|
"loss": 0.7492, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"grad_norm": 0.07561648690728726, |
|
"learning_rate": 0.00018361203090591071, |
|
"loss": 0.7877, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"grad_norm": 0.0788936704645067, |
|
"learning_rate": 0.00018065723119410884, |
|
"loss": 0.7929, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 0.0884884554192815, |
|
"learning_rate": 0.000177485710710289, |
|
"loss": 0.7806, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 0.06410506327210315, |
|
"learning_rate": 0.0001741059911251997, |
|
"loss": 0.7893, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"grad_norm": 0.07945762656708655, |
|
"learning_rate": 0.00017052715352713075, |
|
"loss": 0.7685, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 0.07017317816275172, |
|
"learning_rate": 0.00016675881402161536, |
|
"loss": 0.7965, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"grad_norm": 0.07892803161399417, |
|
"learning_rate": 0.0001628110978935756, |
|
"loss": 0.7909, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 0.07247197366009367, |
|
"learning_rate": 0.0001586946124013354, |
|
"loss": 0.7779, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"grad_norm": 0.0794954580794324, |
|
"learning_rate": 0.00015442041827560274, |
|
"loss": 0.7758, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.0781410228289432, |
|
"learning_rate": 0.00015000000000000001, |
|
"loss": 0.7411, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"grad_norm": 0.08217583726026866, |
|
"learning_rate": 0.00014544523495299842, |
|
"loss": 0.7923, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 0.07210384035280004, |
|
"learning_rate": 0.00014076836149416887, |
|
"loss": 0.7653, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"grad_norm": 0.06743506313322914, |
|
"learning_rate": 0.0001359819460805001, |
|
"loss": 0.7475, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.06929443024377009, |
|
"learning_rate": 0.00013109884950114007, |
|
"loss": 0.7431, |
|
"step": 155 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"grad_norm": 0.07591018161762071, |
|
"learning_rate": 0.00012613219232128608, |
|
"loss": 0.7871, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.08369691300496385, |
|
"learning_rate": 0.00012109531962807332, |
|
"loss": 0.7729, |
|
"step": 165 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"grad_norm": 0.08089436127546482, |
|
"learning_rate": 0.00011600176517318741, |
|
"loss": 0.7707, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"grad_norm": 0.0706473017125327, |
|
"learning_rate": 0.00011086521500854745, |
|
"loss": 0.7557, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 0.07845062376590645, |
|
"learning_rate": 0.00010569947071276847, |
|
"loss": 0.7878, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.08099917146584029, |
|
"learning_rate": 0.00010051841230721065, |
|
"loss": 0.7961, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"grad_norm": 0.08620445155100019, |
|
"learning_rate": 9.533596096125825e-05, |
|
"loss": 0.7524, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 0.08862656104000743, |
|
"learning_rate": 9.016604158703654e-05, |
|
"loss": 0.757, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 0.08502821196989717, |
|
"learning_rate": 8.502254542407186e-05, |
|
"loss": 0.787, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 0.07651964157136036, |
|
"learning_rate": 7.991929271442817e-05, |
|
"loss": 0.7617, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"grad_norm": 0.09084246580451502, |
|
"learning_rate": 7.48699955686089e-05, |
|
"loss": 0.7808, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"grad_norm": 0.0732902588933958, |
|
"learning_rate": 6.988822112200156e-05, |
|
"loss": 0.7814, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 0.09648121763622272, |
|
"learning_rate": 6.498735508086093e-05, |
|
"loss": 0.7292, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 0.07551393443193218, |
|
"learning_rate": 6.018056575578075e-05, |
|
"loss": 0.7748, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"grad_norm": 0.09161304263255975, |
|
"learning_rate": 5.54807686792933e-05, |
|
"loss": 0.7923, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 0.09059355787494829, |
|
"learning_rate": 5.090059190266779e-05, |
|
"loss": 0.7905, |
|
"step": 235 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 0.08085206031151301, |
|
"learning_rate": 4.645234206515171e-05, |
|
"loss": 0.8124, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"grad_norm": 0.0797379987288294, |
|
"learning_rate": 4.2147971326825966e-05, |
|
"loss": 0.7692, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"grad_norm": 0.09502633909434978, |
|
"learning_rate": 3.79990452539225e-05, |
|
"loss": 0.8073, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"grad_norm": 0.08150518384411078, |
|
"learning_rate": 3.401671174289469e-05, |
|
"loss": 0.7554, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 0.08598678237692878, |
|
"learning_rate": 3.021167106673928e-05, |
|
"loss": 0.7474, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"grad_norm": 0.07819439855358523, |
|
"learning_rate": 2.659414712405398e-05, |
|
"loss": 0.7576, |
|
"step": 265 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.08023891142221302, |
|
"learning_rate": 2.3173859968081944e-05, |
|
"loss": 0.7608, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"grad_norm": 0.09088106944567358, |
|
"learning_rate": 1.995999968955641e-05, |
|
"loss": 0.7616, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"grad_norm": 0.09414696185806744, |
|
"learning_rate": 1.696120172352025e-05, |
|
"loss": 0.7986, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"grad_norm": 0.0827414529922064, |
|
"learning_rate": 1.4185523646469822e-05, |
|
"loss": 0.7681, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 0.09854416536124955, |
|
"learning_rate": 1.1640423526166988e-05, |
|
"loss": 0.7776, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"grad_norm": 0.08611779666327021, |
|
"learning_rate": 9.332739882292752e-06, |
|
"loss": 0.7628, |
|
"step": 295 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 0.08869045110724108, |
|
"learning_rate": 7.2686733117863784e-06, |
|
"loss": 0.7729, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 0.07919765304647886, |
|
"learning_rate": 5.453769828241872e-06, |
|
"loss": 0.7709, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"grad_norm": 0.08336346515163112, |
|
"learning_rate": 3.892905960127546e-06, |
|
"loss": 0.7703, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 0.08196958012246881, |
|
"learning_rate": 2.590275647868867e-06, |
|
"loss": 0.7622, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"grad_norm": 0.07692088564904832, |
|
"learning_rate": 1.5493789750014031e-06, |
|
"loss": 0.7952, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"grad_norm": 0.08235291945346827, |
|
"learning_rate": 7.730127636723539e-07, |
|
"loss": 0.7722, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"grad_norm": 0.07764082388173033, |
|
"learning_rate": 2.6326305976001055e-07, |
|
"loss": 0.7769, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 0.08728825013568611, |
|
"learning_rate": 2.1499527803214846e-08, |
|
"loss": 0.7585, |
|
"step": 335 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.9679195284843445, |
|
"eval_runtime": 175.419, |
|
"eval_samples_per_second": 13.168, |
|
"eval_steps_per_second": 0.827, |
|
"step": 337 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 337, |
|
"total_flos": 3852908399427584.0, |
|
"train_loss": 0.08478880317699308, |
|
"train_runtime": 740.7528, |
|
"train_samples_per_second": 29.151, |
|
"train_steps_per_second": 0.455 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 337, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 100, |
|
"total_flos": 3852908399427584.0, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|