|
{ |
|
"best_metric": 0.4443598985671997, |
|
"best_model_checkpoint": "Model-Meme/Typhoon/Typhoon1.5-Topic-Fold1/checkpoint-3824", |
|
"epoch": 10.0, |
|
"eval_steps": 500, |
|
"global_step": 19120, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.2615062761506276, |
|
"grad_norm": 1.392564296722412, |
|
"learning_rate": 8.693379790940767e-06, |
|
"loss": 0.9047, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.5230125523012552, |
|
"grad_norm": 1.0788474082946777, |
|
"learning_rate": 9.987048218238367e-06, |
|
"loss": 0.5001, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.7845188284518828, |
|
"grad_norm": 1.1558558940887451, |
|
"learning_rate": 9.938746068460573e-06, |
|
"loss": 0.5031, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.45572471618652344, |
|
"eval_runtime": 94.2827, |
|
"eval_samples_per_second": 10.14, |
|
"eval_steps_per_second": 10.14, |
|
"step": 1912 |
|
}, |
|
{ |
|
"epoch": 1.0460251046025104, |
|
"grad_norm": 2.1169137954711914, |
|
"learning_rate": 9.855036334031515e-06, |
|
"loss": 0.4826, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.3075313807531381, |
|
"grad_norm": 1.3102385997772217, |
|
"learning_rate": 9.736519159090385e-06, |
|
"loss": 0.4551, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.5690376569037658, |
|
"grad_norm": 1.6843914985656738, |
|
"learning_rate": 9.584044234397988e-06, |
|
"loss": 0.4469, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.8305439330543933, |
|
"grad_norm": 1.686008095741272, |
|
"learning_rate": 9.39870470560879e-06, |
|
"loss": 0.4534, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.4443598985671997, |
|
"eval_runtime": 93.0198, |
|
"eval_samples_per_second": 10.277, |
|
"eval_steps_per_second": 10.277, |
|
"step": 3824 |
|
}, |
|
{ |
|
"epoch": 2.092050209205021, |
|
"grad_norm": 1.8056789636611938, |
|
"learning_rate": 9.181829336130319e-06, |
|
"loss": 0.4372, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.3535564853556483, |
|
"grad_norm": 2.5060336589813232, |
|
"learning_rate": 8.935495472734113e-06, |
|
"loss": 0.3924, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 2.6150627615062763, |
|
"grad_norm": 2.0644381046295166, |
|
"learning_rate": 8.660482449775261e-06, |
|
"loss": 0.3869, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 2.8765690376569037, |
|
"grad_norm": 3.004842758178711, |
|
"learning_rate": 8.359226157693385e-06, |
|
"loss": 0.4131, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 0.448446661233902, |
|
"eval_runtime": 93.8596, |
|
"eval_samples_per_second": 10.185, |
|
"eval_steps_per_second": 10.185, |
|
"step": 5736 |
|
}, |
|
{ |
|
"epoch": 3.1380753138075312, |
|
"grad_norm": 3.1978588104248047, |
|
"learning_rate": 8.033886407403191e-06, |
|
"loss": 0.3767, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 3.399581589958159, |
|
"grad_norm": 4.149834156036377, |
|
"learning_rate": 7.687509931691041e-06, |
|
"loss": 0.3503, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 3.6610878661087867, |
|
"grad_norm": 4.2465620040893555, |
|
"learning_rate": 7.321942712373164e-06, |
|
"loss": 0.3455, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 3.922594142259414, |
|
"grad_norm": 3.6009790897369385, |
|
"learning_rate": 6.939014497791281e-06, |
|
"loss": 0.3487, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_loss": 0.46537432074546814, |
|
"eval_runtime": 89.9144, |
|
"eval_samples_per_second": 10.632, |
|
"eval_steps_per_second": 10.632, |
|
"step": 7648 |
|
}, |
|
{ |
|
"epoch": 4.184100418410042, |
|
"grad_norm": 3.0657713413238525, |
|
"learning_rate": 6.5421848153025935e-06, |
|
"loss": 0.3282, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 4.445606694560669, |
|
"grad_norm": 5.434764385223389, |
|
"learning_rate": 6.134298674628205e-06, |
|
"loss": 0.295, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 4.707112970711297, |
|
"grad_norm": 3.690535545349121, |
|
"learning_rate": 5.718280353075316e-06, |
|
"loss": 0.2963, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 4.968619246861925, |
|
"grad_norm": 4.333564281463623, |
|
"learning_rate": 5.297112430378265e-06, |
|
"loss": 0.2983, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_loss": 0.49299123883247375, |
|
"eval_runtime": 93.404, |
|
"eval_samples_per_second": 10.235, |
|
"eval_steps_per_second": 10.235, |
|
"step": 9560 |
|
}, |
|
{ |
|
"epoch": 5.2301255230125525, |
|
"grad_norm": 4.405161380767822, |
|
"learning_rate": 4.873814405549272e-06, |
|
"loss": 0.2595, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 5.49163179916318, |
|
"grad_norm": 4.975759029388428, |
|
"learning_rate": 4.452262916840334e-06, |
|
"loss": 0.2529, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 5.7531380753138075, |
|
"grad_norm": 6.338866710662842, |
|
"learning_rate": 4.0337916402351766e-06, |
|
"loss": 0.2727, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_loss": 0.5367496013641357, |
|
"eval_runtime": 94.3356, |
|
"eval_samples_per_second": 10.134, |
|
"eval_steps_per_second": 10.134, |
|
"step": 11472 |
|
}, |
|
{ |
|
"epoch": 6.014644351464435, |
|
"grad_norm": 6.288851261138916, |
|
"learning_rate": 3.6222474467023006e-06, |
|
"loss": 0.2505, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 6.2761506276150625, |
|
"grad_norm": 5.509110450744629, |
|
"learning_rate": 3.220580839399443e-06, |
|
"loss": 0.2118, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 6.53765690376569, |
|
"grad_norm": 6.196655750274658, |
|
"learning_rate": 2.8316715056404507e-06, |
|
"loss": 0.2221, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 6.799163179916318, |
|
"grad_norm": 4.003260135650635, |
|
"learning_rate": 2.4583076714174803e-06, |
|
"loss": 0.22, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_loss": 0.5626735687255859, |
|
"eval_runtime": 93.9459, |
|
"eval_samples_per_second": 10.176, |
|
"eval_steps_per_second": 10.176, |
|
"step": 13384 |
|
}, |
|
{ |
|
"epoch": 7.060669456066946, |
|
"grad_norm": 7.2679290771484375, |
|
"learning_rate": 2.103166111641151e-06, |
|
"loss": 0.2174, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 7.322175732217573, |
|
"grad_norm": 4.275331974029541, |
|
"learning_rate": 1.768792959412184e-06, |
|
"loss": 0.1885, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 7.583682008368201, |
|
"grad_norm": 5.202544212341309, |
|
"learning_rate": 1.4575854519095167e-06, |
|
"loss": 0.1949, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 7.845188284518828, |
|
"grad_norm": 4.865263938903809, |
|
"learning_rate": 1.1723196253920394e-06, |
|
"loss": 0.1941, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_loss": 0.5918557643890381, |
|
"eval_runtime": 95.08, |
|
"eval_samples_per_second": 10.055, |
|
"eval_steps_per_second": 10.055, |
|
"step": 15296 |
|
}, |
|
{ |
|
"epoch": 8.106694560669457, |
|
"grad_norm": 6.864102363586426, |
|
"learning_rate": 9.138979845169372e-07, |
|
"loss": 0.1773, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 8.368200836820083, |
|
"grad_norm": 6.523439884185791, |
|
"learning_rate": 6.847710271508462e-07, |
|
"loss": 0.1716, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 8.629707112970712, |
|
"grad_norm": 7.287766933441162, |
|
"learning_rate": 4.865814439740219e-07, |
|
"loss": 0.1764, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 8.891213389121338, |
|
"grad_norm": 8.745458602905273, |
|
"learning_rate": 3.207501248937539e-07, |
|
"loss": 0.173, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_loss": 0.6098597049713135, |
|
"eval_runtime": 93.4663, |
|
"eval_samples_per_second": 10.228, |
|
"eval_steps_per_second": 10.228, |
|
"step": 17208 |
|
}, |
|
{ |
|
"epoch": 9.152719665271967, |
|
"grad_norm": 4.589247703552246, |
|
"learning_rate": 1.886963892774535e-07, |
|
"loss": 0.164, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 9.414225941422593, |
|
"grad_norm": 4.173027038574219, |
|
"learning_rate": 9.083801976963458e-08, |
|
"loss": 0.1639, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 9.675732217573222, |
|
"grad_norm": 6.311884880065918, |
|
"learning_rate": 2.8175136316832e-08, |
|
"loss": 0.163, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 9.93723849372385, |
|
"grad_norm": 5.87669563293457, |
|
"learning_rate": 1.1569908788744423e-09, |
|
"loss": 0.1599, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_loss": 0.6195107102394104, |
|
"eval_runtime": 92.0592, |
|
"eval_samples_per_second": 10.385, |
|
"eval_steps_per_second": 10.385, |
|
"step": 19120 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 19120, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 4.0596421727340134e+17, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|