| { | |
| "best_metric": 0.5405722260475159, | |
| "best_model_checkpoint": "Model-Meme/Typhoon/Typhoon1.5-Offensive-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": 1.978398084640503, | |
| "learning_rate": 8.710801393728223e-06, | |
| "loss": 0.9518, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5230125523012552, | |
| "grad_norm": 1.655224323272705, | |
| "learning_rate": 9.986987223454062e-06, | |
| "loss": 0.6078, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7845188284518828, | |
| "grad_norm": 1.4709585905075073, | |
| "learning_rate": 9.93861382775312e-06, | |
| "loss": 0.6103, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5545303821563721, | |
| "eval_runtime": 106.2407, | |
| "eval_samples_per_second": 8.998, | |
| "eval_steps_per_second": 8.998, | |
| "step": 1912 | |
| }, | |
| { | |
| "epoch": 1.0460251046025104, | |
| "grad_norm": 2.412283420562744, | |
| "learning_rate": 9.854833795480441e-06, | |
| "loss": 0.587, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.3075313807531381, | |
| "grad_norm": 1.6763008832931519, | |
| "learning_rate": 9.736519159090385e-06, | |
| "loss": 0.5516, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5690376569037658, | |
| "grad_norm": 2.1787030696868896, | |
| "learning_rate": 9.584044234397988e-06, | |
| "loss": 0.5432, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.8305439330543933, | |
| "grad_norm": 2.15987229347229, | |
| "learning_rate": 9.39870470560879e-06, | |
| "loss": 0.5497, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5405722260475159, | |
| "eval_runtime": 107.6233, | |
| "eval_samples_per_second": 8.883, | |
| "eval_steps_per_second": 8.883, | |
| "step": 3824 | |
| }, | |
| { | |
| "epoch": 2.092050209205021, | |
| "grad_norm": 2.2572760581970215, | |
| "learning_rate": 9.181829336130319e-06, | |
| "loss": 0.5301, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.3535564853556483, | |
| "grad_norm": 3.1669037342071533, | |
| "learning_rate": 8.934972980757115e-06, | |
| "loss": 0.4753, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.6150627615062763, | |
| "grad_norm": 2.6985182762145996, | |
| "learning_rate": 8.660482449775261e-06, | |
| "loss": 0.4701, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.8765690376569037, | |
| "grad_norm": 3.5524895191192627, | |
| "learning_rate": 8.35985345532633e-06, | |
| "loss": 0.499, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5466441512107849, | |
| "eval_runtime": 105.7924, | |
| "eval_samples_per_second": 9.037, | |
| "eval_steps_per_second": 9.037, | |
| "step": 5736 | |
| }, | |
| { | |
| "epoch": 3.1380753138075312, | |
| "grad_norm": 3.800245523452759, | |
| "learning_rate": 8.034559598992307e-06, | |
| "loss": 0.4554, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.399581589958159, | |
| "grad_norm": 5.512406826019287, | |
| "learning_rate": 7.687509931691041e-06, | |
| "loss": 0.4241, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.6610878661087867, | |
| "grad_norm": 4.886234283447266, | |
| "learning_rate": 7.32119257293766e-06, | |
| "loss": 0.4183, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.922594142259414, | |
| "grad_norm": 4.367607593536377, | |
| "learning_rate": 6.939014497791281e-06, | |
| "loss": 0.4227, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.5681700110435486, | |
| "eval_runtime": 109.707, | |
| "eval_samples_per_second": 8.714, | |
| "eval_steps_per_second": 8.714, | |
| "step": 7648 | |
| }, | |
| { | |
| "epoch": 4.184100418410042, | |
| "grad_norm": 5.268321514129639, | |
| "learning_rate": 6.5421848153025935e-06, | |
| "loss": 0.394, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.445606694560669, | |
| "grad_norm": 6.094832897186279, | |
| "learning_rate": 6.134298674628205e-06, | |
| "loss": 0.3551, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.707112970711297, | |
| "grad_norm": 4.6014723777771, | |
| "learning_rate": 5.718280353075316e-06, | |
| "loss": 0.3585, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.968619246861925, | |
| "grad_norm": 5.809424877166748, | |
| "learning_rate": 5.297112430378265e-06, | |
| "loss": 0.3608, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.5995907187461853, | |
| "eval_runtime": 104.4657, | |
| "eval_samples_per_second": 9.151, | |
| "eval_steps_per_second": 9.151, | |
| "step": 9560 | |
| }, | |
| { | |
| "epoch": 5.2301255230125525, | |
| "grad_norm": 5.566999435424805, | |
| "learning_rate": 4.873814405549272e-06, | |
| "loss": 0.3129, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.49163179916318, | |
| "grad_norm": 5.675551891326904, | |
| "learning_rate": 4.451421049042024e-06, | |
| "loss": 0.3044, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.7531380753138075, | |
| "grad_norm": 7.751038551330566, | |
| "learning_rate": 4.032960645429033e-06, | |
| "loss": 0.3276, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6560445427894592, | |
| "eval_runtime": 109.4544, | |
| "eval_samples_per_second": 8.734, | |
| "eval_steps_per_second": 8.734, | |
| "step": 11472 | |
| }, | |
| { | |
| "epoch": 6.014644351464435, | |
| "grad_norm": 8.045265197753906, | |
| "learning_rate": 3.6222474467023006e-06, | |
| "loss": 0.3008, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.2761506276150625, | |
| "grad_norm": 6.55618953704834, | |
| "learning_rate": 3.220580839399443e-06, | |
| "loss": 0.2541, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.53765690376569, | |
| "grad_norm": 6.615233421325684, | |
| "learning_rate": 2.8324347220558857e-06, | |
| "loss": 0.2642, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.799163179916318, | |
| "grad_norm": 4.452328681945801, | |
| "learning_rate": 2.4590370850090855e-06, | |
| "loss": 0.2655, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.687631368637085, | |
| "eval_runtime": 106.5678, | |
| "eval_samples_per_second": 8.971, | |
| "eval_steps_per_second": 8.971, | |
| "step": 13384 | |
| }, | |
| { | |
| "epoch": 7.060669456066946, | |
| "grad_norm": 9.082950592041016, | |
| "learning_rate": 2.1038564929897326e-06, | |
| "loss": 0.2613, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.322175732217573, | |
| "grad_norm": 4.5683488845825195, | |
| "learning_rate": 1.7694393589342428e-06, | |
| "loss": 0.225, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.583682008368201, | |
| "grad_norm": 6.728757381439209, | |
| "learning_rate": 1.4581832353425335e-06, | |
| "loss": 0.2322, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.845188284518828, | |
| "grad_norm": 5.570503234863281, | |
| "learning_rate": 1.172864616852632e-06, | |
| "loss": 0.2316, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7278637290000916, | |
| "eval_runtime": 103.7809, | |
| "eval_samples_per_second": 9.212, | |
| "eval_steps_per_second": 9.212, | |
| "step": 15296 | |
| }, | |
| { | |
| "epoch": 8.106694560669457, | |
| "grad_norm": 8.535571098327637, | |
| "learning_rate": 9.143861751413785e-07, | |
| "loss": 0.2112, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.368200836820083, | |
| "grad_norm": 6.845192909240723, | |
| "learning_rate": 6.856269305226026e-07, | |
| "loss": 0.2052, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.629707112970712, | |
| "grad_norm": 8.833856582641602, | |
| "learning_rate": 4.873106159655255e-07, | |
| "loss": 0.2105, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.891213389121338, | |
| "grad_norm": 10.798727035522461, | |
| "learning_rate": 3.2134733781792273e-07, | |
| "loss": 0.2057, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.7465147376060486, | |
| "eval_runtime": 103.4261, | |
| "eval_samples_per_second": 9.243, | |
| "eval_steps_per_second": 9.243, | |
| "step": 17208 | |
| }, | |
| { | |
| "epoch": 9.152719665271967, | |
| "grad_norm": 5.5902252197265625, | |
| "learning_rate": 1.8892694442157356e-07, | |
| "loss": 0.1954, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.414225941422593, | |
| "grad_norm": 5.076449871063232, | |
| "learning_rate": 9.099880353576951e-08, | |
| "loss": 0.1949, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.675732217573222, | |
| "grad_norm": 7.189806938171387, | |
| "learning_rate": 2.835499832552657e-08, | |
| "loss": 0.1941, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.93723849372385, | |
| "grad_norm": 5.884722709655762, | |
| "learning_rate": 1.1937170093084816e-09, | |
| "loss": 0.1904, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7588898539543152, | |
| "eval_runtime": 103.3856, | |
| "eval_samples_per_second": 9.247, | |
| "eval_steps_per_second": 9.247, | |
| "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": 3.3672019664225894e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |