| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "global_step": 89859, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0002983307181250626, | |
| "loss": 3.5314, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00029666143625012515, | |
| "loss": 3.2698, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00029499215437518776, | |
| "loss": 3.1741, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0002933228725002504, | |
| "loss": 3.1591, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000291653590625313, | |
| "loss": 3.0795, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00028998430875037555, | |
| "loss": 3.0526, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00028831502687543816, | |
| "loss": 3.0228, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0002866457450005008, | |
| "loss": 2.9966, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0002849764631255634, | |
| "loss": 2.9799, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00028330718125062595, | |
| "loss": 2.9632, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_gen_len": 18.7655, | |
| "eval_loss": 2.559084892272949, | |
| "eval_rouge1": 25.8521, | |
| "eval_rouge2": 11.6024, | |
| "eval_rougeL": 21.0887, | |
| "eval_rougeLsum": 22.857, | |
| "eval_runtime": 625.5093, | |
| "eval_samples_per_second": 20.521, | |
| "eval_steps_per_second": 2.566, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028163789937568856, | |
| "loss": 2.9402, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0002799686175007512, | |
| "loss": 2.9157, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00027829933562581373, | |
| "loss": 2.9143, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00027663005375087635, | |
| "loss": 2.8869, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002749607718759389, | |
| "loss": 2.9195, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002732914900010015, | |
| "loss": 2.8558, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00027162220812606414, | |
| "loss": 2.8594, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00026995292625112675, | |
| "loss": 2.8485, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.0002682836443761893, | |
| "loss": 2.8491, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002666143625012519, | |
| "loss": 2.8327, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_gen_len": 18.8331, | |
| "eval_loss": 2.4890213012695312, | |
| "eval_rouge1": 26.508, | |
| "eval_rouge2": 12.2564, | |
| "eval_rougeL": 21.7467, | |
| "eval_rougeLsum": 23.5474, | |
| "eval_runtime": 624.7506, | |
| "eval_samples_per_second": 20.546, | |
| "eval_steps_per_second": 2.569, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00026494508062631454, | |
| "loss": 2.8157, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00026327579875137715, | |
| "loss": 2.8001, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0002616065168764397, | |
| "loss": 2.84, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002599372350015023, | |
| "loss": 2.8051, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00025826795312656494, | |
| "loss": 2.8138, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00025659867125162755, | |
| "loss": 2.7677, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0002549293893766901, | |
| "loss": 2.6905, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002532601075017527, | |
| "loss": 2.709, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00025159082562681534, | |
| "loss": 2.7136, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00024992154375187795, | |
| "loss": 2.6873, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_gen_len": 18.8424, | |
| "eval_loss": 2.4520416259765625, | |
| "eval_rouge1": 26.8614, | |
| "eval_rouge2": 12.7032, | |
| "eval_rougeL": 22.0965, | |
| "eval_rougeLsum": 23.832, | |
| "eval_runtime": 623.6597, | |
| "eval_samples_per_second": 20.582, | |
| "eval_steps_per_second": 2.574, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.0002482522618769405, | |
| "loss": 2.6962, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0002465829800020031, | |
| "loss": 2.6779, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00024491369812706574, | |
| "loss": 2.6964, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00024324441625212832, | |
| "loss": 2.7048, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00024157513437719094, | |
| "loss": 2.6687, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00023990585250225352, | |
| "loss": 2.6869, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00023823657062731608, | |
| "loss": 2.678, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0002365672887523787, | |
| "loss": 2.6544, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00023489800687744128, | |
| "loss": 2.6591, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002332287250025039, | |
| "loss": 2.6572, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_gen_len": 18.727, | |
| "eval_loss": 2.4031243324279785, | |
| "eval_rouge1": 27.0114, | |
| "eval_rouge2": 12.8148, | |
| "eval_rougeL": 22.2407, | |
| "eval_rougeLsum": 24.0602, | |
| "eval_runtime": 626.2755, | |
| "eval_samples_per_second": 20.496, | |
| "eval_steps_per_second": 2.563, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00023155944312756648, | |
| "loss": 2.628, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.0002298901612526291, | |
| "loss": 2.6544, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00022822087937769168, | |
| "loss": 2.6792, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0002265515975027543, | |
| "loss": 2.6493, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00022488231562781688, | |
| "loss": 2.6385, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0002232130337528795, | |
| "loss": 2.6368, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00022154375187794208, | |
| "loss": 2.6319, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002198744700030047, | |
| "loss": 2.6764, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00021820518812806729, | |
| "loss": 2.6729, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0002165359062531299, | |
| "loss": 2.6461, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_gen_len": 18.7416, | |
| "eval_loss": 2.391615629196167, | |
| "eval_rouge1": 27.2287, | |
| "eval_rouge2": 12.9935, | |
| "eval_rougeL": 22.4718, | |
| "eval_rougeLsum": 24.2517, | |
| "eval_runtime": 622.8089, | |
| "eval_samples_per_second": 20.61, | |
| "eval_steps_per_second": 2.577, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00021486662437819249, | |
| "loss": 2.6244, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.0002131973425032551, | |
| "loss": 2.5805, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00021152806062831769, | |
| "loss": 2.5407, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.00020985877875338027, | |
| "loss": 2.5537, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.00020818949687844289, | |
| "loss": 2.5253, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.00020652021500350547, | |
| "loss": 2.5401, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.00020485093312856809, | |
| "loss": 2.5245, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.00020318165125363067, | |
| "loss": 2.547, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.00020151236937869329, | |
| "loss": 2.5377, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00019984308750375587, | |
| "loss": 2.5374, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_gen_len": 18.8003, | |
| "eval_loss": 2.3686139583587646, | |
| "eval_rouge1": 27.5061, | |
| "eval_rouge2": 13.241, | |
| "eval_rougeL": 22.6877, | |
| "eval_rougeLsum": 24.4465, | |
| "eval_runtime": 629.1948, | |
| "eval_samples_per_second": 20.401, | |
| "eval_steps_per_second": 2.551, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.00019817380562881846, | |
| "loss": 2.5457, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00019650452375388105, | |
| "loss": 2.5105, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.00019483524187894366, | |
| "loss": 2.5367, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00019316596000400625, | |
| "loss": 2.5471, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.00019149667812906886, | |
| "loss": 2.5488, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.00018982739625413145, | |
| "loss": 2.5402, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00018815811437919406, | |
| "loss": 2.5437, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.00018648883250425665, | |
| "loss": 2.5244, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00018481955062931923, | |
| "loss": 2.5389, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00018315026875438185, | |
| "loss": 2.5081, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_gen_len": 18.7821, | |
| "eval_loss": 2.3596315383911133, | |
| "eval_rouge1": 27.4715, | |
| "eval_rouge2": 13.2862, | |
| "eval_rougeL": 22.7022, | |
| "eval_rougeLsum": 24.4252, | |
| "eval_runtime": 624.9994, | |
| "eval_samples_per_second": 20.538, | |
| "eval_steps_per_second": 2.568, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.00018148098687944443, | |
| "loss": 2.5425, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.00017981170500450705, | |
| "loss": 2.5245, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00017814242312956963, | |
| "loss": 2.506, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00017647314125463225, | |
| "loss": 2.542, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00017480385937969483, | |
| "loss": 2.5252, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00017313457750475745, | |
| "loss": 2.5175, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00017146529562982003, | |
| "loss": 2.5091, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.00016979601375488265, | |
| "loss": 2.4468, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.00016812673187994523, | |
| "loss": 2.447, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00016645745000500785, | |
| "loss": 2.4152, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_gen_len": 18.8203, | |
| "eval_loss": 2.326728105545044, | |
| "eval_rouge1": 27.9595, | |
| "eval_rouge2": 13.5813, | |
| "eval_rougeL": 23.0493, | |
| "eval_rougeLsum": 24.9203, | |
| "eval_runtime": 625.9414, | |
| "eval_samples_per_second": 20.507, | |
| "eval_steps_per_second": 2.564, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.00016478816813007043, | |
| "loss": 2.4379, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.00016311888625513305, | |
| "loss": 2.4489, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00016144960438019564, | |
| "loss": 2.4212, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.00015978032250525825, | |
| "loss": 2.4203, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0001581110406303208, | |
| "loss": 2.433, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0001564417587553834, | |
| "loss": 2.432, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.000154772476880446, | |
| "loss": 2.4299, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0001531031950055086, | |
| "loss": 2.4201, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.0001514339131305712, | |
| "loss": 2.4314, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.00014976463125563382, | |
| "loss": 2.4387, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_gen_len": 18.8376, | |
| "eval_loss": 2.317692518234253, | |
| "eval_rouge1": 28.1616, | |
| "eval_rouge2": 13.668, | |
| "eval_rougeL": 23.1738, | |
| "eval_rougeLsum": 25.0342, | |
| "eval_runtime": 622.4609, | |
| "eval_samples_per_second": 20.621, | |
| "eval_steps_per_second": 2.578, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0001480953493806964, | |
| "loss": 2.4389, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 0.000146426067505759, | |
| "loss": 2.4363, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.0001447567856308216, | |
| "loss": 2.4395, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0001430875037558842, | |
| "loss": 2.4418, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0001414182218809468, | |
| "loss": 2.4201, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.0001397489400060094, | |
| "loss": 2.4169, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.000138079658131072, | |
| "loss": 2.4009, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0001364103762561346, | |
| "loss": 2.4279, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 0.0001347410943811972, | |
| "loss": 2.4307, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.0001330718125062598, | |
| "loss": 2.4387, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_gen_len": 18.7872, | |
| "eval_loss": 2.3016672134399414, | |
| "eval_rouge1": 28.1111, | |
| "eval_rouge2": 13.569, | |
| "eval_rougeL": 23.1312, | |
| "eval_rougeLsum": 24.9912, | |
| "eval_runtime": 628.1204, | |
| "eval_samples_per_second": 20.436, | |
| "eval_steps_per_second": 2.555, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0001314025306313224, | |
| "loss": 2.4143, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 0.000129733248756385, | |
| "loss": 2.4197, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 0.00012806396688144758, | |
| "loss": 2.4084, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.00012639468500651017, | |
| "loss": 2.3544, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.00012472540313157278, | |
| "loss": 2.332, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.00012305612125663537, | |
| "loss": 2.3597, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 0.00012138683938169798, | |
| "loss": 2.3492, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 0.00011971755750676058, | |
| "loss": 2.3679, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00011804827563182318, | |
| "loss": 2.3633, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.00011637899375688578, | |
| "loss": 2.3467, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "eval_gen_len": 18.8334, | |
| "eval_loss": 2.3123602867126465, | |
| "eval_rouge1": 28.0679, | |
| "eval_rouge2": 13.7123, | |
| "eval_rougeL": 23.1516, | |
| "eval_rougeLsum": 25.0002, | |
| "eval_runtime": 622.7792, | |
| "eval_samples_per_second": 20.611, | |
| "eval_steps_per_second": 2.577, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.00011470971188194838, | |
| "loss": 2.3579, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.00011304043000701098, | |
| "loss": 2.3501, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.00011137114813207358, | |
| "loss": 2.3592, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.00010970186625713618, | |
| "loss": 2.344, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00010803258438219876, | |
| "loss": 2.3578, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.00010636330250726136, | |
| "loss": 2.3407, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 0.00010469402063232396, | |
| "loss": 2.3452, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 0.00010302473875738656, | |
| "loss": 2.3465, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.00010135545688244916, | |
| "loss": 2.3687, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 9.968617500751176e-05, | |
| "loss": 2.3367, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_gen_len": 18.835, | |
| "eval_loss": 2.2966153621673584, | |
| "eval_rouge1": 28.293, | |
| "eval_rouge2": 13.9084, | |
| "eval_rougeL": 23.3359, | |
| "eval_rougeLsum": 25.1789, | |
| "eval_runtime": 625.9275, | |
| "eval_samples_per_second": 20.507, | |
| "eval_steps_per_second": 2.564, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 9.801689313257436e-05, | |
| "loss": 2.3306, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 9.634761125763696e-05, | |
| "loss": 2.3497, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 9.467832938269956e-05, | |
| "loss": 2.3313, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 9.300904750776216e-05, | |
| "loss": 2.3427, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 9.133976563282476e-05, | |
| "loss": 2.3259, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 8.967048375788736e-05, | |
| "loss": 2.3544, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 8.800120188294995e-05, | |
| "loss": 2.3307, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 8.633192000801255e-05, | |
| "loss": 2.3477, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 8.466263813307513e-05, | |
| "loss": 2.3064, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 8.299335625813773e-05, | |
| "loss": 2.2882, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_gen_len": 18.7974, | |
| "eval_loss": 2.2921857833862305, | |
| "eval_rouge1": 28.3828, | |
| "eval_rouge2": 14.0129, | |
| "eval_rougeL": 23.443, | |
| "eval_rougeLsum": 25.3001, | |
| "eval_runtime": 621.9943, | |
| "eval_samples_per_second": 20.637, | |
| "eval_steps_per_second": 2.58, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 8.132407438320033e-05, | |
| "loss": 2.2733, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.965479250826293e-05, | |
| "loss": 2.2869, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 7.798551063332553e-05, | |
| "loss": 2.2904, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.631622875838813e-05, | |
| "loss": 2.2516, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 7.464694688345073e-05, | |
| "loss": 2.2835, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 7.297766500851333e-05, | |
| "loss": 2.2802, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 7.130838313357593e-05, | |
| "loss": 2.2779, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 6.963910125863853e-05, | |
| "loss": 2.274, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 6.796981938370112e-05, | |
| "loss": 2.2908, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 6.630053750876372e-05, | |
| "loss": 2.2782, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "eval_gen_len": 18.8495, | |
| "eval_loss": 2.2865357398986816, | |
| "eval_rouge1": 28.3987, | |
| "eval_rouge2": 13.9705, | |
| "eval_rougeL": 23.4227, | |
| "eval_rougeLsum": 25.3263, | |
| "eval_runtime": 625.1399, | |
| "eval_samples_per_second": 20.533, | |
| "eval_steps_per_second": 2.567, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 6.463125563382632e-05, | |
| "loss": 2.2882, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 6.296197375888892e-05, | |
| "loss": 2.2802, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 6.129269188395152e-05, | |
| "loss": 2.2774, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 5.9623410009014114e-05, | |
| "loss": 2.31, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 5.7954128134076714e-05, | |
| "loss": 2.2829, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 5.6284846259139314e-05, | |
| "loss": 2.2971, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "learning_rate": 5.4615564384201915e-05, | |
| "loss": 2.2824, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 5.294628250926451e-05, | |
| "loss": 2.2754, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 5.127700063432711e-05, | |
| "loss": 2.2893, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 4.960771875938971e-05, | |
| "loss": 2.2788, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "eval_gen_len": 18.7967, | |
| "eval_loss": 2.278130531311035, | |
| "eval_rouge1": 28.4256, | |
| "eval_rouge2": 14.0668, | |
| "eval_rougeL": 23.4947, | |
| "eval_rougeLsum": 25.3403, | |
| "eval_runtime": 622.5919, | |
| "eval_samples_per_second": 20.617, | |
| "eval_steps_per_second": 2.578, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 4.793843688445231e-05, | |
| "loss": 2.2898, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 4.62691550095149e-05, | |
| "loss": 2.281, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 4.45998731345775e-05, | |
| "loss": 2.28, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.2930591259640095e-05, | |
| "loss": 2.269, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 4.1261309384702695e-05, | |
| "loss": 2.2428, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.9592027509765295e-05, | |
| "loss": 2.2627, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.7922745634827896e-05, | |
| "loss": 2.2219, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 3.625346375989049e-05, | |
| "loss": 2.218, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 3.458418188495309e-05, | |
| "loss": 2.2205, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 3.291490001001569e-05, | |
| "loss": 2.239, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "eval_gen_len": 18.8384, | |
| "eval_loss": 2.2816860675811768, | |
| "eval_rouge1": 28.4681, | |
| "eval_rouge2": 14.0835, | |
| "eval_rougeL": 23.5141, | |
| "eval_rougeLsum": 25.4023, | |
| "eval_runtime": 624.366, | |
| "eval_samples_per_second": 20.558, | |
| "eval_steps_per_second": 2.571, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 3.124561813507828e-05, | |
| "loss": 2.2095, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 2.9576336260140883e-05, | |
| "loss": 2.2441, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 2.7907054385203483e-05, | |
| "loss": 2.2292, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 2.623777251026608e-05, | |
| "loss": 2.2482, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 2.456849063532868e-05, | |
| "loss": 2.2243, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 2.289920876039128e-05, | |
| "loss": 2.2177, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 2.1229926885453873e-05, | |
| "loss": 2.2317, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 1.9560645010516473e-05, | |
| "loss": 2.2311, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 1.7891363135579073e-05, | |
| "loss": 2.2446, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 1.622208126064167e-05, | |
| "loss": 2.2324, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "eval_gen_len": 18.8369, | |
| "eval_loss": 2.27907395362854, | |
| "eval_rouge1": 28.5356, | |
| "eval_rouge2": 14.1871, | |
| "eval_rougeL": 23.5477, | |
| "eval_rougeLsum": 25.4279, | |
| "eval_runtime": 621.3714, | |
| "eval_samples_per_second": 20.658, | |
| "eval_steps_per_second": 2.583, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 1.4552799385704269e-05, | |
| "loss": 2.241, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 1.2883517510766869e-05, | |
| "loss": 2.2262, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 1.1214235635829465e-05, | |
| "loss": 2.2456, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 9.544953760892062e-06, | |
| "loss": 2.2237, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 7.875671885954662e-06, | |
| "loss": 2.2294, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 6.20639001101726e-06, | |
| "loss": 2.2184, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 4.537108136079858e-06, | |
| "loss": 2.2167, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 2.867826261142456e-06, | |
| "loss": 2.2478, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 1.1985443862050543e-06, | |
| "loss": 2.2596, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 89859, | |
| "total_flos": 2.6510315204815258e+17, | |
| "train_loss": 2.493601185993256, | |
| "train_runtime": 53594.8094, | |
| "train_samples_per_second": 13.413, | |
| "train_steps_per_second": 1.677 | |
| } | |
| ], | |
| "max_steps": 89859, | |
| "num_train_epochs": 7, | |
| "total_flos": 2.6510315204815258e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |