| { | |
| "best_metric": 54.4377, | |
| "best_model_checkpoint": "/uoa/home/s02sd1/Desktop/Project data/pytorch_project/facebook_28.03/checkpoint-53000", | |
| "epoch": 78.98448519040903, | |
| "global_step": 56000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.978843441466855e-06, | |
| "loss": 0.6224, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.9576868829337095e-06, | |
| "loss": 0.5049, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_gen_len": 167.2707, | |
| "eval_loss": 0.75015789270401, | |
| "eval_rouge1": 61.2286, | |
| "eval_rouge2": 43.2506, | |
| "eval_rougeL": 42.2051, | |
| "eval_rougeLsum": 46.3795, | |
| "eval_runtime": 5804.0101, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.9365303244005645e-06, | |
| "loss": 0.4992, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 2.915373765867419e-06, | |
| "loss": 0.4523, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_gen_len": 169.4872, | |
| "eval_loss": 0.7313751578330994, | |
| "eval_rouge1": 63.5466, | |
| "eval_rouge2": 45.7831, | |
| "eval_rougeL": 44.4885, | |
| "eval_rougeLsum": 48.6313, | |
| "eval_runtime": 4918.4156, | |
| "eval_samples_per_second": 0.071, | |
| "eval_steps_per_second": 0.004, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.894217207334274e-06, | |
| "loss": 0.4422, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 2.8730606488011285e-06, | |
| "loss": 0.4036, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "eval_gen_len": 160.4074, | |
| "eval_loss": 0.7353909611701965, | |
| "eval_rouge1": 64.7439, | |
| "eval_rouge2": 47.1723, | |
| "eval_rougeL": 45.9479, | |
| "eval_rougeLsum": 50.2341, | |
| "eval_runtime": 4241.5946, | |
| "eval_samples_per_second": 0.083, | |
| "eval_steps_per_second": 0.004, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 2.851904090267983e-06, | |
| "loss": 0.4028, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 2.830747531734838e-06, | |
| "loss": 0.3722, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "eval_gen_len": 162.0085, | |
| "eval_loss": 0.72761070728302, | |
| "eval_rouge1": 65.6958, | |
| "eval_rouge2": 48.5007, | |
| "eval_rougeL": 46.6737, | |
| "eval_rougeLsum": 50.9835, | |
| "eval_runtime": 4232.4341, | |
| "eval_samples_per_second": 0.083, | |
| "eval_steps_per_second": 0.004, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 2.8095909732016925e-06, | |
| "loss": 0.3749, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 2.7884344146685474e-06, | |
| "loss": 0.3482, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_gen_len": 163.9231, | |
| "eval_loss": 0.7417466640472412, | |
| "eval_rouge1": 65.933, | |
| "eval_rouge2": 48.921, | |
| "eval_rougeL": 47.5877, | |
| "eval_rougeLsum": 52.0114, | |
| "eval_runtime": 4836.5569, | |
| "eval_samples_per_second": 0.073, | |
| "eval_steps_per_second": 0.004, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 2.767277856135402e-06, | |
| "loss": 0.3312, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 2.746121297602257e-06, | |
| "loss": 0.3174, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "eval_gen_len": 163.5527, | |
| "eval_loss": 0.7546955943107605, | |
| "eval_rouge1": 65.9445, | |
| "eval_rouge2": 48.9117, | |
| "eval_rougeL": 48.312, | |
| "eval_rougeLsum": 52.4923, | |
| "eval_runtime": 5417.8435, | |
| "eval_samples_per_second": 0.065, | |
| "eval_steps_per_second": 0.003, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 2.7249647390691114e-06, | |
| "loss": 0.3074, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "learning_rate": 2.7038081805359663e-06, | |
| "loss": 0.2922, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "eval_gen_len": 163.1311, | |
| "eval_loss": 0.754240095615387, | |
| "eval_rouge1": 66.6619, | |
| "eval_rouge2": 49.9256, | |
| "eval_rougeL": 48.7757, | |
| "eval_rougeLsum": 53.1916, | |
| "eval_runtime": 5340.7776, | |
| "eval_samples_per_second": 0.066, | |
| "eval_steps_per_second": 0.003, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 2.6826516220028213e-06, | |
| "loss": 0.281, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "learning_rate": 2.6614950634696754e-06, | |
| "loss": 0.276, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "eval_gen_len": 165.0826, | |
| "eval_loss": 0.7793148756027222, | |
| "eval_rouge1": 66.8911, | |
| "eval_rouge2": 50.331, | |
| "eval_rougeL": 49.8396, | |
| "eval_rougeLsum": 54.0617, | |
| "eval_runtime": 6037.4348, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 2.6403385049365303e-06, | |
| "loss": 0.2557, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 2.6191819464033853e-06, | |
| "loss": 0.2464, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "eval_gen_len": 164.9943, | |
| "eval_loss": 0.7877324819564819, | |
| "eval_rouge1": 67.1466, | |
| "eval_rouge2": 50.939, | |
| "eval_rougeL": 50.2255, | |
| "eval_rougeLsum": 54.5145, | |
| "eval_runtime": 5837.726, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 2.5980253878702398e-06, | |
| "loss": 0.2369, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 2.5768688293370947e-06, | |
| "loss": 0.2265, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "eval_gen_len": 162.1966, | |
| "eval_loss": 0.806422233581543, | |
| "eval_rouge1": 67.5137, | |
| "eval_rouge2": 51.2801, | |
| "eval_rougeL": 51.2323, | |
| "eval_rougeLsum": 55.426, | |
| "eval_runtime": 5856.0164, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 2.5557122708039492e-06, | |
| "loss": 0.2215, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 15.51, | |
| "learning_rate": 2.534555712270804e-06, | |
| "loss": 0.2004, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.51, | |
| "eval_gen_len": 161.2365, | |
| "eval_loss": 0.8361101746559143, | |
| "eval_rouge1": 67.9515, | |
| "eval_rouge2": 51.5857, | |
| "eval_rougeL": 51.5796, | |
| "eval_rougeLsum": 55.9738, | |
| "eval_runtime": 5461.4431, | |
| "eval_samples_per_second": 0.064, | |
| "eval_steps_per_second": 0.003, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 2.5133991537376587e-06, | |
| "loss": 0.2039, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 16.93, | |
| "learning_rate": 2.4922425952045136e-06, | |
| "loss": 0.1903, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.93, | |
| "eval_gen_len": 163.8234, | |
| "eval_loss": 0.8543212413787842, | |
| "eval_rouge1": 67.8403, | |
| "eval_rouge2": 51.7839, | |
| "eval_rougeL": 51.8869, | |
| "eval_rougeLsum": 56.2681, | |
| "eval_runtime": 5935.4542, | |
| "eval_samples_per_second": 0.059, | |
| "eval_steps_per_second": 0.003, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "learning_rate": 2.471086036671368e-06, | |
| "loss": 0.1816, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 2.4499294781382227e-06, | |
| "loss": 0.1782, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "eval_gen_len": 161.2137, | |
| "eval_loss": 0.8672023415565491, | |
| "eval_rouge1": 68.2574, | |
| "eval_rouge2": 52.2649, | |
| "eval_rougeL": 52.3755, | |
| "eval_rougeLsum": 56.4509, | |
| "eval_runtime": 5698.6716, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 2.4287729196050776e-06, | |
| "loss": 0.1686, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "learning_rate": 2.407616361071932e-06, | |
| "loss": 0.1581, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "eval_gen_len": 163.0057, | |
| "eval_loss": 0.9022919535636902, | |
| "eval_rouge1": 67.8924, | |
| "eval_rouge2": 51.8802, | |
| "eval_rougeL": 52.5074, | |
| "eval_rougeLsum": 56.5946, | |
| "eval_runtime": 5838.2781, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 2.386459802538787e-06, | |
| "loss": 0.1525, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 21.16, | |
| "learning_rate": 2.3653032440056416e-06, | |
| "loss": 0.1516, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 21.16, | |
| "eval_gen_len": 163.0883, | |
| "eval_loss": 0.9338411092758179, | |
| "eval_rouge1": 68.0101, | |
| "eval_rouge2": 51.7535, | |
| "eval_rougeL": 52.2777, | |
| "eval_rougeLsum": 56.692, | |
| "eval_runtime": 6063.036, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 2.3441466854724966e-06, | |
| "loss": 0.143, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "learning_rate": 2.3229901269393515e-06, | |
| "loss": 0.1358, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "eval_gen_len": 162.3504, | |
| "eval_loss": 0.9403561353683472, | |
| "eval_rouge1": 68.4997, | |
| "eval_rouge2": 52.385, | |
| "eval_rougeL": 52.7137, | |
| "eval_rougeLsum": 57.2165, | |
| "eval_runtime": 5678.9898, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 23.27, | |
| "learning_rate": 2.301833568406206e-06, | |
| "loss": 0.1268, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 23.98, | |
| "learning_rate": 2.280677009873061e-06, | |
| "loss": 0.1256, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 23.98, | |
| "eval_gen_len": 163.6467, | |
| "eval_loss": 0.9707176685333252, | |
| "eval_rouge1": 68.3426, | |
| "eval_rouge2": 52.1326, | |
| "eval_rougeL": 52.6279, | |
| "eval_rougeLsum": 57.0803, | |
| "eval_runtime": 6169.909, | |
| "eval_samples_per_second": 0.057, | |
| "eval_steps_per_second": 0.003, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 24.68, | |
| "learning_rate": 2.2595204513399155e-06, | |
| "loss": 0.1158, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 25.39, | |
| "learning_rate": 2.23836389280677e-06, | |
| "loss": 0.1119, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 25.39, | |
| "eval_gen_len": 160.9316, | |
| "eval_loss": 1.0022860765457153, | |
| "eval_rouge1": 69.1266, | |
| "eval_rouge2": 52.9311, | |
| "eval_rougeL": 53.9465, | |
| "eval_rougeLsum": 58.0804, | |
| "eval_runtime": 5772.1926, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 2.217207334273625e-06, | |
| "loss": 0.1125, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "learning_rate": 2.1960507757404795e-06, | |
| "loss": 0.1038, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "eval_gen_len": 161.8746, | |
| "eval_loss": 1.019147515296936, | |
| "eval_rouge1": 68.6629, | |
| "eval_rouge2": 52.6518, | |
| "eval_rougeL": 53.1635, | |
| "eval_rougeLsum": 57.4514, | |
| "eval_runtime": 5939.4826, | |
| "eval_samples_per_second": 0.059, | |
| "eval_steps_per_second": 0.003, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 2.1748942172073344e-06, | |
| "loss": 0.098, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 28.21, | |
| "learning_rate": 2.153737658674189e-06, | |
| "loss": 0.0944, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 28.21, | |
| "eval_gen_len": 161.4359, | |
| "eval_loss": 1.0427296161651611, | |
| "eval_rouge1": 69.1063, | |
| "eval_rouge2": 53.0226, | |
| "eval_rougeL": 53.5979, | |
| "eval_rougeLsum": 57.8602, | |
| "eval_runtime": 6011.7516, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 28.91, | |
| "learning_rate": 2.132581100141044e-06, | |
| "loss": 0.0929, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 29.62, | |
| "learning_rate": 2.1114245416078984e-06, | |
| "loss": 0.0843, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 29.62, | |
| "eval_gen_len": 161.849, | |
| "eval_loss": 1.0581331253051758, | |
| "eval_rouge1": 68.0982, | |
| "eval_rouge2": 52.0153, | |
| "eval_rougeL": 53.1935, | |
| "eval_rougeLsum": 57.443, | |
| "eval_runtime": 5905.725, | |
| "eval_samples_per_second": 0.059, | |
| "eval_steps_per_second": 0.003, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 30.32, | |
| "learning_rate": 2.0902679830747533e-06, | |
| "loss": 0.0846, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 31.03, | |
| "learning_rate": 2.069111424541608e-06, | |
| "loss": 0.082, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 31.03, | |
| "eval_gen_len": 161.0684, | |
| "eval_loss": 1.0757286548614502, | |
| "eval_rouge1": 68.6318, | |
| "eval_rouge2": 52.8093, | |
| "eval_rougeL": 53.7214, | |
| "eval_rougeLsum": 57.775, | |
| "eval_runtime": 6004.4709, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 31.73, | |
| "learning_rate": 2.047954866008463e-06, | |
| "loss": 0.0735, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 32.44, | |
| "learning_rate": 2.0267983074753173e-06, | |
| "loss": 0.0756, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 32.44, | |
| "eval_gen_len": 162.4245, | |
| "eval_loss": 1.0798823833465576, | |
| "eval_rouge1": 68.557, | |
| "eval_rouge2": 52.4644, | |
| "eval_rougeL": 53.7606, | |
| "eval_rougeLsum": 57.8967, | |
| "eval_runtime": 6130.0883, | |
| "eval_samples_per_second": 0.057, | |
| "eval_steps_per_second": 0.003, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 33.15, | |
| "learning_rate": 2.005641748942172e-06, | |
| "loss": 0.0706, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 33.85, | |
| "learning_rate": 1.9844851904090268e-06, | |
| "loss": 0.0674, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 33.85, | |
| "eval_gen_len": 162.963, | |
| "eval_loss": 1.1047182083129883, | |
| "eval_rouge1": 68.6765, | |
| "eval_rouge2": 52.713, | |
| "eval_rougeL": 53.5339, | |
| "eval_rougeLsum": 57.7006, | |
| "eval_runtime": 6102.6451, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 34.56, | |
| "learning_rate": 1.9633286318758817e-06, | |
| "loss": 0.0647, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 35.26, | |
| "learning_rate": 1.9421720733427362e-06, | |
| "loss": 0.063, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 35.26, | |
| "eval_gen_len": 161.9003, | |
| "eval_loss": 1.1172865629196167, | |
| "eval_rouge1": 69.1911, | |
| "eval_rouge2": 53.1382, | |
| "eval_rougeL": 53.8971, | |
| "eval_rougeLsum": 58.2079, | |
| "eval_runtime": 5980.899, | |
| "eval_samples_per_second": 0.059, | |
| "eval_steps_per_second": 0.003, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 35.97, | |
| "learning_rate": 1.921015514809591e-06, | |
| "loss": 0.0599, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "learning_rate": 1.8998589562764457e-06, | |
| "loss": 0.0566, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "eval_gen_len": 162.3362, | |
| "eval_loss": 1.131187081336975, | |
| "eval_rouge1": 69.0624, | |
| "eval_rouge2": 53.101, | |
| "eval_rougeL": 54.2012, | |
| "eval_rougeLsum": 58.5665, | |
| "eval_runtime": 5792.9256, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 37.38, | |
| "learning_rate": 1.8787023977433004e-06, | |
| "loss": 0.0557, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "learning_rate": 1.8575458392101554e-06, | |
| "loss": 0.0532, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "eval_gen_len": 161.7009, | |
| "eval_loss": 1.1492334604263306, | |
| "eval_rouge1": 68.8232, | |
| "eval_rouge2": 52.7547, | |
| "eval_rougeL": 53.914, | |
| "eval_rougeLsum": 58.1061, | |
| "eval_runtime": 6017.4685, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 38.79, | |
| "learning_rate": 1.8363892806770101e-06, | |
| "loss": 0.0518, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 39.49, | |
| "learning_rate": 1.8152327221438644e-06, | |
| "loss": 0.0482, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 39.49, | |
| "eval_gen_len": 161.4046, | |
| "eval_loss": 1.1425426006317139, | |
| "eval_rouge1": 68.8641, | |
| "eval_rouge2": 52.7579, | |
| "eval_rougeL": 54.0604, | |
| "eval_rougeLsum": 58.3338, | |
| "eval_runtime": 5870.7466, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 40.2, | |
| "learning_rate": 1.7940761636107192e-06, | |
| "loss": 0.0459, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 40.9, | |
| "learning_rate": 1.772919605077574e-06, | |
| "loss": 0.0458, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 40.9, | |
| "eval_gen_len": 156.8718, | |
| "eval_loss": 1.1709474325180054, | |
| "eval_rouge1": 69.9022, | |
| "eval_rouge2": 53.8817, | |
| "eval_rougeL": 55.3029, | |
| "eval_rougeLsum": 59.6658, | |
| "eval_runtime": 5805.0163, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 41.61, | |
| "learning_rate": 1.7517630465444288e-06, | |
| "loss": 0.0417, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 42.31, | |
| "learning_rate": 1.7306064880112836e-06, | |
| "loss": 0.0421, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 42.31, | |
| "eval_gen_len": 161.1595, | |
| "eval_loss": 1.1962451934814453, | |
| "eval_rouge1": 69.8385, | |
| "eval_rouge2": 53.9521, | |
| "eval_rougeL": 55.043, | |
| "eval_rougeLsum": 59.4319, | |
| "eval_runtime": 5621.1651, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "learning_rate": 1.7094499294781383e-06, | |
| "loss": 0.0415, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 43.72, | |
| "learning_rate": 1.688293370944993e-06, | |
| "loss": 0.0378, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 43.72, | |
| "eval_gen_len": 161.1368, | |
| "eval_loss": 1.1894584894180298, | |
| "eval_rouge1": 69.2784, | |
| "eval_rouge2": 53.3896, | |
| "eval_rougeL": 54.7309, | |
| "eval_rougeLsum": 58.9909, | |
| "eval_runtime": 5848.8931, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 44.43, | |
| "learning_rate": 1.6671368124118478e-06, | |
| "loss": 0.0382, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 45.13, | |
| "learning_rate": 1.6459802538787025e-06, | |
| "loss": 0.0351, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 45.13, | |
| "eval_gen_len": 161.0627, | |
| "eval_loss": 1.2170298099517822, | |
| "eval_rouge1": 69.2182, | |
| "eval_rouge2": 53.2021, | |
| "eval_rougeL": 54.8192, | |
| "eval_rougeLsum": 58.858, | |
| "eval_runtime": 5747.8165, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 45.84, | |
| "learning_rate": 1.6248236953455572e-06, | |
| "loss": 0.0348, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 46.54, | |
| "learning_rate": 1.6036671368124117e-06, | |
| "loss": 0.0328, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 46.54, | |
| "eval_gen_len": 162.3647, | |
| "eval_loss": 1.2242525815963745, | |
| "eval_rouge1": 69.2151, | |
| "eval_rouge2": 53.5006, | |
| "eval_rougeL": 54.6811, | |
| "eval_rougeLsum": 58.899, | |
| "eval_runtime": 5815.6916, | |
| "eval_samples_per_second": 0.06, | |
| "eval_steps_per_second": 0.003, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 47.25, | |
| "learning_rate": 1.5825105782792665e-06, | |
| "loss": 0.0334, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 47.95, | |
| "learning_rate": 1.5613540197461212e-06, | |
| "loss": 0.0316, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 47.95, | |
| "eval_gen_len": 163.4872, | |
| "eval_loss": 1.2345460653305054, | |
| "eval_rouge1": 68.9079, | |
| "eval_rouge2": 52.7923, | |
| "eval_rougeL": 54.7024, | |
| "eval_rougeLsum": 59.1333, | |
| "eval_runtime": 5738.4019, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 48.66, | |
| "learning_rate": 1.540197461212976e-06, | |
| "loss": 0.0303, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 49.37, | |
| "learning_rate": 1.5190409026798307e-06, | |
| "loss": 0.0289, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 49.37, | |
| "eval_gen_len": 161.2165, | |
| "eval_loss": 1.2448240518569946, | |
| "eval_rouge1": 69.3435, | |
| "eval_rouge2": 53.5007, | |
| "eval_rougeL": 54.6771, | |
| "eval_rougeLsum": 59.067, | |
| "eval_runtime": 5523.8322, | |
| "eval_samples_per_second": 0.064, | |
| "eval_steps_per_second": 0.003, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 50.07, | |
| "learning_rate": 1.4978843441466856e-06, | |
| "loss": 0.0287, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 50.78, | |
| "learning_rate": 1.4767277856135403e-06, | |
| "loss": 0.0279, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 50.78, | |
| "eval_gen_len": 163.7664, | |
| "eval_loss": 1.254445195198059, | |
| "eval_rouge1": 68.8472, | |
| "eval_rouge2": 53.1054, | |
| "eval_rougeL": 54.3767, | |
| "eval_rougeLsum": 58.4479, | |
| "eval_runtime": 5626.7741, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 51.48, | |
| "learning_rate": 1.455571227080395e-06, | |
| "loss": 0.0266, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 52.19, | |
| "learning_rate": 1.4344146685472496e-06, | |
| "loss": 0.0266, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 52.19, | |
| "eval_gen_len": 163.3675, | |
| "eval_loss": 1.267912745475769, | |
| "eval_rouge1": 69.1523, | |
| "eval_rouge2": 53.3283, | |
| "eval_rougeL": 54.8059, | |
| "eval_rougeLsum": 58.9981, | |
| "eval_runtime": 5635.0806, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 52.89, | |
| "learning_rate": 1.4132581100141043e-06, | |
| "loss": 0.025, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 53.6, | |
| "learning_rate": 1.392101551480959e-06, | |
| "loss": 0.0237, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 53.6, | |
| "eval_gen_len": 161.943, | |
| "eval_loss": 1.278545618057251, | |
| "eval_rouge1": 69.7623, | |
| "eval_rouge2": 53.9089, | |
| "eval_rougeL": 55.4188, | |
| "eval_rougeLsum": 59.7174, | |
| "eval_runtime": 5517.6551, | |
| "eval_samples_per_second": 0.064, | |
| "eval_steps_per_second": 0.003, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 54.3, | |
| "learning_rate": 1.3709449929478138e-06, | |
| "loss": 0.0243, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 55.01, | |
| "learning_rate": 1.3497884344146687e-06, | |
| "loss": 0.0225, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 55.01, | |
| "eval_gen_len": 162.8234, | |
| "eval_loss": 1.2910034656524658, | |
| "eval_rouge1": 69.4568, | |
| "eval_rouge2": 53.5993, | |
| "eval_rougeL": 55.3547, | |
| "eval_rougeLsum": 59.7055, | |
| "eval_runtime": 5493.2949, | |
| "eval_samples_per_second": 0.064, | |
| "eval_steps_per_second": 0.003, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 55.71, | |
| "learning_rate": 1.3286318758815233e-06, | |
| "loss": 0.0222, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 56.42, | |
| "learning_rate": 1.307475317348378e-06, | |
| "loss": 0.0207, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 56.42, | |
| "eval_gen_len": 161.9145, | |
| "eval_loss": 1.302707314491272, | |
| "eval_rouge1": 69.3521, | |
| "eval_rouge2": 53.657, | |
| "eval_rougeL": 55.4158, | |
| "eval_rougeLsum": 59.7595, | |
| "eval_runtime": 5493.618, | |
| "eval_samples_per_second": 0.064, | |
| "eval_steps_per_second": 0.003, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 57.12, | |
| "learning_rate": 1.2863187588152327e-06, | |
| "loss": 0.0216, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 57.83, | |
| "learning_rate": 1.2651622002820875e-06, | |
| "loss": 0.0207, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 57.83, | |
| "eval_gen_len": 161.4017, | |
| "eval_loss": 1.2961533069610596, | |
| "eval_rouge1": 69.5378, | |
| "eval_rouge2": 53.6568, | |
| "eval_rougeL": 55.4833, | |
| "eval_rougeLsum": 59.8609, | |
| "eval_runtime": 5729.7323, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 58.53, | |
| "learning_rate": 1.2440056417489422e-06, | |
| "loss": 0.0202, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 59.24, | |
| "learning_rate": 1.222849083215797e-06, | |
| "loss": 0.019, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 59.24, | |
| "eval_gen_len": 161.5954, | |
| "eval_loss": 1.3159691095352173, | |
| "eval_rouge1": 69.6692, | |
| "eval_rouge2": 53.8824, | |
| "eval_rougeL": 56.153, | |
| "eval_rougeLsum": 60.5724, | |
| "eval_runtime": 5770.1856, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 59.94, | |
| "learning_rate": 1.2016925246826517e-06, | |
| "loss": 0.0193, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 60.65, | |
| "learning_rate": 1.1805359661495064e-06, | |
| "loss": 0.0183, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 60.65, | |
| "eval_gen_len": 159.6097, | |
| "eval_loss": 1.3009204864501953, | |
| "eval_rouge1": 69.8056, | |
| "eval_rouge2": 54.3016, | |
| "eval_rougeL": 56.2054, | |
| "eval_rougeLsum": 60.4544, | |
| "eval_runtime": 5689.5789, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 61.35, | |
| "learning_rate": 1.1593794076163611e-06, | |
| "loss": 0.0186, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 62.06, | |
| "learning_rate": 1.1382228490832158e-06, | |
| "loss": 0.0179, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 62.06, | |
| "eval_gen_len": 159.4046, | |
| "eval_loss": 1.3091576099395752, | |
| "eval_rouge1": 69.9596, | |
| "eval_rouge2": 54.1221, | |
| "eval_rougeL": 56.3196, | |
| "eval_rougeLsum": 60.6711, | |
| "eval_runtime": 5735.0694, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 62.76, | |
| "learning_rate": 1.1170662905500706e-06, | |
| "loss": 0.0179, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 63.47, | |
| "learning_rate": 1.0959097320169253e-06, | |
| "loss": 0.0162, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 63.47, | |
| "eval_gen_len": 161.7721, | |
| "eval_loss": 1.3298344612121582, | |
| "eval_rouge1": 69.7464, | |
| "eval_rouge2": 54.0211, | |
| "eval_rougeL": 56.0314, | |
| "eval_rougeLsum": 60.4478, | |
| "eval_runtime": 5746.6407, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 64.17, | |
| "learning_rate": 1.07475317348378e-06, | |
| "loss": 0.0177, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 64.88, | |
| "learning_rate": 1.0535966149506348e-06, | |
| "loss": 0.0161, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 64.88, | |
| "eval_gen_len": 161.3932, | |
| "eval_loss": 1.3238565921783447, | |
| "eval_rouge1": 69.6076, | |
| "eval_rouge2": 53.7339, | |
| "eval_rougeL": 55.9025, | |
| "eval_rougeLsum": 60.3293, | |
| "eval_runtime": 5658.4636, | |
| "eval_samples_per_second": 0.062, | |
| "eval_steps_per_second": 0.003, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 65.59, | |
| "learning_rate": 1.0324400564174895e-06, | |
| "loss": 0.0157, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 66.29, | |
| "learning_rate": 1.011283497884344e-06, | |
| "loss": 0.0152, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 66.29, | |
| "eval_gen_len": 162.8547, | |
| "eval_loss": 1.3345621824264526, | |
| "eval_rouge1": 69.2641, | |
| "eval_rouge2": 53.4141, | |
| "eval_rougeL": 55.3861, | |
| "eval_rougeLsum": 59.6768, | |
| "eval_runtime": 5757.5178, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "learning_rate": 9.901269393511988e-07, | |
| "loss": 0.0157, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 67.7, | |
| "learning_rate": 9.689703808180537e-07, | |
| "loss": 0.0149, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 67.7, | |
| "eval_gen_len": 163.0684, | |
| "eval_loss": 1.3516221046447754, | |
| "eval_rouge1": 69.3874, | |
| "eval_rouge2": 53.707, | |
| "eval_rougeL": 56.0318, | |
| "eval_rougeLsum": 60.1116, | |
| "eval_runtime": 5778.0873, | |
| "eval_samples_per_second": 0.061, | |
| "eval_steps_per_second": 0.003, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 68.41, | |
| "learning_rate": 9.478138222849083e-07, | |
| "loss": 0.014, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 69.11, | |
| "learning_rate": 9.266572637517632e-07, | |
| "loss": 0.0143, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 69.11, | |
| "eval_gen_len": 163.4387, | |
| "eval_loss": 1.3576686382293701, | |
| "eval_rouge1": 69.5713, | |
| "eval_rouge2": 53.8362, | |
| "eval_rougeL": 55.9625, | |
| "eval_rougeLsum": 60.2185, | |
| "eval_runtime": 6313.8347, | |
| "eval_samples_per_second": 0.056, | |
| "eval_steps_per_second": 0.003, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 69.82, | |
| "learning_rate": 9.055007052186177e-07, | |
| "loss": 0.0141, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 70.52, | |
| "learning_rate": 8.843441466854725e-07, | |
| "loss": 0.0139, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 70.52, | |
| "eval_gen_len": 161.9259, | |
| "eval_loss": 1.346762776374817, | |
| "eval_rouge1": 69.5609, | |
| "eval_rouge2": 53.9709, | |
| "eval_rougeL": 56.0658, | |
| "eval_rougeLsum": 60.0989, | |
| "eval_runtime": 5992.2258, | |
| "eval_samples_per_second": 0.059, | |
| "eval_steps_per_second": 0.003, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 71.23, | |
| "learning_rate": 8.631875881523273e-07, | |
| "loss": 0.0136, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 71.93, | |
| "learning_rate": 8.42031029619182e-07, | |
| "loss": 0.0132, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 71.93, | |
| "eval_gen_len": 161.208, | |
| "eval_loss": 1.3650970458984375, | |
| "eval_rouge1": 69.5418, | |
| "eval_rouge2": 53.7154, | |
| "eval_rougeL": 55.7361, | |
| "eval_rougeLsum": 60.0613, | |
| "eval_runtime": 7093.4012, | |
| "eval_samples_per_second": 0.049, | |
| "eval_steps_per_second": 0.003, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "learning_rate": 8.208744710860367e-07, | |
| "loss": 0.0131, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 73.34, | |
| "learning_rate": 7.997179125528913e-07, | |
| "loss": 0.0127, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 73.34, | |
| "eval_gen_len": 162.2849, | |
| "eval_loss": 1.3561229705810547, | |
| "eval_rouge1": 69.4318, | |
| "eval_rouge2": 53.5887, | |
| "eval_rougeL": 55.4159, | |
| "eval_rougeLsum": 59.6162, | |
| "eval_runtime": 6162.4321, | |
| "eval_samples_per_second": 0.057, | |
| "eval_steps_per_second": 0.003, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 74.05, | |
| "learning_rate": 7.785613540197461e-07, | |
| "loss": 0.0127, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 74.75, | |
| "learning_rate": 7.574047954866008e-07, | |
| "loss": 0.0123, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 74.75, | |
| "eval_gen_len": 159.567, | |
| "eval_loss": 1.3640151023864746, | |
| "eval_rouge1": 69.9555, | |
| "eval_rouge2": 54.4377, | |
| "eval_rougeL": 56.2914, | |
| "eval_rougeLsum": 60.5544, | |
| "eval_runtime": 6099.51, | |
| "eval_samples_per_second": 0.058, | |
| "eval_steps_per_second": 0.003, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 75.46, | |
| "learning_rate": 7.362482369534556e-07, | |
| "loss": 0.0123, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 76.16, | |
| "learning_rate": 7.150916784203103e-07, | |
| "loss": 0.0119, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 76.16, | |
| "eval_gen_len": 162.0228, | |
| "eval_loss": 1.3737815618515015, | |
| "eval_rouge1": 69.6472, | |
| "eval_rouge2": 53.9313, | |
| "eval_rougeL": 55.9586, | |
| "eval_rougeLsum": 59.9692, | |
| "eval_runtime": 6462.5491, | |
| "eval_samples_per_second": 0.054, | |
| "eval_steps_per_second": 0.003, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 76.87, | |
| "learning_rate": 6.93935119887165e-07, | |
| "loss": 0.0118, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 77.57, | |
| "learning_rate": 6.727785613540198e-07, | |
| "loss": 0.0117, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 77.57, | |
| "eval_gen_len": 162.3789, | |
| "eval_loss": 1.3787622451782227, | |
| "eval_rouge1": 69.5463, | |
| "eval_rouge2": 53.7206, | |
| "eval_rougeL": 56.0721, | |
| "eval_rougeLsum": 60.3372, | |
| "eval_runtime": 6234.7587, | |
| "eval_samples_per_second": 0.056, | |
| "eval_steps_per_second": 0.003, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 78.28, | |
| "learning_rate": 6.516220028208745e-07, | |
| "loss": 0.0114, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 78.98, | |
| "learning_rate": 6.304654442877292e-07, | |
| "loss": 0.0114, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 78.98, | |
| "eval_gen_len": 162.1083, | |
| "eval_loss": 1.3712390661239624, | |
| "eval_rouge1": 69.5879, | |
| "eval_rouge2": 53.9682, | |
| "eval_rougeL": 56.2965, | |
| "eval_rougeLsum": 60.5547, | |
| "eval_runtime": 6229.6909, | |
| "eval_samples_per_second": 0.056, | |
| "eval_steps_per_second": 0.003, | |
| "step": 56000 | |
| } | |
| ], | |
| "max_steps": 70900, | |
| "num_train_epochs": 100, | |
| "total_flos": 9.43193629959635e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |