| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 91107, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.983535842470941e-05, | |
| "loss": 4.3991, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.9670716849418816e-05, | |
| "loss": 1.3497, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.9506075274128224e-05, | |
| "loss": 1.0917, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.934143369883763e-05, | |
| "loss": 0.8889, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_exact_match": 18.4, | |
| "eval_f1": 18.8036, | |
| "eval_gen_len": 5.368, | |
| "eval_loss": 0.6919174194335938, | |
| "eval_rouge1": 18.1786, | |
| "eval_rouge2": 8.6387, | |
| "eval_rougeL": 18.1534, | |
| "eval_rougeLsum": 18.0938, | |
| "eval_runtime": 75.0695, | |
| "eval_samples_per_second": 13.321, | |
| "eval_steps_per_second": 4.449, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.917679212354704e-05, | |
| "loss": 0.8433, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.9012150548256447e-05, | |
| "loss": 0.8052, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.8847508972965855e-05, | |
| "loss": 0.8465, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 2.8682867397675263e-05, | |
| "loss": 0.7535, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_exact_match": 13.0, | |
| "eval_f1": 21.7903, | |
| "eval_gen_len": 5.483, | |
| "eval_loss": 0.5836785435676575, | |
| "eval_rouge1": 21.3055, | |
| "eval_rouge2": 9.2894, | |
| "eval_rougeL": 21.2178, | |
| "eval_rougeLsum": 21.3015, | |
| "eval_runtime": 76.4558, | |
| "eval_samples_per_second": 13.079, | |
| "eval_steps_per_second": 4.369, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.851822582238467e-05, | |
| "loss": 0.8144, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.8353584247094075e-05, | |
| "loss": 0.7393, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8188942671803483e-05, | |
| "loss": 0.7273, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.802430109651289e-05, | |
| "loss": 0.6731, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_exact_match": 11.9, | |
| "eval_f1": 21.595, | |
| "eval_gen_len": 5.467, | |
| "eval_loss": 0.5862796306610107, | |
| "eval_rouge1": 21.739, | |
| "eval_rouge2": 9.5334, | |
| "eval_rougeL": 21.7018, | |
| "eval_rougeLsum": 21.6884, | |
| "eval_runtime": 74.4712, | |
| "eval_samples_per_second": 13.428, | |
| "eval_steps_per_second": 4.485, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.7859659521222298e-05, | |
| "loss": 0.6948, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.769501794593171e-05, | |
| "loss": 0.6698, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.7530376370641117e-05, | |
| "loss": 0.6466, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.7365734795350525e-05, | |
| "loss": 0.6879, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_exact_match": 6.1, | |
| "eval_f1": 22.3924, | |
| "eval_gen_len": 5.95, | |
| "eval_loss": 0.5380619168281555, | |
| "eval_rouge1": 25.3106, | |
| "eval_rouge2": 11.8582, | |
| "eval_rougeL": 25.2639, | |
| "eval_rougeLsum": 25.2861, | |
| "eval_runtime": 76.6994, | |
| "eval_samples_per_second": 13.038, | |
| "eval_steps_per_second": 4.355, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.720109322005993e-05, | |
| "loss": 0.6504, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.7036451644769337e-05, | |
| "loss": 0.6834, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 2.6871810069478745e-05, | |
| "loss": 0.6759, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.6707168494188152e-05, | |
| "loss": 0.6459, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_exact_match": 8.1, | |
| "eval_f1": 22.3008, | |
| "eval_gen_len": 5.583, | |
| "eval_loss": 0.536933958530426, | |
| "eval_rouge1": 25.1248, | |
| "eval_rouge2": 11.4572, | |
| "eval_rougeL": 25.13, | |
| "eval_rougeLsum": 25.1636, | |
| "eval_runtime": 74.8336, | |
| "eval_samples_per_second": 13.363, | |
| "eval_steps_per_second": 4.463, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.654252691889756e-05, | |
| "loss": 0.6555, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.6377885343606968e-05, | |
| "loss": 0.622, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.6213243768316376e-05, | |
| "loss": 0.6434, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.6048602193025783e-05, | |
| "loss": 0.5865, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_exact_match": 7.8, | |
| "eval_f1": 22.5664, | |
| "eval_gen_len": 5.502, | |
| "eval_loss": 0.5340938568115234, | |
| "eval_rouge1": 25.6756, | |
| "eval_rouge2": 11.0984, | |
| "eval_rougeL": 25.6163, | |
| "eval_rougeLsum": 25.6348, | |
| "eval_runtime": 74.1837, | |
| "eval_samples_per_second": 13.48, | |
| "eval_steps_per_second": 4.502, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.588396061773519e-05, | |
| "loss": 0.573, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 2.57193190424446e-05, | |
| "loss": 0.6825, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 2.5554677467154007e-05, | |
| "loss": 0.5871, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 2.5390035891863414e-05, | |
| "loss": 0.6442, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_exact_match": 8.0, | |
| "eval_f1": 22.9873, | |
| "eval_gen_len": 5.694, | |
| "eval_loss": 0.5320589542388916, | |
| "eval_rouge1": 25.6004, | |
| "eval_rouge2": 11.31, | |
| "eval_rougeL": 25.6195, | |
| "eval_rougeLsum": 25.6027, | |
| "eval_runtime": 75.839, | |
| "eval_samples_per_second": 13.186, | |
| "eval_steps_per_second": 4.404, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.5225394316572822e-05, | |
| "loss": 0.6252, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 2.5060752741282227e-05, | |
| "loss": 0.613, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.4896111165991638e-05, | |
| "loss": 0.5827, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.4731469590701045e-05, | |
| "loss": 0.5819, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_exact_match": 6.6, | |
| "eval_f1": 23.22, | |
| "eval_gen_len": 5.789, | |
| "eval_loss": 0.5230505466461182, | |
| "eval_rouge1": 25.3937, | |
| "eval_rouge2": 11.1344, | |
| "eval_rougeL": 25.3304, | |
| "eval_rougeLsum": 25.3461, | |
| "eval_runtime": 75.9506, | |
| "eval_samples_per_second": 13.166, | |
| "eval_steps_per_second": 4.398, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.4566828015410453e-05, | |
| "loss": 0.6175, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.440218644011986e-05, | |
| "loss": 0.6132, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.423754486482927e-05, | |
| "loss": 0.6048, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.4072903289538673e-05, | |
| "loss": 0.5908, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_exact_match": 7.3, | |
| "eval_f1": 23.0102, | |
| "eval_gen_len": 5.769, | |
| "eval_loss": 0.5285995006561279, | |
| "eval_rouge1": 25.898, | |
| "eval_rouge2": 11.687, | |
| "eval_rougeL": 25.8492, | |
| "eval_rougeLsum": 25.8886, | |
| "eval_runtime": 76.3215, | |
| "eval_samples_per_second": 13.102, | |
| "eval_steps_per_second": 4.376, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.390826171424808e-05, | |
| "loss": 0.5972, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.374362013895749e-05, | |
| "loss": 0.5995, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.3578978563666896e-05, | |
| "loss": 0.6133, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 2.3414336988376307e-05, | |
| "loss": 0.5814, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_exact_match": 6.8, | |
| "eval_f1": 23.0215, | |
| "eval_gen_len": 5.956, | |
| "eval_loss": 0.5266488790512085, | |
| "eval_rouge1": 25.7545, | |
| "eval_rouge2": 11.6372, | |
| "eval_rougeL": 25.7025, | |
| "eval_rougeLsum": 25.6994, | |
| "eval_runtime": 77.5697, | |
| "eval_samples_per_second": 12.892, | |
| "eval_steps_per_second": 4.306, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 2.3249695413085715e-05, | |
| "loss": 0.5501, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 2.3085053837795123e-05, | |
| "loss": 0.5771, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.2920412262504527e-05, | |
| "loss": 0.5856, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 2.2755770687213935e-05, | |
| "loss": 0.5969, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_exact_match": 6.6, | |
| "eval_f1": 22.3945, | |
| "eval_gen_len": 5.784, | |
| "eval_loss": 0.5150488018989563, | |
| "eval_rouge1": 26.376, | |
| "eval_rouge2": 11.4531, | |
| "eval_rougeL": 26.3401, | |
| "eval_rougeLsum": 26.3444, | |
| "eval_runtime": 76.9076, | |
| "eval_samples_per_second": 13.003, | |
| "eval_steps_per_second": 4.343, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 2.2591129111923343e-05, | |
| "loss": 0.6051, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 2.242648753663275e-05, | |
| "loss": 0.6252, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 2.226184596134216e-05, | |
| "loss": 0.6002, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.2097204386051566e-05, | |
| "loss": 0.6071, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_exact_match": 7.5, | |
| "eval_f1": 22.429, | |
| "eval_gen_len": 5.577, | |
| "eval_loss": 0.5234242081642151, | |
| "eval_rouge1": 26.0675, | |
| "eval_rouge2": 11.2086, | |
| "eval_rougeL": 26.062, | |
| "eval_rougeLsum": 26.0623, | |
| "eval_runtime": 74.7901, | |
| "eval_samples_per_second": 13.371, | |
| "eval_steps_per_second": 4.466, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.1932562810760974e-05, | |
| "loss": 0.5639, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.176792123547038e-05, | |
| "loss": 0.5438, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.160327966017979e-05, | |
| "loss": 0.5473, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.1438638084889197e-05, | |
| "loss": 0.5743, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_exact_match": 6.1, | |
| "eval_f1": 22.4076, | |
| "eval_gen_len": 5.714, | |
| "eval_loss": 0.5406718850135803, | |
| "eval_rouge1": 27.3834, | |
| "eval_rouge2": 11.9993, | |
| "eval_rougeL": 27.3736, | |
| "eval_rougeLsum": 27.3494, | |
| "eval_runtime": 75.5929, | |
| "eval_samples_per_second": 13.229, | |
| "eval_steps_per_second": 4.418, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.1273996509598605e-05, | |
| "loss": 0.6123, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.1109354934308013e-05, | |
| "loss": 0.5416, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 2.094471335901742e-05, | |
| "loss": 0.5794, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.0780071783726825e-05, | |
| "loss": 0.5513, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_exact_match": 5.6, | |
| "eval_f1": 23.3636, | |
| "eval_gen_len": 5.841, | |
| "eval_loss": 0.516300618648529, | |
| "eval_rouge1": 27.5598, | |
| "eval_rouge2": 12.4259, | |
| "eval_rougeL": 27.494, | |
| "eval_rougeLsum": 27.5139, | |
| "eval_runtime": 75.7363, | |
| "eval_samples_per_second": 13.204, | |
| "eval_steps_per_second": 4.41, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.0615430208436232e-05, | |
| "loss": 0.5198, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.0450788633145644e-05, | |
| "loss": 0.5419, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.028614705785505e-05, | |
| "loss": 0.594, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.012150548256446e-05, | |
| "loss": 0.5916, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_exact_match": 6.5, | |
| "eval_f1": 23.2412, | |
| "eval_gen_len": 5.805, | |
| "eval_loss": 0.524455726146698, | |
| "eval_rouge1": 26.6244, | |
| "eval_rouge2": 11.8352, | |
| "eval_rougeL": 26.5432, | |
| "eval_rougeLsum": 26.5964, | |
| "eval_runtime": 76.4476, | |
| "eval_samples_per_second": 13.081, | |
| "eval_steps_per_second": 4.369, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.9956863907273867e-05, | |
| "loss": 0.5885, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.979222233198327e-05, | |
| "loss": 0.5536, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.962758075669268e-05, | |
| "loss": 0.5119, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.9462939181402087e-05, | |
| "loss": 0.5046, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_exact_match": 8.6, | |
| "eval_f1": 22.4289, | |
| "eval_gen_len": 5.578, | |
| "eval_loss": 0.5430198907852173, | |
| "eval_rouge1": 26.57, | |
| "eval_rouge2": 11.7677, | |
| "eval_rougeL": 26.6056, | |
| "eval_rougeLsum": 26.5575, | |
| "eval_runtime": 74.902, | |
| "eval_samples_per_second": 13.351, | |
| "eval_steps_per_second": 4.459, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.9298297606111494e-05, | |
| "loss": 0.5077, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.9133656030820902e-05, | |
| "loss": 0.5206, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.8969014455530313e-05, | |
| "loss": 0.5079, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.880437288023972e-05, | |
| "loss": 0.5349, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_exact_match": 5.4, | |
| "eval_f1": 23.2951, | |
| "eval_gen_len": 5.744, | |
| "eval_loss": 0.517284631729126, | |
| "eval_rouge1": 28.0046, | |
| "eval_rouge2": 12.2802, | |
| "eval_rougeL": 28.0025, | |
| "eval_rougeLsum": 27.9706, | |
| "eval_runtime": 75.9315, | |
| "eval_samples_per_second": 13.17, | |
| "eval_steps_per_second": 4.399, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.8639731304949125e-05, | |
| "loss": 0.4871, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.8475089729658533e-05, | |
| "loss": 0.541, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.831044815436794e-05, | |
| "loss": 0.5659, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.814580657907735e-05, | |
| "loss": 0.5231, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_exact_match": 6.1, | |
| "eval_f1": 23.2074, | |
| "eval_gen_len": 5.814, | |
| "eval_loss": 0.5330841541290283, | |
| "eval_rouge1": 27.4091, | |
| "eval_rouge2": 12.3926, | |
| "eval_rougeL": 27.3466, | |
| "eval_rougeLsum": 27.3708, | |
| "eval_runtime": 77.6684, | |
| "eval_samples_per_second": 12.875, | |
| "eval_steps_per_second": 4.3, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.7981165003786756e-05, | |
| "loss": 0.5209, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.7816523428496164e-05, | |
| "loss": 0.5561, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.7651881853205572e-05, | |
| "loss": 0.5269, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.748724027791498e-05, | |
| "loss": 0.5144, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_exact_match": 6.6, | |
| "eval_f1": 22.9561, | |
| "eval_gen_len": 5.764, | |
| "eval_loss": 0.533650815486908, | |
| "eval_rouge1": 27.7074, | |
| "eval_rouge2": 12.5122, | |
| "eval_rougeL": 27.6729, | |
| "eval_rougeLsum": 27.6909, | |
| "eval_runtime": 76.8102, | |
| "eval_samples_per_second": 13.019, | |
| "eval_steps_per_second": 4.348, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.7322598702624387e-05, | |
| "loss": 0.4877, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.7157957127333795e-05, | |
| "loss": 0.5182, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.6993315552043203e-05, | |
| "loss": 0.4982, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.682867397675261e-05, | |
| "loss": 0.5576, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_exact_match": 7.0, | |
| "eval_f1": 22.646, | |
| "eval_gen_len": 5.801, | |
| "eval_loss": 0.5152503848075867, | |
| "eval_rouge1": 27.8201, | |
| "eval_rouge2": 12.335, | |
| "eval_rougeL": 27.806, | |
| "eval_rougeLsum": 27.7887, | |
| "eval_runtime": 77.4695, | |
| "eval_samples_per_second": 12.908, | |
| "eval_steps_per_second": 4.311, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.666403240146202e-05, | |
| "loss": 0.5032, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.6499390826171423e-05, | |
| "loss": 0.5397, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.633474925088083e-05, | |
| "loss": 0.4944, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.617010767559024e-05, | |
| "loss": 0.5184, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_exact_match": 4.8, | |
| "eval_f1": 23.3492, | |
| "eval_gen_len": 5.988, | |
| "eval_loss": 0.5299638509750366, | |
| "eval_rouge1": 28.1638, | |
| "eval_rouge2": 12.5269, | |
| "eval_rougeL": 28.1245, | |
| "eval_rougeLsum": 28.1199, | |
| "eval_runtime": 77.8204, | |
| "eval_samples_per_second": 12.85, | |
| "eval_steps_per_second": 4.292, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.600546610029965e-05, | |
| "loss": 0.488, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.5840824525009057e-05, | |
| "loss": 0.5176, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.5676182949718465e-05, | |
| "loss": 0.4941, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.5511541374427873e-05, | |
| "loss": 0.5321, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_exact_match": 5.2, | |
| "eval_f1": 23.1651, | |
| "eval_gen_len": 5.997, | |
| "eval_loss": 0.5239735245704651, | |
| "eval_rouge1": 27.9328, | |
| "eval_rouge2": 12.5264, | |
| "eval_rougeL": 27.9238, | |
| "eval_rougeLsum": 27.9206, | |
| "eval_runtime": 78.5078, | |
| "eval_samples_per_second": 12.738, | |
| "eval_steps_per_second": 4.254, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.5346899799137277e-05, | |
| "loss": 0.5, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.5182258223846685e-05, | |
| "loss": 0.5279, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.5017616648556093e-05, | |
| "loss": 0.5223, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.4852975073265502e-05, | |
| "loss": 0.478, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_exact_match": 6.4, | |
| "eval_f1": 22.7142, | |
| "eval_gen_len": 5.824, | |
| "eval_loss": 0.5349439382553101, | |
| "eval_rouge1": 27.9589, | |
| "eval_rouge2": 12.5606, | |
| "eval_rougeL": 27.902, | |
| "eval_rougeLsum": 27.8978, | |
| "eval_runtime": 76.8435, | |
| "eval_samples_per_second": 13.013, | |
| "eval_steps_per_second": 4.346, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.4688333497974908e-05, | |
| "loss": 0.482, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.4523691922684316e-05, | |
| "loss": 0.5004, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.4359050347393724e-05, | |
| "loss": 0.5482, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.4194408772103133e-05, | |
| "loss": 0.497, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_exact_match": 5.5, | |
| "eval_f1": 22.6861, | |
| "eval_gen_len": 5.864, | |
| "eval_loss": 0.5399107933044434, | |
| "eval_rouge1": 28.2327, | |
| "eval_rouge2": 12.5728, | |
| "eval_rougeL": 28.1968, | |
| "eval_rougeLsum": 28.21, | |
| "eval_runtime": 77.9, | |
| "eval_samples_per_second": 12.837, | |
| "eval_steps_per_second": 4.288, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.4029767196812539e-05, | |
| "loss": 0.4476, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1.3865125621521947e-05, | |
| "loss": 0.5754, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.3700484046231355e-05, | |
| "loss": 0.5296, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.3535842470940762e-05, | |
| "loss": 0.5475, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_exact_match": 8.0, | |
| "eval_f1": 22.2882, | |
| "eval_gen_len": 5.651, | |
| "eval_loss": 0.5568350553512573, | |
| "eval_rouge1": 27.6, | |
| "eval_rouge2": 12.259, | |
| "eval_rougeL": 27.6081, | |
| "eval_rougeLsum": 27.561, | |
| "eval_runtime": 76.4106, | |
| "eval_samples_per_second": 13.087, | |
| "eval_steps_per_second": 4.371, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.337120089565017e-05, | |
| "loss": 0.492, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1.3206559320359578e-05, | |
| "loss": 0.5221, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 1.3041917745068984e-05, | |
| "loss": 0.5491, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1.2877276169778392e-05, | |
| "loss": 0.4643, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_exact_match": 8.9, | |
| "eval_f1": 21.3228, | |
| "eval_gen_len": 5.566, | |
| "eval_loss": 0.5837388038635254, | |
| "eval_rouge1": 27.0269, | |
| "eval_rouge2": 11.9442, | |
| "eval_rougeL": 27.0247, | |
| "eval_rougeLsum": 26.9566, | |
| "eval_runtime": 76.2944, | |
| "eval_samples_per_second": 13.107, | |
| "eval_steps_per_second": 4.378, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.2712634594487801e-05, | |
| "loss": 0.5079, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 1.2547993019197209e-05, | |
| "loss": 0.4905, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.2383351443906615e-05, | |
| "loss": 0.4939, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.2218709868616023e-05, | |
| "loss": 0.5333, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_exact_match": 5.5, | |
| "eval_f1": 23.1489, | |
| "eval_gen_len": 5.764, | |
| "eval_loss": 0.5246403813362122, | |
| "eval_rouge1": 27.9369, | |
| "eval_rouge2": 12.2044, | |
| "eval_rougeL": 27.8886, | |
| "eval_rougeLsum": 27.9162, | |
| "eval_runtime": 76.3987, | |
| "eval_samples_per_second": 13.089, | |
| "eval_steps_per_second": 4.372, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.2054068293325432e-05, | |
| "loss": 0.5067, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.1889426718034838e-05, | |
| "loss": 0.4915, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.1724785142744246e-05, | |
| "loss": 0.4855, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.1560143567453654e-05, | |
| "loss": 0.5356, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_exact_match": 7.5, | |
| "eval_f1": 21.9492, | |
| "eval_gen_len": 5.628, | |
| "eval_loss": 0.559809684753418, | |
| "eval_rouge1": 27.8669, | |
| "eval_rouge2": 12.3269, | |
| "eval_rougeL": 27.8231, | |
| "eval_rougeLsum": 27.8187, | |
| "eval_runtime": 75.9003, | |
| "eval_samples_per_second": 13.175, | |
| "eval_steps_per_second": 4.401, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.1395501992163061e-05, | |
| "loss": 0.5376, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.123086041687247e-05, | |
| "loss": 0.5152, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.1066218841581877e-05, | |
| "loss": 0.5355, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.0901577266291283e-05, | |
| "loss": 0.5154, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_exact_match": 5.5, | |
| "eval_f1": 22.8791, | |
| "eval_gen_len": 5.859, | |
| "eval_loss": 0.5113182663917542, | |
| "eval_rouge1": 28.1362, | |
| "eval_rouge2": 12.2055, | |
| "eval_rougeL": 28.0813, | |
| "eval_rougeLsum": 28.0901, | |
| "eval_runtime": 78.3738, | |
| "eval_samples_per_second": 12.759, | |
| "eval_steps_per_second": 4.262, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 1.073693569100069e-05, | |
| "loss": 0.5391, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.05722941157101e-05, | |
| "loss": 0.4996, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.0407652540419508e-05, | |
| "loss": 0.5221, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.0243010965128914e-05, | |
| "loss": 0.4974, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_exact_match": 7.7, | |
| "eval_f1": 22.1627, | |
| "eval_gen_len": 5.622, | |
| "eval_loss": 0.529927670955658, | |
| "eval_rouge1": 27.4181, | |
| "eval_rouge2": 11.8982, | |
| "eval_rougeL": 27.3595, | |
| "eval_rougeLsum": 27.362, | |
| "eval_runtime": 76.1366, | |
| "eval_samples_per_second": 13.134, | |
| "eval_steps_per_second": 4.387, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.0078369389838322e-05, | |
| "loss": 0.4769, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.913727814547731e-06, | |
| "loss": 0.5019, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.749086239257137e-06, | |
| "loss": 0.4819, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.584444663966545e-06, | |
| "loss": 0.4368, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_exact_match": 4.9, | |
| "eval_f1": 23.2707, | |
| "eval_gen_len": 5.85, | |
| "eval_loss": 0.533755362033844, | |
| "eval_rouge1": 28.3591, | |
| "eval_rouge2": 12.5844, | |
| "eval_rougeL": 28.2897, | |
| "eval_rougeLsum": 28.3187, | |
| "eval_runtime": 77.3013, | |
| "eval_samples_per_second": 12.936, | |
| "eval_steps_per_second": 4.321, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.419803088675953e-06, | |
| "loss": 0.5444, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.255161513385359e-06, | |
| "loss": 0.4618, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.090519938094768e-06, | |
| "loss": 0.4942, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 8.925878362804176e-06, | |
| "loss": 0.4685, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_exact_match": 5.4, | |
| "eval_f1": 23.1173, | |
| "eval_gen_len": 5.868, | |
| "eval_loss": 0.5359737277030945, | |
| "eval_rouge1": 28.1371, | |
| "eval_rouge2": 12.7219, | |
| "eval_rougeL": 28.0627, | |
| "eval_rougeLsum": 28.1156, | |
| "eval_runtime": 77.0613, | |
| "eval_samples_per_second": 12.977, | |
| "eval_steps_per_second": 4.334, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 8.761236787513582e-06, | |
| "loss": 0.4914, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 8.59659521222299e-06, | |
| "loss": 0.519, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 8.4319536369324e-06, | |
| "loss": 0.4865, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 8.267312061641807e-06, | |
| "loss": 0.4889, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_exact_match": 5.8, | |
| "eval_f1": 23.0678, | |
| "eval_gen_len": 5.791, | |
| "eval_loss": 0.550064206123352, | |
| "eval_rouge1": 28.0394, | |
| "eval_rouge2": 12.3157, | |
| "eval_rougeL": 27.9714, | |
| "eval_rougeLsum": 27.9793, | |
| "eval_runtime": 76.4779, | |
| "eval_samples_per_second": 13.076, | |
| "eval_steps_per_second": 4.367, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 8.102670486351213e-06, | |
| "loss": 0.4684, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 7.938028911060621e-06, | |
| "loss": 0.4699, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 7.773387335770029e-06, | |
| "loss": 0.5216, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 7.608745760479437e-06, | |
| "loss": 0.4384, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_exact_match": 5.4, | |
| "eval_f1": 23.1485, | |
| "eval_gen_len": 5.799, | |
| "eval_loss": 0.5489073991775513, | |
| "eval_rouge1": 28.3405, | |
| "eval_rouge2": 12.6721, | |
| "eval_rougeL": 28.3062, | |
| "eval_rougeLsum": 28.3124, | |
| "eval_runtime": 76.6915, | |
| "eval_samples_per_second": 13.039, | |
| "eval_steps_per_second": 4.355, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.444104185188844e-06, | |
| "loss": 0.48, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.279462609898251e-06, | |
| "loss": 0.4582, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.11482103460766e-06, | |
| "loss": 0.4575, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 6.9501794593170666e-06, | |
| "loss": 0.4588, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_exact_match": 5.6, | |
| "eval_f1": 23.2198, | |
| "eval_gen_len": 5.809, | |
| "eval_loss": 0.5535086393356323, | |
| "eval_rouge1": 28.2308, | |
| "eval_rouge2": 12.4573, | |
| "eval_rougeL": 28.1886, | |
| "eval_rougeLsum": 28.1639, | |
| "eval_runtime": 76.7297, | |
| "eval_samples_per_second": 13.033, | |
| "eval_steps_per_second": 4.353, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 6.785537884026475e-06, | |
| "loss": 0.4737, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 6.620896308735882e-06, | |
| "loss": 0.4799, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 6.45625473344529e-06, | |
| "loss": 0.4545, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 6.2916131581546976e-06, | |
| "loss": 0.5076, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_exact_match": 6.7, | |
| "eval_f1": 22.6072, | |
| "eval_gen_len": 5.739, | |
| "eval_loss": 0.5515512824058533, | |
| "eval_rouge1": 28.0709, | |
| "eval_rouge2": 12.4522, | |
| "eval_rougeL": 27.9586, | |
| "eval_rougeLsum": 27.9825, | |
| "eval_runtime": 76.3124, | |
| "eval_samples_per_second": 13.104, | |
| "eval_steps_per_second": 4.377, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 6.126971582864105e-06, | |
| "loss": 0.442, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 5.962330007573513e-06, | |
| "loss": 0.4464, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 5.79768843228292e-06, | |
| "loss": 0.4756, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 5.633046856992328e-06, | |
| "loss": 0.4467, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_exact_match": 5.2, | |
| "eval_f1": 22.9357, | |
| "eval_gen_len": 5.876, | |
| "eval_loss": 0.5473840832710266, | |
| "eval_rouge1": 28.2468, | |
| "eval_rouge2": 12.4971, | |
| "eval_rougeL": 28.1601, | |
| "eval_rougeLsum": 28.1898, | |
| "eval_runtime": 77.1004, | |
| "eval_samples_per_second": 12.97, | |
| "eval_steps_per_second": 4.332, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 5.4684052817017355e-06, | |
| "loss": 0.5136, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 5.303763706411143e-06, | |
| "loss": 0.4626, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 5.13912213112055e-06, | |
| "loss": 0.5138, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.974480555829959e-06, | |
| "loss": 0.4262, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_exact_match": 5.5, | |
| "eval_f1": 23.0092, | |
| "eval_gen_len": 5.883, | |
| "eval_loss": 0.5560488700866699, | |
| "eval_rouge1": 28.3505, | |
| "eval_rouge2": 12.633, | |
| "eval_rougeL": 28.2747, | |
| "eval_rougeLsum": 28.2859, | |
| "eval_runtime": 77.2929, | |
| "eval_samples_per_second": 12.938, | |
| "eval_steps_per_second": 4.321, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 4.809838980539366e-06, | |
| "loss": 0.4773, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 4.645197405248774e-06, | |
| "loss": 0.4813, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.480555829958181e-06, | |
| "loss": 0.4833, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.315914254667588e-06, | |
| "loss": 0.4812, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_exact_match": 5.5, | |
| "eval_f1": 23.0871, | |
| "eval_gen_len": 5.812, | |
| "eval_loss": 0.5406239032745361, | |
| "eval_rouge1": 28.1686, | |
| "eval_rouge2": 12.4512, | |
| "eval_rougeL": 28.1366, | |
| "eval_rougeLsum": 28.1438, | |
| "eval_runtime": 76.6212, | |
| "eval_samples_per_second": 13.051, | |
| "eval_steps_per_second": 4.359, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.151272679376997e-06, | |
| "loss": 0.4662, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 3.9866311040864035e-06, | |
| "loss": 0.4939, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.821989528795812e-06, | |
| "loss": 0.5221, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3.6573479535052195e-06, | |
| "loss": 0.4243, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_exact_match": 5.2, | |
| "eval_f1": 22.7384, | |
| "eval_gen_len": 5.908, | |
| "eval_loss": 0.5526044964790344, | |
| "eval_rouge1": 28.0823, | |
| "eval_rouge2": 12.5204, | |
| "eval_rougeL": 28.0028, | |
| "eval_rougeLsum": 28.0533, | |
| "eval_runtime": 77.3998, | |
| "eval_samples_per_second": 12.92, | |
| "eval_steps_per_second": 4.315, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3.492706378214627e-06, | |
| "loss": 0.4704, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.328064802924034e-06, | |
| "loss": 0.4731, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.163423227633442e-06, | |
| "loss": 0.4775, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 2.9987816523428496e-06, | |
| "loss": 0.4828, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_exact_match": 6.7, | |
| "eval_f1": 22.2536, | |
| "eval_gen_len": 5.75, | |
| "eval_loss": 0.5606855154037476, | |
| "eval_rouge1": 27.9073, | |
| "eval_rouge2": 12.4149, | |
| "eval_rougeL": 27.8389, | |
| "eval_rougeLsum": 27.8483, | |
| "eval_runtime": 76.3805, | |
| "eval_samples_per_second": 13.092, | |
| "eval_steps_per_second": 4.373, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.8341400770522574e-06, | |
| "loss": 0.4479, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 2.669498501761665e-06, | |
| "loss": 0.4923, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 2.5048569264710725e-06, | |
| "loss": 0.5143, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.3402153511804802e-06, | |
| "loss": 0.4827, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_exact_match": 4.8, | |
| "eval_f1": 23.1203, | |
| "eval_gen_len": 5.883, | |
| "eval_loss": 0.5355603694915771, | |
| "eval_rouge1": 28.5515, | |
| "eval_rouge2": 12.5974, | |
| "eval_rougeL": 28.4917, | |
| "eval_rougeLsum": 28.4884, | |
| "eval_runtime": 77.1315, | |
| "eval_samples_per_second": 12.965, | |
| "eval_steps_per_second": 4.33, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2.175573775889888e-06, | |
| "loss": 0.453, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.0109322005992957e-06, | |
| "loss": 0.4689, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.846290625308703e-06, | |
| "loss": 0.4628, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.6816490500181106e-06, | |
| "loss": 0.4599, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_exact_match": 6.0, | |
| "eval_f1": 22.775, | |
| "eval_gen_len": 5.75, | |
| "eval_loss": 0.5486857891082764, | |
| "eval_rouge1": 28.1728, | |
| "eval_rouge2": 12.3968, | |
| "eval_rougeL": 28.1105, | |
| "eval_rougeLsum": 28.1316, | |
| "eval_runtime": 76.2649, | |
| "eval_samples_per_second": 13.112, | |
| "eval_steps_per_second": 4.379, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.5170074747275183e-06, | |
| "loss": 0.4783, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 1.3523658994369257e-06, | |
| "loss": 0.4793, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 1.1877243241463334e-06, | |
| "loss": 0.4457, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.0230827488557412e-06, | |
| "loss": 0.4505, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_exact_match": 5.3, | |
| "eval_f1": 23.063, | |
| "eval_gen_len": 5.832, | |
| "eval_loss": 0.5459786057472229, | |
| "eval_rouge1": 28.381, | |
| "eval_rouge2": 12.6567, | |
| "eval_rougeL": 28.3519, | |
| "eval_rougeLsum": 28.3424, | |
| "eval_runtime": 76.7382, | |
| "eval_samples_per_second": 13.031, | |
| "eval_steps_per_second": 4.352, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 8.584411735651487e-07, | |
| "loss": 0.4594, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 6.937995982745563e-07, | |
| "loss": 0.4719, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 5.291580229839639e-07, | |
| "loss": 0.4538, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 3.6451644769337155e-07, | |
| "loss": 0.4741, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_exact_match": 5.3, | |
| "eval_f1": 23.0565, | |
| "eval_gen_len": 5.819, | |
| "eval_loss": 0.5445623993873596, | |
| "eval_rouge1": 28.2709, | |
| "eval_rouge2": 12.4988, | |
| "eval_rougeL": 28.1962, | |
| "eval_rougeLsum": 28.2203, | |
| "eval_runtime": 76.5919, | |
| "eval_samples_per_second": 13.056, | |
| "eval_steps_per_second": 4.361, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.9987487240277914e-07, | |
| "loss": 0.441, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.5233297112186775e-08, | |
| "loss": 0.4993, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 91107, | |
| "total_flos": 1.4035529182858445e+17, | |
| "train_loss": 0.5681342689636171, | |
| "train_runtime": 25047.2432, | |
| "train_samples_per_second": 10.912, | |
| "train_steps_per_second": 3.637 | |
| } | |
| ], | |
| "max_steps": 91107, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.4035529182858445e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |