| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 90000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.9889333333333336e-05, | |
| "loss": 0.6763, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.9778222222222224e-05, | |
| "loss": 0.511, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.9667333333333334e-05, | |
| "loss": 0.4773, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.9556222222222222e-05, | |
| "loss": 0.4519, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.9445111111111114e-05, | |
| "loss": 0.4367, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.9334e-05, | |
| "loss": 0.4295, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.922288888888889e-05, | |
| "loss": 0.4105, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.911177777777778e-05, | |
| "loss": 0.4063, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.9000666666666666e-05, | |
| "loss": 0.3997, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.8889555555555557e-05, | |
| "loss": 0.3921, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.8778444444444445e-05, | |
| "loss": 0.3857, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.8667333333333337e-05, | |
| "loss": 0.3852, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.8556222222222225e-05, | |
| "loss": 0.3754, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.8445111111111113e-05, | |
| "loss": 0.373, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.8334222222222226e-05, | |
| "loss": 0.3731, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8223111111111114e-05, | |
| "loss": 0.3745, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.8112222222222224e-05, | |
| "loss": 0.3698, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.8001111111111115e-05, | |
| "loss": 0.3717, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 14.0401, | |
| "eval_loss": 0.30105069279670715, | |
| "eval_rouge1": 0.8165, | |
| "eval_rouge2": 0.7413, | |
| "eval_rougeL": 0.8143, | |
| "eval_rougeLsum": 0.8143, | |
| "eval_runtime": 2090.1403, | |
| "eval_samples_per_second": 17.224, | |
| "eval_steps_per_second": 4.306, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.789e-05, | |
| "loss": 0.359, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.777888888888889e-05, | |
| "loss": 0.3557, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.766777777777778e-05, | |
| "loss": 0.3485, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.7556666666666667e-05, | |
| "loss": 0.3501, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.744555555555556e-05, | |
| "loss": 0.3504, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.7334444444444444e-05, | |
| "loss": 0.3558, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.7223555555555557e-05, | |
| "loss": 0.3557, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.7112444444444445e-05, | |
| "loss": 0.3425, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.7001333333333333e-05, | |
| "loss": 0.3353, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.6890222222222224e-05, | |
| "loss": 0.3356, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1.6779333333333334e-05, | |
| "loss": 0.3397, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 1.6668222222222222e-05, | |
| "loss": 0.3327, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1.6557111111111113e-05, | |
| "loss": 0.343, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.6446e-05, | |
| "loss": 0.3357, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.6334888888888893e-05, | |
| "loss": 0.3359, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.6224000000000003e-05, | |
| "loss": 0.3307, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.611288888888889e-05, | |
| "loss": 0.337, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6002000000000004e-05, | |
| "loss": 0.3255, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 13.8117, | |
| "eval_loss": 0.2776668965816498, | |
| "eval_rouge1": 0.8238, | |
| "eval_rouge2": 0.7501, | |
| "eval_rougeL": 0.8217, | |
| "eval_rougeLsum": 0.8217, | |
| "eval_runtime": 1917.0267, | |
| "eval_samples_per_second": 18.779, | |
| "eval_steps_per_second": 4.695, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.5890888888888892e-05, | |
| "loss": 0.3292, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.577977777777778e-05, | |
| "loss": 0.3246, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.5668666666666668e-05, | |
| "loss": 0.3213, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.5557555555555556e-05, | |
| "loss": 0.3247, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.5446444444444447e-05, | |
| "loss": 0.3199, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.5335555555555557e-05, | |
| "loss": 0.3255, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.5224444444444447e-05, | |
| "loss": 0.3236, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.5113333333333335e-05, | |
| "loss": 0.3151, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.5002222222222223e-05, | |
| "loss": 0.3174, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 1.4891111111111111e-05, | |
| "loss": 0.3197, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.478e-05, | |
| "loss": 0.3206, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.466888888888889e-05, | |
| "loss": 0.3182, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.4558e-05, | |
| "loss": 0.3156, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.444688888888889e-05, | |
| "loss": 0.321, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.433577777777778e-05, | |
| "loss": 0.3166, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.422466666666667e-05, | |
| "loss": 0.3139, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.4113555555555556e-05, | |
| "loss": 0.314, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.4002444444444446e-05, | |
| "loss": 0.3163, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 13.8253, | |
| "eval_loss": 0.2683161795139313, | |
| "eval_rouge1": 0.8272, | |
| "eval_rouge2": 0.7541, | |
| "eval_rougeL": 0.8252, | |
| "eval_rougeLsum": 0.8251, | |
| "eval_runtime": 2009.2942, | |
| "eval_samples_per_second": 17.917, | |
| "eval_steps_per_second": 4.479, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 1.3891333333333334e-05, | |
| "loss": 0.3092, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 1.3780222222222223e-05, | |
| "loss": 0.3068, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.3669333333333335e-05, | |
| "loss": 0.3124, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 1.3558222222222223e-05, | |
| "loss": 0.3059, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 1.3447111111111113e-05, | |
| "loss": 0.3098, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.3336e-05, | |
| "loss": 0.3053, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.3224888888888889e-05, | |
| "loss": 0.3083, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.3113777777777779e-05, | |
| "loss": 0.3095, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 1.3002666666666668e-05, | |
| "loss": 0.3059, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.2891555555555556e-05, | |
| "loss": 0.3038, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.2780666666666668e-05, | |
| "loss": 0.3116, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.2669555555555558e-05, | |
| "loss": 0.3032, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.2558666666666667e-05, | |
| "loss": 0.3046, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.2447555555555557e-05, | |
| "loss": 0.3021, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 1.2336444444444447e-05, | |
| "loss": 0.3044, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.2225333333333333e-05, | |
| "loss": 0.3034, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 1.2114222222222223e-05, | |
| "loss": 0.3026, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.2003111111111113e-05, | |
| "loss": 0.3001, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 13.8589, | |
| "eval_loss": 0.2630792558193207, | |
| "eval_rouge1": 0.8295, | |
| "eval_rouge2": 0.7568, | |
| "eval_rougeL": 0.8275, | |
| "eval_rougeLsum": 0.8275, | |
| "eval_runtime": 2033.7364, | |
| "eval_samples_per_second": 17.701, | |
| "eval_steps_per_second": 4.425, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 1.1892e-05, | |
| "loss": 0.2956, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 1.178088888888889e-05, | |
| "loss": 0.3012, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.1670000000000002e-05, | |
| "loss": 0.2974, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 1.155888888888889e-05, | |
| "loss": 0.2904, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.1447777777777778e-05, | |
| "loss": 0.298, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 1.1336666666666668e-05, | |
| "loss": 0.2957, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 1.1225777777777777e-05, | |
| "loss": 0.296, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 1.1114666666666667e-05, | |
| "loss": 0.2955, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 1.1003555555555557e-05, | |
| "loss": 0.2973, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 1.0892444444444445e-05, | |
| "loss": 0.2963, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 1.0781555555555556e-05, | |
| "loss": 0.2977, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 1.0670444444444446e-05, | |
| "loss": 0.2956, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0559555555555556e-05, | |
| "loss": 0.2966, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 1.0448444444444445e-05, | |
| "loss": 0.2969, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 1.0337333333333335e-05, | |
| "loss": 0.2969, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.0226222222222223e-05, | |
| "loss": 0.2945, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 1.0115111111111111e-05, | |
| "loss": 0.2923, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.0004000000000001e-05, | |
| "loss": 0.2957, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 13.8742, | |
| "eval_loss": 0.25948867201805115, | |
| "eval_rouge1": 0.8307, | |
| "eval_rouge2": 0.759, | |
| "eval_rougeL": 0.8288, | |
| "eval_rougeLsum": 0.8288, | |
| "eval_runtime": 2061.1131, | |
| "eval_samples_per_second": 17.466, | |
| "eval_steps_per_second": 4.367, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 9.89288888888889e-06, | |
| "loss": 0.2965, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 9.781777777777779e-06, | |
| "loss": 0.2858, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 9.67088888888889e-06, | |
| "loss": 0.2933, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 9.559777777777778e-06, | |
| "loss": 0.2913, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 9.448666666666666e-06, | |
| "loss": 0.2896, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 9.337555555555556e-06, | |
| "loss": 0.2918, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 9.226888888888889e-06, | |
| "loss": 0.2908, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 9.115777777777779e-06, | |
| "loss": 0.2916, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 9.004666666666667e-06, | |
| "loss": 0.284, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 8.893555555555557e-06, | |
| "loss": 0.2853, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 8.782444444444446e-06, | |
| "loss": 0.2885, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 8.671555555555556e-06, | |
| "loss": 0.2878, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 8.560444444444446e-06, | |
| "loss": 0.2909, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 8.449333333333334e-06, | |
| "loss": 0.2845, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 8.338222222222224e-06, | |
| "loss": 0.2879, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 8.227111111111112e-06, | |
| "loss": 0.2843, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 8.116e-06, | |
| "loss": 0.289, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.00488888888889e-06, | |
| "loss": 0.2879, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 13.932, | |
| "eval_loss": 0.2565445601940155, | |
| "eval_rouge1": 0.8315, | |
| "eval_rouge2": 0.7594, | |
| "eval_rougeL": 0.8296, | |
| "eval_rougeLsum": 0.8296, | |
| "eval_runtime": 2068.5889, | |
| "eval_samples_per_second": 17.403, | |
| "eval_steps_per_second": 4.351, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 7.893777777777778e-06, | |
| "loss": 0.2886, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 7.782888888888889e-06, | |
| "loss": 0.2886, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 7.671777777777779e-06, | |
| "loss": 0.2844, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 7.560666666666667e-06, | |
| "loss": 0.2814, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 7.449555555555557e-06, | |
| "loss": 0.2846, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 7.338666666666667e-06, | |
| "loss": 0.2782, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 7.227555555555557e-06, | |
| "loss": 0.2793, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 7.116444444444445e-06, | |
| "loss": 0.2822, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 7.005333333333334e-06, | |
| "loss": 0.2838, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 6.894444444444444e-06, | |
| "loss": 0.2871, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 6.783333333333334e-06, | |
| "loss": 0.2815, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 6.672222222222223e-06, | |
| "loss": 0.2819, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 6.561333333333334e-06, | |
| "loss": 0.2874, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 6.450222222222223e-06, | |
| "loss": 0.2911, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 6.339111111111111e-06, | |
| "loss": 0.279, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 6.228e-06, | |
| "loss": 0.2855, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 6.116888888888889e-06, | |
| "loss": 0.2887, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 6.005777777777778e-06, | |
| "loss": 0.2824, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 13.8921, | |
| "eval_loss": 0.2553006708621979, | |
| "eval_rouge1": 0.8322, | |
| "eval_rouge2": 0.7607, | |
| "eval_rougeL": 0.8302, | |
| "eval_rougeLsum": 0.8303, | |
| "eval_runtime": 2073.9344, | |
| "eval_samples_per_second": 17.358, | |
| "eval_steps_per_second": 4.34, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 5.894666666666667e-06, | |
| "loss": 0.2793, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 5.783555555555556e-06, | |
| "loss": 0.2815, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 5.672666666666667e-06, | |
| "loss": 0.279, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 5.5615555555555554e-06, | |
| "loss": 0.2853, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 5.450444444444445e-06, | |
| "loss": 0.2798, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 5.339333333333334e-06, | |
| "loss": 0.276, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 5.228444444444445e-06, | |
| "loss": 0.2772, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 5.117333333333334e-06, | |
| "loss": 0.2815, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 5.0062222222222224e-06, | |
| "loss": 0.2816, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 4.895111111111111e-06, | |
| "loss": 0.2817, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "learning_rate": 4.784e-06, | |
| "loss": 0.2801, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 4.673111111111112e-06, | |
| "loss": 0.2816, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 4.5620000000000005e-06, | |
| "loss": 0.2849, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 4.451111111111112e-06, | |
| "loss": 0.277, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 4.34e-06, | |
| "loss": 0.2722, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 4.228888888888889e-06, | |
| "loss": 0.2811, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 4.117777777777779e-06, | |
| "loss": 0.2754, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.006666666666667e-06, | |
| "loss": 0.2842, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 13.8792, | |
| "eval_loss": 0.2534540593624115, | |
| "eval_rouge1": 0.8333, | |
| "eval_rouge2": 0.762, | |
| "eval_rougeL": 0.8313, | |
| "eval_rougeLsum": 0.8313, | |
| "eval_runtime": 2051.2088, | |
| "eval_samples_per_second": 17.551, | |
| "eval_steps_per_second": 4.388, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.895555555555556e-06, | |
| "loss": 0.279, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 3.784444444444445e-06, | |
| "loss": 0.2792, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 3.673333333333334e-06, | |
| "loss": 0.281, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 3.5624444444444444e-06, | |
| "loss": 0.2793, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 3.4513333333333337e-06, | |
| "loss": 0.2761, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3.340666666666667e-06, | |
| "loss": 0.2775, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 3.229555555555556e-06, | |
| "loss": 0.2801, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 3.1184444444444445e-06, | |
| "loss": 0.2767, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 3.007333333333334e-06, | |
| "loss": 0.2761, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 2.8964444444444443e-06, | |
| "loss": 0.2766, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 2.7853333333333337e-06, | |
| "loss": 0.2786, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 2.6742222222222226e-06, | |
| "loss": 0.2716, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 2.563111111111111e-06, | |
| "loss": 0.2791, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 2.4520000000000004e-06, | |
| "loss": 0.2769, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 2.3408888888888893e-06, | |
| "loss": 0.2797, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 2.229777777777778e-06, | |
| "loss": 0.2746, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 2.1186666666666667e-06, | |
| "loss": 0.2727, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.0075555555555557e-06, | |
| "loss": 0.2809, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 13.8298, | |
| "eval_loss": 0.25257840752601624, | |
| "eval_rouge1": 0.8331, | |
| "eval_rouge2": 0.7617, | |
| "eval_rougeL": 0.8312, | |
| "eval_rougeLsum": 0.8312, | |
| "eval_runtime": 2075.7734, | |
| "eval_samples_per_second": 17.343, | |
| "eval_steps_per_second": 4.336, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 1.8964444444444446e-06, | |
| "loss": 0.275, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 1.7853333333333333e-06, | |
| "loss": 0.275, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 1.6742222222222224e-06, | |
| "loss": 0.2727, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 1.5631111111111113e-06, | |
| "loss": 0.2769, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 1.452e-06, | |
| "loss": 0.2783, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 1.3408888888888892e-06, | |
| "loss": 0.2736, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 1.2297777777777779e-06, | |
| "loss": 0.2755, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 1.118888888888889e-06, | |
| "loss": 0.2797, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 1.0077777777777777e-06, | |
| "loss": 0.2786, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 8.966666666666668e-07, | |
| "loss": 0.2778, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 7.857777777777778e-07, | |
| "loss": 0.2759, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 6.746666666666667e-07, | |
| "loss": 0.2732, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 5.635555555555556e-07, | |
| "loss": 0.2715, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 4.5244444444444445e-07, | |
| "loss": 0.278, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 9.83, | |
| "learning_rate": 3.4133333333333337e-07, | |
| "loss": 0.2768, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 2.3022222222222224e-07, | |
| "loss": 0.2748, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 1.1911111111111113e-07, | |
| "loss": 0.2761, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 8e-09, | |
| "loss": 0.2738, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 12.9728, | |
| "eval_loss": 0.248253840752601624, | |
| "eval_rouge1": 0.8425, | |
| "eval_rouge2": 0.7724, | |
| "eval_rougeL": 0.8431, | |
| "eval_rougeLsum": 0.8461, | |
| "eval_runtime": 2075.7734, | |
| "eval_samples_per_second": 17.343, | |
| "eval_steps_per_second": 4.336, | |
| "step": 90000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 90000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.748847816389427e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |