| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7811457624923088, | |
| "eval_steps": 1000, | |
| "global_step": 55000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016192234204475534, | |
| "grad_norm": 2.899235963821411, | |
| "learning_rate": 4.9919038828977625e-05, | |
| "loss": 7.6546, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03238446840895107, | |
| "grad_norm": 3.1383473873138428, | |
| "learning_rate": 4.983807765795525e-05, | |
| "loss": 7.0771, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03238446840895107, | |
| "eval_loss": 10.114262580871582, | |
| "eval_runtime": 20.3013, | |
| "eval_samples_per_second": 2704.02, | |
| "eval_steps_per_second": 169.004, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.0485767026134266, | |
| "grad_norm": 3.265001058578491, | |
| "learning_rate": 4.975711648693287e-05, | |
| "loss": 6.9745, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.06476893681790213, | |
| "grad_norm": 4.482234477996826, | |
| "learning_rate": 4.967615531591049e-05, | |
| "loss": 6.8578, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06476893681790213, | |
| "eval_loss": 10.50239086151123, | |
| "eval_runtime": 20.2747, | |
| "eval_samples_per_second": 2707.566, | |
| "eval_steps_per_second": 169.226, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08096117102237767, | |
| "grad_norm": 3.455843925476074, | |
| "learning_rate": 4.9595194144888116e-05, | |
| "loss": 6.8529, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.0971534052268532, | |
| "grad_norm": 4.888940811157227, | |
| "learning_rate": 4.951423297386574e-05, | |
| "loss": 6.7865, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.0971534052268532, | |
| "eval_loss": 11.206120491027832, | |
| "eval_runtime": 20.2787, | |
| "eval_samples_per_second": 2707.023, | |
| "eval_steps_per_second": 169.192, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.11334563943132873, | |
| "grad_norm": 2.9272000789642334, | |
| "learning_rate": 4.9433271802843354e-05, | |
| "loss": 6.7582, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.12953787363580427, | |
| "grad_norm": 3.68345046043396, | |
| "learning_rate": 4.9352310631820984e-05, | |
| "loss": 6.6886, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.12953787363580427, | |
| "eval_loss": 11.919740676879883, | |
| "eval_runtime": 20.2985, | |
| "eval_samples_per_second": 2704.382, | |
| "eval_steps_per_second": 169.027, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.14573010784027982, | |
| "grad_norm": 3.9165866374969482, | |
| "learning_rate": 4.927151138314065e-05, | |
| "loss": 6.657, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.16192234204475534, | |
| "grad_norm": 4.2097015380859375, | |
| "learning_rate": 4.9190550212118266e-05, | |
| "loss": 6.6567, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.16192234204475534, | |
| "eval_loss": 11.381248474121094, | |
| "eval_runtime": 20.3121, | |
| "eval_samples_per_second": 2702.582, | |
| "eval_steps_per_second": 168.914, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.17811457624923088, | |
| "grad_norm": 3.525871753692627, | |
| "learning_rate": 4.9109589041095895e-05, | |
| "loss": 6.6297, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.1943068104537064, | |
| "grad_norm": 5.512066841125488, | |
| "learning_rate": 4.902862787007351e-05, | |
| "loss": 6.6049, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.1943068104537064, | |
| "eval_loss": 11.815380096435547, | |
| "eval_runtime": 20.3269, | |
| "eval_samples_per_second": 2700.605, | |
| "eval_steps_per_second": 168.791, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.21049904465818195, | |
| "grad_norm": 6.878340244293213, | |
| "learning_rate": 4.8947828621393185e-05, | |
| "loss": 6.5587, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.22669127886265747, | |
| "grad_norm": 3.512578248977661, | |
| "learning_rate": 4.886686745037081e-05, | |
| "loss": 6.5451, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.22669127886265747, | |
| "eval_loss": 11.993084907531738, | |
| "eval_runtime": 20.3217, | |
| "eval_samples_per_second": 2701.3, | |
| "eval_steps_per_second": 168.834, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.24288351306713302, | |
| "grad_norm": 4.394603729248047, | |
| "learning_rate": 4.8786068201690474e-05, | |
| "loss": 6.5283, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.25907574727160854, | |
| "grad_norm": 3.8106327056884766, | |
| "learning_rate": 4.8705107030668096e-05, | |
| "loss": 6.5246, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.25907574727160854, | |
| "eval_loss": 12.192233085632324, | |
| "eval_runtime": 20.3069, | |
| "eval_samples_per_second": 2703.263, | |
| "eval_steps_per_second": 168.957, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.27526798147608406, | |
| "grad_norm": 4.158311367034912, | |
| "learning_rate": 4.862414585964571e-05, | |
| "loss": 6.4727, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.29146021568055963, | |
| "grad_norm": 3.797197103500366, | |
| "learning_rate": 4.854318468862334e-05, | |
| "loss": 6.4594, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.29146021568055963, | |
| "eval_loss": 12.250760078430176, | |
| "eval_runtime": 20.2943, | |
| "eval_samples_per_second": 2704.945, | |
| "eval_steps_per_second": 169.062, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.30765244988503515, | |
| "grad_norm": 4.4851908683776855, | |
| "learning_rate": 4.8462223517600964e-05, | |
| "loss": 6.4441, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.32384468408951067, | |
| "grad_norm": 5.158078670501709, | |
| "learning_rate": 4.838142426892063e-05, | |
| "loss": 6.4537, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.32384468408951067, | |
| "eval_loss": 12.235699653625488, | |
| "eval_runtime": 20.3426, | |
| "eval_samples_per_second": 2698.522, | |
| "eval_steps_per_second": 168.661, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.3400369182939862, | |
| "grad_norm": 4.838212966918945, | |
| "learning_rate": 4.8300463097898254e-05, | |
| "loss": 6.4313, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.35622915249846177, | |
| "grad_norm": 4.873743534088135, | |
| "learning_rate": 4.821950192687587e-05, | |
| "loss": 6.4375, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.35622915249846177, | |
| "eval_loss": 13.167688369750977, | |
| "eval_runtime": 20.3858, | |
| "eval_samples_per_second": 2692.805, | |
| "eval_steps_per_second": 168.303, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.3724213867029373, | |
| "grad_norm": 5.022457122802734, | |
| "learning_rate": 4.813854075585349e-05, | |
| "loss": 6.4249, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.3886136209074128, | |
| "grad_norm": 4.3410964012146, | |
| "learning_rate": 4.805774150717316e-05, | |
| "loss": 6.3944, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.3886136209074128, | |
| "eval_loss": 12.753792762756348, | |
| "eval_runtime": 20.3234, | |
| "eval_samples_per_second": 2701.08, | |
| "eval_steps_per_second": 168.821, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.4048058551118883, | |
| "grad_norm": 4.282736301422119, | |
| "learning_rate": 4.797678033615078e-05, | |
| "loss": 6.3343, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.4209980893163639, | |
| "grad_norm": 4.2941484451293945, | |
| "learning_rate": 4.789581916512841e-05, | |
| "loss": 6.3576, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.4209980893163639, | |
| "eval_loss": 14.182137489318848, | |
| "eval_runtime": 20.2915, | |
| "eval_samples_per_second": 2705.322, | |
| "eval_steps_per_second": 169.086, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.4371903235208394, | |
| "grad_norm": 4.584780693054199, | |
| "learning_rate": 4.7814857994106027e-05, | |
| "loss": 6.3133, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.45338255772531494, | |
| "grad_norm": 6.5496134757995605, | |
| "learning_rate": 4.77340587454257e-05, | |
| "loss": 6.3042, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.45338255772531494, | |
| "eval_loss": 13.6593656539917, | |
| "eval_runtime": 20.2751, | |
| "eval_samples_per_second": 2707.503, | |
| "eval_steps_per_second": 169.222, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.46957479192979046, | |
| "grad_norm": 4.188933849334717, | |
| "learning_rate": 4.7653097574403316e-05, | |
| "loss": 6.2764, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.48576702613426603, | |
| "grad_norm": 4.540693283081055, | |
| "learning_rate": 4.757213640338094e-05, | |
| "loss": 6.261, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.48576702613426603, | |
| "eval_loss": 13.340327262878418, | |
| "eval_runtime": 20.2957, | |
| "eval_samples_per_second": 2704.755, | |
| "eval_steps_per_second": 169.05, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5019592603387415, | |
| "grad_norm": 4.090181827545166, | |
| "learning_rate": 4.749117523235857e-05, | |
| "loss": 6.2382, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.5181514945432171, | |
| "grad_norm": 5.604825973510742, | |
| "learning_rate": 4.7410214061336184e-05, | |
| "loss": 6.2225, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.5181514945432171, | |
| "eval_loss": 14.243009567260742, | |
| "eval_runtime": 20.3274, | |
| "eval_samples_per_second": 2700.537, | |
| "eval_steps_per_second": 168.787, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.5343437287476926, | |
| "grad_norm": 5.334819316864014, | |
| "learning_rate": 4.732941481265585e-05, | |
| "loss": 6.2051, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.5505359629521681, | |
| "grad_norm": 5.907509803771973, | |
| "learning_rate": 4.724845364163347e-05, | |
| "loss": 6.1475, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.5505359629521681, | |
| "eval_loss": 14.208978652954102, | |
| "eval_runtime": 20.3636, | |
| "eval_samples_per_second": 2695.742, | |
| "eval_steps_per_second": 168.487, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.5667281971566437, | |
| "grad_norm": 5.113221645355225, | |
| "learning_rate": 4.7167492470611096e-05, | |
| "loss": 6.1765, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.5829204313611193, | |
| "grad_norm": 5.8240461349487305, | |
| "learning_rate": 4.708653129958872e-05, | |
| "loss": 6.1567, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.5829204313611193, | |
| "eval_loss": 14.480341911315918, | |
| "eval_runtime": 20.3098, | |
| "eval_samples_per_second": 2702.888, | |
| "eval_steps_per_second": 168.934, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.5991126655655947, | |
| "grad_norm": 4.948818683624268, | |
| "learning_rate": 4.7005732050908385e-05, | |
| "loss": 6.1198, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.6153048997700703, | |
| "grad_norm": 5.95670223236084, | |
| "learning_rate": 4.692477087988601e-05, | |
| "loss": 6.1038, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.6153048997700703, | |
| "eval_loss": 14.150490760803223, | |
| "eval_runtime": 20.2957, | |
| "eval_samples_per_second": 2704.759, | |
| "eval_steps_per_second": 169.051, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.6314971339745458, | |
| "grad_norm": 7.957947731018066, | |
| "learning_rate": 4.684380970886363e-05, | |
| "loss": 6.0658, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.6476893681790213, | |
| "grad_norm": 6.398547649383545, | |
| "learning_rate": 4.676284853784125e-05, | |
| "loss": 6.007, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.6476893681790213, | |
| "eval_loss": 13.682233810424805, | |
| "eval_runtime": 20.2919, | |
| "eval_samples_per_second": 2705.271, | |
| "eval_steps_per_second": 169.083, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.6638816023834969, | |
| "grad_norm": 7.564326763153076, | |
| "learning_rate": 4.6681887366818875e-05, | |
| "loss": 6.0063, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.6800738365879724, | |
| "grad_norm": 6.847624778747559, | |
| "learning_rate": 4.66009261957965e-05, | |
| "loss": 5.9528, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.6800738365879724, | |
| "eval_loss": 13.830253601074219, | |
| "eval_runtime": 20.3183, | |
| "eval_samples_per_second": 2701.748, | |
| "eval_steps_per_second": 168.862, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.696266070792448, | |
| "grad_norm": 6.5545125007629395, | |
| "learning_rate": 4.651996502477412e-05, | |
| "loss": 5.9257, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.7124583049969235, | |
| "grad_norm": 8.006608009338379, | |
| "learning_rate": 4.643900385375174e-05, | |
| "loss": 5.9066, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.7124583049969235, | |
| "eval_loss": 14.178547859191895, | |
| "eval_runtime": 20.3334, | |
| "eval_samples_per_second": 2699.748, | |
| "eval_steps_per_second": 168.737, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.728650539201399, | |
| "grad_norm": 6.9318766593933105, | |
| "learning_rate": 4.635820460507141e-05, | |
| "loss": 5.8562, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.7448427734058746, | |
| "grad_norm": 6.815021991729736, | |
| "learning_rate": 4.627724343404903e-05, | |
| "loss": 5.7984, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.7448427734058746, | |
| "eval_loss": 14.005717277526855, | |
| "eval_runtime": 20.3208, | |
| "eval_samples_per_second": 2701.413, | |
| "eval_steps_per_second": 168.841, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.76103500761035, | |
| "grad_norm": 7.4206976890563965, | |
| "learning_rate": 4.6196282263026655e-05, | |
| "loss": 5.7574, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.7772272418148256, | |
| "grad_norm": 7.273144721984863, | |
| "learning_rate": 4.611532109200428e-05, | |
| "loss": 5.7247, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.7772272418148256, | |
| "eval_loss": 13.36209774017334, | |
| "eval_runtime": 20.3187, | |
| "eval_samples_per_second": 2701.693, | |
| "eval_steps_per_second": 168.859, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.7934194760193012, | |
| "grad_norm": 8.667037963867188, | |
| "learning_rate": 4.6034521843323944e-05, | |
| "loss": 5.682, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.8096117102237766, | |
| "grad_norm": 9.467761039733887, | |
| "learning_rate": 4.595356067230157e-05, | |
| "loss": 5.6262, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.8096117102237766, | |
| "eval_loss": 14.30241584777832, | |
| "eval_runtime": 20.3265, | |
| "eval_samples_per_second": 2700.664, | |
| "eval_steps_per_second": 168.795, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.8258039444282522, | |
| "grad_norm": 7.378128528594971, | |
| "learning_rate": 4.587259950127919e-05, | |
| "loss": 5.5776, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.8419961786327278, | |
| "grad_norm": 8.246053695678711, | |
| "learning_rate": 4.5791638330256806e-05, | |
| "loss": 5.5251, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.8419961786327278, | |
| "eval_loss": 14.098834037780762, | |
| "eval_runtime": 20.3189, | |
| "eval_samples_per_second": 2701.67, | |
| "eval_steps_per_second": 168.857, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.8581884128372033, | |
| "grad_norm": 6.545131206512451, | |
| "learning_rate": 4.571083908157648e-05, | |
| "loss": 5.4848, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.8743806470416788, | |
| "grad_norm": 7.735929489135742, | |
| "learning_rate": 4.56298779105541e-05, | |
| "loss": 5.4504, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.8743806470416788, | |
| "eval_loss": 13.383395195007324, | |
| "eval_runtime": 20.3058, | |
| "eval_samples_per_second": 2703.417, | |
| "eval_steps_per_second": 168.967, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.8905728812461543, | |
| "grad_norm": 7.53142786026001, | |
| "learning_rate": 4.5548916739531724e-05, | |
| "loss": 5.4094, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.9067651154506299, | |
| "grad_norm": 6.753902912139893, | |
| "learning_rate": 4.546795556850935e-05, | |
| "loss": 5.382, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.9067651154506299, | |
| "eval_loss": 13.9662446975708, | |
| "eval_runtime": 20.3511, | |
| "eval_samples_per_second": 2697.397, | |
| "eval_steps_per_second": 168.59, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.9229573496551055, | |
| "grad_norm": 7.170286655426025, | |
| "learning_rate": 4.538715631982901e-05, | |
| "loss": 5.3786, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.9391495838595809, | |
| "grad_norm": 7.388674736022949, | |
| "learning_rate": 4.5306195148806636e-05, | |
| "loss": 5.3468, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.9391495838595809, | |
| "eval_loss": 14.00700855255127, | |
| "eval_runtime": 20.2945, | |
| "eval_samples_per_second": 2704.924, | |
| "eval_steps_per_second": 169.061, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.9553418180640565, | |
| "grad_norm": 8.680002212524414, | |
| "learning_rate": 4.522523397778426e-05, | |
| "loss": 5.3159, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.9715340522685321, | |
| "grad_norm": 6.966347694396973, | |
| "learning_rate": 4.514427280676188e-05, | |
| "loss": 5.3321, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.9715340522685321, | |
| "eval_loss": 13.2510986328125, | |
| "eval_runtime": 20.2734, | |
| "eval_samples_per_second": 2707.742, | |
| "eval_steps_per_second": 169.237, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.9877262864730075, | |
| "grad_norm": 7.675726413726807, | |
| "learning_rate": 4.506347355808155e-05, | |
| "loss": 5.2547, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 1.003918520677483, | |
| "grad_norm": 9.16435432434082, | |
| "learning_rate": 4.4982512387059164e-05, | |
| "loss": 5.2597, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.003918520677483, | |
| "eval_loss": 13.722345352172852, | |
| "eval_runtime": 20.3123, | |
| "eval_samples_per_second": 2702.549, | |
| "eval_steps_per_second": 168.912, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.0201107548819586, | |
| "grad_norm": 8.053180694580078, | |
| "learning_rate": 4.490155121603679e-05, | |
| "loss": 5.2168, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 1.0363029890864341, | |
| "grad_norm": 7.8324432373046875, | |
| "learning_rate": 4.4820590045014416e-05, | |
| "loss": 5.2229, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.0363029890864341, | |
| "eval_loss": 13.410867691040039, | |
| "eval_runtime": 20.3235, | |
| "eval_samples_per_second": 2701.057, | |
| "eval_steps_per_second": 168.819, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.0524952232909097, | |
| "grad_norm": 6.322490692138672, | |
| "learning_rate": 4.473979079633408e-05, | |
| "loss": 5.1849, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 1.0686874574953853, | |
| "grad_norm": 7.511089324951172, | |
| "learning_rate": 4.4658829625311705e-05, | |
| "loss": 5.1838, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.0686874574953853, | |
| "eval_loss": 13.21849250793457, | |
| "eval_runtime": 20.3207, | |
| "eval_samples_per_second": 2701.433, | |
| "eval_steps_per_second": 168.843, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.0848796916998606, | |
| "grad_norm": 7.725784778594971, | |
| "learning_rate": 4.457786845428932e-05, | |
| "loss": 5.1687, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 1.1010719259043362, | |
| "grad_norm": 7.167893886566162, | |
| "learning_rate": 4.449690728326695e-05, | |
| "loss": 5.157, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.1010719259043362, | |
| "eval_loss": 13.764806747436523, | |
| "eval_runtime": 20.3095, | |
| "eval_samples_per_second": 2702.924, | |
| "eval_steps_per_second": 168.936, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.1172641601088118, | |
| "grad_norm": 9.298649787902832, | |
| "learning_rate": 4.441610803458661e-05, | |
| "loss": 5.1152, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 1.1334563943132874, | |
| "grad_norm": 7.694969654083252, | |
| "learning_rate": 4.433514686356424e-05, | |
| "loss": 5.0736, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.1334563943132874, | |
| "eval_loss": 13.989611625671387, | |
| "eval_runtime": 20.324, | |
| "eval_samples_per_second": 2700.988, | |
| "eval_steps_per_second": 168.815, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.149648628517763, | |
| "grad_norm": 7.617096424102783, | |
| "learning_rate": 4.425418569254186e-05, | |
| "loss": 5.0596, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 1.1658408627222383, | |
| "grad_norm": 6.822023391723633, | |
| "learning_rate": 4.417322452151948e-05, | |
| "loss": 5.0544, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.1658408627222383, | |
| "eval_loss": 14.167996406555176, | |
| "eval_runtime": 20.2953, | |
| "eval_samples_per_second": 2704.815, | |
| "eval_steps_per_second": 169.054, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.1820330969267139, | |
| "grad_norm": 6.709229946136475, | |
| "learning_rate": 4.40922633504971e-05, | |
| "loss": 5.0675, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 1.1982253311311895, | |
| "grad_norm": 8.224237442016602, | |
| "learning_rate": 4.401162602415881e-05, | |
| "loss": 4.9965, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.1982253311311895, | |
| "eval_loss": 14.570647239685059, | |
| "eval_runtime": 20.3304, | |
| "eval_samples_per_second": 2700.144, | |
| "eval_steps_per_second": 168.762, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.214417565335665, | |
| "grad_norm": 8.49647045135498, | |
| "learning_rate": 4.393066485313644e-05, | |
| "loss": 4.9891, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 1.2306097995401406, | |
| "grad_norm": 6.622528553009033, | |
| "learning_rate": 4.384970368211406e-05, | |
| "loss": 4.9896, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.2306097995401406, | |
| "eval_loss": 14.330415725708008, | |
| "eval_runtime": 20.3174, | |
| "eval_samples_per_second": 2701.865, | |
| "eval_steps_per_second": 168.87, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.2468020337446162, | |
| "grad_norm": 7.867303371429443, | |
| "learning_rate": 4.376874251109168e-05, | |
| "loss": 4.9851, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 1.2629942679490918, | |
| "grad_norm": 7.329415321350098, | |
| "learning_rate": 4.368778134006931e-05, | |
| "loss": 4.969, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.2629942679490918, | |
| "eval_loss": 14.989124298095703, | |
| "eval_runtime": 20.3265, | |
| "eval_samples_per_second": 2700.663, | |
| "eval_steps_per_second": 168.795, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.279186502153567, | |
| "grad_norm": 8.647744178771973, | |
| "learning_rate": 4.3606820169046924e-05, | |
| "loss": 4.9547, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 1.2953787363580427, | |
| "grad_norm": 8.578520774841309, | |
| "learning_rate": 4.352585899802455e-05, | |
| "loss": 4.9569, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.2953787363580427, | |
| "eval_loss": 14.283174514770508, | |
| "eval_runtime": 20.3168, | |
| "eval_samples_per_second": 2701.954, | |
| "eval_steps_per_second": 168.875, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.3115709705625183, | |
| "grad_norm": 8.312915802001953, | |
| "learning_rate": 4.3444897827002176e-05, | |
| "loss": 4.9647, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 1.3277632047669938, | |
| "grad_norm": 9.59231185913086, | |
| "learning_rate": 4.3364098578321836e-05, | |
| "loss": 4.9462, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.3277632047669938, | |
| "eval_loss": 14.299793243408203, | |
| "eval_runtime": 20.2951, | |
| "eval_samples_per_second": 2704.839, | |
| "eval_steps_per_second": 169.055, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.3439554389714692, | |
| "grad_norm": 7.844625949859619, | |
| "learning_rate": 4.328313740729946e-05, | |
| "loss": 4.9147, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 1.3601476731759448, | |
| "grad_norm": 6.845381736755371, | |
| "learning_rate": 4.320217623627708e-05, | |
| "loss": 4.914, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.3601476731759448, | |
| "eval_loss": 14.049765586853027, | |
| "eval_runtime": 20.2836, | |
| "eval_samples_per_second": 2706.371, | |
| "eval_steps_per_second": 169.151, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.3763399073804203, | |
| "grad_norm": 7.644216537475586, | |
| "learning_rate": 4.3121215065254704e-05, | |
| "loss": 4.9066, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.392532141584896, | |
| "grad_norm": 7.950870513916016, | |
| "learning_rate": 4.304041581657438e-05, | |
| "loss": 4.8716, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.392532141584896, | |
| "eval_loss": 14.33547306060791, | |
| "eval_runtime": 20.3142, | |
| "eval_samples_per_second": 2702.295, | |
| "eval_steps_per_second": 168.897, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.4087243757893715, | |
| "grad_norm": 7.638858318328857, | |
| "learning_rate": 4.295945464555199e-05, | |
| "loss": 4.8557, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.424916609993847, | |
| "grad_norm": 9.173840522766113, | |
| "learning_rate": 4.2878493474529616e-05, | |
| "loss": 4.8394, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.424916609993847, | |
| "eval_loss": 13.498611450195312, | |
| "eval_runtime": 20.3548, | |
| "eval_samples_per_second": 2696.904, | |
| "eval_steps_per_second": 168.56, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.4411088441983224, | |
| "grad_norm": 8.224470138549805, | |
| "learning_rate": 4.279753230350724e-05, | |
| "loss": 4.8216, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.457301078402798, | |
| "grad_norm": 7.497710704803467, | |
| "learning_rate": 4.2716733054826905e-05, | |
| "loss": 4.8237, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.457301078402798, | |
| "eval_loss": 13.823565483093262, | |
| "eval_runtime": 20.2695, | |
| "eval_samples_per_second": 2708.257, | |
| "eval_steps_per_second": 169.269, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.4734933126072736, | |
| "grad_norm": 6.961141586303711, | |
| "learning_rate": 4.2635771883804535e-05, | |
| "loss": 4.8119, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.4896855468117491, | |
| "grad_norm": 6.934360504150391, | |
| "learning_rate": 4.255481071278215e-05, | |
| "loss": 4.7972, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.4896855468117491, | |
| "eval_loss": 14.0670804977417, | |
| "eval_runtime": 20.2732, | |
| "eval_samples_per_second": 2707.759, | |
| "eval_steps_per_second": 169.238, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.5058777810162245, | |
| "grad_norm": 7.600714206695557, | |
| "learning_rate": 4.247384954175977e-05, | |
| "loss": 4.8131, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.5220700152207, | |
| "grad_norm": 8.766816139221191, | |
| "learning_rate": 4.239305029307944e-05, | |
| "loss": 4.778, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.5220700152207, | |
| "eval_loss": 13.546905517578125, | |
| "eval_runtime": 20.3029, | |
| "eval_samples_per_second": 2703.795, | |
| "eval_steps_per_second": 168.99, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.5382622494251756, | |
| "grad_norm": 8.299079895019531, | |
| "learning_rate": 4.231208912205706e-05, | |
| "loss": 4.7336, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.5544544836296512, | |
| "grad_norm": 8.333426475524902, | |
| "learning_rate": 4.2231127951034685e-05, | |
| "loss": 4.7541, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.5544544836296512, | |
| "eval_loss": 13.711235046386719, | |
| "eval_runtime": 20.3064, | |
| "eval_samples_per_second": 2703.338, | |
| "eval_steps_per_second": 168.962, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.5706467178341268, | |
| "grad_norm": 7.705515384674072, | |
| "learning_rate": 4.215016678001231e-05, | |
| "loss": 4.7901, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 1.5868389520386024, | |
| "grad_norm": 6.669991493225098, | |
| "learning_rate": 4.206920560898993e-05, | |
| "loss": 4.7544, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.5868389520386024, | |
| "eval_loss": 13.069415092468262, | |
| "eval_runtime": 20.3123, | |
| "eval_samples_per_second": 2702.548, | |
| "eval_steps_per_second": 168.912, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.603031186243078, | |
| "grad_norm": 6.958492279052734, | |
| "learning_rate": 4.19884063603096e-05, | |
| "loss": 4.7175, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.6192234204475535, | |
| "grad_norm": 9.244942665100098, | |
| "learning_rate": 4.190744518928722e-05, | |
| "loss": 4.7092, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.6192234204475535, | |
| "eval_loss": 13.0497465133667, | |
| "eval_runtime": 20.3211, | |
| "eval_samples_per_second": 2701.382, | |
| "eval_steps_per_second": 168.839, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.6354156546520289, | |
| "grad_norm": 8.372320175170898, | |
| "learning_rate": 4.182648401826484e-05, | |
| "loss": 4.7401, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 1.6516078888565044, | |
| "grad_norm": 7.373213768005371, | |
| "learning_rate": 4.1745522847242465e-05, | |
| "loss": 4.7252, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.6516078888565044, | |
| "eval_loss": 13.589052200317383, | |
| "eval_runtime": 20.3142, | |
| "eval_samples_per_second": 2702.302, | |
| "eval_steps_per_second": 168.897, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.6678001230609798, | |
| "grad_norm": 9.22149658203125, | |
| "learning_rate": 4.166472359856213e-05, | |
| "loss": 4.6795, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 1.6839923572654554, | |
| "grad_norm": 5.994777202606201, | |
| "learning_rate": 4.1583762427539754e-05, | |
| "loss": 4.7021, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.6839923572654554, | |
| "eval_loss": 13.31000804901123, | |
| "eval_runtime": 20.2902, | |
| "eval_samples_per_second": 2705.498, | |
| "eval_steps_per_second": 169.097, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.700184591469931, | |
| "grad_norm": 8.134748458862305, | |
| "learning_rate": 4.150280125651738e-05, | |
| "loss": 4.7013, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 1.7163768256744065, | |
| "grad_norm": 8.395650863647461, | |
| "learning_rate": 4.1421840085495e-05, | |
| "loss": 4.6976, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.7163768256744065, | |
| "eval_loss": 13.707457542419434, | |
| "eval_runtime": 20.3052, | |
| "eval_samples_per_second": 2703.49, | |
| "eval_steps_per_second": 168.971, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.732569059878882, | |
| "grad_norm": 9.668176651000977, | |
| "learning_rate": 4.1341040836814666e-05, | |
| "loss": 4.6889, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 1.7487612940833577, | |
| "grad_norm": 7.348442554473877, | |
| "learning_rate": 4.126007966579229e-05, | |
| "loss": 4.6465, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.7487612940833577, | |
| "eval_loss": 13.285728454589844, | |
| "eval_runtime": 20.3053, | |
| "eval_samples_per_second": 2703.483, | |
| "eval_steps_per_second": 168.971, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.7649535282878333, | |
| "grad_norm": 7.7653889656066895, | |
| "learning_rate": 4.117911849476991e-05, | |
| "loss": 4.6537, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 1.7811457624923088, | |
| "grad_norm": 8.290285110473633, | |
| "learning_rate": 4.1098157323747534e-05, | |
| "loss": 4.6468, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.7811457624923088, | |
| "eval_loss": 13.134732246398926, | |
| "eval_runtime": 20.3013, | |
| "eval_samples_per_second": 2704.018, | |
| "eval_steps_per_second": 169.004, | |
| "step": 55000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 308790, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 5000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4320645675220992.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |