| { | |
| "best_metric": 2.51826810836792, | |
| "best_model_checkpoint": "./results/models/checkpoint-60605", | |
| "epoch": 31.0, | |
| "eval_steps": 500, | |
| "global_step": 60605, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2557544757033248, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.0009948849104859335, | |
| "loss": 3.8573, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5115089514066496, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.000989769820971867, | |
| "loss": 3.7698, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7672634271099744, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.0009846547314578005, | |
| "loss": 3.7481, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.7266085147857666, | |
| "eval_runtime": 1.0357, | |
| "eval_samples_per_second": 482.766, | |
| "eval_steps_per_second": 0.966, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 1.0230179028132993, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.000979539641943734, | |
| "loss": 3.7319, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.278772378516624, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 0.0009744245524296675, | |
| "loss": 3.7118, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5345268542199488, | |
| "grad_norm": 0.25, | |
| "learning_rate": 0.000969309462915601, | |
| "loss": 3.6851, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7902813299232738, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.0009641943734015346, | |
| "loss": 3.628, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.5190136432647705, | |
| "eval_runtime": 1.0477, | |
| "eval_samples_per_second": 477.242, | |
| "eval_steps_per_second": 0.954, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 2.0460358056265986, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0009590792838874681, | |
| "loss": 3.5437, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.3017902813299234, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.0009539641943734016, | |
| "loss": 3.456, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.557544757033248, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.0009488491048593351, | |
| "loss": 3.3486, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.813299232736573, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0009437340153452686, | |
| "loss": 3.2918, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.189316511154175, | |
| "eval_runtime": 1.0569, | |
| "eval_samples_per_second": 473.077, | |
| "eval_steps_per_second": 0.946, | |
| "step": 5865 | |
| }, | |
| { | |
| "epoch": 3.0690537084398977, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.0009386189258312021, | |
| "loss": 3.2061, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.3248081841432224, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0009335038363171356, | |
| "loss": 3.1412, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.580562659846547, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.0009283887468030691, | |
| "loss": 3.0956, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.836317135549872, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.0009232736572890026, | |
| "loss": 3.0555, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.0206379890441895, | |
| "eval_runtime": 1.0312, | |
| "eval_samples_per_second": 484.854, | |
| "eval_steps_per_second": 0.97, | |
| "step": 7820 | |
| }, | |
| { | |
| "epoch": 4.092071611253197, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.0009181585677749361, | |
| "loss": 3.014, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0009130434782608695, | |
| "loss": 2.9683, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.603580562659847, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.0009079283887468031, | |
| "loss": 2.9468, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.859335038363171, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.0009028132992327366, | |
| "loss": 2.9253, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.9162545204162598, | |
| "eval_runtime": 1.0514, | |
| "eval_samples_per_second": 475.536, | |
| "eval_steps_per_second": 0.951, | |
| "step": 9775 | |
| }, | |
| { | |
| "epoch": 5.115089514066496, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.0008976982097186701, | |
| "loss": 2.8851, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.370843989769821, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.0008925831202046036, | |
| "loss": 2.8639, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.626598465473146, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.000887468030690537, | |
| "loss": 2.8469, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.26171875, | |
| "learning_rate": 0.0008823529411764706, | |
| "loss": 2.8301, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.8372726440429688, | |
| "eval_runtime": 1.0374, | |
| "eval_samples_per_second": 481.997, | |
| "eval_steps_per_second": 0.964, | |
| "step": 11730 | |
| }, | |
| { | |
| "epoch": 6.138107416879795, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.0008772378516624041, | |
| "loss": 2.7998, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.3938618925831205, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 0.0008721227621483376, | |
| "loss": 2.7802, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.649616368286445, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0008670076726342711, | |
| "loss": 2.7704, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.90537084398977, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0008618925831202045, | |
| "loss": 2.7605, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.7794747352600098, | |
| "eval_runtime": 1.0401, | |
| "eval_samples_per_second": 480.743, | |
| "eval_steps_per_second": 0.961, | |
| "step": 13685 | |
| }, | |
| { | |
| "epoch": 7.161125319693094, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.0008567774936061381, | |
| "loss": 2.7314, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.41687979539642, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0008516624040920716, | |
| "loss": 2.7203, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.672634271099744, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0008465473145780051, | |
| "loss": 2.713, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.928388746803069, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.0008414322250639387, | |
| "loss": 2.7029, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.7313828468322754, | |
| "eval_runtime": 1.0364, | |
| "eval_samples_per_second": 482.427, | |
| "eval_steps_per_second": 0.965, | |
| "step": 15640 | |
| }, | |
| { | |
| "epoch": 8.184143222506394, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.000836317135549872, | |
| "loss": 2.6727, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.43989769820972, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.0008312020460358057, | |
| "loss": 2.6709, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.0008260869565217392, | |
| "loss": 2.6632, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.951406649616368, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.0008209718670076727, | |
| "loss": 2.6586, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.697967767715454, | |
| "eval_runtime": 1.5778, | |
| "eval_samples_per_second": 316.897, | |
| "eval_steps_per_second": 0.634, | |
| "step": 17595 | |
| }, | |
| { | |
| "epoch": 9.207161125319693, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.0008158567774936062, | |
| "loss": 2.6339, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.462915601023019, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.0008107416879795396, | |
| "loss": 2.6225, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.718670076726342, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.0008056265984654732, | |
| "loss": 2.6241, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 9.974424552429667, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0008005115089514067, | |
| "loss": 2.6182, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.673384189605713, | |
| "eval_runtime": 1.0686, | |
| "eval_samples_per_second": 467.914, | |
| "eval_steps_per_second": 0.936, | |
| "step": 19550 | |
| }, | |
| { | |
| "epoch": 10.230179028132993, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.0007953964194373402, | |
| "loss": 2.5897, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.485933503836318, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.0007902813299232737, | |
| "loss": 2.5888, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 10.741687979539641, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0007851662404092071, | |
| "loss": 2.5878, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 10.997442455242966, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.0007800511508951407, | |
| "loss": 2.5869, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 2.6476247310638428, | |
| "eval_runtime": 1.0522, | |
| "eval_samples_per_second": 475.199, | |
| "eval_steps_per_second": 0.95, | |
| "step": 21505 | |
| }, | |
| { | |
| "epoch": 11.253196930946292, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0007749360613810742, | |
| "loss": 2.5571, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.508951406649617, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.0007698209718670077, | |
| "loss": 2.5603, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0007647058823529411, | |
| "loss": 2.5602, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 2.6259398460388184, | |
| "eval_runtime": 1.0589, | |
| "eval_samples_per_second": 472.168, | |
| "eval_steps_per_second": 0.944, | |
| "step": 23460 | |
| }, | |
| { | |
| "epoch": 12.020460358056265, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.0007595907928388746, | |
| "loss": 2.554, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 12.27621483375959, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0007544757033248082, | |
| "loss": 2.532, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.531969309462916, | |
| "grad_norm": 0.314453125, | |
| "learning_rate": 0.0007493606138107417, | |
| "loss": 2.53, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 12.787723785166241, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0007442455242966752, | |
| "loss": 2.5295, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 2.6141037940979004, | |
| "eval_runtime": 1.0318, | |
| "eval_samples_per_second": 484.612, | |
| "eval_steps_per_second": 0.969, | |
| "step": 25415 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 0.0007391304347826086, | |
| "loss": 2.5263, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 13.29923273657289, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.0007340153452685422, | |
| "loss": 2.5082, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.554987212276215, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.0007289002557544757, | |
| "loss": 2.5115, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 13.81074168797954, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0007237851662404093, | |
| "loss": 2.5082, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 2.5942933559417725, | |
| "eval_runtime": 1.0321, | |
| "eval_samples_per_second": 484.463, | |
| "eval_steps_per_second": 0.969, | |
| "step": 27370 | |
| }, | |
| { | |
| "epoch": 14.066496163682864, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0007186700767263428, | |
| "loss": 2.4949, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 14.322250639386189, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0007135549872122762, | |
| "loss": 2.4873, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.578005115089514, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.0007084398976982098, | |
| "loss": 2.4895, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 14.83375959079284, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0007033248081841433, | |
| "loss": 2.4875, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 2.5941433906555176, | |
| "eval_runtime": 1.0612, | |
| "eval_samples_per_second": 471.171, | |
| "eval_steps_per_second": 0.942, | |
| "step": 29325 | |
| }, | |
| { | |
| "epoch": 15.089514066496164, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0006982097186700768, | |
| "loss": 2.4796, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 15.345268542199488, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.0006930946291560103, | |
| "loss": 2.4671, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.601023017902813, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.0006879795396419437, | |
| "loss": 2.4647, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 15.856777493606138, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0006828644501278773, | |
| "loss": 2.4729, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 2.5811538696289062, | |
| "eval_runtime": 1.0329, | |
| "eval_samples_per_second": 484.078, | |
| "eval_steps_per_second": 0.968, | |
| "step": 31280 | |
| }, | |
| { | |
| "epoch": 16.11253196930946, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.0006777493606138108, | |
| "loss": 2.4574, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 16.36828644501279, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0006726342710997443, | |
| "loss": 2.4535, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 16.624040920716112, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0006675191815856778, | |
| "loss": 2.452, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 16.87979539641944, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0006624040920716112, | |
| "loss": 2.4508, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 2.5716660022735596, | |
| "eval_runtime": 1.0296, | |
| "eval_samples_per_second": 485.606, | |
| "eval_steps_per_second": 0.971, | |
| "step": 33235 | |
| }, | |
| { | |
| "epoch": 17.135549872122763, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0006572890025575448, | |
| "loss": 2.4415, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0006521739130434783, | |
| "loss": 2.4275, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0006470588235294118, | |
| "loss": 2.4398, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 17.902813299232736, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0006419437340153452, | |
| "loss": 2.4446, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.5545594692230225, | |
| "eval_runtime": 1.0484, | |
| "eval_samples_per_second": 476.904, | |
| "eval_steps_per_second": 0.954, | |
| "step": 35190 | |
| }, | |
| { | |
| "epoch": 18.15856777493606, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.0006368286445012787, | |
| "loss": 2.4267, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 18.414322250639387, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.0006317135549872123, | |
| "loss": 2.4183, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 18.67007672634271, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.0006265984654731458, | |
| "loss": 2.4242, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 18.925831202046037, | |
| "grad_norm": 0.34375, | |
| "learning_rate": 0.0006214833759590793, | |
| "loss": 2.4312, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.5581214427948, | |
| "eval_runtime": 1.082, | |
| "eval_samples_per_second": 462.11, | |
| "eval_steps_per_second": 0.924, | |
| "step": 37145 | |
| }, | |
| { | |
| "epoch": 19.18158567774936, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0006163682864450127, | |
| "loss": 2.4121, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 19.437340153452684, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.0006112531969309462, | |
| "loss": 2.4054, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 19.69309462915601, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0006061381074168799, | |
| "loss": 2.4158, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 19.948849104859335, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0006010230179028134, | |
| "loss": 2.4166, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.545928478240967, | |
| "eval_runtime": 1.0301, | |
| "eval_samples_per_second": 485.371, | |
| "eval_steps_per_second": 0.971, | |
| "step": 39100 | |
| }, | |
| { | |
| "epoch": 20.20460358056266, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0005959079283887469, | |
| "loss": 2.4008, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 20.460358056265985, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0005907928388746803, | |
| "loss": 2.3997, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 20.71611253196931, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.0005856777493606138, | |
| "loss": 2.4061, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 20.971867007672635, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.0005805626598465474, | |
| "loss": 2.4055, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 2.5422630310058594, | |
| "eval_runtime": 1.0775, | |
| "eval_samples_per_second": 464.02, | |
| "eval_steps_per_second": 0.928, | |
| "step": 41055 | |
| }, | |
| { | |
| "epoch": 21.22762148337596, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.0005754475703324809, | |
| "loss": 2.3843, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 21.483375959079282, | |
| "grad_norm": 0.314453125, | |
| "learning_rate": 0.0005703324808184144, | |
| "loss": 2.3914, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.0005652173913043478, | |
| "loss": 2.3954, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 21.994884910485933, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0005601023017902813, | |
| "loss": 2.3992, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 2.538604736328125, | |
| "eval_runtime": 1.0718, | |
| "eval_samples_per_second": 466.508, | |
| "eval_steps_per_second": 0.933, | |
| "step": 43010 | |
| }, | |
| { | |
| "epoch": 22.25063938618926, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.0005549872122762149, | |
| "loss": 2.3769, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 22.506393861892583, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.0005498721227621484, | |
| "loss": 2.3806, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 22.762148337595907, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.0005447570332480819, | |
| "loss": 2.3905, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 2.5324785709381104, | |
| "eval_runtime": 1.0484, | |
| "eval_samples_per_second": 476.898, | |
| "eval_steps_per_second": 0.954, | |
| "step": 44965 | |
| }, | |
| { | |
| "epoch": 23.017902813299234, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.0005396419437340153, | |
| "loss": 2.3874, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 23.273657289002557, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.0005345268542199488, | |
| "loss": 2.3744, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.0005294117647058824, | |
| "loss": 2.3785, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 23.785166240409207, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.0005242966751918159, | |
| "loss": 2.3759, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 2.5388312339782715, | |
| "eval_runtime": 1.0338, | |
| "eval_samples_per_second": 483.635, | |
| "eval_steps_per_second": 0.967, | |
| "step": 46920 | |
| }, | |
| { | |
| "epoch": 24.04092071611253, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.0005191815856777494, | |
| "loss": 2.3774, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 24.296675191815858, | |
| "grad_norm": 0.37109375, | |
| "learning_rate": 0.0005140664961636828, | |
| "loss": 2.367, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 24.55242966751918, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.0005089514066496163, | |
| "loss": 2.3701, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 24.808184143222505, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.0005038363171355499, | |
| "loss": 2.3722, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.528256893157959, | |
| "eval_runtime": 1.03, | |
| "eval_samples_per_second": 485.431, | |
| "eval_steps_per_second": 0.971, | |
| "step": 48875 | |
| }, | |
| { | |
| "epoch": 25.06393861892583, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0004987212276214833, | |
| "loss": 2.3668, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 25.319693094629155, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0004936061381074169, | |
| "loss": 2.3593, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 25.575447570332482, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 0.0004884910485933504, | |
| "loss": 2.3647, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 25.831202046035806, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.0004833759590792839, | |
| "loss": 2.3704, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 2.5277769565582275, | |
| "eval_runtime": 1.6093, | |
| "eval_samples_per_second": 310.697, | |
| "eval_steps_per_second": 0.621, | |
| "step": 50830 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.0004782608695652174, | |
| "loss": 2.3599, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 26.342710997442456, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0004731457800511509, | |
| "loss": 2.3516, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 26.59846547314578, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.0004680306905370844, | |
| "loss": 2.3585, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 26.854219948849106, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00046291560102301786, | |
| "loss": 2.3591, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 2.5286903381347656, | |
| "eval_runtime": 1.0654, | |
| "eval_samples_per_second": 469.299, | |
| "eval_steps_per_second": 0.939, | |
| "step": 52785 | |
| }, | |
| { | |
| "epoch": 27.10997442455243, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.0004578005115089514, | |
| "loss": 2.3548, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 27.365728900255753, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0004526854219948849, | |
| "loss": 2.3496, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 27.62148337595908, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.00044757033248081843, | |
| "loss": 2.3506, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 27.877237851662404, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00044245524296675193, | |
| "loss": 2.3552, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 2.523589849472046, | |
| "eval_runtime": 1.0502, | |
| "eval_samples_per_second": 476.079, | |
| "eval_steps_per_second": 0.952, | |
| "step": 54740 | |
| }, | |
| { | |
| "epoch": 28.132992327365727, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.0004373401534526854, | |
| "loss": 2.3445, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 28.388746803069054, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.00043222506393861894, | |
| "loss": 2.3447, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 28.644501278772378, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.00042710997442455245, | |
| "loss": 2.3485, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 28.900255754475705, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.00042199488491048595, | |
| "loss": 2.3509, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 2.5270164012908936, | |
| "eval_runtime": 1.0578, | |
| "eval_samples_per_second": 472.684, | |
| "eval_steps_per_second": 0.945, | |
| "step": 56695 | |
| }, | |
| { | |
| "epoch": 29.156010230179028, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.00041687979539641946, | |
| "loss": 2.3418, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 29.41176470588235, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.0004117647058823529, | |
| "loss": 2.3364, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 29.66751918158568, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00040664961636828646, | |
| "loss": 2.3473, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 29.923273657289002, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 0.00040153452685421997, | |
| "loss": 2.3423, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 2.5237414836883545, | |
| "eval_runtime": 1.0886, | |
| "eval_samples_per_second": 459.31, | |
| "eval_steps_per_second": 0.919, | |
| "step": 58650 | |
| }, | |
| { | |
| "epoch": 30.17902813299233, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.00039641943734015347, | |
| "loss": 2.3403, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 30.434782608695652, | |
| "grad_norm": 0.34375, | |
| "learning_rate": 0.000391304347826087, | |
| "loss": 2.3339, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 30.690537084398976, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.0003861892583120204, | |
| "loss": 2.3387, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 30.946291560102303, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.000381074168797954, | |
| "loss": 2.3386, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 2.51826810836792, | |
| "eval_runtime": 1.0097, | |
| "eval_samples_per_second": 495.204, | |
| "eval_steps_per_second": 0.99, | |
| "step": 60605 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 97750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.461299764647219e+18, | |
| "train_batch_size": 512, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |