| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9963898916967509, | |
| "eval_steps": 500, | |
| "global_step": 69, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01444043321299639, | |
| "grad_norm": 0.9324310421943665, | |
| "learning_rate": 4.9974091841168195e-05, | |
| "loss": 0.8402, | |
| "num_input_tokens_seen": 2097152, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02888086642599278, | |
| "grad_norm": 0.736538290977478, | |
| "learning_rate": 4.9896421063288286e-05, | |
| "loss": 0.7686, | |
| "num_input_tokens_seen": 4194304, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.04332129963898917, | |
| "grad_norm": 0.6694996953010559, | |
| "learning_rate": 4.976714865090827e-05, | |
| "loss": 0.7563, | |
| "num_input_tokens_seen": 6291456, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05776173285198556, | |
| "grad_norm": 0.5700488090515137, | |
| "learning_rate": 4.958654254084355e-05, | |
| "loss": 0.7139, | |
| "num_input_tokens_seen": 8388608, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07220216606498195, | |
| "grad_norm": 0.47626349329948425, | |
| "learning_rate": 4.9354977066836986e-05, | |
| "loss": 0.6793, | |
| "num_input_tokens_seen": 10485760, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08664259927797834, | |
| "grad_norm": 0.4310459494590759, | |
| "learning_rate": 4.907293218369499e-05, | |
| "loss": 0.6631, | |
| "num_input_tokens_seen": 12582912, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10108303249097472, | |
| "grad_norm": 0.41724854707717896, | |
| "learning_rate": 4.874099247250798e-05, | |
| "loss": 0.64, | |
| "num_input_tokens_seen": 14680064, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11552346570397112, | |
| "grad_norm": 0.37938904762268066, | |
| "learning_rate": 4.835984592901678e-05, | |
| "loss": 0.6145, | |
| "num_input_tokens_seen": 16777216, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1299638989169675, | |
| "grad_norm": 0.30510303378105164, | |
| "learning_rate": 4.793028253763633e-05, | |
| "loss": 0.5997, | |
| "num_input_tokens_seen": 18874368, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1444043321299639, | |
| "grad_norm": 0.17869406938552856, | |
| "learning_rate": 4.74531926340924e-05, | |
| "loss": 0.5896, | |
| "num_input_tokens_seen": 20971520, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1588447653429603, | |
| "grad_norm": 0.1084759458899498, | |
| "learning_rate": 4.6929565060064864e-05, | |
| "loss": 0.6025, | |
| "num_input_tokens_seen": 23068672, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.17328519855595667, | |
| "grad_norm": 0.09031977504491806, | |
| "learning_rate": 4.6360485113662216e-05, | |
| "loss": 0.5644, | |
| "num_input_tokens_seen": 25165824, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18772563176895307, | |
| "grad_norm": 0.08057376742362976, | |
| "learning_rate": 4.574713229997563e-05, | |
| "loss": 0.5558, | |
| "num_input_tokens_seen": 27262976, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.20216606498194944, | |
| "grad_norm": 0.0670362040400505, | |
| "learning_rate": 4.509077788637446e-05, | |
| "loss": 0.5705, | |
| "num_input_tokens_seen": 29360128, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.21660649819494585, | |
| "grad_norm": 0.06539376825094223, | |
| "learning_rate": 4.43927822676105e-05, | |
| "loss": 0.5694, | |
| "num_input_tokens_seen": 31457280, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.23104693140794225, | |
| "grad_norm": 0.05930742993950844, | |
| "learning_rate": 4.365459214619214e-05, | |
| "loss": 0.559, | |
| "num_input_tokens_seen": 33554432, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.24548736462093862, | |
| "grad_norm": 0.054464634507894516, | |
| "learning_rate": 4.2877737533872485e-05, | |
| "loss": 0.5628, | |
| "num_input_tokens_seen": 35651584, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.259927797833935, | |
| "grad_norm": 0.053172189742326736, | |
| "learning_rate": 4.206382858046636e-05, | |
| "loss": 0.5553, | |
| "num_input_tokens_seen": 37748736, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2743682310469314, | |
| "grad_norm": 0.04865848645567894, | |
| "learning_rate": 4.12145522365689e-05, | |
| "loss": 0.5401, | |
| "num_input_tokens_seen": 39845888, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2888086642599278, | |
| "grad_norm": 0.04852156713604927, | |
| "learning_rate": 4.033166875709291e-05, | |
| "loss": 0.575, | |
| "num_input_tokens_seen": 41943040, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.30324909747292417, | |
| "grad_norm": 0.046296387910842896, | |
| "learning_rate": 3.941700805287168e-05, | |
| "loss": 0.5398, | |
| "num_input_tokens_seen": 44040192, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3176895306859206, | |
| "grad_norm": 0.04640813171863556, | |
| "learning_rate": 3.8472465897889394e-05, | |
| "loss": 0.5389, | |
| "num_input_tokens_seen": 46137344, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.33212996389891697, | |
| "grad_norm": 0.043006811290979385, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.5451, | |
| "num_input_tokens_seen": 48234496, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.34657039711191334, | |
| "grad_norm": 0.04406141862273216, | |
| "learning_rate": 3.6501625943278805e-05, | |
| "loss": 0.5558, | |
| "num_input_tokens_seen": 50331648, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.36101083032490977, | |
| "grad_norm": 0.03994145616889, | |
| "learning_rate": 3.547941301041661e-05, | |
| "loss": 0.5403, | |
| "num_input_tokens_seen": 52428800, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.37545126353790614, | |
| "grad_norm": 0.039348017424345016, | |
| "learning_rate": 3.443547989381536e-05, | |
| "loss": 0.5405, | |
| "num_input_tokens_seen": 54525952, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3898916967509025, | |
| "grad_norm": 0.03861572593450546, | |
| "learning_rate": 3.3371990304274656e-05, | |
| "loss": 0.5577, | |
| "num_input_tokens_seen": 56623104, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4043321299638989, | |
| "grad_norm": 0.03878667205572128, | |
| "learning_rate": 3.2291148486370626e-05, | |
| "loss": 0.5345, | |
| "num_input_tokens_seen": 58720256, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.4187725631768953, | |
| "grad_norm": 0.03664080426096916, | |
| "learning_rate": 3.11951946498225e-05, | |
| "loss": 0.5484, | |
| "num_input_tokens_seen": 60817408, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.4332129963898917, | |
| "grad_norm": 0.036824408918619156, | |
| "learning_rate": 3.008640032631585e-05, | |
| "loss": 0.5485, | |
| "num_input_tokens_seen": 62914560, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.44765342960288806, | |
| "grad_norm": 0.037150438874959946, | |
| "learning_rate": 2.8967063661406285e-05, | |
| "loss": 0.5299, | |
| "num_input_tokens_seen": 65011712, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4620938628158845, | |
| "grad_norm": 0.03484778478741646, | |
| "learning_rate": 2.7839504651261872e-05, | |
| "loss": 0.539, | |
| "num_input_tokens_seen": 67108864, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.47653429602888087, | |
| "grad_norm": 0.035446375608444214, | |
| "learning_rate": 2.6706060334116777e-05, | |
| "loss": 0.5248, | |
| "num_input_tokens_seen": 69206016, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.49097472924187724, | |
| "grad_norm": 0.0350475013256073, | |
| "learning_rate": 2.556907994640264e-05, | |
| "loss": 0.5342, | |
| "num_input_tokens_seen": 71303168, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.5054151624548736, | |
| "grad_norm": 0.036621786653995514, | |
| "learning_rate": 2.4430920053597356e-05, | |
| "loss": 0.5431, | |
| "num_input_tokens_seen": 73400320, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.51985559566787, | |
| "grad_norm": 0.0347721092402935, | |
| "learning_rate": 2.329393966588323e-05, | |
| "loss": 0.5471, | |
| "num_input_tokens_seen": 75497472, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5342960288808665, | |
| "grad_norm": 0.03457929939031601, | |
| "learning_rate": 2.2160495348738123e-05, | |
| "loss": 0.542, | |
| "num_input_tokens_seen": 77594624, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5487364620938628, | |
| "grad_norm": 0.035683248192071915, | |
| "learning_rate": 2.1032936338593718e-05, | |
| "loss": 0.542, | |
| "num_input_tokens_seen": 79691776, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5631768953068592, | |
| "grad_norm": 0.03531257063150406, | |
| "learning_rate": 1.991359967368416e-05, | |
| "loss": 0.542, | |
| "num_input_tokens_seen": 81788928, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5776173285198556, | |
| "grad_norm": 0.034907545894384384, | |
| "learning_rate": 1.8804805350177505e-05, | |
| "loss": 0.5425, | |
| "num_input_tokens_seen": 83886080, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.592057761732852, | |
| "grad_norm": 0.03420661389827728, | |
| "learning_rate": 1.7708851513629377e-05, | |
| "loss": 0.5482, | |
| "num_input_tokens_seen": 85983232, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.6064981949458483, | |
| "grad_norm": 0.03401198983192444, | |
| "learning_rate": 1.6628009695725346e-05, | |
| "loss": 0.5598, | |
| "num_input_tokens_seen": 88080384, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6209386281588448, | |
| "grad_norm": 0.03434673324227333, | |
| "learning_rate": 1.5564520106184644e-05, | |
| "loss": 0.5402, | |
| "num_input_tokens_seen": 90177536, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6353790613718412, | |
| "grad_norm": 0.033791348338127136, | |
| "learning_rate": 1.4520586989583406e-05, | |
| "loss": 0.5413, | |
| "num_input_tokens_seen": 92274688, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6498194945848376, | |
| "grad_norm": 0.03330031782388687, | |
| "learning_rate": 1.3498374056721197e-05, | |
| "loss": 0.556, | |
| "num_input_tokens_seen": 94371840, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6642599277978339, | |
| "grad_norm": 0.03298752009868622, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 0.5341, | |
| "num_input_tokens_seen": 96468992, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.6787003610108303, | |
| "grad_norm": 0.03363870084285736, | |
| "learning_rate": 1.1527534102110612e-05, | |
| "loss": 0.5436, | |
| "num_input_tokens_seen": 98566144, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6931407942238267, | |
| "grad_norm": 0.032934173941612244, | |
| "learning_rate": 1.0582991947128324e-05, | |
| "loss": 0.5393, | |
| "num_input_tokens_seen": 100663296, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7075812274368231, | |
| "grad_norm": 0.03438662365078926, | |
| "learning_rate": 9.668331242907089e-06, | |
| "loss": 0.5598, | |
| "num_input_tokens_seen": 102760448, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7220216606498195, | |
| "grad_norm": 0.03351249918341637, | |
| "learning_rate": 8.785447763431101e-06, | |
| "loss": 0.5329, | |
| "num_input_tokens_seen": 104857600, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7364620938628159, | |
| "grad_norm": 0.0330129936337471, | |
| "learning_rate": 7.936171419533653e-06, | |
| "loss": 0.5384, | |
| "num_input_tokens_seen": 106954752, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7509025270758123, | |
| "grad_norm": 0.03433903306722641, | |
| "learning_rate": 7.122262466127514e-06, | |
| "loss": 0.5447, | |
| "num_input_tokens_seen": 109051904, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.7653429602888087, | |
| "grad_norm": 0.03314002603292465, | |
| "learning_rate": 6.3454078538078635e-06, | |
| "loss": 0.5291, | |
| "num_input_tokens_seen": 111149056, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.779783393501805, | |
| "grad_norm": 0.032993488013744354, | |
| "learning_rate": 5.607217732389503e-06, | |
| "loss": 0.5258, | |
| "num_input_tokens_seen": 113246208, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7942238267148014, | |
| "grad_norm": 0.032629404217004776, | |
| "learning_rate": 4.9092221136255444e-06, | |
| "loss": 0.5508, | |
| "num_input_tokens_seen": 115343360, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8086642599277978, | |
| "grad_norm": 0.03205695375800133, | |
| "learning_rate": 4.252867700024374e-06, | |
| "loss": 0.5439, | |
| "num_input_tokens_seen": 117440512, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.8231046931407943, | |
| "grad_norm": 0.03285016119480133, | |
| "learning_rate": 3.6395148863377858e-06, | |
| "loss": 0.5261, | |
| "num_input_tokens_seen": 119537664, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8375451263537906, | |
| "grad_norm": 0.035063523799180984, | |
| "learning_rate": 3.0704349399351435e-06, | |
| "loss": 0.5518, | |
| "num_input_tokens_seen": 121634816, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.851985559566787, | |
| "grad_norm": 0.03116844967007637, | |
| "learning_rate": 2.5468073659076e-06, | |
| "loss": 0.5444, | |
| "num_input_tokens_seen": 123731968, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.8664259927797834, | |
| "grad_norm": 0.03255166485905647, | |
| "learning_rate": 2.0697174623636794e-06, | |
| "loss": 0.5469, | |
| "num_input_tokens_seen": 125829120, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8808664259927798, | |
| "grad_norm": 0.03249025344848633, | |
| "learning_rate": 1.6401540709832242e-06, | |
| "loss": 0.5335, | |
| "num_input_tokens_seen": 127926272, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.8953068592057761, | |
| "grad_norm": 0.03506583720445633, | |
| "learning_rate": 1.2590075274920205e-06, | |
| "loss": 0.5367, | |
| "num_input_tokens_seen": 130023424, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.9097472924187726, | |
| "grad_norm": 0.03293128311634064, | |
| "learning_rate": 9.270678163050217e-07, | |
| "loss": 0.551, | |
| "num_input_tokens_seen": 132120576, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.924187725631769, | |
| "grad_norm": 0.03347219526767731, | |
| "learning_rate": 6.450229331630253e-07, | |
| "loss": 0.5597, | |
| "num_input_tokens_seen": 134217728, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.9386281588447654, | |
| "grad_norm": 0.03135489672422409, | |
| "learning_rate": 4.134574591564494e-07, | |
| "loss": 0.5369, | |
| "num_input_tokens_seen": 136314880, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9530685920577617, | |
| "grad_norm": 0.03153960779309273, | |
| "learning_rate": 2.3285134909173112e-07, | |
| "loss": 0.5451, | |
| "num_input_tokens_seen": 138412032, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.9675090252707581, | |
| "grad_norm": 0.03191647306084633, | |
| "learning_rate": 1.0357893671171792e-07, | |
| "loss": 0.5237, | |
| "num_input_tokens_seen": 140509184, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.9819494584837545, | |
| "grad_norm": 0.03295959159731865, | |
| "learning_rate": 2.590815883181108e-08, | |
| "loss": 0.5604, | |
| "num_input_tokens_seen": 142606336, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.9963898916967509, | |
| "grad_norm": 0.03191553056240082, | |
| "learning_rate": 0.0, | |
| "loss": 0.534, | |
| "num_input_tokens_seen": 144703488, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.9963898916967509, | |
| "num_input_tokens_seen": 144703488, | |
| "step": 69, | |
| "total_flos": 5.635565866281075e+18, | |
| "train_loss": 0.5665888682655666, | |
| "train_runtime": 10913.0245, | |
| "train_samples_per_second": 3.247, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 69, | |
| "num_input_tokens_seen": 144703488, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.635565866281075e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |