| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0425260472039124, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008505209440782479, | |
| "grad_norm": 7.646979331970215, | |
| "learning_rate": 5e-06, | |
| "loss": 11.2339, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008505209440782479, | |
| "eval_loss": 2.182969570159912, | |
| "eval_runtime": 18.1957, | |
| "eval_samples_per_second": 27.204, | |
| "eval_steps_per_second": 13.63, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017010418881564958, | |
| "grad_norm": 2.7476022243499756, | |
| "learning_rate": 1e-05, | |
| "loss": 8.1504, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0025515628322347436, | |
| "grad_norm": 4.530780792236328, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.479, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0034020837763129915, | |
| "grad_norm": 4.3689985275268555, | |
| "learning_rate": 2e-05, | |
| "loss": 8.6752, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00425260472039124, | |
| "grad_norm": 4.250224590301514, | |
| "learning_rate": 2.5e-05, | |
| "loss": 8.7246, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.005103125664469487, | |
| "grad_norm": 3.909888505935669, | |
| "learning_rate": 3e-05, | |
| "loss": 8.8007, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005953646608547736, | |
| "grad_norm": 4.2319536209106445, | |
| "learning_rate": 3.5e-05, | |
| "loss": 8.6001, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006804167552625983, | |
| "grad_norm": 3.7226130962371826, | |
| "learning_rate": 4e-05, | |
| "loss": 8.6729, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007654688496704231, | |
| "grad_norm": 4.978116512298584, | |
| "learning_rate": 4.5e-05, | |
| "loss": 9.7467, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00850520944078248, | |
| "grad_norm": 7.846011161804199, | |
| "learning_rate": 5e-05, | |
| "loss": 10.2529, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009355730384860728, | |
| "grad_norm": 7.632503032684326, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 10.6994, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.010206251328938975, | |
| "grad_norm": 6.259686470031738, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 9.2648, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.011056772273017223, | |
| "grad_norm": 4.645693302154541, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 8.8708, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011056772273017223, | |
| "eval_loss": 2.120561122894287, | |
| "eval_runtime": 18.1798, | |
| "eval_samples_per_second": 27.228, | |
| "eval_steps_per_second": 13.642, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011907293217095471, | |
| "grad_norm": 3.9054300785064697, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 7.6945, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01275781416117372, | |
| "grad_norm": 6.3420305252075195, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 8.6295, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013608335105251966, | |
| "grad_norm": 6.098008155822754, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 7.3762, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.014458856049330214, | |
| "grad_norm": 6.397308826446533, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 8.6222, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.015309376993408463, | |
| "grad_norm": 4.59157133102417, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 7.8273, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01615989793748671, | |
| "grad_norm": 5.998552322387695, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 9.8392, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.01701041888156496, | |
| "grad_norm": 4.757502555847168, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 8.913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.017860939825643206, | |
| "grad_norm": 5.794950485229492, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 8.4942, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.018711460769721456, | |
| "grad_norm": 7.268848419189453, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 9.6959, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.019561981713799702, | |
| "grad_norm": 4.623784065246582, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 7.3824, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02041250265787795, | |
| "grad_norm": 4.572867393493652, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 8.0812, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0212630236019562, | |
| "grad_norm": 6.01804780960083, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 8.0338, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.022113544546034446, | |
| "grad_norm": 4.490617275238037, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 7.893, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.022113544546034446, | |
| "eval_loss": 1.961138367652893, | |
| "eval_runtime": 18.1047, | |
| "eval_samples_per_second": 27.341, | |
| "eval_steps_per_second": 13.698, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.022964065490112696, | |
| "grad_norm": 3.840120553970337, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 6.86, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.023814586434190942, | |
| "grad_norm": 5.417391777038574, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 7.7778, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02466510737826919, | |
| "grad_norm": 3.8475501537323, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 5.9285, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.02551562832234744, | |
| "grad_norm": 4.8878374099731445, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.3934, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.026366149266425686, | |
| "grad_norm": 4.846117973327637, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 7.3275, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.027216670210503932, | |
| "grad_norm": 3.879229784011841, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 6.676, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.028067191154582182, | |
| "grad_norm": 5.315415859222412, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 7.6324, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02891771209866043, | |
| "grad_norm": 4.311729907989502, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 6.9114, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02976823304273868, | |
| "grad_norm": 6.797874450683594, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 8.9163, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.030618753986816925, | |
| "grad_norm": 6.546699047088623, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 9.0227, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.03146927493089517, | |
| "grad_norm": 4.277498722076416, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 7.4829, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.03231979587497342, | |
| "grad_norm": 5.2016143798828125, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 7.472, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03317031681905167, | |
| "grad_norm": 4.864084243774414, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 8.2183, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03317031681905167, | |
| "eval_loss": 1.9271198511123657, | |
| "eval_runtime": 18.0882, | |
| "eval_samples_per_second": 27.366, | |
| "eval_steps_per_second": 13.711, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03402083776312992, | |
| "grad_norm": 4.204705715179443, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 7.1296, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.034871358707208165, | |
| "grad_norm": 6.853778839111328, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 7.9356, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03572187965128641, | |
| "grad_norm": 6.098489284515381, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 9.2043, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03657240059536466, | |
| "grad_norm": 6.261697292327881, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 9.5531, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03742292153944291, | |
| "grad_norm": 5.243067741394043, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 8.3141, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03827344248352116, | |
| "grad_norm": 5.846104145050049, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 8.37, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.039123963427599405, | |
| "grad_norm": 8.3400239944458, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 8.7011, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03997448437167765, | |
| "grad_norm": 4.996219158172607, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 7.4367, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0408250053157559, | |
| "grad_norm": 4.701981067657471, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 7.4361, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.04167552625983415, | |
| "grad_norm": 5.824567794799805, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 9.0153, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0425260472039124, | |
| "grad_norm": 5.769680023193359, | |
| "learning_rate": 0.0, | |
| "loss": 8.7723, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8387975649951744.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |