| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 678, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.778737860918045, |
| "epoch": 0.03690036900369004, |
| "grad_norm": 0.06591796875, |
| "learning_rate": 0.0001999897109785537, |
| "loss": 1.9253872680664061, |
| "mean_token_accuracy": 0.5836515012383461, |
| "num_tokens": 158319.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.3310806779563427, |
| "epoch": 0.07380073800738007, |
| "grad_norm": 0.10546875, |
| "learning_rate": 0.00019910503467261722, |
| "loss": 1.2929591369628906, |
| "mean_token_accuracy": 0.6938765767216682, |
| "num_tokens": 315084.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9595038790814578, |
| "epoch": 0.11070110701107011, |
| "grad_norm": 0.1552734375, |
| "learning_rate": 0.00019680577845157155, |
| "loss": 0.9015888214111328, |
| "mean_token_accuracy": 0.7816851404309273, |
| "num_tokens": 470867.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 0.8337875098735094, |
| "epoch": 0.14760147601476015, |
| "grad_norm": 0.12255859375, |
| "learning_rate": 0.0001931247608465915, |
| "loss": 0.7552251434326172, |
| "mean_token_accuracy": 0.812552462220192, |
| "num_tokens": 626254.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.6474238294456154, |
| "epoch": 0.18450184501845018, |
| "grad_norm": 0.189453125, |
| "learning_rate": 0.0001881145230185612, |
| "loss": 0.5701670455932617, |
| "mean_token_accuracy": 0.8557327410578728, |
| "num_tokens": 786488.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 0.5202154207415879, |
| "epoch": 0.22140221402214022, |
| "grad_norm": 0.1259765625, |
| "learning_rate": 0.00018184657880958635, |
| "loss": 0.46115348815917967, |
| "mean_token_accuracy": 0.8857239605486393, |
| "num_tokens": 943912.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.4175653001386672, |
| "epoch": 0.25830258302583026, |
| "grad_norm": 0.09033203125, |
| "learning_rate": 0.00017441039398713608, |
| "loss": 0.3705097579956055, |
| "mean_token_accuracy": 0.9097741657495498, |
| "num_tokens": 1101961.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 0.3828558912500739, |
| "epoch": 0.2952029520295203, |
| "grad_norm": 0.11181640625, |
| "learning_rate": 0.0001659121092506171, |
| "loss": 0.3168667984008789, |
| "mean_token_accuracy": 0.9167497155070304, |
| "num_tokens": 1258551.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.3276692613959312, |
| "epoch": 0.33210332103321033, |
| "grad_norm": 0.09228515625, |
| "learning_rate": 0.00015647302522759962, |
| "loss": 0.26206918716430666, |
| "mean_token_accuracy": 0.9325725418329239, |
| "num_tokens": 1414182.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 0.2656678088847548, |
| "epoch": 0.36900369003690037, |
| "grad_norm": 0.0986328125, |
| "learning_rate": 0.00014622787108416584, |
| "loss": 0.2154819107055664, |
| "mean_token_accuracy": 0.9447659783065319, |
| "num_tokens": 1571401.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.23459753211122006, |
| "epoch": 0.4059040590405904, |
| "grad_norm": 0.09375, |
| "learning_rate": 0.00013532288146244446, |
| "loss": 0.18326162338256835, |
| "mean_token_accuracy": 0.9511218988895416, |
| "num_tokens": 1727812.0, |
| "step": 275 |
| }, |
| { |
| "entropy": 0.17023452829569577, |
| "epoch": 0.44280442804428044, |
| "grad_norm": 0.1201171875, |
| "learning_rate": 0.00012391370919424485, |
| "loss": 0.12261626243591309, |
| "mean_token_accuracy": 0.9681580965220928, |
| "num_tokens": 1885993.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.17609536536969245, |
| "epoch": 0.4797047970479705, |
| "grad_norm": 0.1162109375, |
| "learning_rate": 0.00011216320358376157, |
| "loss": 0.137310152053833, |
| "mean_token_accuracy": 0.9655524456501007, |
| "num_tokens": 2040863.0, |
| "step": 325 |
| }, |
| { |
| "entropy": 0.12500815264414997, |
| "epoch": 0.5166051660516605, |
| "grad_norm": 0.1328125, |
| "learning_rate": 0.00010023908597112514, |
| "loss": 0.08844621658325195, |
| "mean_token_accuracy": 0.9776459574699402, |
| "num_tokens": 2193836.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.13451697141863406, |
| "epoch": 0.5535055350553506, |
| "grad_norm": 0.126953125, |
| "learning_rate": 8.831155575474102e-05, |
| "loss": 0.09469686508178711, |
| "mean_token_accuracy": 0.9756594082713127, |
| "num_tokens": 2352735.0, |
| "step": 375 |
| }, |
| { |
| "entropy": 0.11615390982478857, |
| "epoch": 0.5904059040590406, |
| "grad_norm": 0.08251953125, |
| "learning_rate": 7.655086104295904e-05, |
| "loss": 0.08285388946533204, |
| "mean_token_accuracy": 0.9785897681117057, |
| "num_tokens": 2506252.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.09932436655741185, |
| "epoch": 0.6273062730627307, |
| "grad_norm": 0.0247802734375, |
| "learning_rate": 6.512486861047911e-05, |
| "loss": 0.07087704181671142, |
| "mean_token_accuracy": 0.9825253394246102, |
| "num_tokens": 2661297.0, |
| "step": 425 |
| }, |
| { |
| "entropy": 0.07707942618988455, |
| "epoch": 0.6642066420664207, |
| "grad_norm": 0.1318359375, |
| "learning_rate": 5.419666784482398e-05, |
| "loss": 0.05760700225830078, |
| "mean_token_accuracy": 0.9852602830529213, |
| "num_tokens": 2816646.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.07893491809722036, |
| "epoch": 0.7011070110701108, |
| "grad_norm": 0.06591796875, |
| "learning_rate": 4.3922242883050224e-05, |
| "loss": 0.048433661460876465, |
| "mean_token_accuracy": 0.9880265095829963, |
| "num_tokens": 2973546.0, |
| "step": 475 |
| }, |
| { |
| "entropy": 0.07320529426448047, |
| "epoch": 0.7380073800738007, |
| "grad_norm": 0.03271484375, |
| "learning_rate": 3.444824616555346e-05, |
| "loss": 0.04846723079681396, |
| "mean_token_accuracy": 0.9885234755277633, |
| "num_tokens": 3129376.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.06474811009131372, |
| "epoch": 0.7749077490774908, |
| "grad_norm": 0.05859375, |
| "learning_rate": 2.590990518624098e-05, |
| "loss": 0.04171095371246338, |
| "mean_token_accuracy": 0.9884812113642693, |
| "num_tokens": 3284644.0, |
| "step": 525 |
| }, |
| { |
| "entropy": 0.06364937928505242, |
| "epoch": 0.8118081180811808, |
| "grad_norm": 0.1259765625, |
| "learning_rate": 1.8429092317163245e-05, |
| "loss": 0.047798018455505374, |
| "mean_token_accuracy": 0.9897640451788903, |
| "num_tokens": 3437690.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.07899157676845789, |
| "epoch": 0.8487084870848709, |
| "grad_norm": 0.064453125, |
| "learning_rate": 1.2112585258042964e-05, |
| "loss": 0.052181754112243656, |
| "mean_token_accuracy": 0.9872020599246025, |
| "num_tokens": 3596227.0, |
| "step": 575 |
| }, |
| { |
| "entropy": 0.08390225057024508, |
| "epoch": 0.8856088560885609, |
| "grad_norm": 0.0225830078125, |
| "learning_rate": 7.050542940250271e-06, |
| "loss": 0.05765305519104004, |
| "mean_token_accuracy": 0.9856913402676583, |
| "num_tokens": 3752314.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.06875003776978701, |
| "epoch": 0.922509225092251, |
| "grad_norm": 0.05517578125, |
| "learning_rate": 3.3152186394722505e-06, |
| "loss": 0.04065555095672607, |
| "mean_token_accuracy": 0.9894865137338639, |
| "num_tokens": 3908910.0, |
| "step": 625 |
| }, |
| { |
| "entropy": 0.07515250355005264, |
| "epoch": 0.959409594095941, |
| "grad_norm": 0.012451171875, |
| "learning_rate": 9.599286655185502e-07, |
| "loss": 0.046056933403015136, |
| "mean_token_accuracy": 0.9879890367388725, |
| "num_tokens": 4065917.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.07211493622511625, |
| "epoch": 0.996309963099631, |
| "grad_norm": 0.06298828125, |
| "learning_rate": 1.8291349712273864e-08, |
| "loss": 0.040222091674804686, |
| "mean_token_accuracy": 0.9893409749865532, |
| "num_tokens": 4217268.0, |
| "step": 675 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 678, |
| "num_input_tokens_seen": 0, |
| "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": 1.8059462405563085e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
