| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.025288179883252902, | |
| "eval_steps": 13, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016858786588835268, | |
| "eval_loss": 0.9273973107337952, | |
| "eval_runtime": 305.4938, | |
| "eval_samples_per_second": 32.701, | |
| "eval_steps_per_second": 8.177, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005057635976650581, | |
| "grad_norm": 0.5028714537620544, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9043, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010115271953301161, | |
| "grad_norm": 0.5011371970176697, | |
| "learning_rate": 6e-05, | |
| "loss": 0.8721, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0015172907929951742, | |
| "grad_norm": 0.8638129234313965, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8597, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0020230543906602323, | |
| "grad_norm": 0.5769167542457581, | |
| "learning_rate": 9.994965332706573e-05, | |
| "loss": 0.8598, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002191642256548585, | |
| "eval_loss": 0.8061522245407104, | |
| "eval_runtime": 306.4617, | |
| "eval_samples_per_second": 32.598, | |
| "eval_steps_per_second": 8.151, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00252881798832529, | |
| "grad_norm": 0.3618827164173126, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.7392, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0030345815859903484, | |
| "grad_norm": 0.3715335428714752, | |
| "learning_rate": 9.919647942993148e-05, | |
| "loss": 0.77, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0035403451836554063, | |
| "grad_norm": 0.35133299231529236, | |
| "learning_rate": 9.848447601883435e-05, | |
| "loss": 0.7486, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004046108781320465, | |
| "grad_norm": 0.39814493060112, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.7931, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00438328451309717, | |
| "eval_loss": 0.779468834400177, | |
| "eval_runtime": 306.7465, | |
| "eval_samples_per_second": 32.568, | |
| "eval_steps_per_second": 8.144, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.004551872378985523, | |
| "grad_norm": 0.35872843861579895, | |
| "learning_rate": 9.640574942595196e-05, | |
| "loss": 0.7735, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00505763597665058, | |
| "grad_norm": 0.334773987531662, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.7485, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005563399574315639, | |
| "grad_norm": 0.3601718544960022, | |
| "learning_rate": 9.348705665778478e-05, | |
| "loss": 0.8453, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.006069163171980697, | |
| "grad_norm": 0.3790546655654907, | |
| "learning_rate": 9.172866268606513e-05, | |
| "loss": 0.7661, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006574926769645755, | |
| "grad_norm": 0.2799323499202728, | |
| "learning_rate": 8.978122744408906e-05, | |
| "loss": 0.7374, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.006574926769645755, | |
| "eval_loss": 0.7738218903541565, | |
| "eval_runtime": 306.4708, | |
| "eval_samples_per_second": 32.597, | |
| "eval_steps_per_second": 8.151, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.007080690367310813, | |
| "grad_norm": 0.3466595411300659, | |
| "learning_rate": 8.765357330018056e-05, | |
| "loss": 0.7734, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.007586453964975871, | |
| "grad_norm": 0.33398839831352234, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7472, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00809221756264093, | |
| "grad_norm": 0.3427945673465729, | |
| "learning_rate": 8.289693629698564e-05, | |
| "loss": 0.7724, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.008597981160305987, | |
| "grad_norm": 0.33757373690605164, | |
| "learning_rate": 8.0289502192041e-05, | |
| "loss": 0.7546, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.00876656902619434, | |
| "eval_loss": 0.7691455483436584, | |
| "eval_runtime": 306.7906, | |
| "eval_samples_per_second": 32.563, | |
| "eval_steps_per_second": 8.142, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.009103744757971046, | |
| "grad_norm": 0.32756149768829346, | |
| "learning_rate": 7.754484907260513e-05, | |
| "loss": 0.7358, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.009609508355636103, | |
| "grad_norm": 0.3512526750564575, | |
| "learning_rate": 7.467541090321735e-05, | |
| "loss": 0.8495, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01011527195330116, | |
| "grad_norm": 0.3337990939617157, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.7544, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01062103555096622, | |
| "grad_norm": 0.2693156599998474, | |
| "learning_rate": 6.861468292009727e-05, | |
| "loss": 0.7507, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.010958211282742925, | |
| "eval_loss": 0.7673783302307129, | |
| "eval_runtime": 306.4834, | |
| "eval_samples_per_second": 32.596, | |
| "eval_steps_per_second": 8.151, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.011126799148631277, | |
| "grad_norm": 0.34625276923179626, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.8253, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.011632562746296335, | |
| "grad_norm": 0.32037216424942017, | |
| "learning_rate": 6.22170203068947e-05, | |
| "loss": 0.7514, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.012138326343961394, | |
| "grad_norm": 0.2649615406990051, | |
| "learning_rate": 5.8927844739931834e-05, | |
| "loss": 0.7331, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.012644089941626451, | |
| "grad_norm": 0.3438839018344879, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.7573, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01314985353929151, | |
| "grad_norm": 0.3098960518836975, | |
| "learning_rate": 5.2243241517525754e-05, | |
| "loss": 0.7062, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.01314985353929151, | |
| "eval_loss": 0.7658212184906006, | |
| "eval_runtime": 306.7841, | |
| "eval_samples_per_second": 32.564, | |
| "eval_steps_per_second": 8.143, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.013655617136956568, | |
| "grad_norm": 0.32385092973709106, | |
| "learning_rate": 4.887809678520976e-05, | |
| "loss": 0.7489, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.014161380734621625, | |
| "grad_norm": 0.3278336822986603, | |
| "learning_rate": 4.551803455482833e-05, | |
| "loss": 0.7979, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.014667144332286684, | |
| "grad_norm": 0.2766069769859314, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.7464, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.015172907929951742, | |
| "grad_norm": 0.3032001256942749, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.6977, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.015341495795840094, | |
| "eval_loss": 0.7647122740745544, | |
| "eval_runtime": 306.5924, | |
| "eval_samples_per_second": 32.584, | |
| "eval_steps_per_second": 8.148, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.0156786715276168, | |
| "grad_norm": 0.3444535434246063, | |
| "learning_rate": 3.562003362839914e-05, | |
| "loss": 0.7746, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.01618443512528186, | |
| "grad_norm": 0.3450552225112915, | |
| "learning_rate": 3.243125879593286e-05, | |
| "loss": 0.76, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.016690198722946917, | |
| "grad_norm": 0.3032662868499756, | |
| "learning_rate": 2.932207475167398e-05, | |
| "loss": 0.7742, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.017195962320611973, | |
| "grad_norm": 0.34202754497528076, | |
| "learning_rate": 2.630656687635007e-05, | |
| "loss": 0.7749, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.01753313805238868, | |
| "eval_loss": 0.7638227343559265, | |
| "eval_runtime": 306.6036, | |
| "eval_samples_per_second": 32.583, | |
| "eval_steps_per_second": 8.147, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.017701725918277032, | |
| "grad_norm": 0.3035350739955902, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.7123, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.01820748951594209, | |
| "grad_norm": 0.2928071916103363, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.7931, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.018713253113607147, | |
| "grad_norm": 0.3145976960659027, | |
| "learning_rate": 1.7956219300748793e-05, | |
| "loss": 0.7774, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.019219016711272206, | |
| "grad_norm": 0.32552438974380493, | |
| "learning_rate": 1.544686755065677e-05, | |
| "loss": 0.7435, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.019724780308937265, | |
| "grad_norm": 0.304483562707901, | |
| "learning_rate": 1.3094050125632972e-05, | |
| "loss": 0.7509, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.019724780308937265, | |
| "eval_loss": 0.7630516886711121, | |
| "eval_runtime": 306.6709, | |
| "eval_samples_per_second": 32.576, | |
| "eval_steps_per_second": 8.146, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.02023054390660232, | |
| "grad_norm": 0.3333122432231903, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.8215, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.02073630750426738, | |
| "grad_norm": 0.3269135355949402, | |
| "learning_rate": 8.899896227604509e-06, | |
| "loss": 0.8011, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.02124207110193244, | |
| "grad_norm": 0.279905766248703, | |
| "learning_rate": 7.077560319906695e-06, | |
| "loss": 0.7328, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.021747834699597495, | |
| "grad_norm": 0.2915093004703522, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.7201, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.02191642256548585, | |
| "eval_loss": 0.7627005577087402, | |
| "eval_runtime": 306.6146, | |
| "eval_samples_per_second": 32.582, | |
| "eval_steps_per_second": 8.147, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.022253598297262554, | |
| "grad_norm": 0.2774188816547394, | |
| "learning_rate": 4.023611372427471e-06, | |
| "loss": 0.6875, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.022759361894927613, | |
| "grad_norm": 0.3083136975765228, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.7822, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.02326512549259267, | |
| "grad_norm": 0.2817324101924896, | |
| "learning_rate": 1.8018569652073381e-06, | |
| "loss": 0.7197, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.02377088909025773, | |
| "grad_norm": 0.2889103293418884, | |
| "learning_rate": 1.016230078838226e-06, | |
| "loss": 0.7419, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.024108064822034436, | |
| "eval_loss": 0.7625572085380554, | |
| "eval_runtime": 306.5341, | |
| "eval_samples_per_second": 32.59, | |
| "eval_steps_per_second": 8.149, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.024276652687922787, | |
| "grad_norm": 0.30902647972106934, | |
| "learning_rate": 4.52511911603265e-07, | |
| "loss": 0.734, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.024782416285587843, | |
| "grad_norm": 0.31932878494262695, | |
| "learning_rate": 1.132562476771959e-07, | |
| "loss": 0.7666, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.025288179883252902, | |
| "grad_norm": 0.31999635696411133, | |
| "learning_rate": 0.0, | |
| "loss": 0.7732, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 150, | |
| "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": 4.26927046262784e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |