| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01465684657945843, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002931369315891686, | |
| "grad_norm": 0.016307968646287918, | |
| "learning_rate": 2e-05, | |
| "loss": 11.9301, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002931369315891686, | |
| "eval_loss": 11.929496765136719, | |
| "eval_runtime": 24.7836, | |
| "eval_samples_per_second": 115.923, | |
| "eval_steps_per_second": 57.982, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005862738631783372, | |
| "grad_norm": 0.01964600197970867, | |
| "learning_rate": 4e-05, | |
| "loss": 11.9297, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008794107947675058, | |
| "grad_norm": 0.014348401688039303, | |
| "learning_rate": 6e-05, | |
| "loss": 11.9295, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0011725477263566744, | |
| "grad_norm": 0.019164111465215683, | |
| "learning_rate": 8e-05, | |
| "loss": 11.9306, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0014656846579458429, | |
| "grad_norm": 0.019603189080953598, | |
| "learning_rate": 0.0001, | |
| "loss": 11.93, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0014656846579458429, | |
| "eval_loss": 11.929431915283203, | |
| "eval_runtime": 24.8869, | |
| "eval_samples_per_second": 115.442, | |
| "eval_steps_per_second": 57.741, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0017588215895350116, | |
| "grad_norm": 0.016719115898013115, | |
| "learning_rate": 0.00012, | |
| "loss": 11.9301, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0020519585211241803, | |
| "grad_norm": 0.019666332751512527, | |
| "learning_rate": 0.00014, | |
| "loss": 11.9303, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0023450954527133487, | |
| "grad_norm": 0.02202892117202282, | |
| "learning_rate": 0.00016, | |
| "loss": 11.93, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0026382323843025172, | |
| "grad_norm": 0.02416735514998436, | |
| "learning_rate": 0.00018, | |
| "loss": 11.9272, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0029313693158916857, | |
| "grad_norm": 0.020293623208999634, | |
| "learning_rate": 0.0002, | |
| "loss": 11.9288, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0029313693158916857, | |
| "eval_loss": 11.929161071777344, | |
| "eval_runtime": 24.9042, | |
| "eval_samples_per_second": 115.362, | |
| "eval_steps_per_second": 57.701, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0032245062474808546, | |
| "grad_norm": 0.019572442397475243, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 11.9261, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.003517643179070023, | |
| "grad_norm": 0.017557360231876373, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 11.9293, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0038107801106591916, | |
| "grad_norm": 0.016772069036960602, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 11.9312, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0041039170422483605, | |
| "grad_norm": 0.024416593834757805, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 11.9293, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004397053973837529, | |
| "grad_norm": 0.025569766759872437, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 11.9288, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004397053973837529, | |
| "eval_loss": 11.928751945495605, | |
| "eval_runtime": 24.8857, | |
| "eval_samples_per_second": 115.448, | |
| "eval_steps_per_second": 57.744, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0046901909054266975, | |
| "grad_norm": 0.018450522795319557, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 11.929, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004983327837015866, | |
| "grad_norm": 0.02540157176554203, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 11.9273, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0052764647686050345, | |
| "grad_norm": 0.0236028004437685, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 11.9276, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005569601700194203, | |
| "grad_norm": 0.02670389786362648, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 11.9295, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.005862738631783371, | |
| "grad_norm": 0.02101299911737442, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 11.9294, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005862738631783371, | |
| "eval_loss": 11.928288459777832, | |
| "eval_runtime": 24.834, | |
| "eval_samples_per_second": 115.688, | |
| "eval_steps_per_second": 57.864, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00615587556337254, | |
| "grad_norm": 0.02786540612578392, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 11.9298, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006449012494961709, | |
| "grad_norm": 0.024821242317557335, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 11.928, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.006742149426550877, | |
| "grad_norm": 0.026225125417113304, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 11.9285, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.007035286358140046, | |
| "grad_norm": 0.026419425383210182, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 11.9279, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007328423289729215, | |
| "grad_norm": 0.027492975816130638, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 11.9283, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.007328423289729215, | |
| "eval_loss": 11.927799224853516, | |
| "eval_runtime": 24.8718, | |
| "eval_samples_per_second": 115.512, | |
| "eval_steps_per_second": 57.776, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.007621560221318383, | |
| "grad_norm": 0.03293245658278465, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 11.9269, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.007914697152907551, | |
| "grad_norm": 0.027991851791739464, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 11.9285, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008207834084496721, | |
| "grad_norm": 0.030205611139535904, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 11.9266, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.008500971016085889, | |
| "grad_norm": 0.02978375367820263, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 11.928, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.008794107947675057, | |
| "grad_norm": 0.03635377064347267, | |
| "learning_rate": 0.0001, | |
| "loss": 11.9275, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008794107947675057, | |
| "eval_loss": 11.927339553833008, | |
| "eval_runtime": 24.8896, | |
| "eval_samples_per_second": 115.43, | |
| "eval_steps_per_second": 57.735, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009087244879264227, | |
| "grad_norm": 0.03566306084394455, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 11.9261, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.009380381810853395, | |
| "grad_norm": 0.03900522366166115, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 11.9265, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.009673518742442563, | |
| "grad_norm": 0.04783564433455467, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 11.9261, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.009966655674031733, | |
| "grad_norm": 0.03774084150791168, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 11.9276, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.010259792605620901, | |
| "grad_norm": 0.0395960696041584, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 11.9252, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.010259792605620901, | |
| "eval_loss": 11.926952362060547, | |
| "eval_runtime": 24.7995, | |
| "eval_samples_per_second": 115.849, | |
| "eval_steps_per_second": 57.945, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.010552929537210069, | |
| "grad_norm": 0.031245063990354538, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 11.9256, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010846066468799239, | |
| "grad_norm": 0.03537562116980553, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 11.9264, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.011139203400388407, | |
| "grad_norm": 0.03238970413804054, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 11.9279, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.011432340331977575, | |
| "grad_norm": 0.03814040124416351, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 11.9251, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.011725477263566743, | |
| "grad_norm": 0.03654216229915619, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 11.9248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.011725477263566743, | |
| "eval_loss": 11.926708221435547, | |
| "eval_runtime": 24.7998, | |
| "eval_samples_per_second": 115.848, | |
| "eval_steps_per_second": 57.944, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.012018614195155913, | |
| "grad_norm": 0.04135998338460922, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 11.9255, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.01231175112674508, | |
| "grad_norm": 0.03332056105136871, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 11.9267, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.012604888058334249, | |
| "grad_norm": 0.038812343031167984, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 11.9276, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.012898024989923419, | |
| "grad_norm": 0.04861786961555481, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 11.9272, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.013191161921512587, | |
| "grad_norm": 0.036478109657764435, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 11.9273, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013191161921512587, | |
| "eval_loss": 11.926604270935059, | |
| "eval_runtime": 24.9172, | |
| "eval_samples_per_second": 115.302, | |
| "eval_steps_per_second": 57.671, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013484298853101755, | |
| "grad_norm": 0.047678880393505096, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 11.925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.013777435784690924, | |
| "grad_norm": 0.03523854911327362, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 11.9267, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.014070572716280092, | |
| "grad_norm": 0.041060738265514374, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 11.9272, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01436370964786926, | |
| "grad_norm": 0.039238590747117996, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 11.9262, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.01465684657945843, | |
| "grad_norm": 0.03868459165096283, | |
| "learning_rate": 0.0, | |
| "loss": 11.9278, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01465684657945843, | |
| "eval_loss": 11.92658519744873, | |
| "eval_runtime": 24.9438, | |
| "eval_samples_per_second": 115.179, | |
| "eval_steps_per_second": 57.61, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 74946969600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |