| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01345985597954102, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00026919711959082036, | |
| "grad_norm": 7.305245876312256, | |
| "learning_rate": 2e-05, | |
| "loss": 6.0527, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00026919711959082036, | |
| "eval_loss": 6.004375457763672, | |
| "eval_runtime": 186.7841, | |
| "eval_samples_per_second": 8.379, | |
| "eval_steps_per_second": 4.192, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005383942391816407, | |
| "grad_norm": 6.773212432861328, | |
| "learning_rate": 4e-05, | |
| "loss": 6.151, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008075913587724611, | |
| "grad_norm": 7.554437160491943, | |
| "learning_rate": 6e-05, | |
| "loss": 6.398, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010767884783632815, | |
| "grad_norm": 6.408179759979248, | |
| "learning_rate": 8e-05, | |
| "loss": 5.7049, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0010767884783632815, | |
| "eval_loss": 5.3220696449279785, | |
| "eval_runtime": 187.1601, | |
| "eval_samples_per_second": 8.362, | |
| "eval_steps_per_second": 4.184, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0013459855979541019, | |
| "grad_norm": 6.113898754119873, | |
| "learning_rate": 0.0001, | |
| "loss": 5.5139, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016151827175449223, | |
| "grad_norm": 4.3670973777771, | |
| "learning_rate": 0.00012, | |
| "loss": 4.8011, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0018843798371357427, | |
| "grad_norm": 8.779766082763672, | |
| "learning_rate": 0.00014, | |
| "loss": 4.9951, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002153576956726563, | |
| "grad_norm": 8.467183113098145, | |
| "learning_rate": 0.00016, | |
| "loss": 4.8208, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.002153576956726563, | |
| "eval_loss": 4.797426700592041, | |
| "eval_runtime": 187.0275, | |
| "eval_samples_per_second": 8.368, | |
| "eval_steps_per_second": 4.187, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0024227740763173833, | |
| "grad_norm": 6.028713226318359, | |
| "learning_rate": 0.00018, | |
| "loss": 4.9807, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0026919711959082038, | |
| "grad_norm": 3.5283517837524414, | |
| "learning_rate": 0.0002, | |
| "loss": 4.4461, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002961168315499024, | |
| "grad_norm": 3.948577880859375, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 4.0834, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "grad_norm": 3.531851053237915, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 4.6573, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0032303654350898446, | |
| "eval_loss": 4.350089073181152, | |
| "eval_runtime": 186.8405, | |
| "eval_samples_per_second": 8.376, | |
| "eval_steps_per_second": 4.191, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003499562554680665, | |
| "grad_norm": 3.562119483947754, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 4.4314, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0037687596742714854, | |
| "grad_norm": 3.721893548965454, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 4.4538, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004037956793862305, | |
| "grad_norm": 2.8506362438201904, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 4.2002, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004307153913453126, | |
| "grad_norm": 2.9634995460510254, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 4.3413, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004307153913453126, | |
| "eval_loss": 4.165503978729248, | |
| "eval_runtime": 187.0045, | |
| "eval_samples_per_second": 8.369, | |
| "eval_steps_per_second": 4.187, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004576351033043946, | |
| "grad_norm": 3.4099552631378174, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.133, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.004845548152634767, | |
| "grad_norm": 3.074971914291382, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 3.9521, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005114745272225587, | |
| "grad_norm": 2.5733466148376465, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 4.1701, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0053839423918164075, | |
| "grad_norm": 3.7706105709075928, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.9503, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0053839423918164075, | |
| "eval_loss": 4.0563483238220215, | |
| "eval_runtime": 187.0013, | |
| "eval_samples_per_second": 8.369, | |
| "eval_steps_per_second": 4.187, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005653139511407228, | |
| "grad_norm": 2.7933688163757324, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 3.8037, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.005922336630998048, | |
| "grad_norm": 2.6206228733062744, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 4.1224, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.006191533750588869, | |
| "grad_norm": 2.956394910812378, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 3.8059, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "grad_norm": 2.445277690887451, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 3.8613, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006460730870179689, | |
| "eval_loss": 3.9950814247131348, | |
| "eval_runtime": 186.8191, | |
| "eval_samples_per_second": 8.377, | |
| "eval_steps_per_second": 4.191, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00672992798977051, | |
| "grad_norm": 2.6795260906219482, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 3.646, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00699912510936133, | |
| "grad_norm": 2.925718307495117, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 4.1437, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0072683222289521504, | |
| "grad_norm": 2.81306529045105, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 3.9201, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007537519348542971, | |
| "grad_norm": 2.4982492923736572, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 3.8325, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.007537519348542971, | |
| "eval_loss": 3.943394660949707, | |
| "eval_runtime": 187.1247, | |
| "eval_samples_per_second": 8.363, | |
| "eval_steps_per_second": 4.184, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.007806716468133791, | |
| "grad_norm": 2.46905255317688, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 4.1957, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.00807591358772461, | |
| "grad_norm": 2.6972992420196533, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7234, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008345110707315431, | |
| "grad_norm": 2.946199655532837, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 3.8905, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.008614307826906252, | |
| "grad_norm": 2.631840944290161, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 3.8763, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.008614307826906252, | |
| "eval_loss": 3.9116568565368652, | |
| "eval_runtime": 187.6034, | |
| "eval_samples_per_second": 8.342, | |
| "eval_steps_per_second": 4.174, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.008883504946497072, | |
| "grad_norm": 2.5778937339782715, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 3.8609, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.009152702066087893, | |
| "grad_norm": 2.673618793487549, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 3.7293, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.009421899185678713, | |
| "grad_norm": 2.699422597885132, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 4.1321, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.009691096305269533, | |
| "grad_norm": 2.799333095550537, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 3.8541, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009691096305269533, | |
| "eval_loss": 3.8941519260406494, | |
| "eval_runtime": 187.6155, | |
| "eval_samples_per_second": 8.342, | |
| "eval_steps_per_second": 4.173, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.009960293424860354, | |
| "grad_norm": 2.519052267074585, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 3.8805, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.010229490544451174, | |
| "grad_norm": 2.59364652633667, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 4.0401, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.010498687664041995, | |
| "grad_norm": 2.6932904720306396, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 3.6313, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.010767884783632815, | |
| "grad_norm": 2.4584312438964844, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 3.9238, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.010767884783632815, | |
| "eval_loss": 3.8838305473327637, | |
| "eval_runtime": 188.8048, | |
| "eval_samples_per_second": 8.289, | |
| "eval_steps_per_second": 4.147, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.011037081903223635, | |
| "grad_norm": 2.763805389404297, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 3.7855, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.011306279022814456, | |
| "grad_norm": 2.3128108978271484, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 3.494, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.011575476142405276, | |
| "grad_norm": 3.1583259105682373, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 4.1824, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.011844673261996097, | |
| "grad_norm": 2.4732134342193604, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 3.8976, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.011844673261996097, | |
| "eval_loss": 3.8724188804626465, | |
| "eval_runtime": 187.5642, | |
| "eval_samples_per_second": 8.344, | |
| "eval_steps_per_second": 4.175, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.012113870381586917, | |
| "grad_norm": 2.592027187347412, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 3.7669, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.012383067501177738, | |
| "grad_norm": 2.7980010509490967, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 4.1847, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.012652264620768558, | |
| "grad_norm": 2.249249219894409, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 3.7236, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.012921461740359378, | |
| "grad_norm": 2.2937920093536377, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 4.0901, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.012921461740359378, | |
| "eval_loss": 3.869504928588867, | |
| "eval_runtime": 186.9808, | |
| "eval_samples_per_second": 8.37, | |
| "eval_steps_per_second": 4.188, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.013190658859950199, | |
| "grad_norm": 2.3129959106445312, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 3.8164, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.01345985597954102, | |
| "grad_norm": 2.5126748085021973, | |
| "learning_rate": 0.0, | |
| "loss": 3.8391, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.83447180771328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |