| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.021105951878429716, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00042211903756859433, | |
| "grad_norm": 9.35902214050293, | |
| "learning_rate": 2e-05, | |
| "loss": 10.0563, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00042211903756859433, | |
| "eval_loss": 2.3652446269989014, | |
| "eval_runtime": 134.5359, | |
| "eval_samples_per_second": 7.418, | |
| "eval_steps_per_second": 3.709, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008442380751371887, | |
| "grad_norm": 7.56370735168457, | |
| "learning_rate": 4e-05, | |
| "loss": 7.42, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.001266357112705783, | |
| "grad_norm": 9.696252822875977, | |
| "learning_rate": 6e-05, | |
| "loss": 10.9588, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0016884761502743773, | |
| "grad_norm": 8.566594123840332, | |
| "learning_rate": 8e-05, | |
| "loss": 10.168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.002110595187842972, | |
| "grad_norm": 7.462744235992432, | |
| "learning_rate": 0.0001, | |
| "loss": 8.2985, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002110595187842972, | |
| "eval_loss": 2.258758783340454, | |
| "eval_runtime": 135.8864, | |
| "eval_samples_per_second": 7.344, | |
| "eval_steps_per_second": 3.672, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002532714225411566, | |
| "grad_norm": 11.242934226989746, | |
| "learning_rate": 0.00012, | |
| "loss": 8.9401, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0029548332629801602, | |
| "grad_norm": 13.016233444213867, | |
| "learning_rate": 0.00014, | |
| "loss": 8.0107, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0033769523005487546, | |
| "grad_norm": 16.885713577270508, | |
| "learning_rate": 0.00016, | |
| "loss": 7.0748, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.003799071338117349, | |
| "grad_norm": 12.32886791229248, | |
| "learning_rate": 0.00018, | |
| "loss": 6.1376, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004221190375685944, | |
| "grad_norm": 19.14803123474121, | |
| "learning_rate": 0.0002, | |
| "loss": 6.8045, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004221190375685944, | |
| "eval_loss": 1.6286500692367554, | |
| "eval_runtime": 135.9287, | |
| "eval_samples_per_second": 7.342, | |
| "eval_steps_per_second": 3.671, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004643309413254538, | |
| "grad_norm": 18.249591827392578, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 5.6408, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.005065428450823132, | |
| "grad_norm": 8.969521522521973, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 3.3644, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005487547488391726, | |
| "grad_norm": 9.032382011413574, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 5.2804, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0059096665259603205, | |
| "grad_norm": 11.984638214111328, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 4.695, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.006331785563528915, | |
| "grad_norm": 14.346260070800781, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 6.4391, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006331785563528915, | |
| "eval_loss": 1.3743129968643188, | |
| "eval_runtime": 135.8966, | |
| "eval_samples_per_second": 7.344, | |
| "eval_steps_per_second": 3.672, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006753904601097509, | |
| "grad_norm": 11.886094093322754, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 5.4466, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.007176023638666104, | |
| "grad_norm": 12.336444854736328, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.6035, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.007598142676234698, | |
| "grad_norm": 19.94431495666504, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 5.2473, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008020261713803293, | |
| "grad_norm": 9.52234172821045, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 3.7112, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.008442380751371888, | |
| "grad_norm": 11.920819282531738, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.9923, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008442380751371888, | |
| "eval_loss": 1.2984857559204102, | |
| "eval_runtime": 135.8981, | |
| "eval_samples_per_second": 7.344, | |
| "eval_steps_per_second": 3.672, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008864499788940482, | |
| "grad_norm": 11.649659156799316, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 5.2017, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.009286618826509076, | |
| "grad_norm": 10.896583557128906, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 5.5399, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.009708737864077669, | |
| "grad_norm": 10.63659381866455, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 5.016, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.010130856901646263, | |
| "grad_norm": 11.454913139343262, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 5.6765, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.010552975939214858, | |
| "grad_norm": 10.400849342346191, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 6.1088, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.010552975939214858, | |
| "eval_loss": 1.2727545499801636, | |
| "eval_runtime": 135.9196, | |
| "eval_samples_per_second": 7.343, | |
| "eval_steps_per_second": 3.671, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.010975094976783452, | |
| "grad_norm": 12.741352081298828, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 5.4666, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.011397214014352047, | |
| "grad_norm": 17.090585708618164, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 4.8389, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.011819333051920641, | |
| "grad_norm": 12.481046676635742, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 4.6741, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.012241452089489235, | |
| "grad_norm": 10.3117094039917, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 5.4316, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01266357112705783, | |
| "grad_norm": 7.214309215545654, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7213, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01266357112705783, | |
| "eval_loss": 1.2609025239944458, | |
| "eval_runtime": 135.9083, | |
| "eval_samples_per_second": 7.343, | |
| "eval_steps_per_second": 3.672, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013085690164626424, | |
| "grad_norm": 18.718671798706055, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 5.4732, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.013507809202195019, | |
| "grad_norm": 11.435547828674316, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 4.9203, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.013929928239763613, | |
| "grad_norm": 10.921402931213379, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 4.5804, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.014352047277332207, | |
| "grad_norm": 19.59139633178711, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 5.5784, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.014774166314900802, | |
| "grad_norm": 7.975940704345703, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 5.5038, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.014774166314900802, | |
| "eval_loss": 1.2495229244232178, | |
| "eval_runtime": 135.8846, | |
| "eval_samples_per_second": 7.344, | |
| "eval_steps_per_second": 3.672, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.015196285352469396, | |
| "grad_norm": 9.815423965454102, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 4.815, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01561840439003799, | |
| "grad_norm": 10.220743179321289, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 4.7406, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.016040523427606587, | |
| "grad_norm": 10.354463577270508, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 4.5756, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.01646264246517518, | |
| "grad_norm": 11.346120834350586, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 6.2784, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.016884761502743775, | |
| "grad_norm": 9.713720321655273, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 5.7159, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016884761502743775, | |
| "eval_loss": 1.2519896030426025, | |
| "eval_runtime": 136.0325, | |
| "eval_samples_per_second": 7.336, | |
| "eval_steps_per_second": 3.668, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017306880540312368, | |
| "grad_norm": 13.019607543945312, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 4.0586, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.017728999577880964, | |
| "grad_norm": 9.759472846984863, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 5.9911, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.018151118615449557, | |
| "grad_norm": 9.983514785766602, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 4.5841, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.018573237653018153, | |
| "grad_norm": 11.719050407409668, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 6.0421, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.018995356690586745, | |
| "grad_norm": 11.842796325683594, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 5.632, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018995356690586745, | |
| "eval_loss": 1.2501074075698853, | |
| "eval_runtime": 136.0353, | |
| "eval_samples_per_second": 7.336, | |
| "eval_steps_per_second": 3.668, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.019417475728155338, | |
| "grad_norm": 13.858030319213867, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 4.1389, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.019839594765723934, | |
| "grad_norm": 12.51378345489502, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 4.6694, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.020261713803292527, | |
| "grad_norm": 14.40710163116455, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 4.5467, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.020683832840861123, | |
| "grad_norm": 12.680715560913086, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 4.3332, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.021105951878429716, | |
| "grad_norm": 10.62983512878418, | |
| "learning_rate": 0.0, | |
| "loss": 5.3688, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021105951878429716, | |
| "eval_loss": 1.2478138208389282, | |
| "eval_runtime": 135.9744, | |
| "eval_samples_per_second": 7.34, | |
| "eval_steps_per_second": 3.67, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 80, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.25746627182592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |