| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005266165152205338, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0532330304410676e-05, | |
| "grad_norm": 0.4928787350654602, | |
| "learning_rate": 2e-05, | |
| "loss": 6.3989, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0532330304410676e-05, | |
| "eval_loss": 1.6582940816879272, | |
| "eval_runtime": 2629.8586, | |
| "eval_samples_per_second": 15.202, | |
| "eval_steps_per_second": 7.601, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.1064660608821352e-05, | |
| "grad_norm": 0.6760787963867188, | |
| "learning_rate": 4e-05, | |
| "loss": 6.5621, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.159699091323203e-05, | |
| "grad_norm": 0.540818989276886, | |
| "learning_rate": 6e-05, | |
| "loss": 6.9074, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.2129321217642704e-05, | |
| "grad_norm": 0.5401029586791992, | |
| "learning_rate": 8e-05, | |
| "loss": 6.9058, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.2661651522053385e-05, | |
| "grad_norm": 0.5291933417320251, | |
| "learning_rate": 0.0001, | |
| "loss": 6.7631, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.319398182646406e-05, | |
| "grad_norm": 0.6951285004615784, | |
| "learning_rate": 0.00012, | |
| "loss": 6.7074, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.372631213087474e-05, | |
| "grad_norm": 0.7059808969497681, | |
| "learning_rate": 0.00014, | |
| "loss": 6.6932, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.425864243528541e-05, | |
| "grad_norm": 0.7493267059326172, | |
| "learning_rate": 0.00016, | |
| "loss": 6.2638, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.479097273969609e-05, | |
| "grad_norm": 0.6883407235145569, | |
| "learning_rate": 0.00018, | |
| "loss": 6.3127, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00010532330304410677, | |
| "grad_norm": 0.6710588932037354, | |
| "learning_rate": 0.0002, | |
| "loss": 6.042, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00011585563334851744, | |
| "grad_norm": 0.7773897051811218, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 6.5847, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00012638796365292812, | |
| "grad_norm": 0.7473054528236389, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 6.18, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0001369202939573388, | |
| "grad_norm": 1.030212640762329, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 6.4094, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0001369202939573388, | |
| "eval_loss": 1.5738751888275146, | |
| "eval_runtime": 2636.0843, | |
| "eval_samples_per_second": 15.166, | |
| "eval_steps_per_second": 7.583, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00014745262426174948, | |
| "grad_norm": 1.023868441581726, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 6.3971, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00015798495456616015, | |
| "grad_norm": 0.9858213663101196, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 6.4279, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00016851728487057082, | |
| "grad_norm": 1.029860496520996, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 6.1746, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0001790496151749815, | |
| "grad_norm": 1.2056444883346558, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 6.1612, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.00018958194547939218, | |
| "grad_norm": 1.5094845294952393, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 5.9291, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00020011427578380285, | |
| "grad_norm": 1.4246412515640259, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 6.4666, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.00021064660608821354, | |
| "grad_norm": 1.2746961116790771, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 5.8342, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0002211789363926242, | |
| "grad_norm": 1.5524417161941528, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 6.2008, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00023171126669703488, | |
| "grad_norm": 1.2203606367111206, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 5.9397, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.00024224359700144557, | |
| "grad_norm": 1.179914116859436, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 6.1692, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.00025277592730585624, | |
| "grad_norm": 1.105993628501892, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 6.1435, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0002633082576102669, | |
| "grad_norm": 1.280403733253479, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 6.2619, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0002738405879146776, | |
| "grad_norm": 1.1139580011367798, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 6.0445, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0002738405879146776, | |
| "eval_loss": 1.5127251148223877, | |
| "eval_runtime": 2636.0088, | |
| "eval_samples_per_second": 15.166, | |
| "eval_steps_per_second": 7.583, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0002843729182190883, | |
| "grad_norm": 1.2400450706481934, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 5.663, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00029490524852349896, | |
| "grad_norm": 1.1047593355178833, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 5.7661, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.00030543757882790963, | |
| "grad_norm": 1.3596596717834473, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 6.2018, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0003159699091323203, | |
| "grad_norm": 1.3211848735809326, | |
| "learning_rate": 0.0001, | |
| "loss": 5.9504, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00032650223943673097, | |
| "grad_norm": 1.265624761581421, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 5.4597, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.00033703456974114163, | |
| "grad_norm": 1.4042067527770996, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 6.7814, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.00034756690004555236, | |
| "grad_norm": 1.2178682088851929, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 5.9631, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.000358099230349963, | |
| "grad_norm": 1.5301004648208618, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 5.5926, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0003686315606543737, | |
| "grad_norm": 1.2925530672073364, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 5.6347, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.00037916389095878436, | |
| "grad_norm": 1.50602126121521, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 6.2784, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.000389696221263195, | |
| "grad_norm": 1.2359845638275146, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 6.0117, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0004002285515676057, | |
| "grad_norm": 1.6936525106430054, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 6.1991, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0004107608818720164, | |
| "grad_norm": 1.3919857740402222, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 5.3184, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0004107608818720164, | |
| "eval_loss": 1.4870572090148926, | |
| "eval_runtime": 2636.1749, | |
| "eval_samples_per_second": 15.165, | |
| "eval_steps_per_second": 7.583, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0004212932121764271, | |
| "grad_norm": 1.2668111324310303, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 6.2134, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00043182554248083775, | |
| "grad_norm": 1.4063031673431396, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 6.3267, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0004423578727852484, | |
| "grad_norm": 1.5289639234542847, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 5.8625, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0004528902030896591, | |
| "grad_norm": 1.3586407899856567, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 6.1417, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.00046342253339406975, | |
| "grad_norm": 1.5407192707061768, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 6.695, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0004739548636984805, | |
| "grad_norm": 1.2807283401489258, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 6.1324, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00048448719400289114, | |
| "grad_norm": 1.375052571296692, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 5.5676, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0004950195243073018, | |
| "grad_norm": 1.4286342859268188, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 6.1877, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0005055518546117125, | |
| "grad_norm": 1.4721269607543945, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 6.0579, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0005160841849161232, | |
| "grad_norm": 1.5001436471939087, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 6.0298, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0005266165152205338, | |
| "grad_norm": 1.804200291633606, | |
| "learning_rate": 0.0, | |
| "loss": 6.4854, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "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": 1.963059069321216e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |