| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06521030322791001, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006521030322791001, | |
| "eval_loss": 1.4430840015411377, | |
| "eval_runtime": 35.4265, | |
| "eval_samples_per_second": 72.911, | |
| "eval_steps_per_second": 9.117, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019563090968373, | |
| "grad_norm": 0.13245901465415955, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.3922, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0039126181936746, | |
| "grad_norm": 0.13980551064014435, | |
| "learning_rate": 3e-05, | |
| "loss": 1.35, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0058689272905119005, | |
| "grad_norm": 0.21757693588733673, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4342, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0058689272905119005, | |
| "eval_loss": 1.4398877620697021, | |
| "eval_runtime": 35.4144, | |
| "eval_samples_per_second": 72.936, | |
| "eval_steps_per_second": 9.121, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0078252363873492, | |
| "grad_norm": 0.16242149472236633, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.3931, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0097815454841865, | |
| "grad_norm": 0.15830300748348236, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.367, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011737854581023801, | |
| "grad_norm": 0.16229018568992615, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.4666, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011737854581023801, | |
| "eval_loss": 1.4245867729187012, | |
| "eval_runtime": 35.2817, | |
| "eval_samples_per_second": 73.211, | |
| "eval_steps_per_second": 9.155, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.013694163677861103, | |
| "grad_norm": 0.14549140632152557, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.4435, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0156504727746984, | |
| "grad_norm": 0.12280628830194473, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.4289, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.017606781871535703, | |
| "grad_norm": 0.1361985057592392, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.4174, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.017606781871535703, | |
| "eval_loss": 1.4081462621688843, | |
| "eval_runtime": 35.2327, | |
| "eval_samples_per_second": 73.313, | |
| "eval_steps_per_second": 9.168, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.019563090968373, | |
| "grad_norm": 0.1305856853723526, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.3956, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.021519400065210303, | |
| "grad_norm": 0.13506704568862915, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.4642, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.023475709162047602, | |
| "grad_norm": 0.1571059674024582, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.4487, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.023475709162047602, | |
| "eval_loss": 1.3942054510116577, | |
| "eval_runtime": 35.4024, | |
| "eval_samples_per_second": 72.961, | |
| "eval_steps_per_second": 9.124, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.025432018258884904, | |
| "grad_norm": 0.12497388571500778, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.345, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.027388327355722206, | |
| "grad_norm": 0.1353573054075241, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.3613, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.029344636452559504, | |
| "grad_norm": 0.11737760901451111, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.3878, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.029344636452559504, | |
| "eval_loss": 1.383650302886963, | |
| "eval_runtime": 35.4483, | |
| "eval_samples_per_second": 72.867, | |
| "eval_steps_per_second": 9.112, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0313009455493968, | |
| "grad_norm": 0.11926010251045227, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.4205, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03325725464623411, | |
| "grad_norm": 0.1115192323923111, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.3493, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.035213563743071406, | |
| "grad_norm": 0.11480339616537094, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.3904, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.035213563743071406, | |
| "eval_loss": 1.376178503036499, | |
| "eval_runtime": 35.4933, | |
| "eval_samples_per_second": 72.774, | |
| "eval_steps_per_second": 9.1, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.037169872839908705, | |
| "grad_norm": 0.11822472512722015, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.3313, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.039126181936746, | |
| "grad_norm": 0.1461370587348938, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.3558, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04108249103358331, | |
| "grad_norm": 0.13045533001422882, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.2992, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04108249103358331, | |
| "eval_loss": 1.3712804317474365, | |
| "eval_runtime": 35.3891, | |
| "eval_samples_per_second": 72.989, | |
| "eval_steps_per_second": 9.127, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04303880013042061, | |
| "grad_norm": 0.13522818684577942, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.244, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.044995109227257905, | |
| "grad_norm": 0.1401805877685547, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.3158, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.046951418324095204, | |
| "grad_norm": 0.16507139801979065, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.3923, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.046951418324095204, | |
| "eval_loss": 1.368349313735962, | |
| "eval_runtime": 35.4333, | |
| "eval_samples_per_second": 72.898, | |
| "eval_steps_per_second": 9.116, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04890772742093251, | |
| "grad_norm": 0.1617535650730133, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.4659, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05086403651776981, | |
| "grad_norm": 0.15123188495635986, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.4194, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.052820345614607106, | |
| "grad_norm": 0.1255597174167633, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.3503, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.052820345614607106, | |
| "eval_loss": 1.3665692806243896, | |
| "eval_runtime": 35.4134, | |
| "eval_samples_per_second": 72.938, | |
| "eval_steps_per_second": 9.121, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05477665471144441, | |
| "grad_norm": 0.1431252807378769, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.3153, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.05673296380828171, | |
| "grad_norm": 0.1532163769006729, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.3877, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.05868927290511901, | |
| "grad_norm": 0.15569448471069336, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.4249, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05868927290511901, | |
| "eval_loss": 1.3660001754760742, | |
| "eval_runtime": 35.493, | |
| "eval_samples_per_second": 72.775, | |
| "eval_steps_per_second": 9.1, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06064558200195631, | |
| "grad_norm": 0.1278517246246338, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.4409, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.0626018910987936, | |
| "grad_norm": 0.10912717878818512, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.331, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.06455820019563091, | |
| "grad_norm": 0.143110990524292, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.3777, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.06455820019563091, | |
| "eval_loss": 1.3658305406570435, | |
| "eval_runtime": 35.5981, | |
| "eval_samples_per_second": 72.56, | |
| "eval_steps_per_second": 9.074, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3352559257387008.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |