| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3147128245476003, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003147128245476003, | |
| "eval_loss": 3.6290574073791504, | |
| "eval_runtime": 23.1819, | |
| "eval_samples_per_second": 23.078, | |
| "eval_steps_per_second": 2.89, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00944138473642801, | |
| "grad_norm": 3.3081445693969727, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.6087, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01888276947285602, | |
| "grad_norm": 3.533874034881592, | |
| "learning_rate": 3e-05, | |
| "loss": 3.8267, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02832415420928403, | |
| "grad_norm": 3.927079200744629, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.5564, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02832415420928403, | |
| "eval_loss": 3.374492883682251, | |
| "eval_runtime": 23.4572, | |
| "eval_samples_per_second": 22.808, | |
| "eval_steps_per_second": 2.856, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03776553894571204, | |
| "grad_norm": 3.0393035411834717, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.1098, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04720692368214005, | |
| "grad_norm": 3.247901439666748, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.8185, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05664830841856806, | |
| "grad_norm": 3.1178035736083984, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.3526, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05664830841856806, | |
| "eval_loss": 2.1767749786376953, | |
| "eval_runtime": 23.5223, | |
| "eval_samples_per_second": 22.744, | |
| "eval_steps_per_second": 2.848, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06608969315499606, | |
| "grad_norm": 3.0339815616607666, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 2.0888, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07553107789142408, | |
| "grad_norm": 2.707477569580078, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.9323, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08497246262785209, | |
| "grad_norm": 2.7072622776031494, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.5336, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.08497246262785209, | |
| "eval_loss": 1.420081377029419, | |
| "eval_runtime": 23.535, | |
| "eval_samples_per_second": 22.732, | |
| "eval_steps_per_second": 2.847, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0944138473642801, | |
| "grad_norm": 2.5003926753997803, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.3794, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1038552321007081, | |
| "grad_norm": 2.235706329345703, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.1922, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.11329661683713611, | |
| "grad_norm": 1.9673322439193726, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.9896, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.11329661683713611, | |
| "eval_loss": 1.117264986038208, | |
| "eval_runtime": 23.5414, | |
| "eval_samples_per_second": 22.726, | |
| "eval_steps_per_second": 2.846, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12273800157356413, | |
| "grad_norm": 2.4774956703186035, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.0046, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.13217938630999213, | |
| "grad_norm": 2.5677967071533203, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.1025, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.14162077104642015, | |
| "grad_norm": 1.922568678855896, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.9316, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.14162077104642015, | |
| "eval_loss": 0.97767573595047, | |
| "eval_runtime": 23.5651, | |
| "eval_samples_per_second": 22.703, | |
| "eval_steps_per_second": 2.843, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15106215578284815, | |
| "grad_norm": 1.8747705221176147, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.818, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.16050354051927615, | |
| "grad_norm": 2.5462067127227783, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.0874, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.16994492525570418, | |
| "grad_norm": 2.608792781829834, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.859, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.16994492525570418, | |
| "eval_loss": 0.9035796523094177, | |
| "eval_runtime": 23.5717, | |
| "eval_samples_per_second": 22.697, | |
| "eval_steps_per_second": 2.842, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.17938630999213218, | |
| "grad_norm": 1.9697829484939575, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.7758, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.1888276947285602, | |
| "grad_norm": 1.9587650299072266, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.8138, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1982690794649882, | |
| "grad_norm": 1.9131907224655151, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.8053, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.1982690794649882, | |
| "eval_loss": 0.8657433986663818, | |
| "eval_runtime": 23.5867, | |
| "eval_samples_per_second": 22.682, | |
| "eval_steps_per_second": 2.841, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2077104642014162, | |
| "grad_norm": 2.0765535831451416, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.7012, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.21715184893784423, | |
| "grad_norm": 2.891166925430298, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.7901, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.22659323367427223, | |
| "grad_norm": 2.2231130599975586, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.7215, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.22659323367427223, | |
| "eval_loss": 0.8475407958030701, | |
| "eval_runtime": 23.4992, | |
| "eval_samples_per_second": 22.767, | |
| "eval_steps_per_second": 2.851, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.23603461841070023, | |
| "grad_norm": 2.564457654953003, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.894, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24547600314712825, | |
| "grad_norm": 2.3753879070281982, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.8234, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.2549173878835563, | |
| "grad_norm": 1.8506582975387573, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.7381, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.2549173878835563, | |
| "eval_loss": 0.8388699293136597, | |
| "eval_runtime": 23.5662, | |
| "eval_samples_per_second": 22.702, | |
| "eval_steps_per_second": 2.843, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.26435877261998425, | |
| "grad_norm": 2.3776350021362305, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.7022, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2738001573564123, | |
| "grad_norm": 3.1713807582855225, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.847, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.2832415420928403, | |
| "grad_norm": 2.261918783187866, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.7808, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2832415420928403, | |
| "eval_loss": 0.8342262506484985, | |
| "eval_runtime": 23.5671, | |
| "eval_samples_per_second": 22.701, | |
| "eval_steps_per_second": 2.843, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2926829268292683, | |
| "grad_norm": 2.4710519313812256, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.784, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3021243115656963, | |
| "grad_norm": 2.067655563354492, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.7981, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.31156569630212433, | |
| "grad_norm": 2.4378199577331543, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.8747, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.31156569630212433, | |
| "eval_loss": 0.8323006629943848, | |
| "eval_runtime": 23.5479, | |
| "eval_samples_per_second": 22.72, | |
| "eval_steps_per_second": 2.845, | |
| "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": 5.51427898343424e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |