| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.20171457387796268, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002017145738779627, | |
| "eval_loss": 1.3803871870040894, | |
| "eval_runtime": 46.8345, | |
| "eval_samples_per_second": 17.829, | |
| "eval_steps_per_second": 2.242, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006051437216338881, | |
| "grad_norm": 0.0706619843840599, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.3971, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.012102874432677761, | |
| "grad_norm": 0.060850467532873154, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3178, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018154311649016642, | |
| "grad_norm": 0.0778835117816925, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.387, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.018154311649016642, | |
| "eval_loss": 1.3788137435913086, | |
| "eval_runtime": 47.4343, | |
| "eval_samples_per_second": 17.603, | |
| "eval_steps_per_second": 2.214, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.024205748865355523, | |
| "grad_norm": 0.07908966392278671, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.4137, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.030257186081694403, | |
| "grad_norm": 0.07839653640985489, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.3398, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.036308623298033284, | |
| "grad_norm": 0.08289563655853271, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.4154, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.036308623298033284, | |
| "eval_loss": 1.3672126531600952, | |
| "eval_runtime": 47.5202, | |
| "eval_samples_per_second": 17.571, | |
| "eval_steps_per_second": 2.21, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04236006051437216, | |
| "grad_norm": 0.07574637979269028, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.3475, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.048411497730711045, | |
| "grad_norm": 0.06648290902376175, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.2828, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05446293494704992, | |
| "grad_norm": 0.07902548462152481, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.3281, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05446293494704992, | |
| "eval_loss": 1.3498567342758179, | |
| "eval_runtime": 47.5478, | |
| "eval_samples_per_second": 17.561, | |
| "eval_steps_per_second": 2.208, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.060514372163388806, | |
| "grad_norm": 0.07119950652122498, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.4393, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06656580937972768, | |
| "grad_norm": 0.06678444892168045, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.3122, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.07261724659606657, | |
| "grad_norm": 0.06940136104822159, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.3441, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07261724659606657, | |
| "eval_loss": 1.3346147537231445, | |
| "eval_runtime": 47.518, | |
| "eval_samples_per_second": 17.572, | |
| "eval_steps_per_second": 2.21, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07866868381240545, | |
| "grad_norm": 0.0725693628191948, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.241, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.08472012102874432, | |
| "grad_norm": 0.07808159291744232, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.323, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0907715582450832, | |
| "grad_norm": 0.07631468772888184, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.3608, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0907715582450832, | |
| "eval_loss": 1.3239437341690063, | |
| "eval_runtime": 47.5205, | |
| "eval_samples_per_second": 17.571, | |
| "eval_steps_per_second": 2.21, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09682299546142209, | |
| "grad_norm": 0.080069400370121, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.2464, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.10287443267776097, | |
| "grad_norm": 0.08555633574724197, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.4155, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.10892586989409984, | |
| "grad_norm": 0.082075335085392, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.3722, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.10892586989409984, | |
| "eval_loss": 1.3156965970993042, | |
| "eval_runtime": 47.5204, | |
| "eval_samples_per_second": 17.571, | |
| "eval_steps_per_second": 2.21, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.11497730711043873, | |
| "grad_norm": 0.07414203882217407, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.2905, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.12102874432677761, | |
| "grad_norm": 0.07650715857744217, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.3504, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12708018154311648, | |
| "grad_norm": 0.08604590594768524, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.3348, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.12708018154311648, | |
| "eval_loss": 1.3101333379745483, | |
| "eval_runtime": 47.5325, | |
| "eval_samples_per_second": 17.567, | |
| "eval_steps_per_second": 2.209, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.13313161875945537, | |
| "grad_norm": 0.0818275585770607, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.3878, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.13918305597579425, | |
| "grad_norm": 0.07330054044723511, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.3035, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.14523449319213314, | |
| "grad_norm": 0.08352211117744446, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.2734, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.14523449319213314, | |
| "eval_loss": 1.3066270351409912, | |
| "eval_runtime": 47.5418, | |
| "eval_samples_per_second": 17.564, | |
| "eval_steps_per_second": 2.209, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.15128593040847202, | |
| "grad_norm": 0.07840047776699066, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.3625, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1573373676248109, | |
| "grad_norm": 0.08086764067411423, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.3874, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.16338880484114976, | |
| "grad_norm": 0.06452049314975739, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.3141, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.16338880484114976, | |
| "eval_loss": 1.3048590421676636, | |
| "eval_runtime": 47.4887, | |
| "eval_samples_per_second": 17.583, | |
| "eval_steps_per_second": 2.211, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.16944024205748864, | |
| "grad_norm": 0.07050218433141708, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.3745, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.17549167927382753, | |
| "grad_norm": 0.08755092322826385, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.383, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.1815431164901664, | |
| "grad_norm": 0.07090733200311661, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.2869, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1815431164901664, | |
| "eval_loss": 1.304002285003662, | |
| "eval_runtime": 47.4951, | |
| "eval_samples_per_second": 17.581, | |
| "eval_steps_per_second": 2.211, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1875945537065053, | |
| "grad_norm": 0.07828027755022049, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.3319, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.19364599092284418, | |
| "grad_norm": 0.07713537663221359, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.3671, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.19969742813918306, | |
| "grad_norm": 0.07717742770910263, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.2967, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.19969742813918306, | |
| "eval_loss": 1.3039239645004272, | |
| "eval_runtime": 47.4866, | |
| "eval_samples_per_second": 17.584, | |
| "eval_steps_per_second": 2.211, | |
| "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": 6.40470623453184e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |