| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2881844380403458, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005763688760806916, | |
| "grad_norm": 2.4409587383270264, | |
| "learning_rate": 1e-05, | |
| "loss": 5.5006, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005763688760806916, | |
| "eval_loss": 5.976919174194336, | |
| "eval_runtime": 0.8589, | |
| "eval_samples_per_second": 678.789, | |
| "eval_steps_per_second": 22.122, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011527377521613832, | |
| "grad_norm": 2.5246589183807373, | |
| "learning_rate": 2e-05, | |
| "loss": 5.7539, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01729106628242075, | |
| "grad_norm": 5.136504173278809, | |
| "learning_rate": 3e-05, | |
| "loss": 5.7782, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.023054755043227664, | |
| "grad_norm": 1.6736292839050293, | |
| "learning_rate": 4e-05, | |
| "loss": 5.7622, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02881844380403458, | |
| "grad_norm": 2.1508893966674805, | |
| "learning_rate": 5e-05, | |
| "loss": 5.9264, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0345821325648415, | |
| "grad_norm": 2.719834089279175, | |
| "learning_rate": 6e-05, | |
| "loss": 5.6231, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.040345821325648415, | |
| "grad_norm": 1.862425684928894, | |
| "learning_rate": 7e-05, | |
| "loss": 5.7143, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04610951008645533, | |
| "grad_norm": 2.0583953857421875, | |
| "learning_rate": 8e-05, | |
| "loss": 5.7937, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05187319884726225, | |
| "grad_norm": 3.5244977474212646, | |
| "learning_rate": 9e-05, | |
| "loss": 5.5759, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05187319884726225, | |
| "eval_loss": 5.762571334838867, | |
| "eval_runtime": 0.8404, | |
| "eval_samples_per_second": 693.698, | |
| "eval_steps_per_second": 22.608, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05763688760806916, | |
| "grad_norm": 2.2905080318450928, | |
| "learning_rate": 0.0001, | |
| "loss": 5.6307, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06340057636887608, | |
| "grad_norm": 2.9346985816955566, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 5.5274, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.069164265129683, | |
| "grad_norm": 3.4477925300598145, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 5.4946, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07492795389048991, | |
| "grad_norm": 2.5624496936798096, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 5.5983, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08069164265129683, | |
| "grad_norm": 2.30907940864563, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 4.965, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.08645533141210375, | |
| "grad_norm": 3.060831308364868, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 5.0549, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09221902017291066, | |
| "grad_norm": 1.7213447093963623, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 5.1858, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.09798270893371758, | |
| "grad_norm": 1.8246761560440063, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 5.1216, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1037463976945245, | |
| "grad_norm": 1.6783112287521362, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 5.0034, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1037463976945245, | |
| "eval_loss": 4.902072429656982, | |
| "eval_runtime": 0.8459, | |
| "eval_samples_per_second": 689.167, | |
| "eval_steps_per_second": 22.46, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10951008645533142, | |
| "grad_norm": 1.777565836906433, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 4.9386, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.11527377521613832, | |
| "grad_norm": 2.9292523860931396, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 4.5237, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12103746397694524, | |
| "grad_norm": 1.343632459640503, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 4.4949, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.12680115273775217, | |
| "grad_norm": 1.9242039918899536, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 4.7291, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.13256484149855907, | |
| "grad_norm": 1.7752628326416016, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 4.6048, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.138328530259366, | |
| "grad_norm": 1.608022689819336, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 4.3314, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1440922190201729, | |
| "grad_norm": 1.1817405223846436, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 4.163, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14985590778097982, | |
| "grad_norm": 1.5801705121994019, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 4.3659, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.15561959654178675, | |
| "grad_norm": 3.451111316680908, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 4.3982, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15561959654178675, | |
| "eval_loss": 4.300004005432129, | |
| "eval_runtime": 0.8515, | |
| "eval_samples_per_second": 684.681, | |
| "eval_steps_per_second": 22.314, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.16138328530259366, | |
| "grad_norm": 1.6136767864227295, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4.3207, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.16714697406340057, | |
| "grad_norm": 1.5284059047698975, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 4.2439, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.1729106628242075, | |
| "grad_norm": 1.335671305656433, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.9352, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1786743515850144, | |
| "grad_norm": 2.7042202949523926, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 4.0291, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1844380403458213, | |
| "grad_norm": 2.1285061836242676, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 3.9501, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.19020172910662825, | |
| "grad_norm": 1.8937512636184692, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.7774, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.19596541786743515, | |
| "grad_norm": 1.931766152381897, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 3.7974, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.2017291066282421, | |
| "grad_norm": 1.651713490486145, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 3.8174, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.207492795389049, | |
| "grad_norm": 1.7000616788864136, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.8321, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.207492795389049, | |
| "eval_loss": 3.770657777786255, | |
| "eval_runtime": 0.8553, | |
| "eval_samples_per_second": 681.651, | |
| "eval_steps_per_second": 22.215, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2132564841498559, | |
| "grad_norm": 1.4033437967300415, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.7747, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.21902017291066284, | |
| "grad_norm": 2.2183494567871094, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 3.7118, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.22478386167146974, | |
| "grad_norm": 1.5647684335708618, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.6716, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.23054755043227665, | |
| "grad_norm": 1.6753904819488525, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 3.4879, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.23631123919308358, | |
| "grad_norm": 1.5152138471603394, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 3.4866, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.2420749279538905, | |
| "grad_norm": 1.228210210800171, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 3.4079, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2478386167146974, | |
| "grad_norm": 1.3091694116592407, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 3.7129, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.25360230547550433, | |
| "grad_norm": 1.4786443710327148, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 3.3594, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.25936599423631124, | |
| "grad_norm": 1.6930687427520752, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 3.4402, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.25936599423631124, | |
| "eval_loss": 3.5291929244995117, | |
| "eval_runtime": 0.8391, | |
| "eval_samples_per_second": 694.79, | |
| "eval_steps_per_second": 22.643, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.26512968299711814, | |
| "grad_norm": 1.3388841152191162, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.2864, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.27089337175792505, | |
| "grad_norm": 1.3085576295852661, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 3.5233, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.276657060518732, | |
| "grad_norm": 1.5007169246673584, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.4159, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2824207492795389, | |
| "grad_norm": 1.2105474472045898, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 3.4101, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.2881844380403458, | |
| "grad_norm": 1.4784791469573975, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 3.3902, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2776751721676800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |