| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.015958507879513265, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00015958507879513265, | |
| "eval_loss": 2.050989866256714, | |
| "eval_runtime": 454.429, | |
| "eval_samples_per_second": 23.225, | |
| "eval_steps_per_second": 2.905, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00047875523638539794, | |
| "grad_norm": 0.9697971343994141, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.9725, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0009575104727707959, | |
| "grad_norm": 0.901890754699707, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0065, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001436265709156194, | |
| "grad_norm": 0.9856873154640198, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.931, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001436265709156194, | |
| "eval_loss": 1.9257014989852905, | |
| "eval_runtime": 457.2135, | |
| "eval_samples_per_second": 23.083, | |
| "eval_steps_per_second": 2.887, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0019150209455415918, | |
| "grad_norm": 0.7096590399742126, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.7845, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0023937761819269898, | |
| "grad_norm": 0.726519763469696, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.8313, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.002872531418312388, | |
| "grad_norm": 0.6210381388664246, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.8015, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002872531418312388, | |
| "eval_loss": 1.7303786277770996, | |
| "eval_runtime": 456.9299, | |
| "eval_samples_per_second": 23.098, | |
| "eval_steps_per_second": 2.889, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0033512866546977858, | |
| "grad_norm": 0.6193668246269226, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.7996, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0038300418910831835, | |
| "grad_norm": 0.6268954873085022, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.6603, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.004308797127468581, | |
| "grad_norm": 0.5985816121101379, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.716, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.004308797127468581, | |
| "eval_loss": 1.707382321357727, | |
| "eval_runtime": 456.9479, | |
| "eval_samples_per_second": 23.097, | |
| "eval_steps_per_second": 2.889, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0047875523638539795, | |
| "grad_norm": 0.5452908873558044, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.7306, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005266307600239378, | |
| "grad_norm": 0.5148115754127502, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.658, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.005745062836624776, | |
| "grad_norm": 0.5095010995864868, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.7653, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.005745062836624776, | |
| "eval_loss": 1.6911619901657104, | |
| "eval_runtime": 457.3928, | |
| "eval_samples_per_second": 23.074, | |
| "eval_steps_per_second": 2.886, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006223818073010173, | |
| "grad_norm": 0.480747789144516, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.6363, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0067025733093955715, | |
| "grad_norm": 0.4569284915924072, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.6942, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.00718132854578097, | |
| "grad_norm": 0.5147221684455872, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.7732, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00718132854578097, | |
| "eval_loss": 1.6826040744781494, | |
| "eval_runtime": 457.1352, | |
| "eval_samples_per_second": 23.087, | |
| "eval_steps_per_second": 2.888, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.007660083782166367, | |
| "grad_norm": 0.46845507621765137, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.6482, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.008138839018551766, | |
| "grad_norm": 0.5122058391571045, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.6855, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.008617594254937163, | |
| "grad_norm": 0.46340513229370117, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.5771, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.008617594254937163, | |
| "eval_loss": 1.678298830986023, | |
| "eval_runtime": 457.7537, | |
| "eval_samples_per_second": 23.056, | |
| "eval_steps_per_second": 2.884, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.00909634949132256, | |
| "grad_norm": 0.4864901602268219, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.6974, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.009575104727707959, | |
| "grad_norm": 0.47456270456314087, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.7266, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.010053859964093357, | |
| "grad_norm": 0.44960471987724304, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.6703, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.010053859964093357, | |
| "eval_loss": 1.6748311519622803, | |
| "eval_runtime": 457.1158, | |
| "eval_samples_per_second": 23.088, | |
| "eval_steps_per_second": 2.888, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.010532615200478756, | |
| "grad_norm": 0.5045154690742493, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.6842, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.011011370436864154, | |
| "grad_norm": 0.49012699723243713, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.6247, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.011490125673249552, | |
| "grad_norm": 0.5065352320671082, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.7605, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.011490125673249552, | |
| "eval_loss": 1.6728583574295044, | |
| "eval_runtime": 457.0738, | |
| "eval_samples_per_second": 23.09, | |
| "eval_steps_per_second": 2.888, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.011968880909634948, | |
| "grad_norm": 0.5055304169654846, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.5792, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.012447636146020347, | |
| "grad_norm": 0.46882903575897217, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.714, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.012926391382405745, | |
| "grad_norm": 0.5146296620368958, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.5839, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.012926391382405745, | |
| "eval_loss": 1.6719353199005127, | |
| "eval_runtime": 456.9575, | |
| "eval_samples_per_second": 23.096, | |
| "eval_steps_per_second": 2.889, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.013405146618791143, | |
| "grad_norm": 0.4731264114379883, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.7055, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.013883901855176541, | |
| "grad_norm": 0.5054563879966736, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.5845, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.01436265709156194, | |
| "grad_norm": 0.46672162413597107, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.7207, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.01436265709156194, | |
| "eval_loss": 1.671474575996399, | |
| "eval_runtime": 456.9871, | |
| "eval_samples_per_second": 23.095, | |
| "eval_steps_per_second": 2.888, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.014841412327947338, | |
| "grad_norm": 0.5002412796020508, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.6392, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.015320167564332734, | |
| "grad_norm": 0.47195887565612793, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.6559, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.015798922800718134, | |
| "grad_norm": 0.513578474521637, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.7077, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.015798922800718134, | |
| "eval_loss": 1.6713879108428955, | |
| "eval_runtime": 457.4203, | |
| "eval_samples_per_second": 23.073, | |
| "eval_steps_per_second": 2.886, | |
| "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": 6.75001861668864e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |