| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0689893066574681, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006898930665746809, | |
| "eval_loss": 6.302078723907471, | |
| "eval_runtime": 13.4852, | |
| "eval_samples_per_second": 181.087, | |
| "eval_steps_per_second": 22.692, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002069679199724043, | |
| "grad_norm": 2.097353219985962, | |
| "learning_rate": 1.5e-05, | |
| "loss": 6.2427, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004139358399448086, | |
| "grad_norm": 6.432583332061768, | |
| "learning_rate": 3e-05, | |
| "loss": 6.2662, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.006209037599172128, | |
| "grad_norm": 2.0673086643218994, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.2928, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006209037599172128, | |
| "eval_loss": 6.254395961761475, | |
| "eval_runtime": 13.4386, | |
| "eval_samples_per_second": 181.715, | |
| "eval_steps_per_second": 22.77, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008278716798896172, | |
| "grad_norm": 2.1118319034576416, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 6.1745, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010348395998620214, | |
| "grad_norm": 2.025097131729126, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 6.2417, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012418075198344257, | |
| "grad_norm": 3.1931934356689453, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 6.1336, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012418075198344257, | |
| "eval_loss": 5.968760013580322, | |
| "eval_runtime": 13.4604, | |
| "eval_samples_per_second": 181.421, | |
| "eval_steps_per_second": 22.733, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014487754398068299, | |
| "grad_norm": 2.5991594791412354, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 6.0403, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.016557433597792343, | |
| "grad_norm": 1.8577134609222412, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 5.8059, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.018627112797516384, | |
| "grad_norm": 2.6672468185424805, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 5.8615, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.018627112797516384, | |
| "eval_loss": 5.678951740264893, | |
| "eval_runtime": 13.4044, | |
| "eval_samples_per_second": 182.18, | |
| "eval_steps_per_second": 22.828, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.020696791997240428, | |
| "grad_norm": 1.9493556022644043, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 5.8126, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02276647119696447, | |
| "grad_norm": 1.8085205554962158, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 5.5456, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.024836150396688513, | |
| "grad_norm": 1.9123843908309937, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 5.5222, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.024836150396688513, | |
| "eval_loss": 5.4077911376953125, | |
| "eval_runtime": 13.4002, | |
| "eval_samples_per_second": 182.236, | |
| "eval_steps_per_second": 22.835, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.026905829596412557, | |
| "grad_norm": 2.0299079418182373, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 5.4701, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.028975508796136598, | |
| "grad_norm": 1.8605529069900513, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 5.2851, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.031045187995860642, | |
| "grad_norm": 1.6242057085037231, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 5.2112, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.031045187995860642, | |
| "eval_loss": 5.138970375061035, | |
| "eval_runtime": 13.433, | |
| "eval_samples_per_second": 181.791, | |
| "eval_steps_per_second": 22.78, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03311486719558469, | |
| "grad_norm": 1.7754476070404053, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 5.1463, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03518454639530873, | |
| "grad_norm": 1.9413535594940186, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 5.1448, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.03725422559503277, | |
| "grad_norm": 2.7202603816986084, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 4.8747, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03725422559503277, | |
| "eval_loss": 4.881633281707764, | |
| "eval_runtime": 13.4257, | |
| "eval_samples_per_second": 181.89, | |
| "eval_steps_per_second": 22.792, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.039323904794756816, | |
| "grad_norm": 1.879242181777954, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 4.8443, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.041393583994480856, | |
| "grad_norm": 2.7195351123809814, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 4.7365, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0434632631942049, | |
| "grad_norm": 1.8447564840316772, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.6527, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0434632631942049, | |
| "eval_loss": 4.671598434448242, | |
| "eval_runtime": 13.3945, | |
| "eval_samples_per_second": 182.313, | |
| "eval_steps_per_second": 22.845, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04553294239392894, | |
| "grad_norm": 2.426953077316284, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.6528, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.047602621593652986, | |
| "grad_norm": 1.9133142232894897, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 4.6137, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.049672300793377026, | |
| "grad_norm": 2.306638479232788, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 4.5453, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.049672300793377026, | |
| "eval_loss": 4.519593715667725, | |
| "eval_runtime": 13.3964, | |
| "eval_samples_per_second": 182.288, | |
| "eval_steps_per_second": 22.842, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05174197999310107, | |
| "grad_norm": 2.3796350955963135, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 4.5081, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.053811659192825115, | |
| "grad_norm": 2.1177785396575928, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 4.4774, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.055881338392549156, | |
| "grad_norm": 2.004594087600708, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.4503, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.055881338392549156, | |
| "eval_loss": 4.435978412628174, | |
| "eval_runtime": 13.4709, | |
| "eval_samples_per_second": 181.28, | |
| "eval_steps_per_second": 22.716, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.057951017592273196, | |
| "grad_norm": 1.812859058380127, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 4.4051, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.06002069679199724, | |
| "grad_norm": 2.5786092281341553, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 4.4897, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.062090375991721285, | |
| "grad_norm": 1.9753355979919434, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 4.3793, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.062090375991721285, | |
| "eval_loss": 4.402246952056885, | |
| "eval_runtime": 13.471, | |
| "eval_samples_per_second": 181.279, | |
| "eval_steps_per_second": 22.716, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06416005519144533, | |
| "grad_norm": 2.001936197280884, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 4.4432, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.06622973439116937, | |
| "grad_norm": 2.1391875743865967, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 4.3886, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.0682994135908934, | |
| "grad_norm": 2.535604238510132, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 4.2764, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.0682994135908934, | |
| "eval_loss": 4.3958845138549805, | |
| "eval_runtime": 13.4312, | |
| "eval_samples_per_second": 181.815, | |
| "eval_steps_per_second": 22.783, | |
| "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": 2776751721676800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |