| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7194244604316546, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007194244604316547, | |
| "eval_loss": 1.3505613803863525, | |
| "eval_runtime": 18.9626, | |
| "eval_samples_per_second": 12.34, | |
| "eval_steps_per_second": 1.582, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02158273381294964, | |
| "grad_norm": 7.126910209655762, | |
| "learning_rate": 1.5e-05, | |
| "loss": 5.4633, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04316546762589928, | |
| "grad_norm": 5.5504536628723145, | |
| "learning_rate": 3e-05, | |
| "loss": 5.4233, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06474820143884892, | |
| "grad_norm": 4.061417579650879, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.131, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06474820143884892, | |
| "eval_loss": 1.1897938251495361, | |
| "eval_runtime": 19.3146, | |
| "eval_samples_per_second": 12.115, | |
| "eval_steps_per_second": 1.553, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08633093525179857, | |
| "grad_norm": 2.9950151443481445, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 4.8732, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1079136690647482, | |
| "grad_norm": 3.6927051544189453, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.6017, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12949640287769784, | |
| "grad_norm": 3.0494635105133057, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 4.6042, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12949640287769784, | |
| "eval_loss": 1.0919393301010132, | |
| "eval_runtime": 19.3623, | |
| "eval_samples_per_second": 12.085, | |
| "eval_steps_per_second": 1.549, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1510791366906475, | |
| "grad_norm": 2.4984405040740967, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 4.4278, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.17266187050359713, | |
| "grad_norm": 2.3186981678009033, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 4.4248, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.19424460431654678, | |
| "grad_norm": 2.286144733428955, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 4.4557, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.19424460431654678, | |
| "eval_loss": 1.0573580265045166, | |
| "eval_runtime": 19.3833, | |
| "eval_samples_per_second": 12.072, | |
| "eval_steps_per_second": 1.548, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2158273381294964, | |
| "grad_norm": 2.0901005268096924, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 4.397, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23741007194244604, | |
| "grad_norm": 3.103538751602173, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 4.2221, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2589928057553957, | |
| "grad_norm": 1.9494324922561646, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 4.3688, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2589928057553957, | |
| "eval_loss": 1.0416053533554077, | |
| "eval_runtime": 19.3673, | |
| "eval_samples_per_second": 12.082, | |
| "eval_steps_per_second": 1.549, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2805755395683453, | |
| "grad_norm": 2.292299270629883, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 4.446, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.302158273381295, | |
| "grad_norm": 2.913973808288574, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 4.0109, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.3237410071942446, | |
| "grad_norm": 2.1814777851104736, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 4.2421, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3237410071942446, | |
| "eval_loss": 1.0296789407730103, | |
| "eval_runtime": 19.3399, | |
| "eval_samples_per_second": 12.099, | |
| "eval_steps_per_second": 1.551, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.34532374100719426, | |
| "grad_norm": 2.3433644771575928, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 4.616, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3669064748201439, | |
| "grad_norm": 1.8417110443115234, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 3.9965, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.38848920863309355, | |
| "grad_norm": 2.258572578430176, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 4.2638, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.38848920863309355, | |
| "eval_loss": 1.0222728252410889, | |
| "eval_runtime": 19.3615, | |
| "eval_samples_per_second": 12.086, | |
| "eval_steps_per_second": 1.549, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.41007194244604317, | |
| "grad_norm": 2.020596504211426, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 4.1487, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.4316546762589928, | |
| "grad_norm": 1.7905324697494507, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 4.1138, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.45323741007194246, | |
| "grad_norm": 2.2661585807800293, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.1022, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.45323741007194246, | |
| "eval_loss": 1.0169346332550049, | |
| "eval_runtime": 19.3659, | |
| "eval_samples_per_second": 12.083, | |
| "eval_steps_per_second": 1.549, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.4748201438848921, | |
| "grad_norm": 2.1625380516052246, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.3836, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.49640287769784175, | |
| "grad_norm": 2.1454732418060303, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 4.2899, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.5179856115107914, | |
| "grad_norm": 2.406993865966797, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 3.9958, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.5179856115107914, | |
| "eval_loss": 1.0135488510131836, | |
| "eval_runtime": 19.3546, | |
| "eval_samples_per_second": 12.09, | |
| "eval_steps_per_second": 1.55, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.539568345323741, | |
| "grad_norm": 2.6071054935455322, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 4.2177, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5611510791366906, | |
| "grad_norm": 2.0609230995178223, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 4.2697, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.5827338129496403, | |
| "grad_norm": 2.143162250518799, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.122, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.5827338129496403, | |
| "eval_loss": 1.011610507965088, | |
| "eval_runtime": 19.3757, | |
| "eval_samples_per_second": 12.077, | |
| "eval_steps_per_second": 1.548, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.60431654676259, | |
| "grad_norm": 1.9092354774475098, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 3.9762, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.6258992805755396, | |
| "grad_norm": 2.086115837097168, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 4.1775, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.6474820143884892, | |
| "grad_norm": 2.177617311477661, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 4.2366, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6474820143884892, | |
| "eval_loss": 1.0106626749038696, | |
| "eval_runtime": 19.3765, | |
| "eval_samples_per_second": 12.076, | |
| "eval_steps_per_second": 1.548, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6690647482014388, | |
| "grad_norm": 2.176861047744751, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 4.1401, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.6906474820143885, | |
| "grad_norm": 1.9706419706344604, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 3.9663, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.7122302158273381, | |
| "grad_norm": 2.0750808715820312, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 4.1725, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.7122302158273381, | |
| "eval_loss": 1.0105128288269043, | |
| "eval_runtime": 19.3722, | |
| "eval_samples_per_second": 12.079, | |
| "eval_steps_per_second": 1.549, | |
| "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": 1.708604681795666e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |