| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1379786133149362, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013797861331493618, | |
| "eval_loss": 0.7372766733169556, | |
| "eval_runtime": 168.7844, | |
| "eval_samples_per_second": 7.234, | |
| "eval_steps_per_second": 0.906, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004139358399448086, | |
| "grad_norm": 33.24546432495117, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.0101, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.008278716798896172, | |
| "grad_norm": 17.34926986694336, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2218, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.012418075198344257, | |
| "grad_norm": 5.28987979888916, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4942, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.012418075198344257, | |
| "eval_loss": 0.26399311423301697, | |
| "eval_runtime": 170.9677, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 0.895, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.016557433597792343, | |
| "grad_norm": 4.505770683288574, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.0678, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.020696791997240428, | |
| "grad_norm": 3.744687795639038, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7375, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.024836150396688513, | |
| "grad_norm": 3.299644708633423, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.9198, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.024836150396688513, | |
| "eval_loss": 0.21940629184246063, | |
| "eval_runtime": 170.9547, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 0.895, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028975508796136598, | |
| "grad_norm": 2.6757843494415283, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.8649, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03311486719558469, | |
| "grad_norm": 3.3077943325042725, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.8514, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03725422559503277, | |
| "grad_norm": 3.4696767330169678, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.7777, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03725422559503277, | |
| "eval_loss": 0.20136968791484833, | |
| "eval_runtime": 171.0507, | |
| "eval_samples_per_second": 7.138, | |
| "eval_steps_per_second": 0.894, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.041393583994480856, | |
| "grad_norm": 2.9004974365234375, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.664, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04553294239392894, | |
| "grad_norm": 2.8733339309692383, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.5927, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.049672300793377026, | |
| "grad_norm": 3.473083019256592, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.6759, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.049672300793377026, | |
| "eval_loss": 0.19248369336128235, | |
| "eval_runtime": 170.9492, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 0.895, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.053811659192825115, | |
| "grad_norm": 2.966871738433838, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.8124, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.057951017592273196, | |
| "grad_norm": 2.5523316860198975, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.7702, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.062090375991721285, | |
| "grad_norm": 2.8729777336120605, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.7699, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.062090375991721285, | |
| "eval_loss": 0.18150073289871216, | |
| "eval_runtime": 171.1339, | |
| "eval_samples_per_second": 7.135, | |
| "eval_steps_per_second": 0.894, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06622973439116937, | |
| "grad_norm": 4.041275501251221, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.9123, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07036909279061745, | |
| "grad_norm": 2.8718698024749756, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.6675, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07450845119006554, | |
| "grad_norm": 2.9034252166748047, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.5178, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07450845119006554, | |
| "eval_loss": 0.17112058401107788, | |
| "eval_runtime": 171.0347, | |
| "eval_samples_per_second": 7.139, | |
| "eval_steps_per_second": 0.895, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07864780958951363, | |
| "grad_norm": 3.4419732093811035, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.6105, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08278716798896171, | |
| "grad_norm": 3.259693145751953, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.6081, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0869265263884098, | |
| "grad_norm": 3.1575546264648438, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.6062, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0869265263884098, | |
| "eval_loss": 0.16651968657970428, | |
| "eval_runtime": 171.0519, | |
| "eval_samples_per_second": 7.138, | |
| "eval_steps_per_second": 0.894, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09106588478785788, | |
| "grad_norm": 2.911757230758667, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.528, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09520524318730597, | |
| "grad_norm": 3.0450966358184814, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.651, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09934460158675405, | |
| "grad_norm": 3.033231019973755, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.5592, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09934460158675405, | |
| "eval_loss": 0.16376812756061554, | |
| "eval_runtime": 171.1546, | |
| "eval_samples_per_second": 7.134, | |
| "eval_steps_per_second": 0.894, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10348395998620213, | |
| "grad_norm": 3.280524969100952, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.684, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.10762331838565023, | |
| "grad_norm": 3.2250235080718994, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.572, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.11176267678509831, | |
| "grad_norm": 3.109015941619873, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.6377, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.11176267678509831, | |
| "eval_loss": 0.16160938143730164, | |
| "eval_runtime": 171.1081, | |
| "eval_samples_per_second": 7.136, | |
| "eval_steps_per_second": 0.894, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.11590203518454639, | |
| "grad_norm": 2.425291061401367, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.4991, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.12004139358399447, | |
| "grad_norm": 2.9057319164276123, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.4731, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.12418075198344257, | |
| "grad_norm": 3.536256790161133, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.6766, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12418075198344257, | |
| "eval_loss": 0.16034354269504547, | |
| "eval_runtime": 171.2352, | |
| "eval_samples_per_second": 7.131, | |
| "eval_steps_per_second": 0.894, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.12832011038289065, | |
| "grad_norm": 3.2692079544067383, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.85, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.13245946878233875, | |
| "grad_norm": 3.3666374683380127, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.6383, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.1365988271817868, | |
| "grad_norm": 3.0570571422576904, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.689, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.1365988271817868, | |
| "eval_loss": 0.1600174754858017, | |
| "eval_runtime": 171.0652, | |
| "eval_samples_per_second": 7.138, | |
| "eval_steps_per_second": 0.894, | |
| "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": 2.812517994725376e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |