| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.591715976331361, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005917159763313609, | |
| "eval_loss": 0.7227837443351746, | |
| "eval_runtime": 40.3117, | |
| "eval_samples_per_second": 7.07, | |
| "eval_steps_per_second": 0.893, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01775147928994083, | |
| "grad_norm": 0.3443498909473419, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.7445, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03550295857988166, | |
| "grad_norm": 0.4017130434513092, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8188, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05325443786982249, | |
| "grad_norm": 0.34028515219688416, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7719, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05325443786982249, | |
| "eval_loss": 0.6912192702293396, | |
| "eval_runtime": 41.1071, | |
| "eval_samples_per_second": 6.933, | |
| "eval_steps_per_second": 0.876, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "grad_norm": 0.22598209977149963, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.734, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08875739644970414, | |
| "grad_norm": 0.36029353737831116, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7571, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10650887573964497, | |
| "grad_norm": 0.250407874584198, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.7031, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10650887573964497, | |
| "eval_loss": 0.6377161145210266, | |
| "eval_runtime": 41.1034, | |
| "eval_samples_per_second": 6.934, | |
| "eval_steps_per_second": 0.876, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1242603550295858, | |
| "grad_norm": 0.28463196754455566, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.7, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.14201183431952663, | |
| "grad_norm": 0.2629949450492859, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.6446, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15976331360946747, | |
| "grad_norm": 0.23156414926052094, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.6471, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15976331360946747, | |
| "eval_loss": 0.6183348298072815, | |
| "eval_runtime": 41.1316, | |
| "eval_samples_per_second": 6.929, | |
| "eval_steps_per_second": 0.875, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17751479289940827, | |
| "grad_norm": 0.20715036988258362, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.6749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1952662721893491, | |
| "grad_norm": 0.25301262736320496, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6594, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.21301775147928995, | |
| "grad_norm": 0.25081703066825867, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.6767, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.21301775147928995, | |
| "eval_loss": 0.6061479449272156, | |
| "eval_runtime": 41.1244, | |
| "eval_samples_per_second": 6.93, | |
| "eval_steps_per_second": 0.875, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 0.2390720248222351, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.6084, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2485207100591716, | |
| "grad_norm": 0.25335678458213806, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.72, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.26627218934911245, | |
| "grad_norm": 0.2510438859462738, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.6385, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.26627218934911245, | |
| "eval_loss": 0.597359836101532, | |
| "eval_runtime": 41.1145, | |
| "eval_samples_per_second": 6.932, | |
| "eval_steps_per_second": 0.876, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.28402366863905326, | |
| "grad_norm": 0.22579920291900635, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.6332, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.30177514792899407, | |
| "grad_norm": 0.2414848506450653, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.6541, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.31952662721893493, | |
| "grad_norm": 0.2127198874950409, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.5908, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.31952662721893493, | |
| "eval_loss": 0.5910584330558777, | |
| "eval_runtime": 41.1561, | |
| "eval_samples_per_second": 6.925, | |
| "eval_steps_per_second": 0.875, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.33727810650887574, | |
| "grad_norm": 0.2393001914024353, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.5929, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "grad_norm": 0.3995908200740814, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.5554, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3727810650887574, | |
| "grad_norm": 0.23936037719249725, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.6399, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3727810650887574, | |
| "eval_loss": 0.5867862701416016, | |
| "eval_runtime": 41.1248, | |
| "eval_samples_per_second": 6.93, | |
| "eval_steps_per_second": 0.875, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3905325443786982, | |
| "grad_norm": 0.2615549564361572, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.6358, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.40828402366863903, | |
| "grad_norm": 0.29804727435112, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.6606, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.4260355029585799, | |
| "grad_norm": 0.26720571517944336, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.6308, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4260355029585799, | |
| "eval_loss": 0.5839556455612183, | |
| "eval_runtime": 41.1651, | |
| "eval_samples_per_second": 6.923, | |
| "eval_steps_per_second": 0.875, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4437869822485207, | |
| "grad_norm": 0.30903738737106323, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.6319, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 0.23036371171474457, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.6056, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.47928994082840237, | |
| "grad_norm": 0.24734920263290405, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.5583, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.47928994082840237, | |
| "eval_loss": 0.5826714038848877, | |
| "eval_runtime": 41.157, | |
| "eval_samples_per_second": 6.925, | |
| "eval_steps_per_second": 0.875, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4970414201183432, | |
| "grad_norm": 0.2538912892341614, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.6217, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.514792899408284, | |
| "grad_norm": 0.26953989267349243, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.6606, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5325443786982249, | |
| "grad_norm": 0.30746757984161377, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.5684, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5325443786982249, | |
| "eval_loss": 0.5818790793418884, | |
| "eval_runtime": 41.1339, | |
| "eval_samples_per_second": 6.929, | |
| "eval_steps_per_second": 0.875, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5502958579881657, | |
| "grad_norm": 0.24370835721492767, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.6293, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5680473372781065, | |
| "grad_norm": 0.25430381298065186, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.6013, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5857988165680473, | |
| "grad_norm": 0.2393229901790619, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.6499, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5857988165680473, | |
| "eval_loss": 0.5817607641220093, | |
| "eval_runtime": 41.1312, | |
| "eval_samples_per_second": 6.929, | |
| "eval_steps_per_second": 0.875, | |
| "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": 2.96383969782399e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |