| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3992932862190812, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014134275618374558, | |
| "eval_loss": 2.42022705078125, | |
| "eval_runtime": 3.0081, | |
| "eval_samples_per_second": 39.893, | |
| "eval_steps_per_second": 4.987, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04240282685512368, | |
| "grad_norm": 0.04235639423131943, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.5861, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08480565371024736, | |
| "grad_norm": 0.04095880687236786, | |
| "learning_rate": 3e-05, | |
| "loss": 2.562, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.127208480565371, | |
| "grad_norm": 0.04078473523259163, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.5813, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.127208480565371, | |
| "eval_loss": 2.419612169265747, | |
| "eval_runtime": 2.9968, | |
| "eval_samples_per_second": 40.042, | |
| "eval_steps_per_second": 5.005, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1696113074204947, | |
| "grad_norm": 0.03982920944690704, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.5416, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.21201413427561838, | |
| "grad_norm": 0.05580739304423332, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.6521, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.254416961130742, | |
| "grad_norm": 0.05870726332068443, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.7235, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.254416961130742, | |
| "eval_loss": 2.414585828781128, | |
| "eval_runtime": 3.0217, | |
| "eval_samples_per_second": 39.713, | |
| "eval_steps_per_second": 4.964, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2968197879858657, | |
| "grad_norm": 0.05065983161330223, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 2.6602, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3392226148409894, | |
| "grad_norm": 0.046720851212739944, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.5159, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.38162544169611307, | |
| "grad_norm": 0.04852389171719551, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.4776, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.38162544169611307, | |
| "eval_loss": 2.4045073986053467, | |
| "eval_runtime": 3.0023, | |
| "eval_samples_per_second": 39.969, | |
| "eval_steps_per_second": 4.996, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.42402826855123676, | |
| "grad_norm": 0.06589920073747635, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.6105, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4664310954063604, | |
| "grad_norm": 0.0595686174929142, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.5886, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.508833922261484, | |
| "grad_norm": 0.07061908394098282, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.5692, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.508833922261484, | |
| "eval_loss": 2.3906164169311523, | |
| "eval_runtime": 3.0144, | |
| "eval_samples_per_second": 39.809, | |
| "eval_steps_per_second": 4.976, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5512367491166078, | |
| "grad_norm": 0.06360410153865814, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.5429, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5936395759717314, | |
| "grad_norm": 0.07501388341188431, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.5588, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6360424028268551, | |
| "grad_norm": 0.06902483105659485, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.5453, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6360424028268551, | |
| "eval_loss": 2.3756144046783447, | |
| "eval_runtime": 2.9987, | |
| "eval_samples_per_second": 40.017, | |
| "eval_steps_per_second": 5.002, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6784452296819788, | |
| "grad_norm": 0.06218775361776352, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.4641, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7208480565371025, | |
| "grad_norm": 0.06082496419548988, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.55, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7632508833922261, | |
| "grad_norm": 0.06064057722687721, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.5901, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7632508833922261, | |
| "eval_loss": 2.362664222717285, | |
| "eval_runtime": 2.9987, | |
| "eval_samples_per_second": 40.017, | |
| "eval_steps_per_second": 5.002, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8056537102473498, | |
| "grad_norm": 0.05651354417204857, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.4944, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8480565371024735, | |
| "grad_norm": 0.06623954325914383, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.4146, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8904593639575972, | |
| "grad_norm": 0.059099238365888596, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.5197, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8904593639575972, | |
| "eval_loss": 2.3530490398406982, | |
| "eval_runtime": 3.0239, | |
| "eval_samples_per_second": 39.684, | |
| "eval_steps_per_second": 4.96, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9328621908127208, | |
| "grad_norm": 0.06605750322341919, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.4754, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.9752650176678446, | |
| "grad_norm": 0.05665251612663269, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.5084, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.017667844522968, | |
| "grad_norm": 0.051736362278461456, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.591, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.017667844522968, | |
| "eval_loss": 2.3466386795043945, | |
| "eval_runtime": 2.9973, | |
| "eval_samples_per_second": 40.037, | |
| "eval_steps_per_second": 5.005, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.0600706713780919, | |
| "grad_norm": 0.05056842789053917, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.46, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.1024734982332156, | |
| "grad_norm": 0.0757237896323204, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.5239, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.1448763250883391, | |
| "grad_norm": 0.051590222865343094, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.4761, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.1448763250883391, | |
| "eval_loss": 2.3426692485809326, | |
| "eval_runtime": 3.0085, | |
| "eval_samples_per_second": 39.887, | |
| "eval_steps_per_second": 4.986, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.187279151943463, | |
| "grad_norm": 0.054393671452999115, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.3731, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.2296819787985867, | |
| "grad_norm": 0.061679407954216, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.449, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.2720848056537102, | |
| "grad_norm": 0.06303272396326065, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.4823, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2720848056537102, | |
| "eval_loss": 2.341158151626587, | |
| "eval_runtime": 3.0042, | |
| "eval_samples_per_second": 39.944, | |
| "eval_steps_per_second": 4.993, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.314487632508834, | |
| "grad_norm": 0.06280746310949326, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.4519, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.3568904593639575, | |
| "grad_norm": 0.07523226737976074, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.6474, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.3992932862190812, | |
| "grad_norm": 0.0763295441865921, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.5686, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.3992932862190812, | |
| "eval_loss": 2.3409645557403564, | |
| "eval_runtime": 2.9989, | |
| "eval_samples_per_second": 40.015, | |
| "eval_steps_per_second": 5.002, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.258627528654848e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |