| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1804788213627992, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011786372007366482, | |
| "eval_loss": 3.2576773166656494, | |
| "eval_runtime": 48.3038, | |
| "eval_samples_per_second": 5.921, | |
| "eval_steps_per_second": 2.96, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03535911602209945, | |
| "grad_norm": 37.830448150634766, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 99.4918, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0707182320441989, | |
| "grad_norm": 30.299089431762695, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 95.1039, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10607734806629834, | |
| "grad_norm": 31.264686584472656, | |
| "learning_rate": 6.75e-05, | |
| "loss": 99.6355, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10607734806629834, | |
| "eval_loss": 2.994905948638916, | |
| "eval_runtime": 48.3217, | |
| "eval_samples_per_second": 5.919, | |
| "eval_steps_per_second": 2.959, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1414364640883978, | |
| "grad_norm": 34.47152328491211, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 91.6026, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17679558011049723, | |
| "grad_norm": 26.11569595336914, | |
| "learning_rate": 0.0001125, | |
| "loss": 85.1959, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2121546961325967, | |
| "grad_norm": 38.721710205078125, | |
| "learning_rate": 0.000135, | |
| "loss": 78.4229, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2121546961325967, | |
| "eval_loss": 2.432400703430176, | |
| "eval_runtime": 48.3101, | |
| "eval_samples_per_second": 5.92, | |
| "eval_steps_per_second": 2.96, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.24751381215469614, | |
| "grad_norm": 37.68834686279297, | |
| "learning_rate": 0.00014994217771805422, | |
| "loss": 78.5547, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2828729281767956, | |
| "grad_norm": 36.297088623046875, | |
| "learning_rate": 0.00014907662554463532, | |
| "loss": 71.8931, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.318232044198895, | |
| "grad_norm": 40.15138244628906, | |
| "learning_rate": 0.0001471841427340235, | |
| "loss": 68.7287, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.318232044198895, | |
| "eval_loss": 2.1229209899902344, | |
| "eval_runtime": 48.3187, | |
| "eval_samples_per_second": 5.919, | |
| "eval_steps_per_second": 2.96, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.35359116022099446, | |
| "grad_norm": 22.00638198852539, | |
| "learning_rate": 0.0001442909649383465, | |
| "loss": 65.7473, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3889502762430939, | |
| "grad_norm": 26.196809768676758, | |
| "learning_rate": 0.0001404372005304598, | |
| "loss": 64.8032, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4243093922651934, | |
| "grad_norm": 30.012163162231445, | |
| "learning_rate": 0.00013567627457812106, | |
| "loss": 62.5845, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4243093922651934, | |
| "eval_loss": 1.9335837364196777, | |
| "eval_runtime": 48.3129, | |
| "eval_samples_per_second": 5.92, | |
| "eval_steps_per_second": 2.96, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.45966850828729283, | |
| "grad_norm": 24.18491554260254, | |
| "learning_rate": 0.0001300741882076764, | |
| "loss": 56.7812, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4950276243093923, | |
| "grad_norm": 29.25063133239746, | |
| "learning_rate": 0.00012370860362476374, | |
| "loss": 59.2692, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5303867403314917, | |
| "grad_norm": 26.837539672851562, | |
| "learning_rate": 0.00011666776747647015, | |
| "loss": 57.7333, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5303867403314917, | |
| "eval_loss": 1.8067072629928589, | |
| "eval_runtime": 48.2991, | |
| "eval_samples_per_second": 5.921, | |
| "eval_steps_per_second": 2.961, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5657458563535912, | |
| "grad_norm": 27.746295928955078, | |
| "learning_rate": 0.00010904928748046599, | |
| "loss": 57.6997, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6011049723756906, | |
| "grad_norm": 26.590911865234375, | |
| "learning_rate": 0.00010095877928081196, | |
| "loss": 57.611, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.63646408839779, | |
| "grad_norm": 24.721492767333984, | |
| "learning_rate": 9.25084022891929e-05, | |
| "loss": 54.3467, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.63646408839779, | |
| "eval_loss": 1.7132052183151245, | |
| "eval_runtime": 48.2964, | |
| "eval_samples_per_second": 5.922, | |
| "eval_steps_per_second": 2.961, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6718232044198895, | |
| "grad_norm": 28.768779754638672, | |
| "learning_rate": 8.381530480933783e-05, | |
| "loss": 55.5514, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.7071823204419889, | |
| "grad_norm": 32.7011833190918, | |
| "learning_rate": 7.5e-05, | |
| "loss": 54.9361, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7425414364640884, | |
| "grad_norm": 36.07527542114258, | |
| "learning_rate": 6.618469519066217e-05, | |
| "loss": 53.3471, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7425414364640884, | |
| "eval_loss": 1.646447777748108, | |
| "eval_runtime": 48.3148, | |
| "eval_samples_per_second": 5.92, | |
| "eval_steps_per_second": 2.96, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7779005524861878, | |
| "grad_norm": 31.98736000061035, | |
| "learning_rate": 5.7491597710807114e-05, | |
| "loss": 50.4475, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.8132596685082873, | |
| "grad_norm": 32.868492126464844, | |
| "learning_rate": 4.904122071918801e-05, | |
| "loss": 55.192, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8486187845303867, | |
| "grad_norm": 24.729814529418945, | |
| "learning_rate": 4.095071251953399e-05, | |
| "loss": 54.725, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8486187845303867, | |
| "eval_loss": 1.5971978902816772, | |
| "eval_runtime": 48.3066, | |
| "eval_samples_per_second": 5.921, | |
| "eval_steps_per_second": 2.96, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8839779005524862, | |
| "grad_norm": 27.09636116027832, | |
| "learning_rate": 3.333223252352985e-05, | |
| "loss": 51.9685, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9193370165745857, | |
| "grad_norm": 34.2943000793457, | |
| "learning_rate": 2.6291396375236232e-05, | |
| "loss": 50.2621, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9546961325966851, | |
| "grad_norm": 29.85685157775879, | |
| "learning_rate": 1.99258117923236e-05, | |
| "loss": 49.1029, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9546961325966851, | |
| "eval_loss": 1.5639307498931885, | |
| "eval_runtime": 48.3157, | |
| "eval_samples_per_second": 5.919, | |
| "eval_steps_per_second": 2.96, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9900552486187846, | |
| "grad_norm": 28.34723663330078, | |
| "learning_rate": 1.4323725421878949e-05, | |
| "loss": 46.7876, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.027255985267035, | |
| "grad_norm": 29.410560607910156, | |
| "learning_rate": 9.56279946954021e-06, | |
| "loss": 46.1371, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.0626151012891345, | |
| "grad_norm": 32.85994338989258, | |
| "learning_rate": 5.709035061653494e-06, | |
| "loss": 43.1931, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0626151012891345, | |
| "eval_loss": 1.5428903102874756, | |
| "eval_runtime": 48.3008, | |
| "eval_samples_per_second": 5.921, | |
| "eval_steps_per_second": 2.961, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.097974217311234, | |
| "grad_norm": 27.65435791015625, | |
| "learning_rate": 2.815857265976462e-06, | |
| "loss": 48.2304, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.1333333333333333, | |
| "grad_norm": 22.98744773864746, | |
| "learning_rate": 9.233744553646754e-07, | |
| "loss": 42.7889, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.1686924493554327, | |
| "grad_norm": 29.46680450439453, | |
| "learning_rate": 5.7822281945782424e-08, | |
| "loss": 47.1437, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.1686924493554327, | |
| "eval_loss": 1.5387442111968994, | |
| "eval_runtime": 48.2916, | |
| "eval_samples_per_second": 5.922, | |
| "eval_steps_per_second": 2.961, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.259156315439104e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |