| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.905320108205591, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.028854824165915238, | |
| "eval_loss": 2.2234888076782227, | |
| "eval_runtime": 63.3246, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08656447249774572, | |
| "grad_norm": 11.650296211242676, | |
| "learning_rate": 3e-05, | |
| "loss": 71.8161, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.17312894499549145, | |
| "grad_norm": 12.471053123474121, | |
| "learning_rate": 6e-05, | |
| "loss": 72.0251, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.25969341749323716, | |
| "grad_norm": 11.917525291442871, | |
| "learning_rate": 9e-05, | |
| "loss": 66.3847, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.25969341749323716, | |
| "eval_loss": 1.9188385009765625, | |
| "eval_runtime": 63.2986, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3462578899909829, | |
| "grad_norm": 11.10000991821289, | |
| "learning_rate": 0.00012, | |
| "loss": 59.632, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4328223624887286, | |
| "grad_norm": 12.129708290100098, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 55.7109, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5193868349864743, | |
| "grad_norm": 7.577877044677734, | |
| "learning_rate": 0.00018, | |
| "loss": 51.9351, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5193868349864743, | |
| "eval_loss": 1.5988303422927856, | |
| "eval_runtime": 63.413, | |
| "eval_samples_per_second": 1.845, | |
| "eval_steps_per_second": 0.93, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.60595130748422, | |
| "grad_norm": 6.87202262878418, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 51.0237, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6925157799819658, | |
| "grad_norm": 6.496087551116943, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 48.9338, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7790802524797115, | |
| "grad_norm": 6.911287784576416, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 47.7303, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7790802524797115, | |
| "eval_loss": 1.4810194969177246, | |
| "eval_runtime": 63.4002, | |
| "eval_samples_per_second": 1.845, | |
| "eval_steps_per_second": 0.931, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.8656447249774571, | |
| "grad_norm": 6.416966915130615, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 47.3634, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9522091974752029, | |
| "grad_norm": 6.382626056671143, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 46.6147, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.048692515779982, | |
| "grad_norm": 7.635010719299316, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 46.168, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.048692515779982, | |
| "eval_loss": 1.4284162521362305, | |
| "eval_runtime": 63.4232, | |
| "eval_samples_per_second": 1.845, | |
| "eval_steps_per_second": 0.93, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.1352569882777277, | |
| "grad_norm": 7.26214075088501, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 42.9364, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.2218214607754734, | |
| "grad_norm": 8.307271003723145, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 43.8462, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.308385933273219, | |
| "grad_norm": 8.332019805908203, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 43.2469, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.308385933273219, | |
| "eval_loss": 1.39437735080719, | |
| "eval_runtime": 63.2944, | |
| "eval_samples_per_second": 1.849, | |
| "eval_steps_per_second": 0.932, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.394950405770965, | |
| "grad_norm": 8.096892356872559, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 41.2373, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.4815148782687104, | |
| "grad_norm": 8.60307502746582, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 42.4838, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.5680793507664563, | |
| "grad_norm": 8.435927391052246, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 43.469, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5680793507664563, | |
| "eval_loss": 1.376704454421997, | |
| "eval_runtime": 63.2953, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.654643823264202, | |
| "grad_norm": 9.683271408081055, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 41.1976, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.7412082957619477, | |
| "grad_norm": 8.922101974487305, | |
| "learning_rate": 0.0001, | |
| "loss": 41.2615, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.8277727682596934, | |
| "grad_norm": 8.805673599243164, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 40.953, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.8277727682596934, | |
| "eval_loss": 1.3633700609207153, | |
| "eval_runtime": 63.3102, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.914337240757439, | |
| "grad_norm": 9.028271675109863, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 40.7417, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.010820559062218, | |
| "grad_norm": 9.204501152038574, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 41.2274, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.097385031559964, | |
| "grad_norm": 9.566389083862305, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 37.3547, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.097385031559964, | |
| "eval_loss": 1.3545383214950562, | |
| "eval_runtime": 63.2718, | |
| "eval_samples_per_second": 1.849, | |
| "eval_steps_per_second": 0.932, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.1839495040577095, | |
| "grad_norm": 10.605365753173828, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 36.6109, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.2705139765554554, | |
| "grad_norm": 11.574414253234863, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 37.0909, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.3570784490532013, | |
| "grad_norm": 11.285858154296875, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 38.6229, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.3570784490532013, | |
| "eval_loss": 1.3696061372756958, | |
| "eval_runtime": 63.2953, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.4436429215509468, | |
| "grad_norm": 11.047082901000977, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 37.311, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.5302073940486927, | |
| "grad_norm": 11.326862335205078, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 35.5301, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.616771866546438, | |
| "grad_norm": 10.88636302947998, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 36.4226, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.616771866546438, | |
| "eval_loss": 1.3591108322143555, | |
| "eval_runtime": 63.3023, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.932, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.703336339044184, | |
| "grad_norm": 10.809871673583984, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 37.0722, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.78990081154193, | |
| "grad_norm": 11.037607192993164, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 35.8129, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.8764652840396754, | |
| "grad_norm": 10.600300788879395, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 35.964, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.8764652840396754, | |
| "eval_loss": 1.3585889339447021, | |
| "eval_runtime": 63.2827, | |
| "eval_samples_per_second": 1.849, | |
| "eval_steps_per_second": 0.932, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.020574826496e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |