| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.013875627425387843, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00014015785278169538, | |
| "eval_loss": 4.543891429901123, | |
| "eval_runtime": 2673.8664, | |
| "eval_samples_per_second": 8.988, | |
| "eval_steps_per_second": 4.494, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00042047355834508613, | |
| "grad_norm": 4.4303879737854, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2509, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008409471166901723, | |
| "grad_norm": 4.1902384757995605, | |
| "learning_rate": 6e-05, | |
| "loss": 4.0204, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0012614206750352585, | |
| "grad_norm": 2.7684712409973145, | |
| "learning_rate": 9e-05, | |
| "loss": 2.7809, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0012614206750352585, | |
| "eval_loss": 2.1958885192871094, | |
| "eval_runtime": 2674.8711, | |
| "eval_samples_per_second": 8.985, | |
| "eval_steps_per_second": 4.493, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016818942333803445, | |
| "grad_norm": 2.783182382583618, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9193, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002102367791725431, | |
| "grad_norm": 2.446138381958008, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.6654, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.002522841350070517, | |
| "grad_norm": 1.4197582006454468, | |
| "learning_rate": 0.00018, | |
| "loss": 1.4818, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002522841350070517, | |
| "eval_loss": 1.463165283203125, | |
| "eval_runtime": 2674.6792, | |
| "eval_samples_per_second": 8.986, | |
| "eval_steps_per_second": 4.493, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002943314908415603, | |
| "grad_norm": 1.5759952068328857, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 1.4326, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.003363788466760689, | |
| "grad_norm": 1.0148948431015015, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.5209, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0037842620251057756, | |
| "grad_norm": 1.0281277894973755, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 1.3521, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0037842620251057756, | |
| "eval_loss": 1.3808079957962036, | |
| "eval_runtime": 2674.464, | |
| "eval_samples_per_second": 8.986, | |
| "eval_steps_per_second": 4.493, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.004204735583450862, | |
| "grad_norm": 1.0885648727416992, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.4687, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.004625209141795947, | |
| "grad_norm": 1.1229428052902222, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 1.336, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.005045682700141034, | |
| "grad_norm": 1.0429704189300537, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.3521, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.005045682700141034, | |
| "eval_loss": 1.3701305389404297, | |
| "eval_runtime": 3089.1504, | |
| "eval_samples_per_second": 7.78, | |
| "eval_steps_per_second": 3.89, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.00546615625848612, | |
| "grad_norm": 0.9580327868461609, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 1.3615, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.005886629816831206, | |
| "grad_norm": 0.9180206656455994, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.2992, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0063071033751762925, | |
| "grad_norm": 0.8730207681655884, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.2218, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0063071033751762925, | |
| "eval_loss": 1.368905782699585, | |
| "eval_runtime": 2719.9889, | |
| "eval_samples_per_second": 8.836, | |
| "eval_steps_per_second": 4.418, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.006727576933521378, | |
| "grad_norm": 0.9200890064239502, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 1.3726, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.007148050491866465, | |
| "grad_norm": 0.962693989276886, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 1.3319, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.007568524050211551, | |
| "grad_norm": 0.935663640499115, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.3547, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.007568524050211551, | |
| "eval_loss": 1.3415731191635132, | |
| "eval_runtime": 2673.8225, | |
| "eval_samples_per_second": 8.989, | |
| "eval_steps_per_second": 4.494, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.007988997608556637, | |
| "grad_norm": 0.9458617568016052, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 1.3576, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.008409471166901723, | |
| "grad_norm": 0.8497123122215271, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2875, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.00882994472524681, | |
| "grad_norm": 0.7052412033081055, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.3197, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.00882994472524681, | |
| "eval_loss": 1.3330885171890259, | |
| "eval_runtime": 2673.7616, | |
| "eval_samples_per_second": 8.989, | |
| "eval_steps_per_second": 4.494, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.009250418283591895, | |
| "grad_norm": 0.8267295360565186, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.2522, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.009670891841936981, | |
| "grad_norm": 0.8061164021492004, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 1.2313, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.010091365400282068, | |
| "grad_norm": 0.8370426893234253, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.2828, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.010091365400282068, | |
| "eval_loss": 1.3291566371917725, | |
| "eval_runtime": 2673.7837, | |
| "eval_samples_per_second": 8.989, | |
| "eval_steps_per_second": 4.494, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.010511838958627154, | |
| "grad_norm": 0.9212751984596252, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.3108, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01093231251697224, | |
| "grad_norm": 0.8381961584091187, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.3383, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.011352786075317325, | |
| "grad_norm": 0.8427872657775879, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.3653, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.011352786075317325, | |
| "eval_loss": 1.324036717414856, | |
| "eval_runtime": 2673.9575, | |
| "eval_samples_per_second": 8.988, | |
| "eval_steps_per_second": 4.494, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.011773259633662412, | |
| "grad_norm": 0.9498192667961121, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.331, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.012193733192007498, | |
| "grad_norm": 0.8051778078079224, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.2632, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.012614206750352585, | |
| "grad_norm": 0.9142070412635803, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.3475, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.012614206750352585, | |
| "eval_loss": 1.3223891258239746, | |
| "eval_runtime": 2674.1901, | |
| "eval_samples_per_second": 8.987, | |
| "eval_steps_per_second": 4.494, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.013034680308697671, | |
| "grad_norm": 0.7538347840309143, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.2384, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.013455153867042756, | |
| "grad_norm": 0.7631012201309204, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.2526, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.013875627425387843, | |
| "grad_norm": 1.0042481422424316, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.3432, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.013875627425387843, | |
| "eval_loss": 1.321218729019165, | |
| "eval_runtime": 2673.6321, | |
| "eval_samples_per_second": 8.989, | |
| "eval_steps_per_second": 4.495, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.908402781143695e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |