| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 25, | |
| "global_step": 250, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 219.40835571289062, | |
| "learning_rate": 0.00018801838248448536, | |
| "loss": 115.2915, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 62.35649108886719, | |
| "learning_rate": 0.00018018428321429848, | |
| "loss": 19.8096, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 2.5786311626434326, | |
| "eval_meteor": 0.0, | |
| "eval_rouge": 0.0, | |
| "eval_runtime": 185.8184, | |
| "eval_samples_per_second": 1.345, | |
| "eval_steps_per_second": 1.345, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 76.14645385742188, | |
| "learning_rate": 0.00017235018394411159, | |
| "loss": 9.7015, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.961805820465088, | |
| "learning_rate": 0.00016451608467392469, | |
| "loss": 2.9942, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 43.01859664916992, | |
| "learning_rate": 0.00015668198540373781, | |
| "loss": 1.427, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.28364765644073486, | |
| "eval_meteor": 0.02273246229783531, | |
| "eval_rouge": 0.03521551652763062, | |
| "eval_runtime": 185.2799, | |
| "eval_samples_per_second": 1.349, | |
| "eval_steps_per_second": 1.349, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.4709396362304688, | |
| "learning_rate": 0.00014884788613355092, | |
| "loss": 0.9872, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 1.0122220516204834, | |
| "learning_rate": 0.00014101378686336402, | |
| "loss": 0.7634, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.14952343702316284, | |
| "eval_meteor": 0.07689507069769044, | |
| "eval_rouge": 0.10951199778067923, | |
| "eval_runtime": 183.0423, | |
| "eval_samples_per_second": 1.366, | |
| "eval_steps_per_second": 1.366, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 20.24289321899414, | |
| "learning_rate": 0.00013317968759317715, | |
| "loss": 0.6099, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.6482983827590942, | |
| "learning_rate": 0.00012534558832299025, | |
| "loss": 0.5676, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.0459566116333008, | |
| "learning_rate": 0.00011751148905280335, | |
| "loss": 0.5225, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.12942516803741455, | |
| "eval_meteor": 0.09042579962176768, | |
| "eval_rouge": 0.1346603229762297, | |
| "eval_runtime": 183.3098, | |
| "eval_samples_per_second": 1.364, | |
| "eval_steps_per_second": 1.364, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.8052979111671448, | |
| "learning_rate": 0.00010967738978261648, | |
| "loss": 0.515, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.8996737599372864, | |
| "learning_rate": 0.00010184329051242958, | |
| "loss": 0.5177, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.12226101756095886, | |
| "eval_meteor": 0.10434284222368143, | |
| "eval_rouge": 0.15413964328066987, | |
| "eval_runtime": 184.6588, | |
| "eval_samples_per_second": 1.354, | |
| "eval_steps_per_second": 1.354, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.6967951655387878, | |
| "learning_rate": 9.400919124224268e-05, | |
| "loss": 0.5249, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5539537072181702, | |
| "learning_rate": 8.617509197205579e-05, | |
| "loss": 0.4336, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.6934689283370972, | |
| "learning_rate": 7.834099270186891e-05, | |
| "loss": 0.4581, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.12093699723482132, | |
| "eval_meteor": 0.11141702521015225, | |
| "eval_rouge": 0.16876141956675836, | |
| "eval_runtime": 184.3427, | |
| "eval_samples_per_second": 1.356, | |
| "eval_steps_per_second": 1.356, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 3.469697952270508, | |
| "learning_rate": 7.050689343168201e-05, | |
| "loss": 0.4494, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.0301263332366943, | |
| "learning_rate": 6.267279416149512e-05, | |
| "loss": 0.4498, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.11946863681077957, | |
| "eval_meteor": 0.11051272177679479, | |
| "eval_rouge": 0.16635535722733613, | |
| "eval_runtime": 185.0704, | |
| "eval_samples_per_second": 1.351, | |
| "eval_steps_per_second": 1.351, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.7396179437637329, | |
| "learning_rate": 5.483869489130824e-05, | |
| "loss": 0.7089, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.0098114013671875, | |
| "learning_rate": 4.700459562112134e-05, | |
| "loss": 0.4616, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 119.26329803466797, | |
| "learning_rate": 3.9170496350934454e-05, | |
| "loss": 0.4699, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.11950699985027313, | |
| "eval_meteor": 0.11040707687984884, | |
| "eval_rouge": 0.16702139582886516, | |
| "eval_runtime": 185.055, | |
| "eval_samples_per_second": 1.351, | |
| "eval_steps_per_second": 1.351, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.8761780858039856, | |
| "learning_rate": 3.133639708074756e-05, | |
| "loss": 0.9796, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.6994351148605347, | |
| "learning_rate": 2.350229781056067e-05, | |
| "loss": 0.4515, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.1189429759979248, | |
| "eval_meteor": 0.11115742614650127, | |
| "eval_rouge": 0.17056126666712712, | |
| "eval_runtime": 186.9575, | |
| "eval_samples_per_second": 1.337, | |
| "eval_steps_per_second": 1.337, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.6938163638114929, | |
| "learning_rate": 1.566819854037378e-05, | |
| "loss": 0.4165, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.7089868783950806, | |
| "learning_rate": 7.83409927018689e-06, | |
| "loss": 0.4554, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.742696225643158, | |
| "learning_rate": 0.0, | |
| "loss": 0.461, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.11835635453462601, | |
| "eval_meteor": 0.11444057098479471, | |
| "eval_rouge": 0.1729760215295089, | |
| "eval_runtime": 187.724, | |
| "eval_samples_per_second": 1.332, | |
| "eval_steps_per_second": 1.332, | |
| "step": 250 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1027178496000000.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": { | |
| "gradient_accumulation_steps": 3, | |
| "learning_rate": 0.00019585248175467225, | |
| "num_train_epochs": 1 | |
| } | |
| } | |