| { | |
| "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": 218.56756591796875, | |
| "learning_rate": 0.00020137375226675998, | |
| "loss": 116.2836, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 63.336116790771484, | |
| "learning_rate": 0.000192983179255645, | |
| "loss": 19.7601, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 2.302827835083008, | |
| "eval_meteor": 0.0, | |
| "eval_rouge": 0.0, | |
| "eval_runtime": 185.734, | |
| "eval_samples_per_second": 1.346, | |
| "eval_steps_per_second": 1.346, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 70.50274658203125, | |
| "learning_rate": 0.00018459260624453, | |
| "loss": 9.1734, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.8686909675598145, | |
| "learning_rate": 0.00017620203323341497, | |
| "loss": 2.7262, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 3.4275879859924316, | |
| "learning_rate": 0.0001678114602223, | |
| "loss": 1.414, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.28776201605796814, | |
| "eval_meteor": 0.0321522930829127, | |
| "eval_rouge": 0.04923081784861186, | |
| "eval_runtime": 190.019, | |
| "eval_samples_per_second": 1.316, | |
| "eval_steps_per_second": 1.316, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.1697942018508911, | |
| "learning_rate": 0.000159420887211185, | |
| "loss": 0.9842, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 1.0976201295852661, | |
| "learning_rate": 0.00015103031420006998, | |
| "loss": 0.8046, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.15482668578624725, | |
| "eval_meteor": 0.07836448350060113, | |
| "eval_rouge": 0.11588703439134326, | |
| "eval_runtime": 185.0738, | |
| "eval_samples_per_second": 1.351, | |
| "eval_steps_per_second": 1.351, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.179997682571411, | |
| "learning_rate": 0.000142639741188955, | |
| "loss": 0.6368, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.6928311586380005, | |
| "learning_rate": 0.00013424916817784, | |
| "loss": 0.5702, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.005065679550171, | |
| "learning_rate": 0.00012585859516672497, | |
| "loss": 1.5576, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.1288101226091385, | |
| "eval_meteor": 0.09437373191900951, | |
| "eval_rouge": 0.13950906104693644, | |
| "eval_runtime": 185.6365, | |
| "eval_samples_per_second": 1.347, | |
| "eval_steps_per_second": 1.347, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 161.4024658203125, | |
| "learning_rate": 0.00011746802215561, | |
| "loss": 0.5221, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.9336469173431396, | |
| "learning_rate": 0.00010907744914449499, | |
| "loss": 0.5364, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.12464781105518341, | |
| "eval_meteor": 0.10298030905449183, | |
| "eval_rouge": 0.15341155496038483, | |
| "eval_runtime": 185.1181, | |
| "eval_samples_per_second": 1.35, | |
| "eval_steps_per_second": 1.35, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.0329785346984863, | |
| "learning_rate": 0.00010068687613337999, | |
| "loss": 0.489, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1525.2017822265625, | |
| "learning_rate": 9.2296303122265e-05, | |
| "loss": 0.5098, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.729365885257721, | |
| "learning_rate": 8.390573011115e-05, | |
| "loss": 0.4573, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.12196389585733414, | |
| "eval_meteor": 0.11105957979118833, | |
| "eval_rouge": 0.1671068291083625, | |
| "eval_runtime": 185.6124, | |
| "eval_samples_per_second": 1.347, | |
| "eval_steps_per_second": 1.347, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.80484539270401, | |
| "learning_rate": 7.551515710003499e-05, | |
| "loss": 0.4867, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.1568104028701782, | |
| "learning_rate": 6.712458408892e-05, | |
| "loss": 0.4605, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.12032772600650787, | |
| "eval_meteor": 0.11023612064638272, | |
| "eval_rouge": 0.16584557386508242, | |
| "eval_runtime": 186.0494, | |
| "eval_samples_per_second": 1.344, | |
| "eval_steps_per_second": 1.344, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.0484607219696045, | |
| "learning_rate": 5.8734011077805e-05, | |
| "loss": 0.484, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.6715890169143677, | |
| "learning_rate": 5.0343438066689994e-05, | |
| "loss": 0.5185, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.8641834855079651, | |
| "learning_rate": 4.1952865055575e-05, | |
| "loss": 0.4503, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.11902841925621033, | |
| "eval_meteor": 0.1107724532213342, | |
| "eval_rouge": 0.16928825200269476, | |
| "eval_runtime": 185.5798, | |
| "eval_samples_per_second": 1.347, | |
| "eval_steps_per_second": 1.347, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.8465960025787354, | |
| "learning_rate": 3.356229204446e-05, | |
| "loss": 0.4348, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.7913973331451416, | |
| "learning_rate": 2.5171719033344997e-05, | |
| "loss": 0.4515, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.11855262517929077, | |
| "eval_meteor": 0.11552610996579722, | |
| "eval_rouge": 0.17371654211387, | |
| "eval_runtime": 185.2214, | |
| "eval_samples_per_second": 1.35, | |
| "eval_steps_per_second": 1.35, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.65450519323349, | |
| "learning_rate": 1.678114602223e-05, | |
| "loss": 0.4112, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 5.642087459564209, | |
| "learning_rate": 8.390573011115e-06, | |
| "loss": 0.478, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.7398505806922913, | |
| "learning_rate": 0.0, | |
| "loss": 0.4641, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.11815133690834045, | |
| "eval_meteor": 0.11557259073835732, | |
| "eval_rouge": 0.17417265621215244, | |
| "eval_runtime": 185.0931, | |
| "eval_samples_per_second": 1.351, | |
| "eval_steps_per_second": 1.351, | |
| "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": 0.00020976432527787498, | |
| "num_train_epochs": 1 | |
| } | |
| } | |