| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9863013698630136, | |
| "eval_steps": 500, | |
| "global_step": 54, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0182648401826484, | |
| "grad_norm": 49.48547958368949, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 2.3036, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.091324200913242, | |
| "grad_norm": 2.6538097911936167, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.05, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.182648401826484, | |
| "grad_norm": 1.663364659313015, | |
| "learning_rate": 1.9659258262890683e-05, | |
| "loss": 1.6599, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.273972602739726, | |
| "grad_norm": 1.4600874528540437, | |
| "learning_rate": 1.8314696123025456e-05, | |
| "loss": 1.4826, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.365296803652968, | |
| "grad_norm": 1.4976704886658572, | |
| "learning_rate": 1.608761429008721e-05, | |
| "loss": 1.3547, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.45662100456621, | |
| "grad_norm": 1.5085864559550755, | |
| "learning_rate": 1.3214394653031616e-05, | |
| "loss": 1.2516, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.547945205479452, | |
| "grad_norm": 1.5000381615082468, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1713, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.639269406392694, | |
| "grad_norm": 1.5216804128853407, | |
| "learning_rate": 6.785605346968387e-06, | |
| "loss": 1.1011, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.730593607305936, | |
| "grad_norm": 1.5550519866063839, | |
| "learning_rate": 3.912385709912794e-06, | |
| "loss": 1.0503, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "grad_norm": 1.518319715437106, | |
| "learning_rate": 1.6853038769745466e-06, | |
| "loss": 1.0184, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.91324200913242, | |
| "grad_norm": 1.5109821547066817, | |
| "learning_rate": 3.4074173710931804e-07, | |
| "loss": 1.0034, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9863013698630136, | |
| "eval_loss": 2.2775027751922607, | |
| "eval_runtime": 0.7582, | |
| "eval_samples_per_second": 27.698, | |
| "eval_steps_per_second": 1.319, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9863013698630136, | |
| "step": 54, | |
| "total_flos": 22508312985600.0, | |
| "train_loss": 1.2952520185046725, | |
| "train_runtime": 626.5642, | |
| "train_samples_per_second": 44.6, | |
| "train_steps_per_second": 0.086 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 54, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 22508312985600.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |