| { |
| "best_metric": 0.22571060061454773, |
| "best_model_checkpoint": "./beans_outputs/checkpoint-119", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 119, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.5882352941176471, |
| "grad_norm": 1.37638521194458, |
| "learning_rate": 1.831932773109244e-05, |
| "loss": 0.9859, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.9323308270676691, |
| "eval_loss": 0.7496436834335327, |
| "eval_runtime": 5.5278, |
| "eval_samples_per_second": 24.06, |
| "eval_steps_per_second": 0.543, |
| "step": 17 |
| }, |
| { |
| "epoch": 1.1764705882352942, |
| "grad_norm": 1.1821305751800537, |
| "learning_rate": 1.6638655462184877e-05, |
| "loss": 0.808, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.7647058823529411, |
| "grad_norm": 1.143455982208252, |
| "learning_rate": 1.4957983193277313e-05, |
| "loss": 0.6761, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_loss": 0.5276528000831604, |
| "eval_runtime": 6.5714, |
| "eval_samples_per_second": 20.239, |
| "eval_steps_per_second": 0.457, |
| "step": 34 |
| }, |
| { |
| "epoch": 2.3529411764705883, |
| "grad_norm": 1.2171024084091187, |
| "learning_rate": 1.3277310924369749e-05, |
| "loss": 0.5489, |
| "step": 40 |
| }, |
| { |
| "epoch": 2.9411764705882355, |
| "grad_norm": 1.0112650394439697, |
| "learning_rate": 1.1596638655462186e-05, |
| "loss": 0.4598, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_loss": 0.37166041135787964, |
| "eval_runtime": 5.6136, |
| "eval_samples_per_second": 23.693, |
| "eval_steps_per_second": 0.534, |
| "step": 51 |
| }, |
| { |
| "epoch": 3.5294117647058822, |
| "grad_norm": 1.159828782081604, |
| "learning_rate": 9.915966386554622e-06, |
| "loss": 0.4037, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.2968071401119232, |
| "eval_runtime": 6.783, |
| "eval_samples_per_second": 19.608, |
| "eval_steps_per_second": 0.442, |
| "step": 68 |
| }, |
| { |
| "epoch": 4.117647058823529, |
| "grad_norm": 1.0410431623458862, |
| "learning_rate": 8.23529411764706e-06, |
| "loss": 0.3257, |
| "step": 70 |
| }, |
| { |
| "epoch": 4.705882352941177, |
| "grad_norm": 1.3511030673980713, |
| "learning_rate": 6.5546218487394966e-06, |
| "loss": 0.3161, |
| "step": 80 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.2529054880142212, |
| "eval_runtime": 5.166, |
| "eval_samples_per_second": 25.745, |
| "eval_steps_per_second": 0.581, |
| "step": 85 |
| }, |
| { |
| "epoch": 5.294117647058823, |
| "grad_norm": 1.0630207061767578, |
| "learning_rate": 4.873949579831933e-06, |
| "loss": 0.2763, |
| "step": 90 |
| }, |
| { |
| "epoch": 5.882352941176471, |
| "grad_norm": 0.831298828125, |
| "learning_rate": 3.1932773109243696e-06, |
| "loss": 0.2529, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.2268138825893402, |
| "eval_runtime": 4.7432, |
| "eval_samples_per_second": 28.04, |
| "eval_steps_per_second": 0.632, |
| "step": 102 |
| }, |
| { |
| "epoch": 6.470588235294118, |
| "grad_norm": 1.0963715314865112, |
| "learning_rate": 1.5126050420168068e-06, |
| "loss": 0.2624, |
| "step": 110 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.9699248120300752, |
| "eval_loss": 0.22571060061454773, |
| "eval_runtime": 6.6482, |
| "eval_samples_per_second": 20.005, |
| "eval_steps_per_second": 0.451, |
| "step": 119 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 119, |
| "total_flos": 5.608920478833746e+17, |
| "train_loss": 0.46598047368666706, |
| "train_runtime": 470.7987, |
| "train_samples_per_second": 15.374, |
| "train_steps_per_second": 0.253 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 119, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 7, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.608920478833746e+17, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
