| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.995334370139969, | |
| "eval_steps": 100, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 355.3971033751965, | |
| "epoch": 0.24883359253499224, | |
| "grad_norm": 0.5323885083198547, | |
| "kl": 14.018265570513904, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 0.5607, | |
| "reward": 0.26941965536680074, | |
| "reward_std": 0.20019021732732653, | |
| "rewards/accuracy_reward": 0.05513393102446571, | |
| "rewards/format_reward": 0.2142857245518826, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 997.3060455322266, | |
| "epoch": 0.4976671850699845, | |
| "grad_norm": 0.22486089169979095, | |
| "kl": 0.11681123820599168, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.0047, | |
| "reward": 0.0003348214435391128, | |
| "reward_std": 0.000947018014267087, | |
| "rewards/accuracy_reward": 0.0003348214435391128, | |
| "rewards/format_reward": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "completion_length": 104.72723671197892, | |
| "epoch": 0.7465007776049767, | |
| "grad_norm": 0.828610360622406, | |
| "kl": 1.0758276525884867, | |
| "learning_rate": 3.5721239031346067e-06, | |
| "loss": 0.043, | |
| "reward": 0.0027901787078008057, | |
| "reward_std": 0.006452280096709728, | |
| "rewards/accuracy_reward": 0.0027901787078008057, | |
| "rewards/format_reward": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "completion_length": 14.610380122065544, | |
| "epoch": 0.995334370139969, | |
| "grad_norm": 7.8437886238098145, | |
| "kl": 2.0839827720075847, | |
| "learning_rate": 0.0, | |
| "loss": 0.0834, | |
| "reward": 0.29776787203736604, | |
| "reward_std": 0.21829486889764665, | |
| "rewards/accuracy_reward": 0.29776787203736604, | |
| "rewards/format_reward": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.995334370139969, | |
| "step": 40, | |
| "total_flos": 0.0, | |
| "train_loss": 0.17294886400923132, | |
| "train_runtime": 6524.5156, | |
| "train_samples_per_second": 0.69, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 40, | |
| "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": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |