| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.451612903225806, | |
| "eval_steps": 200, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03225806451612903, | |
| "eval_loss": 2.5740277767181396, | |
| "eval_runtime": 31.6195, | |
| "eval_samples_per_second": 47.471, | |
| "eval_steps_per_second": 1.992, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 2.625, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9901, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 0.00016, | |
| "loss": 2.0923, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 2.5625, | |
| "learning_rate": 0.00019994532573409262, | |
| "loss": 2.3185, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 0.00019950829025450114, | |
| "loss": 1.3841, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 1.2813, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 0.0001973326597248006, | |
| "loss": 1.3828, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.258064516129032, | |
| "grad_norm": 2.046875, | |
| "learning_rate": 0.00019560357815343577, | |
| "loss": 0.8247, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.5806451612903225, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 0.0001934564464599461, | |
| "loss": 0.7192, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.903225806451613, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 0.00019090065350491626, | |
| "loss": 0.7919, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 0.4979, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.5483870967741935, | |
| "grad_norm": 2.140625, | |
| "learning_rate": 0.00018460952524209355, | |
| "loss": 0.435, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.870967741935484, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.4868, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.193548387096774, | |
| "grad_norm": 1.328125, | |
| "learning_rate": 0.00017684011108568592, | |
| "loss": 0.3416, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.516129032258064, | |
| "grad_norm": 2.0, | |
| "learning_rate": 0.00017244252047910892, | |
| "loss": 0.29, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.838709677419355, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 0.3182, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.161290322580645, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.0001627176358473537, | |
| "loss": 0.2422, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.483870967741936, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 0.00015743286626829437, | |
| "loss": 0.1847, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.806451612903226, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.00015189695737812152, | |
| "loss": 0.2181, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.129032258064516, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 0.0001461341162978688, | |
| "loss": 0.1774, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.451612903225806, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 0.1334, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.451612903225806, | |
| "eval_loss": 4.735217571258545, | |
| "eval_runtime": 31.1163, | |
| "eval_samples_per_second": 48.238, | |
| "eval_steps_per_second": 2.025, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 17, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1941092488380416e+17, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |