| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0020815986677768525, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.326394671107411e-05, | |
| "grad_norm": 0.6871480941772461, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1197, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.326394671107411e-05, | |
| "eval_loss": 1.2025511264801025, | |
| "eval_runtime": 1873.6459, | |
| "eval_samples_per_second": 5.398, | |
| "eval_steps_per_second": 0.675, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016652789342214822, | |
| "grad_norm": 0.6065630912780762, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4157, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002497918401332223, | |
| "grad_norm": 0.6395959854125977, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0421, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00033305578684429644, | |
| "grad_norm": 0.7535728812217712, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8854, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00041631973355537054, | |
| "grad_norm": 0.8002839684486389, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1319, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004995836802664446, | |
| "grad_norm": 0.7185359597206116, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1534, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005828476269775188, | |
| "grad_norm": 0.7994174957275391, | |
| "learning_rate": 7e-05, | |
| "loss": 1.1366, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0006661115736885929, | |
| "grad_norm": 0.9843465089797974, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2944, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.000749375520399667, | |
| "grad_norm": 1.1043306589126587, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1426, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.000749375520399667, | |
| "eval_loss": 1.0168187618255615, | |
| "eval_runtime": 1879.978, | |
| "eval_samples_per_second": 5.38, | |
| "eval_steps_per_second": 0.673, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008326394671107411, | |
| "grad_norm": 0.7300062775611877, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8043, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0009159034138218152, | |
| "grad_norm": 0.5794069170951843, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.8879, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0009991673605328892, | |
| "grad_norm": 0.656294047832489, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8947, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0010824313072439634, | |
| "grad_norm": 0.6405145525932312, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.7168, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0011656952539550376, | |
| "grad_norm": 0.7153993248939514, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.8198, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0012489592006661116, | |
| "grad_norm": 0.6253563761711121, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.9583, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0013322231473771858, | |
| "grad_norm": 0.9064929485321045, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7007, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0014154870940882597, | |
| "grad_norm": 0.6331225633621216, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.6507, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.001498751040799334, | |
| "grad_norm": 0.6313807964324951, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.8009, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001498751040799334, | |
| "eval_loss": 0.7656639218330383, | |
| "eval_runtime": 1878.3063, | |
| "eval_samples_per_second": 5.385, | |
| "eval_steps_per_second": 0.673, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001582014987510408, | |
| "grad_norm": 0.5964687466621399, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8406, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0016652789342214821, | |
| "grad_norm": 0.8008723855018616, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.7004, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0017485428809325561, | |
| "grad_norm": 0.7248008251190186, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.6875, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0018318068276436303, | |
| "grad_norm": 0.6736342906951904, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.761, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0019150707743547043, | |
| "grad_norm": 0.7091475129127502, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.1793, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0019983347210657783, | |
| "grad_norm": 0.6263002753257751, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.7844, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0020815986677768525, | |
| "grad_norm": 0.8400290608406067, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.7321, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.941868820135936e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |