| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.007058657443354274, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000282346297734171, | |
| "grad_norm": 15.19815444946289, | |
| "learning_rate": 2e-05, | |
| "loss": 12.8505, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000282346297734171, | |
| "eval_loss": NaN, | |
| "eval_runtime": 129.5074, | |
| "eval_samples_per_second": 23.033, | |
| "eval_steps_per_second": 11.521, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000564692595468342, | |
| "grad_norm": 14.651299476623535, | |
| "learning_rate": 4e-05, | |
| "loss": 13.507, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0008470388932025129, | |
| "grad_norm": 17.63684844970703, | |
| "learning_rate": 6e-05, | |
| "loss": 12.885, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008470388932025129, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.7833, | |
| "eval_samples_per_second": 23.344, | |
| "eval_steps_per_second": 11.676, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001129385190936684, | |
| "grad_norm": 18.415891647338867, | |
| "learning_rate": 8e-05, | |
| "loss": 14.0113, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0014117314886708547, | |
| "grad_norm": 13.788007736206055, | |
| "learning_rate": 0.0001, | |
| "loss": 11.6638, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016940777864050257, | |
| "grad_norm": 8.320622444152832, | |
| "learning_rate": 0.00012, | |
| "loss": 10.1862, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0016940777864050257, | |
| "eval_loss": NaN, | |
| "eval_runtime": 128.1603, | |
| "eval_samples_per_second": 23.276, | |
| "eval_steps_per_second": 11.642, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0019764240841391965, | |
| "grad_norm": 15.466882705688477, | |
| "learning_rate": 0.00014, | |
| "loss": 16.0399, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002258770381873368, | |
| "grad_norm": 8.012770652770996, | |
| "learning_rate": 0.00016, | |
| "loss": 8.4435, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0025411166796075386, | |
| "grad_norm": 10.118023872375488, | |
| "learning_rate": 0.00018, | |
| "loss": 8.8868, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0025411166796075386, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.7698, | |
| "eval_samples_per_second": 23.347, | |
| "eval_steps_per_second": 11.677, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0028234629773417094, | |
| "grad_norm": 11.526106834411621, | |
| "learning_rate": 0.0002, | |
| "loss": 7.4083, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0031058092750758807, | |
| "grad_norm": 29.451435089111328, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 11.2583, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0033881555728100515, | |
| "grad_norm": 20.30539321899414, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 6.8495, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0033881555728100515, | |
| "eval_loss": NaN, | |
| "eval_runtime": 128.0998, | |
| "eval_samples_per_second": 23.287, | |
| "eval_steps_per_second": 11.647, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0036705018705442227, | |
| "grad_norm": 15.124122619628906, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 5.6325, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.003952848168278393, | |
| "grad_norm": 26.152379989624023, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 5.1915, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004235194466012565, | |
| "grad_norm": 10.615568161010742, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.4316, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004235194466012565, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.6488, | |
| "eval_samples_per_second": 23.369, | |
| "eval_steps_per_second": 11.688, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004517540763746736, | |
| "grad_norm": 17.02305793762207, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 5.1498, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.004799887061480906, | |
| "grad_norm": 18.758045196533203, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 2.8793, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.005082233359215077, | |
| "grad_norm": 19.13789176940918, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 3.4841, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005082233359215077, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.5619, | |
| "eval_samples_per_second": 23.385, | |
| "eval_steps_per_second": 11.696, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005364579656949248, | |
| "grad_norm": 12.120722770690918, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 2.6428, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.005646925954683419, | |
| "grad_norm": 13.740382194519043, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.3224, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0059292722524175905, | |
| "grad_norm": 19.52411460876465, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 2.2475, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0059292722524175905, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.529, | |
| "eval_samples_per_second": 23.391, | |
| "eval_steps_per_second": 11.699, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006211618550151761, | |
| "grad_norm": 15.904952049255371, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 2.4317, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.006493964847885932, | |
| "grad_norm": 14.597325325012207, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 1.2465, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.006776311145620103, | |
| "grad_norm": 13.051502227783203, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.67, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.006776311145620103, | |
| "eval_loss": NaN, | |
| "eval_runtime": 127.5334, | |
| "eval_samples_per_second": 23.39, | |
| "eval_steps_per_second": 11.699, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007058657443354274, | |
| "grad_norm": 14.694950103759766, | |
| "learning_rate": 0.0, | |
| "loss": 2.2391, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6531344184115200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |