| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08818342151675485, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003527336860670194, | |
| "grad_norm": 24.732351303100586, | |
| "learning_rate": 1e-05, | |
| "loss": 12.7829, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003527336860670194, | |
| "eval_loss": 6.23192834854126, | |
| "eval_runtime": 37.6613, | |
| "eval_samples_per_second": 6.346, | |
| "eval_steps_per_second": 0.797, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007054673721340388, | |
| "grad_norm": 25.1240234375, | |
| "learning_rate": 2e-05, | |
| "loss": 12.1045, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.010582010582010581, | |
| "grad_norm": 25.76141357421875, | |
| "learning_rate": 3e-05, | |
| "loss": 12.4319, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.014109347442680775, | |
| "grad_norm": 24.581771850585938, | |
| "learning_rate": 4e-05, | |
| "loss": 12.3229, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01763668430335097, | |
| "grad_norm": 25.8974552154541, | |
| "learning_rate": 5e-05, | |
| "loss": 11.0399, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.021164021164021163, | |
| "grad_norm": 25.636842727661133, | |
| "learning_rate": 6e-05, | |
| "loss": 10.5125, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "grad_norm": 22.79175567626953, | |
| "learning_rate": 7e-05, | |
| "loss": 8.734, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02821869488536155, | |
| "grad_norm": 21.83588218688965, | |
| "learning_rate": 8e-05, | |
| "loss": 6.1147, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.031746031746031744, | |
| "grad_norm": 13.920812606811523, | |
| "learning_rate": 9e-05, | |
| "loss": 3.7133, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.031746031746031744, | |
| "eval_loss": 0.9912834763526917, | |
| "eval_runtime": 37.6592, | |
| "eval_samples_per_second": 6.346, | |
| "eval_steps_per_second": 0.797, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03527336860670194, | |
| "grad_norm": 10.068586349487305, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0718, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03880070546737213, | |
| "grad_norm": 6.6233954429626465, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 1.2433, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.042328042328042326, | |
| "grad_norm": 6.429514408111572, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8386, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04585537918871252, | |
| "grad_norm": 5.240244388580322, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.6355, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04938271604938271, | |
| "grad_norm": 6.982906341552734, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.7284, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05291005291005291, | |
| "grad_norm": 5.769637584686279, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.696, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0564373897707231, | |
| "grad_norm": 5.077568054199219, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3887, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.059964726631393295, | |
| "grad_norm": 4.646890640258789, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.6242, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06349206349206349, | |
| "grad_norm": 11.119400024414062, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.9408, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06349206349206349, | |
| "eval_loss": 0.24564185738563538, | |
| "eval_runtime": 37.6951, | |
| "eval_samples_per_second": 6.34, | |
| "eval_steps_per_second": 0.796, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06701940035273368, | |
| "grad_norm": 3.4085452556610107, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.2693, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07054673721340388, | |
| "grad_norm": 3.222567081451416, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.3019, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 2.951718807220459, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.3, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07760141093474426, | |
| "grad_norm": 3.150925397872925, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.1912, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08112874779541446, | |
| "grad_norm": 4.397229194641113, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.3351, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08465608465608465, | |
| "grad_norm": 7.507371425628662, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.6114, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08818342151675485, | |
| "grad_norm": 3.729475736618042, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1746, | |
| "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": 3.173894848512e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |