| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01156871818602499, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00046274872744099955, | |
| "grad_norm": 12.331781387329102, | |
| "learning_rate": 1e-05, | |
| "loss": 5.9018, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00046274872744099955, | |
| "eval_loss": 6.24040412902832, | |
| "eval_runtime": 110.0457, | |
| "eval_samples_per_second": 16.539, | |
| "eval_steps_per_second": 2.072, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009254974548819991, | |
| "grad_norm": 11.983473777770996, | |
| "learning_rate": 2e-05, | |
| "loss": 6.1357, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0013882461823229986, | |
| "grad_norm": 11.926801681518555, | |
| "learning_rate": 3e-05, | |
| "loss": 6.3402, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0018509949097639982, | |
| "grad_norm": 12.546172142028809, | |
| "learning_rate": 4e-05, | |
| "loss": 6.0555, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0023137436372049976, | |
| "grad_norm": 13.246417045593262, | |
| "learning_rate": 5e-05, | |
| "loss": 6.1008, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002776492364645997, | |
| "grad_norm": 11.150693893432617, | |
| "learning_rate": 6e-05, | |
| "loss": 6.134, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003239241092086997, | |
| "grad_norm": 12.205668449401855, | |
| "learning_rate": 7e-05, | |
| "loss": 6.3615, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0037019898195279964, | |
| "grad_norm": 11.150944709777832, | |
| "learning_rate": 8e-05, | |
| "loss": 4.6314, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0041647385469689956, | |
| "grad_norm": 11.63155460357666, | |
| "learning_rate": 9e-05, | |
| "loss": 3.4176, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0041647385469689956, | |
| "eval_loss": 3.59645414352417, | |
| "eval_runtime": 108.8575, | |
| "eval_samples_per_second": 16.719, | |
| "eval_steps_per_second": 2.094, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004627487274409995, | |
| "grad_norm": 8.951762199401855, | |
| "learning_rate": 0.0001, | |
| "loss": 3.0291, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005090236001850995, | |
| "grad_norm": 8.684383392333984, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.5782, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.005552984729291994, | |
| "grad_norm": 8.277400970458984, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.9991, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.006015733456732994, | |
| "grad_norm": 16.194562911987305, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.2676, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.006478482184173994, | |
| "grad_norm": 49.7684440612793, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 1.3732, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.006941230911614993, | |
| "grad_norm": 8.08752155303955, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.1734, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.007403979639055993, | |
| "grad_norm": 5.786002159118652, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.1763, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.007866728366496992, | |
| "grad_norm": 4.976882457733154, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.8316, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.008329477093937991, | |
| "grad_norm": 5.076099395751953, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.658, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008329477093937991, | |
| "eval_loss": 0.7718914747238159, | |
| "eval_runtime": 108.8254, | |
| "eval_samples_per_second": 16.724, | |
| "eval_steps_per_second": 2.095, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008792225821378992, | |
| "grad_norm": 4.327938556671143, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.7712, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.00925497454881999, | |
| "grad_norm": 4.913948059082031, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6246, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00971772327626099, | |
| "grad_norm": 3.194206714630127, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.4449, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01018047200370199, | |
| "grad_norm": 2.053872585296631, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.2754, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01064322073114299, | |
| "grad_norm": 5.223790168762207, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.6558, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.011105969458583989, | |
| "grad_norm": 9.427882194519043, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8059, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01156871818602499, | |
| "grad_norm": 3.001323938369751, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1372, | |
| "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": 4940970105765888.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |