| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.014400443090556632, | |
| "eval_steps": 13, | |
| "global_step": 52, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00027693159789531985, | |
| "eval_loss": 6.243055820465088, | |
| "eval_runtime": 457.8006, | |
| "eval_samples_per_second": 13.285, | |
| "eval_steps_per_second": 1.662, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008307947936859596, | |
| "grad_norm": 27.681867599487305, | |
| "learning_rate": 3e-05, | |
| "loss": 24.6744, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001661589587371919, | |
| "grad_norm": 27.828317642211914, | |
| "learning_rate": 6e-05, | |
| "loss": 22.5088, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0024923843810578787, | |
| "grad_norm": 29.053722381591797, | |
| "learning_rate": 9e-05, | |
| "loss": 17.9623, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003323179174743838, | |
| "grad_norm": 19.01649284362793, | |
| "learning_rate": 9.994965332706573e-05, | |
| "loss": 11.8099, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.003600110772639158, | |
| "eval_loss": 1.6348825693130493, | |
| "eval_runtime": 460.8045, | |
| "eval_samples_per_second": 13.199, | |
| "eval_steps_per_second": 1.651, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004153973968429798, | |
| "grad_norm": 18.268274307250977, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 6.9041, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004984768762115757, | |
| "grad_norm": 9.425353050231934, | |
| "learning_rate": 9.919647942993148e-05, | |
| "loss": 3.1179, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005815563555801717, | |
| "grad_norm": 13.245735168457031, | |
| "learning_rate": 9.848447601883435e-05, | |
| "loss": 2.4225, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006646358349487676, | |
| "grad_norm": 7.058098793029785, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.7562, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007200221545278316, | |
| "eval_loss": 0.40671274065971375, | |
| "eval_runtime": 461.0952, | |
| "eval_samples_per_second": 13.19, | |
| "eval_steps_per_second": 1.65, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.007477153143173636, | |
| "grad_norm": 5.528973579406738, | |
| "learning_rate": 9.640574942595196e-05, | |
| "loss": 1.6548, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008307947936859596, | |
| "grad_norm": 5.370375156402588, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 1.6354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009138742730545555, | |
| "grad_norm": 4.518975257873535, | |
| "learning_rate": 9.348705665778478e-05, | |
| "loss": 1.5657, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.009969537524231515, | |
| "grad_norm": 3.3204827308654785, | |
| "learning_rate": 9.172866268606513e-05, | |
| "loss": 1.445, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010800332317917475, | |
| "grad_norm": 5.033892631530762, | |
| "learning_rate": 8.978122744408906e-05, | |
| "loss": 1.8423, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.010800332317917475, | |
| "eval_loss": 0.37738242745399475, | |
| "eval_runtime": 460.9405, | |
| "eval_samples_per_second": 13.195, | |
| "eval_steps_per_second": 1.651, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.011631127111603435, | |
| "grad_norm": 2.9625065326690674, | |
| "learning_rate": 8.765357330018056e-05, | |
| "loss": 1.5417, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.012461921905289393, | |
| "grad_norm": 2.775251865386963, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.4174, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013292716698975353, | |
| "grad_norm": 3.418978214263916, | |
| "learning_rate": 8.289693629698564e-05, | |
| "loss": 1.444, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.014123511492661313, | |
| "grad_norm": 2.627183437347412, | |
| "learning_rate": 8.0289502192041e-05, | |
| "loss": 1.3904, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.014400443090556632, | |
| "eval_loss": 0.3832974135875702, | |
| "eval_runtime": 461.0753, | |
| "eval_samples_per_second": 13.191, | |
| "eval_steps_per_second": 1.65, | |
| "step": 52 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.249073800419738e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |