| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.176056338028169, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007042253521126761, | |
| "grad_norm": 13.748222351074219, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 8.8316, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007042253521126761, | |
| "eval_loss": 7.815854072570801, | |
| "eval_runtime": 7.8376, | |
| "eval_samples_per_second": 7.655, | |
| "eval_steps_per_second": 1.914, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014084507042253521, | |
| "grad_norm": 15.815945625305176, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 7.4468, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02112676056338028, | |
| "grad_norm": 18.948543548583984, | |
| "learning_rate": 3e-06, | |
| "loss": 7.4097, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.028169014084507043, | |
| "grad_norm": 12.374052047729492, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 7.2361, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.028169014084507043, | |
| "eval_loss": 7.8030195236206055, | |
| "eval_runtime": 7.5, | |
| "eval_samples_per_second": 8.0, | |
| "eval_steps_per_second": 2.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.035211267605633804, | |
| "grad_norm": 12.28042221069336, | |
| "learning_rate": 5e-06, | |
| "loss": 7.465, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04225352112676056, | |
| "grad_norm": 15.298364639282227, | |
| "learning_rate": 6e-06, | |
| "loss": 8.0029, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04929577464788732, | |
| "grad_norm": 15.470710754394531, | |
| "learning_rate": 7e-06, | |
| "loss": 7.5117, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.056338028169014086, | |
| "grad_norm": 15.850247383117676, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 7.5796, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.056338028169014086, | |
| "eval_loss": 7.705232620239258, | |
| "eval_runtime": 7.531, | |
| "eval_samples_per_second": 7.967, | |
| "eval_steps_per_second": 1.992, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06338028169014084, | |
| "grad_norm": 14.885255813598633, | |
| "learning_rate": 9e-06, | |
| "loss": 7.9653, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07042253521126761, | |
| "grad_norm": 17.975255966186523, | |
| "learning_rate": 1e-05, | |
| "loss": 7.2979, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07746478873239436, | |
| "grad_norm": 17.041446685791016, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 7.5555, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.08450704225352113, | |
| "grad_norm": 15.812573432922363, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 7.1048, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08450704225352113, | |
| "eval_loss": 7.370495796203613, | |
| "eval_runtime": 7.5722, | |
| "eval_samples_per_second": 7.924, | |
| "eval_steps_per_second": 1.981, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09154929577464789, | |
| "grad_norm": 16.6043643951416, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 9.138, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.09859154929577464, | |
| "grad_norm": 17.724496841430664, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 7.2715, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1056338028169014, | |
| "grad_norm": 15.051722526550293, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 7.1912, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11267605633802817, | |
| "grad_norm": 15.276244163513184, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 6.932, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.11267605633802817, | |
| "eval_loss": 6.921834468841553, | |
| "eval_runtime": 7.5683, | |
| "eval_samples_per_second": 7.928, | |
| "eval_steps_per_second": 1.982, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.11971830985915492, | |
| "grad_norm": 15.998421669006348, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 6.2836, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1267605633802817, | |
| "grad_norm": 15.332056999206543, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 6.4141, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.13380281690140844, | |
| "grad_norm": 14.301487922668457, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 6.5616, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.14084507042253522, | |
| "grad_norm": 16.059947967529297, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 7.4165, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14084507042253522, | |
| "eval_loss": 6.513575077056885, | |
| "eval_runtime": 7.5941, | |
| "eval_samples_per_second": 7.901, | |
| "eval_steps_per_second": 1.975, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14788732394366197, | |
| "grad_norm": 17.150333404541016, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 5.9722, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.15492957746478872, | |
| "grad_norm": 14.834078788757324, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 7.2155, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.1619718309859155, | |
| "grad_norm": 15.119147300720215, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 7.5088, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.16901408450704225, | |
| "grad_norm": 15.117931365966797, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 6.8602, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.16901408450704225, | |
| "eval_loss": 6.4019856452941895, | |
| "eval_runtime": 7.6048, | |
| "eval_samples_per_second": 7.89, | |
| "eval_steps_per_second": 1.972, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.176056338028169, | |
| "grad_norm": 16.927875518798828, | |
| "learning_rate": 0.0, | |
| "loss": 5.5331, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9273591044505600.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |