| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.001625381964761719, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.501527859046876e-05, | |
| "grad_norm": 1.5031994581222534, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 2.6512, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.501527859046876e-05, | |
| "eval_loss": 2.5592129230499268, | |
| "eval_runtime": 744.3133, | |
| "eval_samples_per_second": 8.702, | |
| "eval_steps_per_second": 2.177, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001300305571809375, | |
| "grad_norm": 1.2966653108596802, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.4145, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001950458357714063, | |
| "grad_norm": 1.2454464435577393, | |
| "learning_rate": 3e-06, | |
| "loss": 2.1895, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.000260061114361875, | |
| "grad_norm": 1.6535450220108032, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.4311, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.000260061114361875, | |
| "eval_loss": 2.5587995052337646, | |
| "eval_runtime": 747.1156, | |
| "eval_samples_per_second": 8.669, | |
| "eval_steps_per_second": 2.168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0003250763929523438, | |
| "grad_norm": 1.460278868675232, | |
| "learning_rate": 5e-06, | |
| "loss": 2.4834, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0003900916715428126, | |
| "grad_norm": 1.3657655715942383, | |
| "learning_rate": 6e-06, | |
| "loss": 2.764, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00045510695013328134, | |
| "grad_norm": 1.2265658378601074, | |
| "learning_rate": 7e-06, | |
| "loss": 2.2487, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00052012222872375, | |
| "grad_norm": 1.4672365188598633, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.7131, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00052012222872375, | |
| "eval_loss": 2.557461738586426, | |
| "eval_runtime": 747.5395, | |
| "eval_samples_per_second": 8.664, | |
| "eval_steps_per_second": 2.167, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0005851375073142189, | |
| "grad_norm": 1.4981539249420166, | |
| "learning_rate": 9e-06, | |
| "loss": 2.5813, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006501527859046876, | |
| "grad_norm": 1.5430346727371216, | |
| "learning_rate": 1e-05, | |
| "loss": 2.8314, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007151680644951564, | |
| "grad_norm": 1.5252623558044434, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 2.8162, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0007801833430856252, | |
| "grad_norm": 1.510179042816162, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 2.5948, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0007801833430856252, | |
| "eval_loss": 2.5526089668273926, | |
| "eval_runtime": 747.8674, | |
| "eval_samples_per_second": 8.661, | |
| "eval_steps_per_second": 2.166, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008451986216760939, | |
| "grad_norm": 1.5021880865097046, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 2.7735, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0009102139002665627, | |
| "grad_norm": 1.5108366012573242, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 2.3945, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0009752291788570314, | |
| "grad_norm": 1.2698441743850708, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 2.123, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0010402444574475, | |
| "grad_norm": 1.502321720123291, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 2.4292, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0010402444574475, | |
| "eval_loss": 2.5462558269500732, | |
| "eval_runtime": 747.5898, | |
| "eval_samples_per_second": 8.664, | |
| "eval_steps_per_second": 2.167, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.001105259736037969, | |
| "grad_norm": 1.6321555376052856, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 2.7363, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0011702750146284377, | |
| "grad_norm": 1.568772315979004, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 2.4517, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012352902932189064, | |
| "grad_norm": 1.418908715248108, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 2.3306, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.001300305571809375, | |
| "grad_norm": 1.7367687225341797, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 2.7322, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001300305571809375, | |
| "eval_loss": 2.5415749549865723, | |
| "eval_runtime": 747.407, | |
| "eval_samples_per_second": 8.666, | |
| "eval_steps_per_second": 2.167, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001365320850399844, | |
| "grad_norm": 1.4429658651351929, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 2.2171, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0014303361289903127, | |
| "grad_norm": 1.568127989768982, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 2.943, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0014953514075807814, | |
| "grad_norm": 1.6385589838027954, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 2.3739, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0015603666861712503, | |
| "grad_norm": 1.5198174715042114, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 2.8015, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0015603666861712503, | |
| "eval_loss": 2.5409297943115234, | |
| "eval_runtime": 747.9115, | |
| "eval_samples_per_second": 8.66, | |
| "eval_steps_per_second": 2.166, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001625381964761719, | |
| "grad_norm": 1.4175868034362793, | |
| "learning_rate": 0.0, | |
| "loss": 2.5874, | |
| "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": 4115865010176000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |