| { | |
| "best_global_step": 50, | |
| "best_metric": 0.39821675419807434, | |
| "best_model_checkpoint": "/kaggle/working/Llama-Factory-out/checkpoint-50", | |
| "epoch": 1.0, | |
| "eval_steps": 50, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.056100981767180924, | |
| "grad_norm": 14.733423233032227, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 1.5407, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11220196353436185, | |
| "grad_norm": 12.783563613891602, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1933, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16830294530154277, | |
| "grad_norm": 9.208017349243164, | |
| "learning_rate": 1.555555555555556e-05, | |
| "loss": 0.6806, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2244039270687237, | |
| "grad_norm": 4.085337162017822, | |
| "learning_rate": 1.999048221581858e-05, | |
| "loss": 0.5145, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2805049088359046, | |
| "grad_norm": 4.095623970031738, | |
| "learning_rate": 1.9659258262890683e-05, | |
| "loss": 0.4157, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.33660589060308554, | |
| "grad_norm": 4.0649638175964355, | |
| "learning_rate": 1.887010833178222e-05, | |
| "loss": 0.3864, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.39270687237026647, | |
| "grad_norm": 2.9867515563964844, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.4244, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4488078541374474, | |
| "grad_norm": 2.810619592666626, | |
| "learning_rate": 1.608761429008721e-05, | |
| "loss": 0.3213, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5049088359046283, | |
| "grad_norm": 3.328486204147339, | |
| "learning_rate": 1.4226182617406996e-05, | |
| "loss": 0.3678, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5610098176718092, | |
| "grad_norm": 4.280186653137207, | |
| "learning_rate": 1.2164396139381029e-05, | |
| "loss": 0.4148, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5610098176718092, | |
| "eval_loss": 0.39821675419807434, | |
| "eval_runtime": 224.7069, | |
| "eval_samples_per_second": 4.232, | |
| "eval_steps_per_second": 1.059, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6171107994389902, | |
| "grad_norm": 2.749398708343506, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3889, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6732117812061711, | |
| "grad_norm": 2.5752906799316406, | |
| "learning_rate": 7.835603860618973e-06, | |
| "loss": 0.3327, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.729312762973352, | |
| "grad_norm": 2.7851309776306152, | |
| "learning_rate": 5.773817382593008e-06, | |
| "loss": 0.4176, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.7854137447405329, | |
| "grad_norm": 3.187135934829712, | |
| "learning_rate": 3.912385709912794e-06, | |
| "loss": 0.3558, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8415147265077139, | |
| "grad_norm": 2.3262107372283936, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.3842, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8976157082748948, | |
| "grad_norm": 3.2473442554473877, | |
| "learning_rate": 1.129891668217783e-06, | |
| "loss": 0.3543, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9537166900420757, | |
| "grad_norm": 3.429651975631714, | |
| "learning_rate": 3.4074173710931804e-07, | |
| "loss": 0.4073, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.828529119491577, | |
| "learning_rate": 9.517784181422018e-09, | |
| "loss": 0.2383, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 90, | |
| "total_flos": 1.3864992889634816e+16, | |
| "train_loss": 0.5076928615570069, | |
| "train_runtime": 2432.8191, | |
| "train_samples_per_second": 1.172, | |
| "train_steps_per_second": 0.037 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3864992889634816e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |