trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018648984666390384,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.9006199836730957,
      "learning_rate": 0.0002,
      "loss": 1.9682,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.7868534326553345,
      "learning_rate": 0.0002,
      "loss": 1.64,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.5012632608413696,
      "learning_rate": 0.0002,
      "loss": 1.2592,
      "step": 15
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.6685112714767456,
      "learning_rate": 0.0002,
      "loss": 1.1207,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7823353409767151,
      "learning_rate": 0.0002,
      "loss": 0.9307,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6528989672660828,
      "learning_rate": 0.0002,
      "loss": 0.8931,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8416471481323242,
      "learning_rate": 0.0002,
      "loss": 0.7583,
      "step": 35
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4832550883293152,
      "learning_rate": 0.0002,
      "loss": 0.5014,
      "step": 40
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6577726602554321,
      "learning_rate": 0.0002,
      "loss": 0.5638,
      "step": 45
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.4030466079711914,
      "learning_rate": 0.0002,
      "loss": 0.4848,
      "step": 50
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3284258842468262,
      "learning_rate": 0.0002,
      "loss": 0.9848,
      "step": 55
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4354955554008484,
      "learning_rate": 0.0002,
      "loss": 0.795,
      "step": 60
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8855576515197754,
      "learning_rate": 0.0002,
      "loss": 0.8178,
      "step": 65
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7113056778907776,
      "learning_rate": 0.0002,
      "loss": 0.8127,
      "step": 70
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5555365681648254,
      "learning_rate": 0.0002,
      "loss": 0.8163,
      "step": 75
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.4705950617790222,
      "learning_rate": 0.0002,
      "loss": 0.5927,
      "step": 80
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.41031908988952637,
      "learning_rate": 0.0002,
      "loss": 0.5747,
      "step": 85
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.43267205357551575,
      "learning_rate": 0.0002,
      "loss": 0.4016,
      "step": 90
    }
  ],
  "logging_steps": 5,
  "max_steps": 4826,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "total_flos": 1254693615894528.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}