last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03163055511624229,
  "eval_steps": 13,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006326111023248458,
      "eval_loss": 0.9087442755699158,
      "eval_runtime": 50.294,
      "eval_samples_per_second": 13.242,
      "eval_steps_per_second": 6.621,
      "step": 1
    },
    {
      "epoch": 0.003163055511624229,
      "grad_norm": 0.1919880360364914,
      "learning_rate": 5e-05,
      "loss": 0.8983,
      "step": 5
    },
    {
      "epoch": 0.006326111023248458,
      "grad_norm": 0.21347008645534515,
      "learning_rate": 0.0001,
      "loss": 0.8824,
      "step": 10
    },
    {
      "epoch": 0.008223944330222995,
      "eval_loss": 0.7723798155784607,
      "eval_runtime": 50.8359,
      "eval_samples_per_second": 13.101,
      "eval_steps_per_second": 6.55,
      "step": 13
    },
    {
      "epoch": 0.009489166534872686,
      "grad_norm": 0.25495731830596924,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.7955,
      "step": 15
    },
    {
      "epoch": 0.012652222046496917,
      "grad_norm": 0.3009412884712219,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7136,
      "step": 20
    },
    {
      "epoch": 0.015815277558121146,
      "grad_norm": 0.28066885471343994,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.659,
      "step": 25
    },
    {
      "epoch": 0.01644788866044599,
      "eval_loss": 0.6159524917602539,
      "eval_runtime": 51.1075,
      "eval_samples_per_second": 13.031,
      "eval_steps_per_second": 6.516,
      "step": 26
    },
    {
      "epoch": 0.018978333069745373,
      "grad_norm": 0.3070535957813263,
      "learning_rate": 5e-05,
      "loss": 0.6121,
      "step": 30
    },
    {
      "epoch": 0.022141388581369603,
      "grad_norm": 0.3568773567676544,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.574,
      "step": 35
    },
    {
      "epoch": 0.024671832990668986,
      "eval_loss": 0.5655958652496338,
      "eval_runtime": 51.1891,
      "eval_samples_per_second": 13.011,
      "eval_steps_per_second": 6.505,
      "step": 39
    },
    {
      "epoch": 0.025304444092993834,
      "grad_norm": 0.35025447607040405,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.5634,
      "step": 40
    },
    {
      "epoch": 0.02846749960461806,
      "grad_norm": 0.36843404173851013,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.5792,
      "step": 45
    },
    {
      "epoch": 0.03163055511624229,
      "grad_norm": 0.3375423848628998,
      "learning_rate": 0.0,
      "loss": 0.5554,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.74263604609024e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}