last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03035413153456998,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003372681281618887,
      "grad_norm": 1.5285249948501587,
      "learning_rate": 2e-05,
      "loss": 3.6562,
      "step": 1
    },
    {
      "epoch": 0.003372681281618887,
      "eval_loss": 3.238746166229248,
      "eval_runtime": 5.2214,
      "eval_samples_per_second": 23.94,
      "eval_steps_per_second": 12.066,
      "step": 1
    },
    {
      "epoch": 0.006745362563237774,
      "grad_norm": 1.1580705642700195,
      "learning_rate": 4e-05,
      "loss": 2.7946,
      "step": 2
    },
    {
      "epoch": 0.01011804384485666,
      "grad_norm": 1.1308095455169678,
      "learning_rate": 6e-05,
      "loss": 2.9512,
      "step": 3
    },
    {
      "epoch": 0.01011804384485666,
      "eval_loss": 3.2365224361419678,
      "eval_runtime": 3.8555,
      "eval_samples_per_second": 32.421,
      "eval_steps_per_second": 16.34,
      "step": 3
    },
    {
      "epoch": 0.013490725126475547,
      "grad_norm": 0.8919812440872192,
      "learning_rate": 8e-05,
      "loss": 2.684,
      "step": 4
    },
    {
      "epoch": 0.016863406408094434,
      "grad_norm": 1.456526279449463,
      "learning_rate": 0.0001,
      "loss": 3.4474,
      "step": 5
    },
    {
      "epoch": 0.02023608768971332,
      "grad_norm": 1.1174732446670532,
      "learning_rate": 0.00012,
      "loss": 2.9132,
      "step": 6
    },
    {
      "epoch": 0.02023608768971332,
      "eval_loss": 3.2028563022613525,
      "eval_runtime": 3.8965,
      "eval_samples_per_second": 32.08,
      "eval_steps_per_second": 16.168,
      "step": 6
    },
    {
      "epoch": 0.023608768971332208,
      "grad_norm": 1.7769813537597656,
      "learning_rate": 0.00014,
      "loss": 3.1908,
      "step": 7
    },
    {
      "epoch": 0.026981450252951095,
      "grad_norm": 1.253227710723877,
      "learning_rate": 0.00016,
      "loss": 3.0612,
      "step": 8
    },
    {
      "epoch": 0.03035413153456998,
      "grad_norm": 1.5583608150482178,
      "learning_rate": 0.00018,
      "loss": 2.9888,
      "step": 9
    },
    {
      "epoch": 0.03035413153456998,
      "eval_loss": 3.009704828262329,
      "eval_runtime": 3.867,
      "eval_samples_per_second": 32.325,
      "eval_steps_per_second": 16.292,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1664607823331328.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}