last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001007582054963601,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010075820549636011,
      "grad_norm": 6.578465480158216e+18,
      "learning_rate": 2e-05,
      "loss": 1.8654,
      "step": 1
    },
    {
      "epoch": 0.00010075820549636011,
      "eval_loss": 1.7779617309570312,
      "eval_runtime": 1454.0406,
      "eval_samples_per_second": 2.874,
      "eval_steps_per_second": 1.437,
      "step": 1
    },
    {
      "epoch": 0.00020151641099272022,
      "grad_norm": 2.1462115130466632e+18,
      "learning_rate": 4e-05,
      "loss": 1.3359,
      "step": 2
    },
    {
      "epoch": 0.0003022746164890803,
      "grad_norm": 2.517414610043142e+18,
      "learning_rate": 6e-05,
      "loss": 1.3138,
      "step": 3
    },
    {
      "epoch": 0.0003022746164890803,
      "eval_loss": 1.7785661220550537,
      "eval_runtime": 1454.8386,
      "eval_samples_per_second": 2.872,
      "eval_steps_per_second": 1.437,
      "step": 3
    },
    {
      "epoch": 0.00040303282198544043,
      "grad_norm": 7.439758018171896e+18,
      "learning_rate": 8e-05,
      "loss": 1.5352,
      "step": 4
    },
    {
      "epoch": 0.0005037910274818005,
      "grad_norm": 3.50965568439006e+18,
      "learning_rate": 0.0001,
      "loss": 1.2199,
      "step": 5
    },
    {
      "epoch": 0.0006045492329781607,
      "grad_norm": 5.017360729099993e+18,
      "learning_rate": 0.00012,
      "loss": 1.6428,
      "step": 6
    },
    {
      "epoch": 0.0006045492329781607,
      "eval_loss": 1.799889326095581,
      "eval_runtime": 1454.8745,
      "eval_samples_per_second": 2.872,
      "eval_steps_per_second": 1.437,
      "step": 6
    },
    {
      "epoch": 0.0007053074384745208,
      "grad_norm": 3.866104434853413e+18,
      "learning_rate": 0.00014,
      "loss": 1.3893,
      "step": 7
    },
    {
      "epoch": 0.0008060656439708809,
      "grad_norm": 6.592969687796023e+18,
      "learning_rate": 0.00016,
      "loss": 1.4081,
      "step": 8
    },
    {
      "epoch": 0.000906823849467241,
      "grad_norm": 1.0697785243865186e+19,
      "learning_rate": 0.00018,
      "loss": 1.3395,
      "step": 9
    },
    {
      "epoch": 0.000906823849467241,
      "eval_loss": 1.7862416505813599,
      "eval_runtime": 1455.8794,
      "eval_samples_per_second": 2.87,
      "eval_steps_per_second": 1.436,
      "step": 9
    },
    {
      "epoch": 0.001007582054963601,
      "grad_norm": 8.756319288584831e+18,
      "learning_rate": 0.0002,
      "loss": 1.5203,
      "step": 10
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2058737544069120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}