last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.015255530129672006,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0015255530129672007,
      "grad_norm": 3.5874860286712646,
      "learning_rate": 2e-05,
      "loss": 1.924,
      "step": 1
    },
    {
      "epoch": 0.0015255530129672007,
      "eval_loss": 3.1886789798736572,
      "eval_runtime": 48.1701,
      "eval_samples_per_second": 5.75,
      "eval_steps_per_second": 2.886,
      "step": 1
    },
    {
      "epoch": 0.0030511060259344014,
      "grad_norm": 4.946148872375488,
      "learning_rate": 4e-05,
      "loss": 3.4698,
      "step": 2
    },
    {
      "epoch": 0.004576659038901602,
      "grad_norm": 7.080003261566162,
      "learning_rate": 6e-05,
      "loss": 4.0679,
      "step": 3
    },
    {
      "epoch": 0.004576659038901602,
      "eval_loss": 3.142275810241699,
      "eval_runtime": 50.3607,
      "eval_samples_per_second": 5.5,
      "eval_steps_per_second": 2.76,
      "step": 3
    },
    {
      "epoch": 0.006102212051868803,
      "grad_norm": 4.932255744934082,
      "learning_rate": 8e-05,
      "loss": 3.0787,
      "step": 4
    },
    {
      "epoch": 0.007627765064836003,
      "grad_norm": 6.312702655792236,
      "learning_rate": 0.0001,
      "loss": 3.8769,
      "step": 5
    },
    {
      "epoch": 0.009153318077803204,
      "grad_norm": 7.200475215911865,
      "learning_rate": 0.00012,
      "loss": 3.5321,
      "step": 6
    },
    {
      "epoch": 0.009153318077803204,
      "eval_loss": 2.6806535720825195,
      "eval_runtime": 51.326,
      "eval_samples_per_second": 5.397,
      "eval_steps_per_second": 2.708,
      "step": 6
    },
    {
      "epoch": 0.010678871090770405,
      "grad_norm": 3.708185911178589,
      "learning_rate": 0.00014,
      "loss": 1.4473,
      "step": 7
    },
    {
      "epoch": 0.012204424103737605,
      "grad_norm": 4.481770038604736,
      "learning_rate": 0.00016,
      "loss": 1.8168,
      "step": 8
    },
    {
      "epoch": 0.013729977116704805,
      "grad_norm": 8.652414321899414,
      "learning_rate": 0.00018,
      "loss": 1.4997,
      "step": 9
    },
    {
      "epoch": 0.013729977116704805,
      "eval_loss": 2.371807336807251,
      "eval_runtime": 50.9338,
      "eval_samples_per_second": 5.438,
      "eval_steps_per_second": 2.729,
      "step": 9
    },
    {
      "epoch": 0.015255530129672006,
      "grad_norm": 15.314831733703613,
      "learning_rate": 0.0002,
      "loss": 1.8957,
      "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": 2245532414115840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}