last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.025789813023855575,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025789813023855577,
      "grad_norm": 1.0424517393112183,
      "learning_rate": 2e-05,
      "loss": 1.0005,
      "step": 1
    },
    {
      "epoch": 0.0025789813023855577,
      "eval_loss": 1.0008567571640015,
      "eval_runtime": 23.9282,
      "eval_samples_per_second": 6.854,
      "eval_steps_per_second": 3.427,
      "step": 1
    },
    {
      "epoch": 0.0051579626047711154,
      "grad_norm": 0.9918249845504761,
      "learning_rate": 4e-05,
      "loss": 0.9353,
      "step": 2
    },
    {
      "epoch": 0.007736943907156673,
      "grad_norm": 1.1058495044708252,
      "learning_rate": 6e-05,
      "loss": 0.9892,
      "step": 3
    },
    {
      "epoch": 0.007736943907156673,
      "eval_loss": 0.9951302409172058,
      "eval_runtime": 23.8207,
      "eval_samples_per_second": 6.885,
      "eval_steps_per_second": 3.442,
      "step": 3
    },
    {
      "epoch": 0.010315925209542231,
      "grad_norm": 0.9536322951316833,
      "learning_rate": 8e-05,
      "loss": 1.1222,
      "step": 4
    },
    {
      "epoch": 0.012894906511927788,
      "grad_norm": 0.6797281503677368,
      "learning_rate": 0.0001,
      "loss": 0.858,
      "step": 5
    },
    {
      "epoch": 0.015473887814313346,
      "grad_norm": 0.8917789459228516,
      "learning_rate": 0.00012,
      "loss": 0.8876,
      "step": 6
    },
    {
      "epoch": 0.015473887814313346,
      "eval_loss": 0.9553524851799011,
      "eval_runtime": 23.8162,
      "eval_samples_per_second": 6.886,
      "eval_steps_per_second": 3.443,
      "step": 6
    },
    {
      "epoch": 0.018052869116698903,
      "grad_norm": 0.7992050051689148,
      "learning_rate": 0.00014,
      "loss": 0.9266,
      "step": 7
    },
    {
      "epoch": 0.020631850419084462,
      "grad_norm": 0.774902880191803,
      "learning_rate": 0.00016,
      "loss": 0.8374,
      "step": 8
    },
    {
      "epoch": 0.02321083172147002,
      "grad_norm": 0.604135274887085,
      "learning_rate": 0.00018,
      "loss": 0.9627,
      "step": 9
    },
    {
      "epoch": 0.02321083172147002,
      "eval_loss": 0.9024636149406433,
      "eval_runtime": 23.8364,
      "eval_samples_per_second": 6.88,
      "eval_steps_per_second": 3.44,
      "step": 9
    },
    {
      "epoch": 0.025789813023855575,
      "grad_norm": 0.654033362865448,
      "learning_rate": 0.0002,
      "loss": 0.8902,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}