last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002203978180616012,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00022039781806160118,
      "grad_norm": 0.6059606075286865,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 1
    },
    {
      "epoch": 0.00022039781806160118,
      "eval_loss": 0.6642674207687378,
      "eval_runtime": 56.6046,
      "eval_samples_per_second": 33.76,
      "eval_steps_per_second": 16.889,
      "step": 1
    },
    {
      "epoch": 0.00044079563612320236,
      "grad_norm": 0.5567238926887512,
      "learning_rate": 4e-05,
      "loss": 0.603,
      "step": 2
    },
    {
      "epoch": 0.0006611934541848036,
      "grad_norm": 0.622408926486969,
      "learning_rate": 6e-05,
      "loss": 0.694,
      "step": 3
    },
    {
      "epoch": 0.0006611934541848036,
      "eval_loss": 0.660050094127655,
      "eval_runtime": 55.3516,
      "eval_samples_per_second": 34.525,
      "eval_steps_per_second": 17.271,
      "step": 3
    },
    {
      "epoch": 0.0008815912722464047,
      "grad_norm": 0.6857208609580994,
      "learning_rate": 8e-05,
      "loss": 0.8926,
      "step": 4
    },
    {
      "epoch": 0.001101989090308006,
      "grad_norm": 0.7746605277061462,
      "learning_rate": 0.0001,
      "loss": 0.5486,
      "step": 5
    },
    {
      "epoch": 0.0013223869083696072,
      "grad_norm": 1.2170764207839966,
      "learning_rate": 0.00012,
      "loss": 0.7053,
      "step": 6
    },
    {
      "epoch": 0.0013223869083696072,
      "eval_loss": 0.5303690433502197,
      "eval_runtime": 55.4034,
      "eval_samples_per_second": 34.492,
      "eval_steps_per_second": 17.255,
      "step": 6
    },
    {
      "epoch": 0.0015427847264312083,
      "grad_norm": 0.7118781208992004,
      "learning_rate": 0.00014,
      "loss": 0.4517,
      "step": 7
    },
    {
      "epoch": 0.0017631825444928094,
      "grad_norm": 0.635513961315155,
      "learning_rate": 0.00016,
      "loss": 0.3173,
      "step": 8
    },
    {
      "epoch": 0.0019835803625544105,
      "grad_norm": 0.5836746096611023,
      "learning_rate": 0.00018,
      "loss": 0.3211,
      "step": 9
    },
    {
      "epoch": 0.0019835803625544105,
      "eval_loss": 0.3485614061355591,
      "eval_runtime": 55.5095,
      "eval_samples_per_second": 34.427,
      "eval_steps_per_second": 17.222,
      "step": 9
    },
    {
      "epoch": 0.002203978180616012,
      "grad_norm": 0.5625272393226624,
      "learning_rate": 0.0002,
      "loss": 0.2329,
      "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": 1742636046090240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}