last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09345794392523364,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009345794392523364,
      "grad_norm": 12.886663436889648,
      "learning_rate": 2e-05,
      "loss": 7.7466,
      "step": 1
    },
    {
      "epoch": 0.009345794392523364,
      "eval_loss": 7.740489482879639,
      "eval_runtime": 1.2519,
      "eval_samples_per_second": 35.946,
      "eval_steps_per_second": 18.372,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 9.37987995147705,
      "learning_rate": 4e-05,
      "loss": 7.6111,
      "step": 2
    },
    {
      "epoch": 0.028037383177570093,
      "grad_norm": 11.580976486206055,
      "learning_rate": 6e-05,
      "loss": 7.5346,
      "step": 3
    },
    {
      "epoch": 0.028037383177570093,
      "eval_loss": 7.704050064086914,
      "eval_runtime": 1.2203,
      "eval_samples_per_second": 36.876,
      "eval_steps_per_second": 18.848,
      "step": 3
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 11.376309394836426,
      "learning_rate": 8e-05,
      "loss": 7.6639,
      "step": 4
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 11.1522855758667,
      "learning_rate": 0.0001,
      "loss": 7.3003,
      "step": 5
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 8.25159740447998,
      "learning_rate": 0.00012,
      "loss": 6.9106,
      "step": 6
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 7.386815071105957,
      "eval_runtime": 1.279,
      "eval_samples_per_second": 35.184,
      "eval_steps_per_second": 17.983,
      "step": 6
    },
    {
      "epoch": 0.06542056074766354,
      "grad_norm": 7.800024509429932,
      "learning_rate": 0.00014,
      "loss": 7.3329,
      "step": 7
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 6.739772319793701,
      "learning_rate": 0.00016,
      "loss": 7.4786,
      "step": 8
    },
    {
      "epoch": 0.08411214953271028,
      "grad_norm": 7.774903297424316,
      "learning_rate": 0.00018,
      "loss": 7.3138,
      "step": 9
    },
    {
      "epoch": 0.08411214953271028,
      "eval_loss": 6.979888439178467,
      "eval_runtime": 1.2754,
      "eval_samples_per_second": 35.282,
      "eval_steps_per_second": 18.033,
      "step": 9
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 7.0427446365356445,
      "learning_rate": 0.0002,
      "loss": 7.1759,
      "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": 5429904015360.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}