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": 2.570714235305786,
      "learning_rate": 2e-05,
      "loss": 1.4585,
      "step": 1
    },
    {
      "epoch": 0.009345794392523364,
      "eval_loss": 1.5077929496765137,
      "eval_runtime": 9.5194,
      "eval_samples_per_second": 4.727,
      "eval_steps_per_second": 2.416,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 2.610569953918457,
      "learning_rate": 4e-05,
      "loss": 1.5572,
      "step": 2
    },
    {
      "epoch": 0.028037383177570093,
      "grad_norm": 2.719212293624878,
      "learning_rate": 6e-05,
      "loss": 1.6284,
      "step": 3
    },
    {
      "epoch": 0.028037383177570093,
      "eval_loss": 1.4604904651641846,
      "eval_runtime": 9.7081,
      "eval_samples_per_second": 4.635,
      "eval_steps_per_second": 2.369,
      "step": 3
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 3.4488842487335205,
      "learning_rate": 8e-05,
      "loss": 1.718,
      "step": 4
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 2.8154029846191406,
      "learning_rate": 0.0001,
      "loss": 1.355,
      "step": 5
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 3.2244303226470947,
      "learning_rate": 0.00012,
      "loss": 1.1568,
      "step": 6
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 0.9131178259849548,
      "eval_runtime": 9.8418,
      "eval_samples_per_second": 4.572,
      "eval_steps_per_second": 2.337,
      "step": 6
    },
    {
      "epoch": 0.06542056074766354,
      "grad_norm": 2.8733081817626953,
      "learning_rate": 0.00014,
      "loss": 0.8193,
      "step": 7
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 2.0992445945739746,
      "learning_rate": 0.00016,
      "loss": 0.5052,
      "step": 8
    },
    {
      "epoch": 0.08411214953271028,
      "grad_norm": 1.654015064239502,
      "learning_rate": 0.00018,
      "loss": 0.2479,
      "step": 9
    },
    {
      "epoch": 0.08411214953271028,
      "eval_loss": 0.18080545961856842,
      "eval_runtime": 9.8815,
      "eval_samples_per_second": 4.554,
      "eval_steps_per_second": 2.328,
      "step": 9
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 1.7849133014678955,
      "learning_rate": 0.0002,
      "loss": 0.1533,
      "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": 1.306916403806208e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}