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": 3.153057813644409,
      "learning_rate": 2e-05,
      "loss": 3.6096,
      "step": 1
    },
    {
      "epoch": 0.009345794392523364,
      "eval_loss": 3.5910470485687256,
      "eval_runtime": 4.8275,
      "eval_samples_per_second": 9.322,
      "eval_steps_per_second": 4.764,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 3.5898184776306152,
      "learning_rate": 4e-05,
      "loss": 3.6859,
      "step": 2
    },
    {
      "epoch": 0.028037383177570093,
      "grad_norm": 3.4159717559814453,
      "learning_rate": 6e-05,
      "loss": 3.6541,
      "step": 3
    },
    {
      "epoch": 0.028037383177570093,
      "eval_loss": 3.5559732913970947,
      "eval_runtime": 3.2553,
      "eval_samples_per_second": 13.823,
      "eval_steps_per_second": 7.065,
      "step": 3
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 3.732574462890625,
      "learning_rate": 8e-05,
      "loss": 3.5935,
      "step": 4
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 4.238545894622803,
      "learning_rate": 0.0001,
      "loss": 3.6626,
      "step": 5
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 4.564755439758301,
      "learning_rate": 0.00012,
      "loss": 3.2426,
      "step": 6
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 2.5475592613220215,
      "eval_runtime": 3.2628,
      "eval_samples_per_second": 13.792,
      "eval_steps_per_second": 7.049,
      "step": 6
    },
    {
      "epoch": 0.06542056074766354,
      "grad_norm": 6.730138301849365,
      "learning_rate": 0.00014,
      "loss": 2.4926,
      "step": 7
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 5.640018939971924,
      "learning_rate": 0.00016,
      "loss": 1.8213,
      "step": 8
    },
    {
      "epoch": 0.08411214953271028,
      "grad_norm": 6.035444259643555,
      "learning_rate": 0.00018,
      "loss": 1.2622,
      "step": 9
    },
    {
      "epoch": 0.08411214953271028,
      "eval_loss": 0.6004507541656494,
      "eval_runtime": 3.3196,
      "eval_samples_per_second": 13.556,
      "eval_steps_per_second": 6.929,
      "step": 9
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 5.222229480743408,
      "learning_rate": 0.0002,
      "loss": 0.6174,
      "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": 1849565254778880.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}