last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0015081250235644535,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00015081250235644535,
      "grad_norm": 0.5317559242248535,
      "learning_rate": 2e-05,
      "loss": 1.3914,
      "step": 1
    },
    {
      "epoch": 0.00015081250235644535,
      "eval_loss": 1.3070451021194458,
      "eval_runtime": 118.3819,
      "eval_samples_per_second": 23.585,
      "eval_steps_per_second": 11.792,
      "step": 1
    },
    {
      "epoch": 0.0003016250047128907,
      "grad_norm": 0.6152631044387817,
      "learning_rate": 4e-05,
      "loss": 1.2828,
      "step": 2
    },
    {
      "epoch": 0.00045243750706933604,
      "grad_norm": 0.4285082221031189,
      "learning_rate": 6e-05,
      "loss": 1.1774,
      "step": 3
    },
    {
      "epoch": 0.00045243750706933604,
      "eval_loss": 1.3028250932693481,
      "eval_runtime": 117.0178,
      "eval_samples_per_second": 23.86,
      "eval_steps_per_second": 11.93,
      "step": 3
    },
    {
      "epoch": 0.0006032500094257814,
      "grad_norm": 0.5396704077720642,
      "learning_rate": 8e-05,
      "loss": 1.4807,
      "step": 4
    },
    {
      "epoch": 0.0007540625117822267,
      "grad_norm": 0.5223986506462097,
      "learning_rate": 0.0001,
      "loss": 1.1522,
      "step": 5
    },
    {
      "epoch": 0.0009048750141386721,
      "grad_norm": 0.5262135863304138,
      "learning_rate": 0.00012,
      "loss": 1.2866,
      "step": 6
    },
    {
      "epoch": 0.0009048750141386721,
      "eval_loss": 1.2506712675094604,
      "eval_runtime": 117.1884,
      "eval_samples_per_second": 23.825,
      "eval_steps_per_second": 11.912,
      "step": 6
    },
    {
      "epoch": 0.0010556875164951175,
      "grad_norm": 0.6160966157913208,
      "learning_rate": 0.00014,
      "loss": 1.3513,
      "step": 7
    },
    {
      "epoch": 0.0012065000188515628,
      "grad_norm": 0.6128122806549072,
      "learning_rate": 0.00016,
      "loss": 1.2196,
      "step": 8
    },
    {
      "epoch": 0.0013573125212080082,
      "grad_norm": 0.4621923565864563,
      "learning_rate": 0.00018,
      "loss": 1.3301,
      "step": 9
    },
    {
      "epoch": 0.0013573125212080082,
      "eval_loss": 1.1254338026046753,
      "eval_runtime": 117.0854,
      "eval_samples_per_second": 23.846,
      "eval_steps_per_second": 11.923,
      "step": 9
    },
    {
      "epoch": 0.0015081250235644535,
      "grad_norm": 0.443646639585495,
      "learning_rate": 0.0002,
      "loss": 1.1217,
      "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": 2389343238881280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}