last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004253056884635832,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004253056884635832,
      "grad_norm": 0.30059701204299927,
      "learning_rate": 2e-05,
      "loss": 1.5456,
      "step": 1
    },
    {
      "epoch": 0.0004253056884635832,
      "eval_loss": 1.6050080060958862,
      "eval_runtime": 35.1324,
      "eval_samples_per_second": 28.179,
      "eval_steps_per_second": 14.09,
      "step": 1
    },
    {
      "epoch": 0.0008506113769271664,
      "grad_norm": 0.3101019263267517,
      "learning_rate": 4e-05,
      "loss": 1.6868,
      "step": 2
    },
    {
      "epoch": 0.0012759170653907496,
      "grad_norm": 0.3087214231491089,
      "learning_rate": 6e-05,
      "loss": 1.5642,
      "step": 3
    },
    {
      "epoch": 0.0012759170653907496,
      "eval_loss": 1.6043040752410889,
      "eval_runtime": 35.6368,
      "eval_samples_per_second": 27.78,
      "eval_steps_per_second": 13.89,
      "step": 3
    },
    {
      "epoch": 0.0017012227538543328,
      "grad_norm": 0.30784478783607483,
      "learning_rate": 8e-05,
      "loss": 1.5413,
      "step": 4
    },
    {
      "epoch": 0.002126528442317916,
      "grad_norm": 0.3248178958892822,
      "learning_rate": 0.0001,
      "loss": 1.7453,
      "step": 5
    },
    {
      "epoch": 0.002551834130781499,
      "grad_norm": 0.3052712082862854,
      "learning_rate": 0.00012,
      "loss": 1.5164,
      "step": 6
    },
    {
      "epoch": 0.002551834130781499,
      "eval_loss": 1.594720721244812,
      "eval_runtime": 36.0722,
      "eval_samples_per_second": 27.445,
      "eval_steps_per_second": 13.722,
      "step": 6
    },
    {
      "epoch": 0.0029771398192450825,
      "grad_norm": 0.3414212465286255,
      "learning_rate": 0.00014,
      "loss": 1.5186,
      "step": 7
    },
    {
      "epoch": 0.0034024455077086655,
      "grad_norm": 0.3427436053752899,
      "learning_rate": 0.00016,
      "loss": 1.4756,
      "step": 8
    },
    {
      "epoch": 0.003827751196172249,
      "grad_norm": 0.3569692373275757,
      "learning_rate": 0.00018,
      "loss": 1.4998,
      "step": 9
    },
    {
      "epoch": 0.003827751196172249,
      "eval_loss": 1.5611094236373901,
      "eval_runtime": 36.2928,
      "eval_samples_per_second": 27.278,
      "eval_steps_per_second": 13.639,
      "step": 9
    },
    {
      "epoch": 0.004253056884635832,
      "grad_norm": 0.41053274273872375,
      "learning_rate": 0.0002,
      "loss": 1.727,
      "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": 511583956500480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}