last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018190086402910415,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018190086402910413,
      "grad_norm": 3.2539453506469727,
      "learning_rate": 2e-05,
      "loss": 2.9626,
      "step": 1
    },
    {
      "epoch": 0.0018190086402910413,
      "eval_loss": 0.7271671295166016,
      "eval_runtime": 17.2405,
      "eval_samples_per_second": 13.457,
      "eval_steps_per_second": 6.728,
      "step": 1
    },
    {
      "epoch": 0.0036380172805820826,
      "grad_norm": 2.2560973167419434,
      "learning_rate": 4e-05,
      "loss": 1.6971,
      "step": 2
    },
    {
      "epoch": 0.005457025920873124,
      "grad_norm": 2.837073564529419,
      "learning_rate": 6e-05,
      "loss": 2.1407,
      "step": 3
    },
    {
      "epoch": 0.005457025920873124,
      "eval_loss": 0.7196894288063049,
      "eval_runtime": 17.3608,
      "eval_samples_per_second": 13.363,
      "eval_steps_per_second": 6.682,
      "step": 3
    },
    {
      "epoch": 0.007276034561164165,
      "grad_norm": 2.9078829288482666,
      "learning_rate": 8e-05,
      "loss": 2.3146,
      "step": 4
    },
    {
      "epoch": 0.009095043201455207,
      "grad_norm": 4.572423934936523,
      "learning_rate": 0.0001,
      "loss": 2.6655,
      "step": 5
    },
    {
      "epoch": 0.010914051841746248,
      "grad_norm": 4.510873317718506,
      "learning_rate": 0.00012,
      "loss": 2.8634,
      "step": 6
    },
    {
      "epoch": 0.010914051841746248,
      "eval_loss": 0.6238253116607666,
      "eval_runtime": 17.4728,
      "eval_samples_per_second": 13.278,
      "eval_steps_per_second": 6.639,
      "step": 6
    },
    {
      "epoch": 0.01273306048203729,
      "grad_norm": 4.861642360687256,
      "learning_rate": 0.00014,
      "loss": 2.3511,
      "step": 7
    },
    {
      "epoch": 0.01455206912232833,
      "grad_norm": 4.970233917236328,
      "learning_rate": 0.00016,
      "loss": 1.7317,
      "step": 8
    },
    {
      "epoch": 0.01637107776261937,
      "grad_norm": 7.024624824523926,
      "learning_rate": 0.00018,
      "loss": 1.8258,
      "step": 9
    },
    {
      "epoch": 0.01637107776261937,
      "eval_loss": 0.31996724009513855,
      "eval_runtime": 17.4948,
      "eval_samples_per_second": 13.261,
      "eval_steps_per_second": 6.631,
      "step": 9
    },
    {
      "epoch": 0.018190086402910415,
      "grad_norm": 5.711084365844727,
      "learning_rate": 0.0002,
      "loss": 1.8195,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3257466271825920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}