last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.017543859649122806,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "grad_norm": 5.400356769561768,
      "learning_rate": 2e-05,
      "loss": 5.6511,
      "step": 1
    },
    {
      "epoch": 0.0017543859649122807,
      "eval_loss": 1.8766847848892212,
      "eval_runtime": 3.4026,
      "eval_samples_per_second": 70.535,
      "eval_steps_per_second": 35.267,
      "step": 1
    },
    {
      "epoch": 0.0035087719298245615,
      "grad_norm": 5.834110736846924,
      "learning_rate": 4e-05,
      "loss": 7.0515,
      "step": 2
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": 14.82625961303711,
      "learning_rate": 6e-05,
      "loss": 9.2793,
      "step": 3
    },
    {
      "epoch": 0.005263157894736842,
      "eval_loss": 1.8716143369674683,
      "eval_runtime": 2.3264,
      "eval_samples_per_second": 103.165,
      "eval_steps_per_second": 51.582,
      "step": 3
    },
    {
      "epoch": 0.007017543859649123,
      "grad_norm": 6.407444953918457,
      "learning_rate": 8e-05,
      "loss": 7.4857,
      "step": 4
    },
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 6.00010347366333,
      "learning_rate": 0.0001,
      "loss": 6.5114,
      "step": 5
    },
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 6.50211763381958,
      "learning_rate": 0.00012,
      "loss": 7.7122,
      "step": 6
    },
    {
      "epoch": 0.010526315789473684,
      "eval_loss": 1.8417414426803589,
      "eval_runtime": 2.4088,
      "eval_samples_per_second": 99.637,
      "eval_steps_per_second": 49.818,
      "step": 6
    },
    {
      "epoch": 0.012280701754385965,
      "grad_norm": 5.6868414878845215,
      "learning_rate": 0.00014,
      "loss": 6.1701,
      "step": 7
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 6.236565589904785,
      "learning_rate": 0.00016,
      "loss": 8.7456,
      "step": 8
    },
    {
      "epoch": 0.015789473684210527,
      "grad_norm": 5.2089314460754395,
      "learning_rate": 0.00018,
      "loss": 6.338,
      "step": 9
    },
    {
      "epoch": 0.015789473684210527,
      "eval_loss": 1.7883986234664917,
      "eval_runtime": 2.3464,
      "eval_samples_per_second": 102.283,
      "eval_steps_per_second": 51.142,
      "step": 9
    },
    {
      "epoch": 0.017543859649122806,
      "grad_norm": 5.971890449523926,
      "learning_rate": 0.0002,
      "loss": 8.793,
      "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": 34497956413440.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}