last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0059907143926913285,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005990714392691329,
      "grad_norm": 0.3094187080860138,
      "learning_rate": 2e-05,
      "loss": 1.0845,
      "step": 1
    },
    {
      "epoch": 0.0005990714392691329,
      "eval_loss": 1.052202820777893,
      "eval_runtime": 81.582,
      "eval_samples_per_second": 8.617,
      "eval_steps_per_second": 4.315,
      "step": 1
    },
    {
      "epoch": 0.0011981428785382657,
      "grad_norm": 0.23705752193927765,
      "learning_rate": 4e-05,
      "loss": 1.0554,
      "step": 2
    },
    {
      "epoch": 0.0017972143178073986,
      "grad_norm": 0.5556271076202393,
      "learning_rate": 6e-05,
      "loss": 1.067,
      "step": 3
    },
    {
      "epoch": 0.0017972143178073986,
      "eval_loss": 1.0498274564743042,
      "eval_runtime": 79.529,
      "eval_samples_per_second": 8.84,
      "eval_steps_per_second": 4.426,
      "step": 3
    },
    {
      "epoch": 0.0023962857570765314,
      "grad_norm": 0.28615206480026245,
      "learning_rate": 8e-05,
      "loss": 1.1168,
      "step": 4
    },
    {
      "epoch": 0.0029953571963456643,
      "grad_norm": 0.26683083176612854,
      "learning_rate": 0.0001,
      "loss": 0.9644,
      "step": 5
    },
    {
      "epoch": 0.003594428635614797,
      "grad_norm": 0.2868586480617523,
      "learning_rate": 0.00012,
      "loss": 0.98,
      "step": 6
    },
    {
      "epoch": 0.003594428635614797,
      "eval_loss": 1.0248427391052246,
      "eval_runtime": 79.5269,
      "eval_samples_per_second": 8.84,
      "eval_steps_per_second": 4.426,
      "step": 6
    },
    {
      "epoch": 0.00419350007488393,
      "grad_norm": 0.3919077515602112,
      "learning_rate": 0.00014,
      "loss": 0.9973,
      "step": 7
    },
    {
      "epoch": 0.004792571514153063,
      "grad_norm": 0.4294110834598541,
      "learning_rate": 0.00016,
      "loss": 0.9599,
      "step": 8
    },
    {
      "epoch": 0.005391642953422196,
      "grad_norm": 0.28537315130233765,
      "learning_rate": 0.00018,
      "loss": 0.9995,
      "step": 9
    },
    {
      "epoch": 0.005391642953422196,
      "eval_loss": 0.9368210434913635,
      "eval_runtime": 79.5623,
      "eval_samples_per_second": 8.836,
      "eval_steps_per_second": 4.424,
      "step": 9
    },
    {
      "epoch": 0.0059907143926913285,
      "grad_norm": 0.3777184784412384,
      "learning_rate": 0.0002,
      "loss": 0.9124,
      "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": 3295645846732800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}