last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.053763440860215055,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035842293906810036,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 1.6449,
      "step": 1
    },
    {
      "epoch": 0.0035842293906810036,
      "eval_loss": 1.371159315109253,
      "eval_runtime": 8.3908,
      "eval_samples_per_second": 56.014,
      "eval_steps_per_second": 7.032,
      "step": 1
    },
    {
      "epoch": 0.007168458781362007,
      "grad_norm": 1.1762983798980713,
      "learning_rate": 2e-05,
      "loss": 1.2685,
      "step": 2
    },
    {
      "epoch": 0.010752688172043012,
      "grad_norm": 0.999072253704071,
      "learning_rate": 4e-05,
      "loss": 1.1567,
      "step": 3
    },
    {
      "epoch": 0.014336917562724014,
      "grad_norm": 1.0890846252441406,
      "learning_rate": 6e-05,
      "loss": 1.2905,
      "step": 4
    },
    {
      "epoch": 0.017921146953405017,
      "grad_norm": 1.1394107341766357,
      "learning_rate": 8e-05,
      "loss": 1.2343,
      "step": 5
    },
    {
      "epoch": 0.017921146953405017,
      "eval_loss": 1.3072917461395264,
      "eval_runtime": 8.5738,
      "eval_samples_per_second": 54.818,
      "eval_steps_per_second": 6.881,
      "step": 5
    },
    {
      "epoch": 0.021505376344086023,
      "grad_norm": 1.2360237836837769,
      "learning_rate": 0.0001,
      "loss": 1.4268,
      "step": 6
    },
    {
      "epoch": 0.025089605734767026,
      "grad_norm": 0.9018695950508118,
      "learning_rate": 0.00012,
      "loss": 1.0908,
      "step": 7
    },
    {
      "epoch": 0.02867383512544803,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 1.4248,
      "step": 8
    },
    {
      "epoch": 0.03225806451612903,
      "grad_norm": 0.8699414134025574,
      "learning_rate": 0.00014,
      "loss": 1.1859,
      "step": 9
    },
    {
      "epoch": 0.035842293906810034,
      "grad_norm": 0.7875003218650818,
      "learning_rate": 0.00016,
      "loss": 1.1898,
      "step": 10
    },
    {
      "epoch": 0.035842293906810034,
      "eval_loss": 1.1698945760726929,
      "eval_runtime": 8.6367,
      "eval_samples_per_second": 54.419,
      "eval_steps_per_second": 6.831,
      "step": 10
    },
    {
      "epoch": 0.03942652329749104,
      "grad_norm": 0.9833978414535522,
      "learning_rate": 0.00018,
      "loss": 1.1755,
      "step": 11
    },
    {
      "epoch": 0.043010752688172046,
      "grad_norm": 0.890007734298706,
      "learning_rate": 0.0002,
      "loss": 1.0835,
      "step": 12
    },
    {
      "epoch": 0.04659498207885305,
      "grad_norm": 0.9671109914779663,
      "learning_rate": 0.00019510565162951537,
      "loss": 1.1763,
      "step": 13
    },
    {
      "epoch": 0.05017921146953405,
      "grad_norm": 0.7680312991142273,
      "learning_rate": 0.00018090169943749476,
      "loss": 1.007,
      "step": 14
    },
    {
      "epoch": 0.053763440860215055,
      "grad_norm": 0.7991800308227539,
      "learning_rate": 0.00015877852522924732,
      "loss": 1.0003,
      "step": 15
    },
    {
      "epoch": 0.053763440860215055,
      "eval_loss": 1.107828140258789,
      "eval_runtime": 8.6971,
      "eval_samples_per_second": 54.041,
      "eval_steps_per_second": 6.784,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.250124588122112e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}