last-checkpoint/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04,
  "eval_steps": 1000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 1.158125638961792,
      "learning_rate": 7.920000000000001e-05,
      "loss": 8.2536,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 0.6474116444587708,
      "learning_rate": 0.00015920000000000002,
      "loss": 6.577,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 0.8440109491348267,
      "learning_rate": 0.00023920000000000001,
      "loss": 6.1636,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.712805986404419,
      "learning_rate": 0.0003192,
      "loss": 5.8686,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6756730675697327,
      "learning_rate": 0.0003992,
      "loss": 5.5639,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 0.8336403965950012,
      "learning_rate": 0.00047920000000000005,
      "loss": 5.3098,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 0.6417033672332764,
      "learning_rate": 0.0005592,
      "loss": 5.0845,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.5867908596992493,
      "learning_rate": 0.0006392,
      "loss": 4.8767,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 0.5321984887123108,
      "learning_rate": 0.0007191999999999999,
      "loss": 4.7211,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.5662059187889099,
      "learning_rate": 0.0007992,
      "loss": 4.6059,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.2709784735812133,
      "eval_loss": 4.504483222961426,
      "eval_runtime": 11.2003,
      "eval_samples_per_second": 89.283,
      "eval_steps_per_second": 0.714,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 0.45525625348091125,
      "learning_rate": 0.0008792,
      "loss": 4.4901,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.4701859652996063,
      "learning_rate": 0.0009592000000000001,
      "loss": 4.4344,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.40884163975715637,
      "learning_rate": 0.0010391999999999999,
      "loss": 4.3707,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.3846769332885742,
      "learning_rate": 0.0011192,
      "loss": 4.336,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3585458993911743,
      "learning_rate": 0.0011992,
      "loss": 4.2832,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.4558027982711792,
      "learning_rate": 0.0012791999999999999,
      "loss": 4.2599,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.33011195063591003,
      "learning_rate": 0.0013592,
      "loss": 4.2287,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.32163774967193604,
      "learning_rate": 0.0014392,
      "loss": 4.1932,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.30342814326286316,
      "learning_rate": 0.0015192,
      "loss": 4.1447,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.32037997245788574,
      "learning_rate": 0.0015992,
      "loss": 4.149,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.30734833659491195,
      "eval_loss": 4.087185859680176,
      "eval_runtime": 11.8275,
      "eval_samples_per_second": 84.549,
      "eval_steps_per_second": 0.676,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}