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": 0.8063191175460815,
      "learning_rate": 7.920000000000001e-05,
      "loss": 8.2537,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 0.6161019206047058,
      "learning_rate": 0.00015920000000000002,
      "loss": 6.5731,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 0.8347169160842896,
      "learning_rate": 0.00023920000000000001,
      "loss": 6.1639,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.9150793552398682,
      "learning_rate": 0.0003192,
      "loss": 5.8682,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6698479652404785,
      "learning_rate": 0.0003992,
      "loss": 5.5708,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 0.5323518514633179,
      "learning_rate": 0.00047920000000000005,
      "loss": 5.3157,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 0.7472386956214905,
      "learning_rate": 0.0005592,
      "loss": 5.0907,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.6033297181129456,
      "learning_rate": 0.0006392,
      "loss": 4.8805,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 0.497113436460495,
      "learning_rate": 0.0007191999999999999,
      "loss": 4.7207,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.47698545455932617,
      "learning_rate": 0.0007992,
      "loss": 4.6068,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.2703913894324853,
      "eval_loss": 4.5053935050964355,
      "eval_runtime": 13.0992,
      "eval_samples_per_second": 76.341,
      "eval_steps_per_second": 0.611,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 0.44341349601745605,
      "learning_rate": 0.0008792,
      "loss": 4.4904,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.4383724331855774,
      "learning_rate": 0.0009592000000000001,
      "loss": 4.439,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.3894674777984619,
      "learning_rate": 0.0010391999999999999,
      "loss": 4.3735,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.41903820633888245,
      "learning_rate": 0.0011192,
      "loss": 4.3406,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.36601418256759644,
      "learning_rate": 0.0011992,
      "loss": 4.2879,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.4282030165195465,
      "learning_rate": 0.0012791999999999999,
      "loss": 4.2625,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.3431214690208435,
      "learning_rate": 0.0013592,
      "loss": 4.2316,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.3303460478782654,
      "learning_rate": 0.0014392,
      "loss": 4.1976,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.3030431866645813,
      "learning_rate": 0.0015192,
      "loss": 4.1494,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3076256513595581,
      "learning_rate": 0.0015992,
      "loss": 4.1688,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.30658512720156555,
      "eval_loss": 4.094621658325195,
      "eval_runtime": 12.4723,
      "eval_samples_per_second": 80.178,
      "eval_steps_per_second": 0.641,
      "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
}