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": 1.2400550842285156,
      "learning_rate": 7.920000000000001e-05,
      "loss": 8.362694702148438,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 2.1569149494171143,
      "learning_rate": 0.00015920000000000002,
      "loss": 6.592593383789063,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 4.070979595184326,
      "learning_rate": 0.00023920000000000001,
      "loss": 6.208985595703125,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 4.392622470855713,
      "learning_rate": 0.0003192,
      "loss": 5.921519775390625,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.653286457061768,
      "learning_rate": 0.0003992,
      "loss": 5.658856201171875,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 2.6610119342803955,
      "learning_rate": 0.00047920000000000005,
      "loss": 5.450134887695312,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 2.6298203468322754,
      "learning_rate": 0.0005592,
      "loss": 5.260469360351562,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 1.681970238685608,
      "learning_rate": 0.0006392,
      "loss": 5.112535095214843,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 1.5673184394836426,
      "learning_rate": 0.0007191999999999999,
      "loss": 5.0076907348632815,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.6654455661773682,
      "learning_rate": 0.0007992,
      "loss": 4.9159228515625,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.2274637964774951,
      "eval_loss": 4.8702616691589355,
      "eval_runtime": 5.1375,
      "eval_samples_per_second": 194.646,
      "eval_steps_per_second": 1.557,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 1.3562606573104858,
      "learning_rate": 0.0008792,
      "loss": 4.811459045410157,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 1.1230770349502563,
      "learning_rate": 0.0009592000000000001,
      "loss": 4.759424438476563,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.9026045203208923,
      "learning_rate": 0.0010391999999999999,
      "loss": 4.691336669921875,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.8914104104042053,
      "learning_rate": 0.0011192,
      "loss": 4.6415283203125,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8752658367156982,
      "learning_rate": 0.0011992,
      "loss": 4.560531311035156,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.7435582280158997,
      "learning_rate": 0.0012791999999999999,
      "loss": 4.516304016113281,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.7041499018669128,
      "learning_rate": 0.0013592,
      "loss": 4.48115234375,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.5327283143997192,
      "learning_rate": 0.0014392,
      "loss": 4.429443969726562,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.45207348465919495,
      "learning_rate": 0.0015192,
      "loss": 4.37362060546875,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.5152547359466553,
      "learning_rate": 0.0015992,
      "loss": 4.377236938476562,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.2694227005870842,
      "eval_loss": 4.3582682609558105,
      "eval_runtime": 5.3878,
      "eval_samples_per_second": 185.606,
      "eval_steps_per_second": 1.485,
      "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
}