last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2701789935832489,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002701789935832489,
      "eval_loss": 1.1311285495758057,
      "eval_runtime": 801.2397,
      "eval_samples_per_second": 7.78,
      "eval_steps_per_second": 0.973,
      "step": 1
    },
    {
      "epoch": 0.013508949679162444,
      "grad_norm": 1.9258880615234375,
      "learning_rate": 0.0001,
      "loss": 0.9863,
      "step": 50
    },
    {
      "epoch": 0.02701789935832489,
      "grad_norm": 2.3394365310668945,
      "learning_rate": 0.0002,
      "loss": 0.9093,
      "step": 100
    },
    {
      "epoch": 0.040526849037487336,
      "grad_norm": 1.3427051305770874,
      "learning_rate": 0.00019848077530122083,
      "loss": 0.8795,
      "step": 150
    },
    {
      "epoch": 0.05403579871664978,
      "grad_norm": 1.533150553703308,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.8673,
      "step": 200
    },
    {
      "epoch": 0.06754474839581223,
      "grad_norm": 1.6213335990905762,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.8694,
      "step": 250
    },
    {
      "epoch": 0.08105369807497467,
      "grad_norm": 1.8890631198883057,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.8616,
      "step": 300
    },
    {
      "epoch": 0.09456264775413711,
      "grad_norm": 1.590556263923645,
      "learning_rate": 0.00016427876096865394,
      "loss": 0.857,
      "step": 350
    },
    {
      "epoch": 0.10807159743329955,
      "grad_norm": 1.1496354341506958,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.8575,
      "step": 400
    },
    {
      "epoch": 0.12158054711246201,
      "grad_norm": 1.5908688306808472,
      "learning_rate": 0.00013420201433256689,
      "loss": 0.8696,
      "step": 450
    },
    {
      "epoch": 0.13508949679162446,
      "grad_norm": 1.1338423490524292,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.8455,
      "step": 500
    },
    {
      "epoch": 0.1485984464707869,
      "grad_norm": 1.2144519090652466,
      "learning_rate": 0.0001,
      "loss": 0.8511,
      "step": 550
    },
    {
      "epoch": 0.16210739614994935,
      "grad_norm": 1.2948704957962036,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.8333,
      "step": 600
    },
    {
      "epoch": 0.1756163458291118,
      "grad_norm": 1.0864368677139282,
      "learning_rate": 6.579798566743314e-05,
      "loss": 0.8449,
      "step": 650
    },
    {
      "epoch": 0.18912529550827423,
      "grad_norm": 1.156383991241455,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.838,
      "step": 700
    },
    {
      "epoch": 0.20263424518743667,
      "grad_norm": 1.3383092880249023,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.842,
      "step": 750
    },
    {
      "epoch": 0.2161431948665991,
      "grad_norm": 1.2148094177246094,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.8341,
      "step": 800
    },
    {
      "epoch": 0.22965214454576158,
      "grad_norm": 1.3319884538650513,
      "learning_rate": 1.339745962155613e-05,
      "loss": 0.8472,
      "step": 850
    },
    {
      "epoch": 0.24316109422492402,
      "grad_norm": 1.5228185653686523,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.8233,
      "step": 900
    },
    {
      "epoch": 0.25667004390408643,
      "grad_norm": 1.154761791229248,
      "learning_rate": 1.5192246987791981e-06,
      "loss": 0.8262,
      "step": 950
    },
    {
      "epoch": 0.2701789935832489,
      "grad_norm": 1.1170167922973633,
      "learning_rate": 0.0,
      "loss": 0.8293,
      "step": 1000
    },
    {
      "epoch": 0.2701789935832489,
      "eval_loss": 0.8305127024650574,
      "eval_runtime": 807.4769,
      "eval_samples_per_second": 7.72,
      "eval_steps_per_second": 0.966,
      "step": 1000
    }
  ],
  "logging_steps": 50,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5573330969388646e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}