checkpoint-14/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9739130434782609,
  "eval_steps": 4,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06956521739130435,
      "grad_norm": 0.24870926141738892,
      "learning_rate": 2e-05,
      "loss": 0.9615,
      "step": 1
    },
    {
      "epoch": 0.06956521739130435,
      "eval_loss": 0.9210608005523682,
      "eval_runtime": 74.7303,
      "eval_samples_per_second": 2.088,
      "eval_steps_per_second": 1.044,
      "step": 1
    },
    {
      "epoch": 0.1391304347826087,
      "grad_norm": 0.2108100801706314,
      "learning_rate": 4e-05,
      "loss": 0.7833,
      "step": 2
    },
    {
      "epoch": 0.20869565217391303,
      "grad_norm": 0.1993255764245987,
      "learning_rate": 6e-05,
      "loss": 0.7999,
      "step": 3
    },
    {
      "epoch": 0.2782608695652174,
      "grad_norm": 0.28567054867744446,
      "learning_rate": 8e-05,
      "loss": 0.8231,
      "step": 4
    },
    {
      "epoch": 0.2782608695652174,
      "eval_loss": 0.9139525294303894,
      "eval_runtime": 75.4634,
      "eval_samples_per_second": 2.067,
      "eval_steps_per_second": 1.034,
      "step": 4
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.2238488644361496,
      "learning_rate": 0.0001,
      "loss": 0.6855,
      "step": 5
    },
    {
      "epoch": 0.41739130434782606,
      "grad_norm": 0.248153418302536,
      "learning_rate": 0.00012,
      "loss": 0.9259,
      "step": 6
    },
    {
      "epoch": 0.48695652173913045,
      "grad_norm": 0.252511590719223,
      "learning_rate": 0.00014,
      "loss": 0.7668,
      "step": 7
    },
    {
      "epoch": 0.5565217391304348,
      "grad_norm": 0.25402677059173584,
      "learning_rate": 0.00016,
      "loss": 0.779,
      "step": 8
    },
    {
      "epoch": 0.5565217391304348,
      "eval_loss": 0.8034127354621887,
      "eval_runtime": 75.4244,
      "eval_samples_per_second": 2.068,
      "eval_steps_per_second": 1.034,
      "step": 8
    },
    {
      "epoch": 0.6260869565217392,
      "grad_norm": 0.1748448610305786,
      "learning_rate": 0.00018,
      "loss": 0.5345,
      "step": 9
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.18634748458862305,
      "learning_rate": 0.0002,
      "loss": 0.5699,
      "step": 10
    },
    {
      "epoch": 0.7652173913043478,
      "grad_norm": 0.21753615140914917,
      "learning_rate": 0.00019848077530122083,
      "loss": 0.5804,
      "step": 11
    },
    {
      "epoch": 0.8347826086956521,
      "grad_norm": 0.23333732783794403,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.545,
      "step": 12
    },
    {
      "epoch": 0.8347826086956521,
      "eval_loss": 0.648797869682312,
      "eval_runtime": 75.308,
      "eval_samples_per_second": 2.071,
      "eval_steps_per_second": 1.036,
      "step": 12
    },
    {
      "epoch": 0.9043478260869565,
      "grad_norm": 0.20313376188278198,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.4392,
      "step": 13
    },
    {
      "epoch": 0.9739130434782609,
      "grad_norm": 0.20026056468486786,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.38,
      "step": 14
    }
  ],
  "logging_steps": 1,
  "max_steps": 28,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 14,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5784968764522496e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}