last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0423728813559322,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00423728813559322,
      "grad_norm": 0.2546420097351074,
      "learning_rate": 2e-05,
      "loss": 1.7026,
      "step": 1
    },
    {
      "epoch": 0.00423728813559322,
      "eval_loss": 1.7572693824768066,
      "eval_runtime": 13.9531,
      "eval_samples_per_second": 7.167,
      "eval_steps_per_second": 3.583,
      "step": 1
    },
    {
      "epoch": 0.00847457627118644,
      "grad_norm": 0.48551133275032043,
      "learning_rate": 4e-05,
      "loss": 1.7611,
      "step": 2
    },
    {
      "epoch": 0.012711864406779662,
      "grad_norm": 0.3137155771255493,
      "learning_rate": 6e-05,
      "loss": 1.7868,
      "step": 3
    },
    {
      "epoch": 0.012711864406779662,
      "eval_loss": 1.7567241191864014,
      "eval_runtime": 12.5769,
      "eval_samples_per_second": 7.951,
      "eval_steps_per_second": 3.976,
      "step": 3
    },
    {
      "epoch": 0.01694915254237288,
      "grad_norm": 0.4075331687927246,
      "learning_rate": 8e-05,
      "loss": 1.7026,
      "step": 4
    },
    {
      "epoch": 0.0211864406779661,
      "grad_norm": 0.29102247953414917,
      "learning_rate": 0.0001,
      "loss": 1.8396,
      "step": 5
    },
    {
      "epoch": 0.025423728813559324,
      "grad_norm": 0.7371453046798706,
      "learning_rate": 0.00012,
      "loss": 1.7147,
      "step": 6
    },
    {
      "epoch": 0.025423728813559324,
      "eval_loss": 1.7477717399597168,
      "eval_runtime": 12.5608,
      "eval_samples_per_second": 7.961,
      "eval_steps_per_second": 3.981,
      "step": 6
    },
    {
      "epoch": 0.029661016949152543,
      "grad_norm": 0.37393781542778015,
      "learning_rate": 0.00014,
      "loss": 2.0564,
      "step": 7
    },
    {
      "epoch": 0.03389830508474576,
      "grad_norm": 0.42129209637641907,
      "learning_rate": 0.00016,
      "loss": 1.9499,
      "step": 8
    },
    {
      "epoch": 0.038135593220338986,
      "grad_norm": 0.3632056713104248,
      "learning_rate": 0.00018,
      "loss": 1.9532,
      "step": 9
    },
    {
      "epoch": 0.038135593220338986,
      "eval_loss": 1.7135396003723145,
      "eval_runtime": 12.5871,
      "eval_samples_per_second": 7.945,
      "eval_steps_per_second": 3.972,
      "step": 9
    },
    {
      "epoch": 0.0423728813559322,
      "grad_norm": 0.5667614936828613,
      "learning_rate": 0.0002,
      "loss": 1.6446,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2348147665797120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}