last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.015594541910331383,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0015594541910331384,
      "grad_norm": 0.3541686236858368,
      "learning_rate": 2e-05,
      "loss": 1.5323,
      "step": 1
    },
    {
      "epoch": 0.0015594541910331384,
      "eval_loss": 1.7419472932815552,
      "eval_runtime": 15.9635,
      "eval_samples_per_second": 16.976,
      "eval_steps_per_second": 8.519,
      "step": 1
    },
    {
      "epoch": 0.0031189083820662767,
      "grad_norm": 0.5283167958259583,
      "learning_rate": 4e-05,
      "loss": 1.8561,
      "step": 2
    },
    {
      "epoch": 0.004678362573099415,
      "grad_norm": 0.4033576250076294,
      "learning_rate": 6e-05,
      "loss": 1.7909,
      "step": 3
    },
    {
      "epoch": 0.004678362573099415,
      "eval_loss": 1.739487886428833,
      "eval_runtime": 13.9046,
      "eval_samples_per_second": 19.49,
      "eval_steps_per_second": 9.781,
      "step": 3
    },
    {
      "epoch": 0.006237816764132553,
      "grad_norm": 0.31834664940834045,
      "learning_rate": 8e-05,
      "loss": 1.4852,
      "step": 4
    },
    {
      "epoch": 0.007797270955165692,
      "grad_norm": 0.3573683500289917,
      "learning_rate": 0.0001,
      "loss": 1.3656,
      "step": 5
    },
    {
      "epoch": 0.00935672514619883,
      "grad_norm": 0.3832826614379883,
      "learning_rate": 0.00012,
      "loss": 1.583,
      "step": 6
    },
    {
      "epoch": 0.00935672514619883,
      "eval_loss": 1.7191375494003296,
      "eval_runtime": 13.9435,
      "eval_samples_per_second": 19.436,
      "eval_steps_per_second": 9.754,
      "step": 6
    },
    {
      "epoch": 0.01091617933723197,
      "grad_norm": 0.3834376037120819,
      "learning_rate": 0.00014,
      "loss": 1.2282,
      "step": 7
    },
    {
      "epoch": 0.012475633528265107,
      "grad_norm": 0.7654609680175781,
      "learning_rate": 0.00016,
      "loss": 2.039,
      "step": 8
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 0.6056851148605347,
      "learning_rate": 0.00018,
      "loss": 1.8312,
      "step": 9
    },
    {
      "epoch": 0.014035087719298246,
      "eval_loss": 1.6474940776824951,
      "eval_runtime": 13.9683,
      "eval_samples_per_second": 19.401,
      "eval_steps_per_second": 9.736,
      "step": 9
    },
    {
      "epoch": 0.015594541910331383,
      "grad_norm": 0.6715533137321472,
      "learning_rate": 0.0002,
      "loss": 1.6847,
      "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": 3176029615030272.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}