last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005623031938821412,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.623031938821413e-05,
      "grad_norm": 0.9738835692405701,
      "learning_rate": 2e-05,
      "loss": 1.3355,
      "step": 1
    },
    {
      "epoch": 5.623031938821413e-05,
      "eval_loss": 1.3673052787780762,
      "eval_runtime": 382.7942,
      "eval_samples_per_second": 19.561,
      "eval_steps_per_second": 9.781,
      "step": 1
    },
    {
      "epoch": 0.00011246063877642825,
      "grad_norm": 1.0294928550720215,
      "learning_rate": 4e-05,
      "loss": 1.5085,
      "step": 2
    },
    {
      "epoch": 0.00016869095816464237,
      "grad_norm": 0.8840482831001282,
      "learning_rate": 6e-05,
      "loss": 1.1577,
      "step": 3
    },
    {
      "epoch": 0.00016869095816464237,
      "eval_loss": 1.3508278131484985,
      "eval_runtime": 382.9201,
      "eval_samples_per_second": 19.555,
      "eval_steps_per_second": 9.777,
      "step": 3
    },
    {
      "epoch": 0.0002249212775528565,
      "grad_norm": 0.9637212753295898,
      "learning_rate": 8e-05,
      "loss": 1.2729,
      "step": 4
    },
    {
      "epoch": 0.0002811515969410706,
      "grad_norm": 1.0110697746276855,
      "learning_rate": 0.0001,
      "loss": 1.3614,
      "step": 5
    },
    {
      "epoch": 0.00033738191632928474,
      "grad_norm": 1.03305983543396,
      "learning_rate": 0.00012,
      "loss": 1.3385,
      "step": 6
    },
    {
      "epoch": 0.00033738191632928474,
      "eval_loss": 1.1737115383148193,
      "eval_runtime": 383.0311,
      "eval_samples_per_second": 19.549,
      "eval_steps_per_second": 9.775,
      "step": 6
    },
    {
      "epoch": 0.00039361223571749885,
      "grad_norm": 0.8017333745956421,
      "learning_rate": 0.00014,
      "loss": 1.1112,
      "step": 7
    },
    {
      "epoch": 0.000449842555105713,
      "grad_norm": 0.7049347162246704,
      "learning_rate": 0.00016,
      "loss": 1.0699,
      "step": 8
    },
    {
      "epoch": 0.0005060728744939271,
      "grad_norm": 0.6511578559875488,
      "learning_rate": 0.00018,
      "loss": 1.0103,
      "step": 9
    },
    {
      "epoch": 0.0005060728744939271,
      "eval_loss": 1.0450962781906128,
      "eval_runtime": 382.8295,
      "eval_samples_per_second": 19.56,
      "eval_steps_per_second": 9.78,
      "step": 9
    },
    {
      "epoch": 0.0005623031938821412,
      "grad_norm": 0.7811943888664246,
      "learning_rate": 0.0002,
      "loss": 0.9902,
      "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": 2103703457759232.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}