last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0027795149746369257,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002779514974636926,
      "grad_norm": 2.0346224308013916,
      "learning_rate": 2e-05,
      "loss": 4.2698,
      "step": 1
    },
    {
      "epoch": 0.0002779514974636926,
      "eval_loss": 4.093295097351074,
      "eval_runtime": 48.0961,
      "eval_samples_per_second": 31.499,
      "eval_steps_per_second": 15.76,
      "step": 1
    },
    {
      "epoch": 0.0005559029949273852,
      "grad_norm": 1.7794543504714966,
      "learning_rate": 4e-05,
      "loss": 4.1508,
      "step": 2
    },
    {
      "epoch": 0.0008338544923910778,
      "grad_norm": 2.4146764278411865,
      "learning_rate": 6e-05,
      "loss": 3.6022,
      "step": 3
    },
    {
      "epoch": 0.0008338544923910778,
      "eval_loss": 4.080883026123047,
      "eval_runtime": 47.0797,
      "eval_samples_per_second": 32.18,
      "eval_steps_per_second": 16.1,
      "step": 3
    },
    {
      "epoch": 0.0011118059898547704,
      "grad_norm": 1.9711511135101318,
      "learning_rate": 8e-05,
      "loss": 3.4461,
      "step": 4
    },
    {
      "epoch": 0.0013897574873184629,
      "grad_norm": 2.1288695335388184,
      "learning_rate": 0.0001,
      "loss": 4.0495,
      "step": 5
    },
    {
      "epoch": 0.0016677089847821555,
      "grad_norm": 2.4953269958496094,
      "learning_rate": 0.00012,
      "loss": 3.8412,
      "step": 6
    },
    {
      "epoch": 0.0016677089847821555,
      "eval_loss": 3.8704497814178467,
      "eval_runtime": 47.1276,
      "eval_samples_per_second": 32.147,
      "eval_steps_per_second": 16.084,
      "step": 6
    },
    {
      "epoch": 0.0019456604822458482,
      "grad_norm": 2.96105694770813,
      "learning_rate": 0.00014,
      "loss": 3.194,
      "step": 7
    },
    {
      "epoch": 0.002223611979709541,
      "grad_norm": 3.311476707458496,
      "learning_rate": 0.00016,
      "loss": 3.4736,
      "step": 8
    },
    {
      "epoch": 0.0025015634771732333,
      "grad_norm": 6.470930099487305,
      "learning_rate": 0.00018,
      "loss": 3.5299,
      "step": 9
    },
    {
      "epoch": 0.0025015634771732333,
      "eval_loss": 3.3130292892456055,
      "eval_runtime": 47.1102,
      "eval_samples_per_second": 32.159,
      "eval_steps_per_second": 16.09,
      "step": 9
    },
    {
      "epoch": 0.0027795149746369257,
      "grad_norm": 2.392509698867798,
      "learning_rate": 0.0002,
      "loss": 3.3454,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}