last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00039440827665768565,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.9440827665768565e-05,
      "grad_norm": 2.0876424312591553,
      "learning_rate": 2e-05,
      "loss": 4.1328,
      "step": 1
    },
    {
      "epoch": 3.9440827665768565e-05,
      "eval_loss": 4.354150295257568,
      "eval_runtime": 6938.5773,
      "eval_samples_per_second": 3.077,
      "eval_steps_per_second": 3.077,
      "step": 1
    },
    {
      "epoch": 7.888165533153713e-05,
      "grad_norm": 1.7746790647506714,
      "learning_rate": 4e-05,
      "loss": 3.6041,
      "step": 2
    },
    {
      "epoch": 0.0001183224829973057,
      "grad_norm": 1.5225496292114258,
      "learning_rate": 6e-05,
      "loss": 3.6342,
      "step": 3
    },
    {
      "epoch": 0.0001183224829973057,
      "eval_loss": 4.327748775482178,
      "eval_runtime": 5818.9698,
      "eval_samples_per_second": 3.669,
      "eval_steps_per_second": 3.669,
      "step": 3
    },
    {
      "epoch": 0.00015776331066307426,
      "grad_norm": 2.0381834506988525,
      "learning_rate": 8e-05,
      "loss": 4.0984,
      "step": 4
    },
    {
      "epoch": 0.00019720413832884282,
      "grad_norm": 1.3214229345321655,
      "learning_rate": 0.0001,
      "loss": 3.4137,
      "step": 5
    },
    {
      "epoch": 0.0002366449659946114,
      "grad_norm": 2.0282857418060303,
      "learning_rate": 0.00012,
      "loss": 3.4421,
      "step": 6
    },
    {
      "epoch": 0.0002366449659946114,
      "eval_loss": 3.8443140983581543,
      "eval_runtime": 5820.5712,
      "eval_samples_per_second": 3.668,
      "eval_steps_per_second": 3.668,
      "step": 6
    },
    {
      "epoch": 0.00027608579366037995,
      "grad_norm": 2.0494728088378906,
      "learning_rate": 0.00014,
      "loss": 3.8704,
      "step": 7
    },
    {
      "epoch": 0.0003155266213261485,
      "grad_norm": 2.3501408100128174,
      "learning_rate": 0.00016,
      "loss": 3.7304,
      "step": 8
    },
    {
      "epoch": 0.0003549674489919171,
      "grad_norm": 2.1949191093444824,
      "learning_rate": 0.00018,
      "loss": 3.1309,
      "step": 9
    },
    {
      "epoch": 0.0003549674489919171,
      "eval_loss": 3.0262250900268555,
      "eval_runtime": 5815.7762,
      "eval_samples_per_second": 3.671,
      "eval_steps_per_second": 3.671,
      "step": 9
    },
    {
      "epoch": 0.00039440827665768565,
      "grad_norm": 1.7487046718597412,
      "learning_rate": 0.0002,
      "loss": 3.0529,
      "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": 2.796154773307392e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}