last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08316008316008316,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008316008316008316,
      "grad_norm": 1.1778547763824463,
      "learning_rate": 2e-05,
      "loss": 3.7524,
      "step": 1
    },
    {
      "epoch": 0.008316008316008316,
      "eval_loss": 0.8911983370780945,
      "eval_runtime": 1.0732,
      "eval_samples_per_second": 47.522,
      "eval_steps_per_second": 24.227,
      "step": 1
    },
    {
      "epoch": 0.016632016632016633,
      "grad_norm": 1.3017749786376953,
      "learning_rate": 4e-05,
      "loss": 4.1642,
      "step": 2
    },
    {
      "epoch": 0.02494802494802495,
      "grad_norm": 1.22097909450531,
      "learning_rate": 6e-05,
      "loss": 3.553,
      "step": 3
    },
    {
      "epoch": 0.02494802494802495,
      "eval_loss": 0.8898816108703613,
      "eval_runtime": 1.0801,
      "eval_samples_per_second": 47.218,
      "eval_steps_per_second": 24.072,
      "step": 3
    },
    {
      "epoch": 0.033264033264033266,
      "grad_norm": 1.1316472291946411,
      "learning_rate": 8e-05,
      "loss": 3.3653,
      "step": 4
    },
    {
      "epoch": 0.04158004158004158,
      "grad_norm": 1.2046618461608887,
      "learning_rate": 0.0001,
      "loss": 3.7439,
      "step": 5
    },
    {
      "epoch": 0.0498960498960499,
      "grad_norm": 1.393701434135437,
      "learning_rate": 0.00012,
      "loss": 3.8004,
      "step": 6
    },
    {
      "epoch": 0.0498960498960499,
      "eval_loss": 0.8759763836860657,
      "eval_runtime": 1.0971,
      "eval_samples_per_second": 46.487,
      "eval_steps_per_second": 23.699,
      "step": 6
    },
    {
      "epoch": 0.058212058212058215,
      "grad_norm": 1.4396038055419922,
      "learning_rate": 0.00014,
      "loss": 3.4585,
      "step": 7
    },
    {
      "epoch": 0.06652806652806653,
      "grad_norm": 2.0053093433380127,
      "learning_rate": 0.00016,
      "loss": 4.133,
      "step": 8
    },
    {
      "epoch": 0.07484407484407485,
      "grad_norm": 1.6646336317062378,
      "learning_rate": 0.00018,
      "loss": 3.3301,
      "step": 9
    },
    {
      "epoch": 0.07484407484407485,
      "eval_loss": 0.789121150970459,
      "eval_runtime": 1.0966,
      "eval_samples_per_second": 46.507,
      "eval_steps_per_second": 23.709,
      "step": 9
    },
    {
      "epoch": 0.08316008316008316,
      "grad_norm": 1.7604037523269653,
      "learning_rate": 0.0002,
      "loss": 3.4752,
      "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": 358475284611072.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}