last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0025143000817147527,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00025143000817147524,
      "grad_norm": 0.4191521406173706,
      "learning_rate": 2e-05,
      "loss": 1.001,
      "step": 1
    },
    {
      "epoch": 0.00025143000817147524,
      "eval_loss": 1.156248927116394,
      "eval_runtime": 75.8462,
      "eval_samples_per_second": 22.084,
      "eval_steps_per_second": 11.049,
      "step": 1
    },
    {
      "epoch": 0.0005028600163429505,
      "grad_norm": 0.3998629152774811,
      "learning_rate": 4e-05,
      "loss": 1.1657,
      "step": 2
    },
    {
      "epoch": 0.0007542900245144258,
      "grad_norm": 0.411793053150177,
      "learning_rate": 6e-05,
      "loss": 0.8879,
      "step": 3
    },
    {
      "epoch": 0.0007542900245144258,
      "eval_loss": 1.1521437168121338,
      "eval_runtime": 73.8309,
      "eval_samples_per_second": 22.687,
      "eval_steps_per_second": 11.35,
      "step": 3
    },
    {
      "epoch": 0.001005720032685901,
      "grad_norm": 0.2495160698890686,
      "learning_rate": 8e-05,
      "loss": 0.6845,
      "step": 4
    },
    {
      "epoch": 0.0012571500408573764,
      "grad_norm": 0.6579082012176514,
      "learning_rate": 0.0001,
      "loss": 1.2625,
      "step": 5
    },
    {
      "epoch": 0.0015085800490288515,
      "grad_norm": 0.3734517991542816,
      "learning_rate": 0.00012,
      "loss": 1.0214,
      "step": 6
    },
    {
      "epoch": 0.0015085800490288515,
      "eval_loss": 1.100870132446289,
      "eval_runtime": 73.8883,
      "eval_samples_per_second": 22.669,
      "eval_steps_per_second": 11.341,
      "step": 6
    },
    {
      "epoch": 0.001760010057200327,
      "grad_norm": 0.33622726798057556,
      "learning_rate": 0.00014,
      "loss": 0.8701,
      "step": 7
    },
    {
      "epoch": 0.002011440065371802,
      "grad_norm": 0.42788028717041016,
      "learning_rate": 0.00016,
      "loss": 1.1302,
      "step": 8
    },
    {
      "epoch": 0.0022628700735432775,
      "grad_norm": 0.3505564033985138,
      "learning_rate": 0.00018,
      "loss": 1.0666,
      "step": 9
    },
    {
      "epoch": 0.0022628700735432775,
      "eval_loss": 1.0334452390670776,
      "eval_runtime": 73.8456,
      "eval_samples_per_second": 22.682,
      "eval_steps_per_second": 11.348,
      "step": 9
    },
    {
      "epoch": 0.0025143000817147527,
      "grad_norm": 0.6299397945404053,
      "learning_rate": 0.0002,
      "loss": 0.8493,
      "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": 2866824584626176.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}