last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1893491124260355,
  "eval_steps": 20,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001183431952662722,
      "eval_loss": 11.5,
      "eval_runtime": 3.3777,
      "eval_samples_per_second": 444.678,
      "eval_steps_per_second": 27.829,
      "step": 1
    },
    {
      "epoch": 0.011834319526627219,
      "grad_norm": 0.004608154296875,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 11.5,
      "step": 10
    },
    {
      "epoch": 0.023668639053254437,
      "grad_norm": 0.0023193359375,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 11.5,
      "step": 20
    },
    {
      "epoch": 0.023668639053254437,
      "eval_loss": 11.5,
      "eval_runtime": 3.3046,
      "eval_samples_per_second": 454.517,
      "eval_steps_per_second": 28.445,
      "step": 20
    },
    {
      "epoch": 0.03550295857988166,
      "grad_norm": 0.00604248046875,
      "learning_rate": 4.8e-05,
      "loss": 11.5,
      "step": 30
    },
    {
      "epoch": 0.047337278106508875,
      "grad_norm": 0.0032501220703125,
      "learning_rate": 6.400000000000001e-05,
      "loss": 11.5,
      "step": 40
    },
    {
      "epoch": 0.047337278106508875,
      "eval_loss": 11.5,
      "eval_runtime": 3.3267,
      "eval_samples_per_second": 451.492,
      "eval_steps_per_second": 28.256,
      "step": 40
    },
    {
      "epoch": 0.05917159763313609,
      "grad_norm": 0.0087890625,
      "learning_rate": 8e-05,
      "loss": 11.5,
      "step": 50
    },
    {
      "epoch": 0.07100591715976332,
      "grad_norm": 0.004241943359375,
      "learning_rate": 9.6e-05,
      "loss": 11.5,
      "step": 60
    },
    {
      "epoch": 0.07100591715976332,
      "eval_loss": 11.5,
      "eval_runtime": 3.3414,
      "eval_samples_per_second": 449.509,
      "eval_steps_per_second": 28.132,
      "step": 60
    },
    {
      "epoch": 0.08284023668639054,
      "grad_norm": 0.00238037109375,
      "learning_rate": 0.00011200000000000001,
      "loss": 11.5,
      "step": 70
    },
    {
      "epoch": 0.09467455621301775,
      "grad_norm": 0.006256103515625,
      "learning_rate": 0.00012800000000000002,
      "loss": 11.5,
      "step": 80
    },
    {
      "epoch": 0.09467455621301775,
      "eval_loss": 11.5,
      "eval_runtime": 3.3003,
      "eval_samples_per_second": 455.107,
      "eval_steps_per_second": 28.482,
      "step": 80
    },
    {
      "epoch": 0.10650887573964497,
      "grad_norm": 0.003936767578125,
      "learning_rate": 0.000144,
      "loss": 11.5,
      "step": 90
    },
    {
      "epoch": 0.11834319526627218,
      "grad_norm": 0.0123291015625,
      "learning_rate": 0.00016,
      "loss": 11.5,
      "step": 100
    },
    {
      "epoch": 0.11834319526627218,
      "eval_loss": 11.5,
      "eval_runtime": 3.3009,
      "eval_samples_per_second": 455.027,
      "eval_steps_per_second": 28.477,
      "step": 100
    },
    {
      "epoch": 0.1301775147928994,
      "grad_norm": 0.00634765625,
      "learning_rate": 0.00017600000000000002,
      "loss": 11.5,
      "step": 110
    },
    {
      "epoch": 0.14201183431952663,
      "grad_norm": 0.005706787109375,
      "learning_rate": 0.000192,
      "loss": 11.5,
      "step": 120
    },
    {
      "epoch": 0.14201183431952663,
      "eval_loss": 11.5,
      "eval_runtime": 3.2885,
      "eval_samples_per_second": 456.742,
      "eval_steps_per_second": 28.584,
      "step": 120
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.01129150390625,
      "learning_rate": 0.0001999978128380225,
      "loss": 11.5,
      "step": 130
    },
    {
      "epoch": 0.16568047337278108,
      "grad_norm": 0.0162353515625,
      "learning_rate": 0.0001999803161162393,
      "loss": 11.5,
      "step": 140
    },
    {
      "epoch": 0.16568047337278108,
      "eval_loss": 11.5,
      "eval_runtime": 3.3618,
      "eval_samples_per_second": 446.782,
      "eval_steps_per_second": 27.961,
      "step": 140
    },
    {
      "epoch": 0.17751479289940827,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.00019994532573409262,
      "loss": 11.5,
      "step": 150
    },
    {
      "epoch": 0.1893491124260355,
      "grad_norm": 0.030029296875,
      "learning_rate": 0.00019989284781388617,
      "loss": 11.5,
      "step": 160
    },
    {
      "epoch": 0.1893491124260355,
      "eval_loss": 11.5,
      "eval_runtime": 3.2697,
      "eval_samples_per_second": 459.367,
      "eval_steps_per_second": 28.749,
      "step": 160
    }
  ],
  "logging_steps": 10,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 40,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 25470836408320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}