last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00026621055924183233,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.6621055924183234e-05,
      "grad_norm": 0.19601939618587494,
      "learning_rate": 2e-05,
      "loss": 2.689,
      "step": 1
    },
    {
      "epoch": 2.6621055924183234e-05,
      "eval_loss": 2.87511944770813,
      "eval_runtime": 468.7111,
      "eval_samples_per_second": 33.746,
      "eval_steps_per_second": 16.874,
      "step": 1
    },
    {
      "epoch": 5.324211184836647e-05,
      "grad_norm": 0.2489328384399414,
      "learning_rate": 4e-05,
      "loss": 2.8836,
      "step": 2
    },
    {
      "epoch": 7.98631677725497e-05,
      "grad_norm": 0.22495155036449432,
      "learning_rate": 6e-05,
      "loss": 2.8706,
      "step": 3
    },
    {
      "epoch": 7.98631677725497e-05,
      "eval_loss": 2.8748691082000732,
      "eval_runtime": 467.3613,
      "eval_samples_per_second": 33.843,
      "eval_steps_per_second": 16.923,
      "step": 3
    },
    {
      "epoch": 0.00010648422369673294,
      "grad_norm": 0.2937033176422119,
      "learning_rate": 8e-05,
      "loss": 2.9725,
      "step": 4
    },
    {
      "epoch": 0.00013310527962091616,
      "grad_norm": 0.6798184514045715,
      "learning_rate": 0.0001,
      "loss": 2.8408,
      "step": 5
    },
    {
      "epoch": 0.0001597263355450994,
      "grad_norm": 0.24778014421463013,
      "learning_rate": 0.00012,
      "loss": 2.8675,
      "step": 6
    },
    {
      "epoch": 0.0001597263355450994,
      "eval_loss": 2.8702588081359863,
      "eval_runtime": 467.2955,
      "eval_samples_per_second": 33.848,
      "eval_steps_per_second": 16.925,
      "step": 6
    },
    {
      "epoch": 0.00018634739146928262,
      "grad_norm": 0.25821998715400696,
      "learning_rate": 0.00014,
      "loss": 2.8108,
      "step": 7
    },
    {
      "epoch": 0.00021296844739346587,
      "grad_norm": 0.27111247181892395,
      "learning_rate": 0.00016,
      "loss": 2.6447,
      "step": 8
    },
    {
      "epoch": 0.0002395895033176491,
      "grad_norm": 0.2788885533809662,
      "learning_rate": 0.00018,
      "loss": 2.7487,
      "step": 9
    },
    {
      "epoch": 0.0002395895033176491,
      "eval_loss": 2.8484585285186768,
      "eval_runtime": 467.5976,
      "eval_samples_per_second": 33.826,
      "eval_steps_per_second": 16.914,
      "step": 9
    },
    {
      "epoch": 0.00026621055924183233,
      "grad_norm": 0.2790322005748749,
      "learning_rate": 0.0002,
      "loss": 2.6885,
      "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": 1873333749547008.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}