trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9883268482490273,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.311284046692607,
      "grad_norm": 1.0218406598109495,
      "learning_rate": 5e-06,
      "loss": 0.5263,
      "step": 10
    },
    {
      "epoch": 0.622568093385214,
      "grad_norm": 0.521829995531278,
      "learning_rate": 5e-06,
      "loss": 0.4298,
      "step": 20
    },
    {
      "epoch": 0.933852140077821,
      "grad_norm": 0.7569470799217258,
      "learning_rate": 5e-06,
      "loss": 0.3929,
      "step": 30
    },
    {
      "epoch": 0.9961089494163424,
      "eval_loss": 0.3781737685203552,
      "eval_runtime": 35.0934,
      "eval_samples_per_second": 24.62,
      "eval_steps_per_second": 0.399,
      "step": 32
    },
    {
      "epoch": 1.245136186770428,
      "grad_norm": 0.5574261830886987,
      "learning_rate": 5e-06,
      "loss": 0.3965,
      "step": 40
    },
    {
      "epoch": 1.556420233463035,
      "grad_norm": 0.5419153412341505,
      "learning_rate": 5e-06,
      "loss": 0.3502,
      "step": 50
    },
    {
      "epoch": 1.867704280155642,
      "grad_norm": 0.5071134368069596,
      "learning_rate": 5e-06,
      "loss": 0.3474,
      "step": 60
    },
    {
      "epoch": 1.9922178988326849,
      "eval_loss": 0.3496624529361725,
      "eval_runtime": 35.195,
      "eval_samples_per_second": 24.549,
      "eval_steps_per_second": 0.398,
      "step": 64
    },
    {
      "epoch": 2.178988326848249,
      "grad_norm": 0.5601606997936093,
      "learning_rate": 5e-06,
      "loss": 0.3615,
      "step": 70
    },
    {
      "epoch": 2.490272373540856,
      "grad_norm": 0.6943931265708301,
      "learning_rate": 5e-06,
      "loss": 0.3185,
      "step": 80
    },
    {
      "epoch": 2.801556420233463,
      "grad_norm": 0.5181184317448785,
      "learning_rate": 5e-06,
      "loss": 0.3178,
      "step": 90
    },
    {
      "epoch": 2.9883268482490273,
      "eval_loss": 0.3399236798286438,
      "eval_runtime": 33.9585,
      "eval_samples_per_second": 25.443,
      "eval_steps_per_second": 0.412,
      "step": 96
    },
    {
      "epoch": 2.9883268482490273,
      "step": 96,
      "total_flos": 160594195906560.0,
      "train_loss": 0.37829776108264923,
      "train_runtime": 5916.3383,
      "train_samples_per_second": 8.317,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 96,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 160594195906560.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}