checkpoint-25/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0044835007173601145,
  "eval_steps": 5,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000896700143472023,
      "grad_norm": 1.124503254890442,
      "learning_rate": 0.00018,
      "loss": 2.3789,
      "step": 5
    },
    {
      "epoch": 0.000896700143472023,
      "eval_loss": 1.9937291145324707,
      "eval_runtime": 65.4243,
      "eval_samples_per_second": 2.736,
      "eval_steps_per_second": 1.376,
      "step": 5
    },
    {
      "epoch": 0.001793400286944046,
      "grad_norm": 0.8716641664505005,
      "learning_rate": 0.00016,
      "loss": 1.7336,
      "step": 10
    },
    {
      "epoch": 0.001793400286944046,
      "eval_loss": 1.590023398399353,
      "eval_runtime": 63.9353,
      "eval_samples_per_second": 2.8,
      "eval_steps_per_second": 1.408,
      "step": 10
    },
    {
      "epoch": 0.002690100430416069,
      "grad_norm": 0.5799658298492432,
      "learning_rate": 0.00014,
      "loss": 1.6249,
      "step": 15
    },
    {
      "epoch": 0.002690100430416069,
      "eval_loss": 1.4548176527023315,
      "eval_runtime": 63.7041,
      "eval_samples_per_second": 2.81,
      "eval_steps_per_second": 1.413,
      "step": 15
    },
    {
      "epoch": 0.003586800573888092,
      "grad_norm": 0.6462221145629883,
      "learning_rate": 0.00012,
      "loss": 1.3909,
      "step": 20
    },
    {
      "epoch": 0.003586800573888092,
      "eval_loss": 1.3800599575042725,
      "eval_runtime": 64.2415,
      "eval_samples_per_second": 2.786,
      "eval_steps_per_second": 1.401,
      "step": 20
    },
    {
      "epoch": 0.0044835007173601145,
      "grad_norm": 0.5445486903190613,
      "learning_rate": 0.0001,
      "loss": 1.3388,
      "step": 25
    },
    {
      "epoch": 0.0044835007173601145,
      "eval_loss": 1.3405883312225342,
      "eval_runtime": 63.9014,
      "eval_samples_per_second": 2.801,
      "eval_steps_per_second": 1.408,
      "step": 25
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3358078906501440.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}