last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03898635477582846,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003898635477582846,
      "grad_norm": 0.2849408686161041,
      "learning_rate": 2e-05,
      "loss": 0.4839,
      "step": 1
    },
    {
      "epoch": 0.003898635477582846,
      "eval_loss": 0.5632493495941162,
      "eval_runtime": 6.9097,
      "eval_samples_per_second": 15.63,
      "eval_steps_per_second": 7.815,
      "step": 1
    },
    {
      "epoch": 0.007797270955165692,
      "grad_norm": 0.2545192837715149,
      "learning_rate": 4e-05,
      "loss": 0.5144,
      "step": 2
    },
    {
      "epoch": 0.011695906432748537,
      "grad_norm": 0.2741990387439728,
      "learning_rate": 6e-05,
      "loss": 0.5664,
      "step": 3
    },
    {
      "epoch": 0.011695906432748537,
      "eval_loss": 0.5611659288406372,
      "eval_runtime": 5.7118,
      "eval_samples_per_second": 18.908,
      "eval_steps_per_second": 9.454,
      "step": 3
    },
    {
      "epoch": 0.015594541910331383,
      "grad_norm": 0.2734798789024353,
      "learning_rate": 8e-05,
      "loss": 0.4709,
      "step": 4
    },
    {
      "epoch": 0.01949317738791423,
      "grad_norm": 0.3190411329269409,
      "learning_rate": 0.0001,
      "loss": 0.5674,
      "step": 5
    },
    {
      "epoch": 0.023391812865497075,
      "grad_norm": 0.40352895855903625,
      "learning_rate": 0.00012,
      "loss": 0.5019,
      "step": 6
    },
    {
      "epoch": 0.023391812865497075,
      "eval_loss": 0.5187922120094299,
      "eval_runtime": 5.7089,
      "eval_samples_per_second": 18.918,
      "eval_steps_per_second": 9.459,
      "step": 6
    },
    {
      "epoch": 0.02729044834307992,
      "grad_norm": 0.47463640570640564,
      "learning_rate": 0.00014,
      "loss": 0.4935,
      "step": 7
    },
    {
      "epoch": 0.031189083820662766,
      "grad_norm": 0.365204393863678,
      "learning_rate": 0.00016,
      "loss": 0.48,
      "step": 8
    },
    {
      "epoch": 0.03508771929824561,
      "grad_norm": 0.41965925693511963,
      "learning_rate": 0.00018,
      "loss": 0.4796,
      "step": 9
    },
    {
      "epoch": 0.03508771929824561,
      "eval_loss": 0.43623507022857666,
      "eval_runtime": 5.7174,
      "eval_samples_per_second": 18.89,
      "eval_steps_per_second": 9.445,
      "step": 9
    },
    {
      "epoch": 0.03898635477582846,
      "grad_norm": 0.40204137563705444,
      "learning_rate": 0.0002,
      "loss": 0.4752,
      "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": 3720289902919680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}