last-checkpoint/trainer_state.json

{
  "best_metric": 19.603507093484392,
  "best_model_checkpoint": "./models/rootflo/fauna-v3.6/checkpoint-516",
  "epoch": 3.9845559845559846,
  "eval_steps": 100,
  "global_step": 516,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7722007722007722,
      "grad_norm": 0.353707492351532,
      "learning_rate": 1.98e-06,
      "loss": 0.3935,
      "step": 100
    },
    {
      "epoch": 0.9961389961389961,
      "eval_bleu": 16.54427447246732,
      "eval_loss": 0.1285262256860733,
      "eval_runtime": 2990.993,
      "eval_samples_per_second": 1.62,
      "eval_steps_per_second": 0.004,
      "step": 129
    },
    {
      "epoch": 1.5444015444015444,
      "grad_norm": 0.3564741611480713,
      "learning_rate": 1.7332964244331777e-06,
      "loss": 0.2695,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_bleu": 18.565923091763654,
      "eval_loss": 0.11729109287261963,
      "eval_runtime": 3235.1892,
      "eval_samples_per_second": 1.497,
      "eval_steps_per_second": 0.004,
      "step": 259
    },
    {
      "epoch": 2.3166023166023164,
      "grad_norm": 0.34130024909973145,
      "learning_rate": 1.0679148314231502e-06,
      "loss": 0.2469,
      "step": 300
    },
    {
      "epoch": 2.9961389961389964,
      "eval_bleu": 11.442911241118665,
      "eval_loss": 0.11346418410539627,
      "eval_runtime": 2891.0098,
      "eval_samples_per_second": 1.676,
      "eval_steps_per_second": 0.004,
      "step": 388
    },
    {
      "epoch": 3.088803088803089,
      "grad_norm": 0.32632899284362793,
      "learning_rate": 3.656067158363546e-07,
      "loss": 0.2347,
      "step": 400
    },
    {
      "epoch": 3.861003861003861,
      "grad_norm": 0.3226456344127655,
      "learning_rate": 8.229706450227802e-09,
      "loss": 0.2285,
      "step": 500
    },
    {
      "epoch": 3.9845559845559846,
      "eval_bleu": 19.603507093484392,
      "eval_loss": 0.11292669177055359,
      "eval_runtime": 2938.0533,
      "eval_samples_per_second": 1.649,
      "eval_steps_per_second": 0.004,
      "step": 516
    }
  ],
  "logging_steps": 100,
  "max_steps": 516,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 1
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3463878146985915e+21,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}