last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008531149359097404,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.531149359097404e-05,
      "grad_norm": 0.9269247055053711,
      "learning_rate": 2e-05,
      "loss": 2.7629,
      "step": 1
    },
    {
      "epoch": 8.531149359097404e-05,
      "eval_loss": 3.126101493835449,
      "eval_runtime": 187.3728,
      "eval_samples_per_second": 26.343,
      "eval_steps_per_second": 13.172,
      "step": 1
    },
    {
      "epoch": 0.00017062298718194808,
      "grad_norm": 1.0083377361297607,
      "learning_rate": 4e-05,
      "loss": 3.0406,
      "step": 2
    },
    {
      "epoch": 0.00025593448077292216,
      "grad_norm": 0.8629823923110962,
      "learning_rate": 6e-05,
      "loss": 3.1489,
      "step": 3
    },
    {
      "epoch": 0.00025593448077292216,
      "eval_loss": 3.122657060623169,
      "eval_runtime": 186.4576,
      "eval_samples_per_second": 26.473,
      "eval_steps_per_second": 13.236,
      "step": 3
    },
    {
      "epoch": 0.00034124597436389616,
      "grad_norm": 0.8757002353668213,
      "learning_rate": 8e-05,
      "loss": 2.9894,
      "step": 4
    },
    {
      "epoch": 0.0004265574679548702,
      "grad_norm": 0.9721984267234802,
      "learning_rate": 0.0001,
      "loss": 3.1353,
      "step": 5
    },
    {
      "epoch": 0.0005118689615458443,
      "grad_norm": 0.9829956889152527,
      "learning_rate": 0.00012,
      "loss": 2.9926,
      "step": 6
    },
    {
      "epoch": 0.0005118689615458443,
      "eval_loss": 3.0818350315093994,
      "eval_runtime": 186.681,
      "eval_samples_per_second": 26.441,
      "eval_steps_per_second": 13.22,
      "step": 6
    },
    {
      "epoch": 0.0005971804551368183,
      "grad_norm": 1.1662498712539673,
      "learning_rate": 0.00014,
      "loss": 2.771,
      "step": 7
    },
    {
      "epoch": 0.0006824919487277923,
      "grad_norm": 1.2197030782699585,
      "learning_rate": 0.00016,
      "loss": 2.583,
      "step": 8
    },
    {
      "epoch": 0.0007678034423187664,
      "grad_norm": 1.4195125102996826,
      "learning_rate": 0.00018,
      "loss": 3.2318,
      "step": 9
    },
    {
      "epoch": 0.0007678034423187664,
      "eval_loss": 2.8649885654449463,
      "eval_runtime": 186.4772,
      "eval_samples_per_second": 26.47,
      "eval_steps_per_second": 13.235,
      "step": 9
    },
    {
      "epoch": 0.0008531149359097404,
      "grad_norm": 1.595644474029541,
      "learning_rate": 0.0002,
      "loss": 2.7923,
      "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": 2059778654208000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}