last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007542900245144258,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00025143000817147524,
      "eval_loss": 0.9352075457572937,
      "eval_runtime": 184.4804,
      "eval_samples_per_second": 9.08,
      "eval_steps_per_second": 4.542,
      "step": 1
    },
    {
      "epoch": 0.0007542900245144258,
      "grad_norm": 0.43845057487487793,
      "learning_rate": 4e-05,
      "loss": 0.9096,
      "step": 3
    },
    {
      "epoch": 0.0012571500408573764,
      "eval_loss": 0.9017426371574402,
      "eval_runtime": 185.2286,
      "eval_samples_per_second": 9.043,
      "eval_steps_per_second": 4.524,
      "step": 5
    },
    {
      "epoch": 0.0015085800490288515,
      "grad_norm": 0.4952695369720459,
      "learning_rate": 8e-05,
      "loss": 0.8794,
      "step": 6
    },
    {
      "epoch": 0.0022628700735432775,
      "grad_norm": 0.4610755145549774,
      "learning_rate": 0.00012,
      "loss": 0.8754,
      "step": 9
    },
    {
      "epoch": 0.0025143000817147527,
      "eval_loss": 0.7889909148216248,
      "eval_runtime": 185.2916,
      "eval_samples_per_second": 9.04,
      "eval_steps_per_second": 4.523,
      "step": 10
    },
    {
      "epoch": 0.003017160098057703,
      "grad_norm": 0.3406424820423126,
      "learning_rate": 0.00016,
      "loss": 0.8257,
      "step": 12
    },
    {
      "epoch": 0.003771450122572129,
      "grad_norm": 0.2703261077404022,
      "learning_rate": 0.0002,
      "loss": 0.7771,
      "step": 15
    },
    {
      "epoch": 0.003771450122572129,
      "eval_loss": 0.6933387517929077,
      "eval_runtime": 185.364,
      "eval_samples_per_second": 9.036,
      "eval_steps_per_second": 4.521,
      "step": 15
    },
    {
      "epoch": 0.004525740147086555,
      "grad_norm": 0.2652558982372284,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.598,
      "step": 18
    },
    {
      "epoch": 0.005028600163429505,
      "eval_loss": 0.6620884537696838,
      "eval_runtime": 185.462,
      "eval_samples_per_second": 9.031,
      "eval_steps_per_second": 4.518,
      "step": 20
    },
    {
      "epoch": 0.005280030171600981,
      "grad_norm": 0.2349843829870224,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.6841,
      "step": 21
    },
    {
      "epoch": 0.006034320196115406,
      "grad_norm": 0.19294080138206482,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.6622,
      "step": 24
    },
    {
      "epoch": 0.006285750204286881,
      "eval_loss": 0.649798572063446,
      "eval_runtime": 185.4412,
      "eval_samples_per_second": 9.033,
      "eval_steps_per_second": 4.519,
      "step": 25
    },
    {
      "epoch": 0.006788610220629833,
      "grad_norm": 0.17021793127059937,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.5969,
      "step": 27
    },
    {
      "epoch": 0.007542900245144258,
      "grad_norm": 0.17728255689144135,
      "learning_rate": 0.0,
      "loss": 0.6806,
      "step": 30
    },
    {
      "epoch": 0.007542900245144258,
      "eval_loss": 0.6474659442901611,
      "eval_runtime": 185.4886,
      "eval_samples_per_second": 9.03,
      "eval_steps_per_second": 4.518,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6001167968501760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}