last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003865854837150865,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003865854837150865,
      "grad_norm": 2.70934796333313,
      "learning_rate": 2e-05,
      "loss": 2.1883,
      "step": 1
    },
    {
      "epoch": 0.0003865854837150865,
      "eval_loss": 2.2865633964538574,
      "eval_runtime": 52.5061,
      "eval_samples_per_second": 20.759,
      "eval_steps_per_second": 10.38,
      "step": 1
    },
    {
      "epoch": 0.000773170967430173,
      "grad_norm": 3.197664976119995,
      "learning_rate": 4e-05,
      "loss": 2.2499,
      "step": 2
    },
    {
      "epoch": 0.0011597564511452595,
      "grad_norm": 3.132007122039795,
      "learning_rate": 6e-05,
      "loss": 2.5349,
      "step": 3
    },
    {
      "epoch": 0.0011597564511452595,
      "eval_loss": 2.2641870975494385,
      "eval_runtime": 52.4717,
      "eval_samples_per_second": 20.773,
      "eval_steps_per_second": 10.387,
      "step": 3
    },
    {
      "epoch": 0.001546341934860346,
      "grad_norm": 2.2434349060058594,
      "learning_rate": 8e-05,
      "loss": 1.9627,
      "step": 4
    },
    {
      "epoch": 0.0019329274185754324,
      "grad_norm": 3.057121992111206,
      "learning_rate": 0.0001,
      "loss": 2.0058,
      "step": 5
    },
    {
      "epoch": 0.002319512902290519,
      "grad_norm": 3.5964255332946777,
      "learning_rate": 0.00012,
      "loss": 2.3888,
      "step": 6
    },
    {
      "epoch": 0.002319512902290519,
      "eval_loss": 1.9325042963027954,
      "eval_runtime": 52.4509,
      "eval_samples_per_second": 20.781,
      "eval_steps_per_second": 10.391,
      "step": 6
    },
    {
      "epoch": 0.0027060983860056056,
      "grad_norm": 3.7439322471618652,
      "learning_rate": 0.00014,
      "loss": 2.0192,
      "step": 7
    },
    {
      "epoch": 0.003092683869720692,
      "grad_norm": 3.7384166717529297,
      "learning_rate": 0.00016,
      "loss": 1.8645,
      "step": 8
    },
    {
      "epoch": 0.0034792693534357786,
      "grad_norm": 8.410094261169434,
      "learning_rate": 0.00018,
      "loss": 1.6249,
      "step": 9
    },
    {
      "epoch": 0.0034792693534357786,
      "eval_loss": 1.3259292840957642,
      "eval_runtime": 52.4965,
      "eval_samples_per_second": 20.763,
      "eval_steps_per_second": 10.382,
      "step": 9
    },
    {
      "epoch": 0.003865854837150865,
      "grad_norm": 2.686150312423706,
      "learning_rate": 0.0002,
      "loss": 1.2755,
      "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": 2052393617326080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}