last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0031338138514572234,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00031338138514572234,
      "grad_norm": 3.2858474254608154,
      "learning_rate": 2e-05,
      "loss": 3.4776,
      "step": 1
    },
    {
      "epoch": 0.00031338138514572234,
      "eval_loss": 3.1475110054016113,
      "eval_runtime": 31.5587,
      "eval_samples_per_second": 42.587,
      "eval_steps_per_second": 21.294,
      "step": 1
    },
    {
      "epoch": 0.0006267627702914447,
      "grad_norm": 3.081576347351074,
      "learning_rate": 4e-05,
      "loss": 2.9316,
      "step": 2
    },
    {
      "epoch": 0.000940144155437167,
      "grad_norm": 2.488064765930176,
      "learning_rate": 6e-05,
      "loss": 3.1302,
      "step": 3
    },
    {
      "epoch": 0.000940144155437167,
      "eval_loss": 3.1216983795166016,
      "eval_runtime": 30.4296,
      "eval_samples_per_second": 44.168,
      "eval_steps_per_second": 22.084,
      "step": 3
    },
    {
      "epoch": 0.0012535255405828893,
      "grad_norm": 3.2934648990631104,
      "learning_rate": 8e-05,
      "loss": 3.6858,
      "step": 4
    },
    {
      "epoch": 0.0015669069257286117,
      "grad_norm": 3.7503325939178467,
      "learning_rate": 0.0001,
      "loss": 3.138,
      "step": 5
    },
    {
      "epoch": 0.001880288310874334,
      "grad_norm": 1.4498717784881592,
      "learning_rate": 0.00012,
      "loss": 3.0642,
      "step": 6
    },
    {
      "epoch": 0.001880288310874334,
      "eval_loss": 2.9685864448547363,
      "eval_runtime": 30.6019,
      "eval_samples_per_second": 43.919,
      "eval_steps_per_second": 21.959,
      "step": 6
    },
    {
      "epoch": 0.0021936696960200563,
      "grad_norm": 0.9387413263320923,
      "learning_rate": 0.00014,
      "loss": 2.8147,
      "step": 7
    },
    {
      "epoch": 0.0025070510811657787,
      "grad_norm": 0.8472381830215454,
      "learning_rate": 0.00016,
      "loss": 2.8266,
      "step": 8
    },
    {
      "epoch": 0.002820432466311501,
      "grad_norm": 0.7642291188240051,
      "learning_rate": 0.00018,
      "loss": 2.8682,
      "step": 9
    },
    {
      "epoch": 0.002820432466311501,
      "eval_loss": 2.9096810817718506,
      "eval_runtime": 30.6428,
      "eval_samples_per_second": 43.86,
      "eval_steps_per_second": 21.93,
      "step": 9
    },
    {
      "epoch": 0.0031338138514572234,
      "grad_norm": 0.768290102481842,
      "learning_rate": 0.0002,
      "loss": 2.9347,
      "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": 153447605207040.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}