last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000112310651261389,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.12310651261389e-05,
      "grad_norm": 1.5523439645767212,
      "learning_rate": 2e-05,
      "loss": 4.8092,
      "step": 1
    },
    {
      "epoch": 1.12310651261389e-05,
      "eval_loss": 1.0306237936019897,
      "eval_runtime": 2598.3026,
      "eval_samples_per_second": 14.429,
      "eval_steps_per_second": 7.214,
      "step": 1
    },
    {
      "epoch": 2.24621302522778e-05,
      "grad_norm": 0.8822906017303467,
      "learning_rate": 4e-05,
      "loss": 3.8404,
      "step": 2
    },
    {
      "epoch": 3.36931953784167e-05,
      "grad_norm": 1.0201040506362915,
      "learning_rate": 6e-05,
      "loss": 3.8607,
      "step": 3
    },
    {
      "epoch": 3.36931953784167e-05,
      "eval_loss": 1.0291807651519775,
      "eval_runtime": 2595.7356,
      "eval_samples_per_second": 14.443,
      "eval_steps_per_second": 7.221,
      "step": 3
    },
    {
      "epoch": 4.49242605045556e-05,
      "grad_norm": 1.056449294090271,
      "learning_rate": 8e-05,
      "loss": 4.9529,
      "step": 4
    },
    {
      "epoch": 5.61553256306945e-05,
      "grad_norm": 1.198383092880249,
      "learning_rate": 0.0001,
      "loss": 4.6338,
      "step": 5
    },
    {
      "epoch": 6.73863907568334e-05,
      "grad_norm": 0.8677435517311096,
      "learning_rate": 0.00012,
      "loss": 3.8863,
      "step": 6
    },
    {
      "epoch": 6.73863907568334e-05,
      "eval_loss": 1.0128856897354126,
      "eval_runtime": 2598.0382,
      "eval_samples_per_second": 14.43,
      "eval_steps_per_second": 7.215,
      "step": 6
    },
    {
      "epoch": 7.86174558829723e-05,
      "grad_norm": 1.3429189920425415,
      "learning_rate": 0.00014,
      "loss": 3.6685,
      "step": 7
    },
    {
      "epoch": 8.98485210091112e-05,
      "grad_norm": 1.129899263381958,
      "learning_rate": 0.00016,
      "loss": 4.254,
      "step": 8
    },
    {
      "epoch": 0.0001010795861352501,
      "grad_norm": 1.9611945152282715,
      "learning_rate": 0.00018,
      "loss": 4.4935,
      "step": 9
    },
    {
      "epoch": 0.0001010795861352501,
      "eval_loss": 0.952491819858551,
      "eval_runtime": 2596.6683,
      "eval_samples_per_second": 14.438,
      "eval_steps_per_second": 7.219,
      "step": 9
    },
    {
      "epoch": 0.000112310651261389,
      "grad_norm": 1.8180216550827026,
      "learning_rate": 0.0002,
      "loss": 3.9938,
      "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": 4986771571998720.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}