last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007778317938745746,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.778317938745747e-05,
      "grad_norm": 1.6048940420150757,
      "learning_rate": 2e-05,
      "loss": 3.5142,
      "step": 1
    },
    {
      "epoch": 7.778317938745747e-05,
      "eval_loss": 3.8527331352233887,
      "eval_runtime": 164.0325,
      "eval_samples_per_second": 33.006,
      "eval_steps_per_second": 16.503,
      "step": 1
    },
    {
      "epoch": 0.00015556635877491494,
      "grad_norm": 2.2154529094696045,
      "learning_rate": 4e-05,
      "loss": 4.0158,
      "step": 2
    },
    {
      "epoch": 0.00023334953816237238,
      "grad_norm": 1.6464277505874634,
      "learning_rate": 6e-05,
      "loss": 3.7409,
      "step": 3
    },
    {
      "epoch": 0.00023334953816237238,
      "eval_loss": 3.837887763977051,
      "eval_runtime": 163.416,
      "eval_samples_per_second": 33.13,
      "eval_steps_per_second": 16.565,
      "step": 3
    },
    {
      "epoch": 0.00031113271754982987,
      "grad_norm": 2.033937454223633,
      "learning_rate": 8e-05,
      "loss": 3.9793,
      "step": 4
    },
    {
      "epoch": 0.0003889158969372873,
      "grad_norm": 1.8927987813949585,
      "learning_rate": 0.0001,
      "loss": 3.3964,
      "step": 5
    },
    {
      "epoch": 0.00046669907632474475,
      "grad_norm": 2.2311577796936035,
      "learning_rate": 0.00012,
      "loss": 3.3525,
      "step": 6
    },
    {
      "epoch": 0.00046669907632474475,
      "eval_loss": 3.505251407623291,
      "eval_runtime": 163.145,
      "eval_samples_per_second": 33.185,
      "eval_steps_per_second": 16.593,
      "step": 6
    },
    {
      "epoch": 0.0005444822557122023,
      "grad_norm": 2.6450042724609375,
      "learning_rate": 0.00014,
      "loss": 3.4371,
      "step": 7
    },
    {
      "epoch": 0.0006222654350996597,
      "grad_norm": 2.665839195251465,
      "learning_rate": 0.00016,
      "loss": 3.1358,
      "step": 8
    },
    {
      "epoch": 0.0007000486144871172,
      "grad_norm": 3.3060426712036133,
      "learning_rate": 0.00018,
      "loss": 2.8456,
      "step": 9
    },
    {
      "epoch": 0.0007000486144871172,
      "eval_loss": 2.617194652557373,
      "eval_runtime": 163.4523,
      "eval_samples_per_second": 33.123,
      "eval_steps_per_second": 16.561,
      "step": 9
    },
    {
      "epoch": 0.0007778317938745746,
      "grad_norm": 2.782524347305298,
      "learning_rate": 0.0002,
      "loss": 2.6517,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}