last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06172839506172839,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00205761316872428,
      "eval_loss": 1.7741729021072388,
      "eval_runtime": 42.4596,
      "eval_samples_per_second": 3.627,
      "eval_steps_per_second": 1.813,
      "step": 1
    },
    {
      "epoch": 0.006172839506172839,
      "grad_norm": 0.3325711786746979,
      "learning_rate": 4e-05,
      "loss": 1.7895,
      "step": 3
    },
    {
      "epoch": 0.012345679012345678,
      "grad_norm": 0.4484289586544037,
      "learning_rate": 8e-05,
      "loss": 1.802,
      "step": 6
    },
    {
      "epoch": 0.01646090534979424,
      "eval_loss": 1.6747422218322754,
      "eval_runtime": 41.9827,
      "eval_samples_per_second": 3.668,
      "eval_steps_per_second": 1.834,
      "step": 8
    },
    {
      "epoch": 0.018518518518518517,
      "grad_norm": 0.5205878615379333,
      "learning_rate": 0.00012,
      "loss": 1.752,
      "step": 9
    },
    {
      "epoch": 0.024691358024691357,
      "grad_norm": 0.28959599137306213,
      "learning_rate": 0.00016,
      "loss": 1.5129,
      "step": 12
    },
    {
      "epoch": 0.030864197530864196,
      "grad_norm": 0.46413400769233704,
      "learning_rate": 0.0002,
      "loss": 1.5706,
      "step": 15
    },
    {
      "epoch": 0.03292181069958848,
      "eval_loss": 1.4741032123565674,
      "eval_runtime": 42.0815,
      "eval_samples_per_second": 3.66,
      "eval_steps_per_second": 1.83,
      "step": 16
    },
    {
      "epoch": 0.037037037037037035,
      "grad_norm": 0.21619834005832672,
      "learning_rate": 0.00018090169943749476,
      "loss": 1.3482,
      "step": 18
    },
    {
      "epoch": 0.043209876543209874,
      "grad_norm": 0.2760312557220459,
      "learning_rate": 0.00013090169943749476,
      "loss": 1.5502,
      "step": 21
    },
    {
      "epoch": 0.04938271604938271,
      "grad_norm": 0.22206786274909973,
      "learning_rate": 6.909830056250527e-05,
      "loss": 1.3306,
      "step": 24
    },
    {
      "epoch": 0.04938271604938271,
      "eval_loss": 1.4186367988586426,
      "eval_runtime": 42.1086,
      "eval_samples_per_second": 3.657,
      "eval_steps_per_second": 1.829,
      "step": 24
    },
    {
      "epoch": 0.05555555555555555,
      "grad_norm": 0.24945999681949615,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 1.3103,
      "step": 27
    },
    {
      "epoch": 0.06172839506172839,
      "grad_norm": 0.24243028461933136,
      "learning_rate": 0.0,
      "loss": 1.2879,
      "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": 2.7771590934528e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}