last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016386055466797755,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016386055466797756,
      "grad_norm": 1.510071039199829,
      "learning_rate": 2e-05,
      "loss": 3.8897,
      "step": 1
    },
    {
      "epoch": 0.00016386055466797756,
      "eval_loss": 5.125027179718018,
      "eval_runtime": 176.3495,
      "eval_samples_per_second": 14.573,
      "eval_steps_per_second": 7.287,
      "step": 1
    },
    {
      "epoch": 0.0003277211093359551,
      "grad_norm": 2.477065086364746,
      "learning_rate": 4e-05,
      "loss": 4.8175,
      "step": 2
    },
    {
      "epoch": 0.0004915816640039326,
      "grad_norm": 1.9006699323654175,
      "learning_rate": 6e-05,
      "loss": 4.273,
      "step": 3
    },
    {
      "epoch": 0.0004915816640039326,
      "eval_loss": 5.103329658508301,
      "eval_runtime": 177.9719,
      "eval_samples_per_second": 14.44,
      "eval_steps_per_second": 7.22,
      "step": 3
    },
    {
      "epoch": 0.0006554422186719102,
      "grad_norm": 2.2485129833221436,
      "learning_rate": 8e-05,
      "loss": 5.1019,
      "step": 4
    },
    {
      "epoch": 0.0008193027733398877,
      "grad_norm": 2.8604471683502197,
      "learning_rate": 0.0001,
      "loss": 4.3341,
      "step": 5
    },
    {
      "epoch": 0.0009831633280078652,
      "grad_norm": 3.7464327812194824,
      "learning_rate": 0.00012,
      "loss": 4.2908,
      "step": 6
    },
    {
      "epoch": 0.0009831633280078652,
      "eval_loss": 4.862543106079102,
      "eval_runtime": 177.6343,
      "eval_samples_per_second": 14.468,
      "eval_steps_per_second": 7.234,
      "step": 6
    },
    {
      "epoch": 0.0011470238826758428,
      "grad_norm": 1.7869113683700562,
      "learning_rate": 0.00014,
      "loss": 4.0132,
      "step": 7
    },
    {
      "epoch": 0.0013108844373438205,
      "grad_norm": 2.2198233604431152,
      "learning_rate": 0.00016,
      "loss": 4.0817,
      "step": 8
    },
    {
      "epoch": 0.001474744992011798,
      "grad_norm": 2.592177391052246,
      "learning_rate": 0.00018,
      "loss": 4.1479,
      "step": 9
    },
    {
      "epoch": 0.001474744992011798,
      "eval_loss": 4.242178916931152,
      "eval_runtime": 177.9632,
      "eval_samples_per_second": 14.441,
      "eval_steps_per_second": 7.221,
      "step": 9
    },
    {
      "epoch": 0.0016386055466797755,
      "grad_norm": 3.6763322353363037,
      "learning_rate": 0.0002,
      "loss": 3.5734,
      "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
}