last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008097165991902834,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008097165991902834,
      "grad_norm": 5.243322849273682,
      "learning_rate": 2e-05,
      "loss": 14.12,
      "step": 1
    },
    {
      "epoch": 0.0008097165991902834,
      "eval_loss": 16.039548873901367,
      "eval_runtime": 48.3021,
      "eval_samples_per_second": 5.403,
      "eval_steps_per_second": 5.403,
      "step": 1
    },
    {
      "epoch": 0.0016194331983805667,
      "grad_norm": 5.311802387237549,
      "learning_rate": 4e-05,
      "loss": 14.9434,
      "step": 2
    },
    {
      "epoch": 0.0024291497975708503,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 17.5635,
      "step": 3
    },
    {
      "epoch": 0.0024291497975708503,
      "eval_loss": 16.010040283203125,
      "eval_runtime": 47.6629,
      "eval_samples_per_second": 5.476,
      "eval_steps_per_second": 5.476,
      "step": 3
    },
    {
      "epoch": 0.0032388663967611335,
      "grad_norm": 4.934993267059326,
      "learning_rate": 6e-05,
      "loss": 15.4629,
      "step": 4
    },
    {
      "epoch": 0.004048582995951417,
      "grad_norm": 6.582982063293457,
      "learning_rate": 8e-05,
      "loss": 14.5189,
      "step": 5
    },
    {
      "epoch": 0.004858299595141701,
      "grad_norm": 7.352993011474609,
      "learning_rate": 0.0001,
      "loss": 14.5364,
      "step": 6
    },
    {
      "epoch": 0.004858299595141701,
      "eval_loss": 15.222670555114746,
      "eval_runtime": 47.7487,
      "eval_samples_per_second": 5.466,
      "eval_steps_per_second": 5.466,
      "step": 6
    },
    {
      "epoch": 0.005668016194331984,
      "grad_norm": 7.531305313110352,
      "learning_rate": 0.00012,
      "loss": 16.2939,
      "step": 7
    },
    {
      "epoch": 0.006477732793522267,
      "grad_norm": 8.360276222229004,
      "learning_rate": 0.00014,
      "loss": 13.913,
      "step": 8
    },
    {
      "epoch": 0.0072874493927125505,
      "grad_norm": 11.318435668945312,
      "learning_rate": 0.00016,
      "loss": 13.087,
      "step": 9
    },
    {
      "epoch": 0.0072874493927125505,
      "eval_loss": 12.104350090026855,
      "eval_runtime": 45.0844,
      "eval_samples_per_second": 5.789,
      "eval_steps_per_second": 5.789,
      "step": 9
    },
    {
      "epoch": 0.008097165991902834,
      "grad_norm": 14.088312149047852,
      "learning_rate": 0.00018,
      "loss": 12.2794,
      "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": 1643164226027520.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}