checkpoint-780/trainer_state.json

{
  "best_global_step": 600,
  "best_metric": 0.2659165561199188,
  "best_model_checkpoint": "codellama_utests_embedded/checkpoint-600",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 780,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2570694087403599,
      "grad_norm": 0.058597221970558167,
      "learning_rate": 9.8e-05,
      "loss": 0.4511,
      "step": 50
    },
    {
      "epoch": 0.5141388174807198,
      "grad_norm": 0.04719100147485733,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.2892,
      "step": 100
    },
    {
      "epoch": 0.7712082262210797,
      "grad_norm": 0.03564457222819328,
      "learning_rate": 0.00018558823529411767,
      "loss": 0.2751,
      "step": 150
    },
    {
      "epoch": 1.025706940874036,
      "grad_norm": 0.04530099779367447,
      "learning_rate": 0.00017088235294117647,
      "loss": 0.2709,
      "step": 200
    },
    {
      "epoch": 1.025706940874036,
      "eval_loss": 0.2812548875808716,
      "eval_runtime": 364.2767,
      "eval_samples_per_second": 0.535,
      "eval_steps_per_second": 0.535,
      "step": 200
    },
    {
      "epoch": 1.2827763496143958,
      "grad_norm": 0.04306064546108246,
      "learning_rate": 0.0001561764705882353,
      "loss": 0.2655,
      "step": 250
    },
    {
      "epoch": 1.5398457583547558,
      "grad_norm": 0.059078045189380646,
      "learning_rate": 0.00014147058823529415,
      "loss": 0.25,
      "step": 300
    },
    {
      "epoch": 1.7969151670951158,
      "grad_norm": 0.058344610035419464,
      "learning_rate": 0.00012676470588235294,
      "loss": 0.2494,
      "step": 350
    },
    {
      "epoch": 2.051413881748072,
      "grad_norm": 0.05032959207892418,
      "learning_rate": 0.00011205882352941177,
      "loss": 0.2397,
      "step": 400
    },
    {
      "epoch": 2.051413881748072,
      "eval_loss": 0.26848915219306946,
      "eval_runtime": 294.6836,
      "eval_samples_per_second": 0.662,
      "eval_steps_per_second": 0.662,
      "step": 400
    },
    {
      "epoch": 2.308483290488432,
      "grad_norm": 0.05623197555541992,
      "learning_rate": 9.73529411764706e-05,
      "loss": 0.2252,
      "step": 450
    },
    {
      "epoch": 2.5655526992287916,
      "grad_norm": 0.07187215238809586,
      "learning_rate": 8.26470588235294e-05,
      "loss": 0.2322,
      "step": 500
    },
    {
      "epoch": 2.822622107969152,
      "grad_norm": 0.05385003983974457,
      "learning_rate": 6.794117647058824e-05,
      "loss": 0.2175,
      "step": 550
    },
    {
      "epoch": 3.077120822622108,
      "grad_norm": 0.06531719863414764,
      "learning_rate": 5.323529411764706e-05,
      "loss": 0.2219,
      "step": 600
    },
    {
      "epoch": 3.077120822622108,
      "eval_loss": 0.2659165561199188,
      "eval_runtime": 477.3584,
      "eval_samples_per_second": 0.408,
      "eval_steps_per_second": 0.408,
      "step": 600
    },
    {
      "epoch": 3.334190231362468,
      "grad_norm": 0.06546808034181595,
      "learning_rate": 3.8529411764705886e-05,
      "loss": 0.1976,
      "step": 650
    },
    {
      "epoch": 3.5912596401028276,
      "grad_norm": 0.08491069823503494,
      "learning_rate": 2.3823529411764707e-05,
      "loss": 0.2109,
      "step": 700
    },
    {
      "epoch": 3.848329048843188,
      "grad_norm": 0.08265674859285355,
      "learning_rate": 9.117647058823529e-06,
      "loss": 0.2079,
      "step": 750
    }
  ],
  "logging_steps": 50,
  "max_steps": 780,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.104705552169042e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}