checkpoint-1500/trainer_state.json

{
  "best_global_step": 500,
  "best_metric": 0.0044,
  "best_model_checkpoint": "./vit-codegpt-cadcoder/checkpoint-500",
  "epoch": 2.820705882352941,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18823529411764706,
      "grad_norm": 2.2795276641845703,
      "learning_rate": 1.85625e-05,
      "loss": 4.284,
      "step": 100
    },
    {
      "epoch": 0.3764705882352941,
      "grad_norm": 2.0265636444091797,
      "learning_rate": 2.918523676880223e-05,
      "loss": 0.6672,
      "step": 200
    },
    {
      "epoch": 0.5647058823529412,
      "grad_norm": 1.4436606168746948,
      "learning_rate": 2.7096100278551535e-05,
      "loss": 0.5342,
      "step": 300
    },
    {
      "epoch": 0.7529411764705882,
      "grad_norm": 1.2276115417480469,
      "learning_rate": 2.5006963788300836e-05,
      "loss": 0.4912,
      "step": 400
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.580186128616333,
      "learning_rate": 2.2917827298050142e-05,
      "loss": 0.4748,
      "step": 500
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 0.41850993037223816,
      "eval_rouge1": 0.085,
      "eval_rouge2": 0.0044,
      "eval_rougeL": 0.0773,
      "eval_runtime": 4240.7834,
      "eval_samples_per_second": 0.354,
      "eval_steps_per_second": 0.088,
      "step": 500
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 1.35711669921875,
      "learning_rate": 2.0828690807799444e-05,
      "loss": 0.4408,
      "step": 600
    },
    {
      "epoch": 1.316235294117647,
      "grad_norm": 0.9837644696235657,
      "learning_rate": 1.8739554317548746e-05,
      "loss": 0.4193,
      "step": 700
    },
    {
      "epoch": 1.5044705882352942,
      "grad_norm": 0.9567480087280273,
      "learning_rate": 1.6650417827298047e-05,
      "loss": 0.4028,
      "step": 800
    },
    {
      "epoch": 1.692705882352941,
      "grad_norm": 0.9603307247161865,
      "learning_rate": 1.4561281337047354e-05,
      "loss": 0.4049,
      "step": 900
    },
    {
      "epoch": 1.8809411764705883,
      "grad_norm": 1.3618443012237549,
      "learning_rate": 1.2472144846796658e-05,
      "loss": 0.4005,
      "step": 1000
    },
    {
      "epoch": 1.8809411764705883,
      "eval_loss": 0.36618879437446594,
      "eval_rouge1": 0.0946,
      "eval_rouge2": 0.0041,
      "eval_rougeL": 0.0861,
      "eval_runtime": 4228.5695,
      "eval_samples_per_second": 0.355,
      "eval_steps_per_second": 0.089,
      "step": 1000
    },
    {
      "epoch": 2.067764705882353,
      "grad_norm": 0.9115932583808899,
      "learning_rate": 1.0383008356545962e-05,
      "loss": 0.4064,
      "step": 1100
    },
    {
      "epoch": 2.2560000000000002,
      "grad_norm": 1.121819019317627,
      "learning_rate": 8.293871866295265e-06,
      "loss": 0.3864,
      "step": 1200
    },
    {
      "epoch": 2.444235294117647,
      "grad_norm": 1.1339781284332275,
      "learning_rate": 6.204735376044569e-06,
      "loss": 0.3595,
      "step": 1300
    },
    {
      "epoch": 2.632470588235294,
      "grad_norm": 0.7861893177032471,
      "learning_rate": 4.1155988857938724e-06,
      "loss": 0.367,
      "step": 1400
    },
    {
      "epoch": 2.820705882352941,
      "grad_norm": 0.9953855872154236,
      "learning_rate": 2.026462395543175e-06,
      "loss": 0.3571,
      "step": 1500
    },
    {
      "epoch": 2.820705882352941,
      "eval_loss": 0.34929922223091125,
      "eval_rouge1": 0.0944,
      "eval_rouge2": 0.004,
      "eval_rougeL": 0.0863,
      "eval_runtime": 4261.9383,
      "eval_samples_per_second": 0.352,
      "eval_steps_per_second": 0.088,
      "step": 1500
    }
  ],
  "logging_steps": 100,
  "max_steps": 1596,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.326802253496189e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}