specialized_llm_8b_base_100/checkpoint-50/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 107.8238553314552,
      "learning_rate": 2e-05,
      "loss": 4.7784,
      "step": 5
    },
    {
      "epoch": 0.2,
      "grad_norm": 116.95906337998,
      "learning_rate": 2e-05,
      "loss": 2.4617,
      "step": 10
    },
    {
      "epoch": 0.3,
      "grad_norm": 53.544280733687785,
      "learning_rate": 2e-05,
      "loss": 1.9213,
      "step": 15
    },
    {
      "epoch": 0.4,
      "grad_norm": 76.58961089404328,
      "learning_rate": 2e-05,
      "loss": 1.4219,
      "step": 20
    },
    {
      "epoch": 0.5,
      "grad_norm": 429.80406194643473,
      "learning_rate": 2e-05,
      "loss": 1.2028,
      "step": 25
    },
    {
      "epoch": 0.6,
      "grad_norm": 36.97523628037835,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 30
    },
    {
      "epoch": 0.7,
      "grad_norm": 141.6321860313076,
      "learning_rate": 2e-05,
      "loss": 1.4171,
      "step": 35
    },
    {
      "epoch": 0.8,
      "grad_norm": 30.152893149931465,
      "learning_rate": 2e-05,
      "loss": 0.907,
      "step": 40
    },
    {
      "epoch": 0.9,
      "grad_norm": 26.493427680758966,
      "learning_rate": 2e-05,
      "loss": 0.9565,
      "step": 45
    },
    {
      "epoch": 1.0,
      "grad_norm": 27.625250842445954,
      "learning_rate": 2e-05,
      "loss": 0.8956,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 40894464000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}