checkpoint-200/trainer_state.json

{
  "best_metric": 0.6097185015678406,
  "best_model_checkpoint": "./llama3/28-08-24-Weni-Pipeline_test_Experiment with SFT and Llama3 70b-2_max_steps-1362_batch_16_2024-08-28/checkpoint-200",
  "epoch": 0.4404073768235618,
  "eval_steps": 100,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02202036884117809,
      "grad_norm": 0.6965782642364502,
      "learning_rate": 7.5e-05,
      "loss": 2.1201,
      "step": 10
    },
    {
      "epoch": 0.04404073768235618,
      "grad_norm": 0.7504029870033264,
      "learning_rate": 0.00015,
      "loss": 0.9392,
      "step": 20
    },
    {
      "epoch": 0.06606110652353427,
      "grad_norm": 0.7269854545593262,
      "learning_rate": 0.000225,
      "loss": 0.7958,
      "step": 30
    },
    {
      "epoch": 0.08808147536471236,
      "grad_norm": 0.15891791880130768,
      "learning_rate": 0.0003,
      "loss": 0.7251,
      "step": 40
    },
    {
      "epoch": 0.11010184420589045,
      "grad_norm": 0.14764881134033203,
      "learning_rate": 0.00029995764763563235,
      "loss": 0.6941,
      "step": 50
    },
    {
      "epoch": 0.13212221304706853,
      "grad_norm": 0.11882930248975754,
      "learning_rate": 0.00029983061445883305,
      "loss": 0.673,
      "step": 60
    },
    {
      "epoch": 0.15414258188824662,
      "grad_norm": 0.15152081847190857,
      "learning_rate": 0.0002996189722050073,
      "loss": 0.6428,
      "step": 70
    },
    {
      "epoch": 0.17616295072942473,
      "grad_norm": 0.1619480848312378,
      "learning_rate": 0.0002993228403881531,
      "loss": 0.6465,
      "step": 80
    },
    {
      "epoch": 0.1981833195706028,
      "grad_norm": 0.15019242465496063,
      "learning_rate": 0.00029894238623337174,
      "loss": 0.6308,
      "step": 90
    },
    {
      "epoch": 0.2202036884117809,
      "grad_norm": 0.14553773403167725,
      "learning_rate": 0.00029847782458243663,
      "loss": 0.6314,
      "step": 100
    },
    {
      "epoch": 0.2202036884117809,
      "eval_accuracy": 1.0,
      "eval_f1": 1.0,
      "eval_f1_macro": 1.0,
      "eval_f1_micro": 1.0,
      "eval_loss": 0.6374139189720154,
      "eval_precision": 1.0,
      "eval_precision_macro": 1.0,
      "eval_precision_micro": 1.0,
      "eval_recall": 1.0,
      "eval_recall_macro": 1.0,
      "eval_recall_micro": 1.0,
      "eval_runtime": 1957.1676,
      "eval_samples_per_second": 0.413,
      "eval_steps_per_second": 0.206,
      "step": 100
    },
    {
      "epoch": 0.24222405725295898,
      "grad_norm": 0.20665155351161957,
      "learning_rate": 0.00029792941777247184,
      "loss": 0.646,
      "step": 110
    },
    {
      "epoch": 0.26424442609413706,
      "grad_norm": 0.18165886402130127,
      "learning_rate": 0.0002972974754878111,
      "loss": 0.6326,
      "step": 120
    },
    {
      "epoch": 0.28626479493531515,
      "grad_norm": 0.13634872436523438,
      "learning_rate": 0.0002965823545851199,
      "loss": 0.5785,
      "step": 130
    },
    {
      "epoch": 0.30828516377649323,
      "grad_norm": 0.14753705263137817,
      "learning_rate": 0.00029578445889187865,
      "loss": 0.639,
      "step": 140
    },
    {
      "epoch": 0.33030553261767137,
      "grad_norm": 0.2674409747123718,
      "learning_rate": 0.00029490423897834234,
      "loss": 0.626,
      "step": 150
    },
    {
      "epoch": 0.35232590145884946,
      "grad_norm": 0.16381941735744476,
      "learning_rate": 0.0002939421919031044,
      "loss": 0.6009,
      "step": 160
    },
    {
      "epoch": 0.37434627030002754,
      "grad_norm": 0.1673993617296219,
      "learning_rate": 0.00029289886093240847,
      "loss": 0.6193,
      "step": 170
    },
    {
      "epoch": 0.3963666391412056,
      "grad_norm": 0.16463495790958405,
      "learning_rate": 0.0002917748352333667,
      "loss": 0.5894,
      "step": 180
    },
    {
      "epoch": 0.4183870079823837,
      "grad_norm": 0.17818249762058258,
      "learning_rate": 0.0002905707495412589,
      "loss": 0.6249,
      "step": 190
    },
    {
      "epoch": 0.4404073768235618,
      "grad_norm": 0.14480963349342346,
      "learning_rate": 0.00028928728380109764,
      "loss": 0.6223,
      "step": 200
    },
    {
      "epoch": 0.4404073768235618,
      "eval_accuracy": 1.0,
      "eval_f1": 1.0,
      "eval_f1_macro": 1.0,
      "eval_f1_micro": 1.0,
      "eval_loss": 0.6097185015678406,
      "eval_precision": 1.0,
      "eval_precision_macro": 1.0,
      "eval_precision_micro": 1.0,
      "eval_recall": 1.0,
      "eval_recall_macro": 1.0,
      "eval_recall_micro": 1.0,
      "eval_runtime": 1863.6968,
      "eval_samples_per_second": 0.434,
      "eval_steps_per_second": 0.217,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 1362,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9495153475584.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}