checkpoint-12000/trainer_state.json

{
  "best_metric": 0.1595447063446045,
  "best_model_checkpoint": "/Users/frapadovani/Desktop/babyLM_controlled/models_trained/convergence_french/random_sentence_french/checkpoint-12000",
  "epoch": 0.021700917948829234,
  "eval_steps": 2000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003616819658138206,
      "grad_norm": 0.14744240045547485,
      "learning_rate": 0.0001,
      "loss": 0.7697,
      "step": 2000
    },
    {
      "epoch": 0.003616819658138206,
      "eval_loss": 0.19154727458953857,
      "eval_runtime": 91.812,
      "eval_samples_per_second": 397.933,
      "eval_steps_per_second": 24.877,
      "step": 2000
    },
    {
      "epoch": 0.007233639316276412,
      "grad_norm": 0.1623348593711853,
      "learning_rate": 0.0001,
      "loss": 0.1884,
      "step": 4000
    },
    {
      "epoch": 0.007233639316276412,
      "eval_loss": 0.17771275341510773,
      "eval_runtime": 99.0721,
      "eval_samples_per_second": 368.772,
      "eval_steps_per_second": 23.054,
      "step": 4000
    },
    {
      "epoch": 0.010850458974414617,
      "grad_norm": 0.13657265901565552,
      "learning_rate": 0.0001,
      "loss": 0.178,
      "step": 6000
    },
    {
      "epoch": 0.010850458974414617,
      "eval_loss": 0.16984443366527557,
      "eval_runtime": 90.3285,
      "eval_samples_per_second": 404.468,
      "eval_steps_per_second": 25.285,
      "step": 6000
    },
    {
      "epoch": 0.014467278632552824,
      "grad_norm": 0.15028506517410278,
      "learning_rate": 0.0001,
      "loss": 0.1714,
      "step": 8000
    },
    {
      "epoch": 0.014467278632552824,
      "eval_loss": 0.16548627614974976,
      "eval_runtime": 90.9787,
      "eval_samples_per_second": 401.578,
      "eval_steps_per_second": 25.105,
      "step": 8000
    },
    {
      "epoch": 0.01808409829069103,
      "grad_norm": 0.17982865869998932,
      "learning_rate": 0.0001,
      "loss": 0.1669,
      "step": 10000
    },
    {
      "epoch": 0.01808409829069103,
      "eval_loss": 0.16203773021697998,
      "eval_runtime": 87.7097,
      "eval_samples_per_second": 416.544,
      "eval_steps_per_second": 26.04,
      "step": 10000
    },
    {
      "epoch": 0.021700917948829234,
      "grad_norm": 0.14700624346733093,
      "learning_rate": 0.0001,
      "loss": 0.1647,
      "step": 12000
    },
    {
      "epoch": 0.021700917948829234,
      "eval_loss": 0.1595447063446045,
      "eval_runtime": 87.4086,
      "eval_samples_per_second": 417.979,
      "eval_steps_per_second": 26.13,
      "step": 12000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 552972,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 233928916992000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}