checkpoint-2240/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 2240,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.464285714285714,
      "grad_norm": 0.21455785632133484,
      "learning_rate": 9.980000000000001e-06,
      "loss": 0.1949,
      "step": 500
    },
    {
      "epoch": 4.464285714285714,
      "eval_runtime": 0.9738,
      "eval_samples_per_second": 0.0,
      "eval_steps_per_second": 0.0,
      "eval_test-eval_cosine_accuracy@1": 0.8519553072625698,
      "eval_test-eval_cosine_accuracy@10": 1.0,
      "eval_test-eval_cosine_accuracy@3": 0.9748603351955307,
      "eval_test-eval_cosine_accuracy@5": 0.9888268156424581,
      "eval_test-eval_cosine_map@100": 0.9115068280571075,
      "eval_test-eval_cosine_mrr@10": 0.9115068280571073,
      "eval_test-eval_cosine_ndcg@10": 0.9336742870997445,
      "eval_test-eval_cosine_precision@1": 0.8519553072625698,
      "eval_test-eval_cosine_precision@10": 0.09999999999999999,
      "eval_test-eval_cosine_precision@3": 0.3249534450651769,
      "eval_test-eval_cosine_precision@5": 0.19776536312849158,
      "eval_test-eval_cosine_recall@1": 0.8519553072625698,
      "eval_test-eval_cosine_recall@10": 1.0,
      "eval_test-eval_cosine_recall@3": 0.9748603351955307,
      "eval_test-eval_cosine_recall@5": 0.9888268156424581,
      "step": 500
    },
    {
      "epoch": 8.928571428571429,
      "grad_norm": 0.14167314767837524,
      "learning_rate": 1.9980000000000002e-05,
      "loss": 0.1347,
      "step": 1000
    },
    {
      "epoch": 8.928571428571429,
      "eval_runtime": 1.0193,
      "eval_samples_per_second": 0.0,
      "eval_steps_per_second": 0.0,
      "eval_test-eval_cosine_accuracy@1": 0.8687150837988827,
      "eval_test-eval_cosine_accuracy@10": 1.0,
      "eval_test-eval_cosine_accuracy@3": 0.9916201117318436,
      "eval_test-eval_cosine_accuracy@5": 1.0,
      "eval_test-eval_cosine_map@100": 0.9263966480446927,
      "eval_test-eval_cosine_mrr@10": 0.9263966480446925,
      "eval_test-eval_cosine_ndcg@10": 0.9452350119743832,
      "eval_test-eval_cosine_precision@1": 0.8687150837988827,
      "eval_test-eval_cosine_precision@10": 0.09999999999999999,
      "eval_test-eval_cosine_precision@3": 0.33054003724394787,
      "eval_test-eval_cosine_precision@5": 0.19999999999999998,
      "eval_test-eval_cosine_recall@1": 0.8687150837988827,
      "eval_test-eval_cosine_recall@10": 1.0,
      "eval_test-eval_cosine_recall@3": 0.9916201117318436,
      "eval_test-eval_cosine_recall@5": 1.0,
      "step": 1000
    },
    {
      "epoch": 13.392857142857142,
      "grad_norm": 0.1641000360250473,
      "learning_rate": 0.0,
      "loss": 0.1193,
      "step": 1500
    },
    {
      "epoch": 13.392857142857142,
      "eval_runtime": 0.9947,
      "eval_samples_per_second": 0.0,
      "eval_steps_per_second": 0.0,
      "eval_test-eval_cosine_accuracy@1": 0.8659217877094972,
      "eval_test-eval_cosine_accuracy@10": 1.0,
      "eval_test-eval_cosine_accuracy@3": 0.9916201117318436,
      "eval_test-eval_cosine_accuracy@5": 0.9972067039106145,
      "eval_test-eval_cosine_map@100": 0.9273743016759777,
      "eval_test-eval_cosine_mrr@10": 0.9273743016759775,
      "eval_test-eval_cosine_ndcg@10": 0.9460695277624867,
      "eval_test-eval_cosine_precision@1": 0.8659217877094972,
      "eval_test-eval_cosine_precision@10": 0.09999999999999999,
      "eval_test-eval_cosine_precision@3": 0.33054003724394787,
      "eval_test-eval_cosine_precision@5": 0.1994413407821229,
      "eval_test-eval_cosine_recall@1": 0.8659217877094972,
      "eval_test-eval_cosine_recall@10": 1.0,
      "eval_test-eval_cosine_recall@3": 0.9916201117318436,
      "eval_test-eval_cosine_recall@5": 0.9972067039106145,
      "step": 1500
    },
    {
      "epoch": 17.857142857142858,
      "grad_norm": 0.13586528599262238,
      "learning_rate": 0.0,
      "loss": 0.117,
      "step": 2000
    },
    {
      "epoch": 17.857142857142858,
      "eval_runtime": 1.0029,
      "eval_samples_per_second": 0.0,
      "eval_steps_per_second": 0.0,
      "eval_test-eval_cosine_accuracy@1": 0.8659217877094972,
      "eval_test-eval_cosine_accuracy@10": 1.0,
      "eval_test-eval_cosine_accuracy@3": 0.9916201117318436,
      "eval_test-eval_cosine_accuracy@5": 0.9972067039106145,
      "eval_test-eval_cosine_map@100": 0.9273743016759777,
      "eval_test-eval_cosine_mrr@10": 0.9273743016759775,
      "eval_test-eval_cosine_ndcg@10": 0.9460695277624867,
      "eval_test-eval_cosine_precision@1": 0.8659217877094972,
      "eval_test-eval_cosine_precision@10": 0.09999999999999999,
      "eval_test-eval_cosine_precision@3": 0.33054003724394787,
      "eval_test-eval_cosine_precision@5": 0.1994413407821229,
      "eval_test-eval_cosine_recall@1": 0.8659217877094972,
      "eval_test-eval_cosine_recall@10": 1.0,
      "eval_test-eval_cosine_recall@3": 0.9916201117318436,
      "eval_test-eval_cosine_recall@5": 0.9972067039106145,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}