checkpoint-50/trainer_state.json

{
  "best_metric": 0.40609535574913025,
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-50",
  "epoch": 0.49504950495049505,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "eval_loss": 0.6292536854743958,
      "eval_runtime": 93.2576,
      "eval_samples_per_second": 30.743,
      "eval_steps_per_second": 0.965,
      "step": 10
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.24918170273303986,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.9604,
      "step": 20
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.46683359146118164,
      "eval_runtime": 93.2154,
      "eval_samples_per_second": 30.757,
      "eval_steps_per_second": 0.966,
      "step": 20
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.4341636002063751,
      "eval_runtime": 93.1765,
      "eval_samples_per_second": 30.77,
      "eval_steps_per_second": 0.966,
      "step": 30
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.16474689543247223,
      "learning_rate": 0.0001529919264233205,
      "loss": 0.4381,
      "step": 40
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.40781059861183167,
      "eval_runtime": 93.1933,
      "eval_samples_per_second": 30.764,
      "eval_steps_per_second": 0.966,
      "step": 40
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.40609535574913025,
      "eval_runtime": 93.1926,
      "eval_samples_per_second": 30.764,
      "eval_steps_per_second": 0.966,
      "step": 50
    }
  ],
  "logging_steps": 20,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 4.5255742262450586e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}