Qwen2-7B_pct_ortho_r16/checkpoint-32/trainer_state.json

{
  "best_metric": 1.9184670448303223,
  "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_pct_ortho_r16/checkpoint-32",
  "epoch": 0.08244766505636071,
  "eval_steps": 8,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025764895330112722,
      "grad_norm": 2.5038321018218994,
      "learning_rate": 1.25e-05,
      "loss": 2.0463,
      "step": 1
    },
    {
      "epoch": 0.010305958132045089,
      "grad_norm": 2.0884063243865967,
      "learning_rate": 5e-05,
      "loss": 2.1379,
      "step": 4
    },
    {
      "epoch": 0.020611916264090178,
      "grad_norm": 2.1778032779693604,
      "learning_rate": 0.0001,
      "loss": 2.0486,
      "step": 8
    },
    {
      "epoch": 0.020611916264090178,
      "eval_loss": 1.997605562210083,
      "eval_runtime": 130.7432,
      "eval_samples_per_second": 1.874,
      "eval_steps_per_second": 0.941,
      "step": 8
    },
    {
      "epoch": 0.030917874396135265,
      "grad_norm": 1.3578734397888184,
      "learning_rate": 9.997266286704631e-05,
      "loss": 2.0017,
      "step": 12
    },
    {
      "epoch": 0.041223832528180356,
      "grad_norm": 1.0907150506973267,
      "learning_rate": 9.989068136093873e-05,
      "loss": 1.9839,
      "step": 16
    },
    {
      "epoch": 0.041223832528180356,
      "eval_loss": 1.9347867965698242,
      "eval_runtime": 133.851,
      "eval_samples_per_second": 1.83,
      "eval_steps_per_second": 0.919,
      "step": 16
    },
    {
      "epoch": 0.05152979066022544,
      "grad_norm": 1.1747814416885376,
      "learning_rate": 9.975414512725057e-05,
      "loss": 1.9421,
      "step": 20
    },
    {
      "epoch": 0.06183574879227053,
      "grad_norm": 1.2741020917892456,
      "learning_rate": 9.956320346634876e-05,
      "loss": 2.0083,
      "step": 24
    },
    {
      "epoch": 0.06183574879227053,
      "eval_loss": 1.9231261014938354,
      "eval_runtime": 123.8213,
      "eval_samples_per_second": 1.979,
      "eval_steps_per_second": 0.993,
      "step": 24
    },
    {
      "epoch": 0.07214170692431562,
      "grad_norm": 1.1227424144744873,
      "learning_rate": 9.931806517013612e-05,
      "loss": 1.9946,
      "step": 28
    },
    {
      "epoch": 0.08244766505636071,
      "grad_norm": 1.14728581905365,
      "learning_rate": 9.901899829374047e-05,
      "loss": 1.923,
      "step": 32
    },
    {
      "epoch": 0.08244766505636071,
      "eval_loss": 1.9184670448303223,
      "eval_runtime": 236.0583,
      "eval_samples_per_second": 1.038,
      "eval_steps_per_second": 0.521,
      "step": 32
    }
  ],
  "logging_steps": 4,
  "max_steps": 388,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.483746134414131e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}