max_seq_length_128_experiments/Qwen/Qwen1.5_1.8B_amazon/eval_results.json

{
    "epoch": 1.0,
    "eval_accuracy": 0.852437417654809,
    "eval_f1_macro": 0.8289709215944268,
    "eval_f1_micro": 0.852437417654809,
    "eval_loss": 0.5073133111000061,
    "eval_runtime": 3.9722,
    "eval_samples": 1518,
    "eval_samples_per_second": 382.155,
    "eval_steps_per_second": 12.084
}