max_seq_length_128_experiments/Qwen/Qwen1.5_1.8B_scotus/all_results.json

{
    "epoch": 3.0,
    "eval_accuracy": 0.515,
    "eval_f1_macro": 0.35688242427831446,
    "eval_f1_micro": 0.515,
    "eval_loss": 1.6402901411056519,
    "eval_runtime": 3.5882,
    "eval_samples": 1400,
    "eval_samples_per_second": 390.167,
    "eval_steps_per_second": 12.262,
    "test_accuracy": 0.46214285714285713,
    "test_f1_macro": 0.2791167222946888,
    "test_f1_micro": 0.46214285714285713,
    "test_loss": 1.7285268306732178,
    "test_runtime": 3.5996,
    "test_samples_per_second": 388.93,
    "test_steps_per_second": 12.224,
    "train_loss": 1.457581732683121,
    "train_runtime": 520.0867,
    "train_samples": 5000,
    "train_samples_per_second": 28.841,
    "train_steps_per_second": 0.906
}