{
    "rl_info/A2G": -0.006648760288953781,
    "rl_info/entropy": 3.1247427463531494,
    "rl_info/total_token": 1710.0,
    "rl_info/advantage_b4_norm": -524.6408081054688,
    "rl_info/advantage_after_gnorm": 1.010803461074829,
    "rl_info/kl_w_ref": 0.0,
    "train/rl_loss": 0.6645635962486267,
    "train/lm_loss": 6.4453349113464355,
    "train/total_loss": 7.109898567199707,
    "samsum/rouge1": 0.07018707411132723,
    "samsum/rouge2": 0.015589278392518668,
    "samsum/rougeL": 0.05694806401174433,
    "samsum/rougeLsum": 0.05097452753746131,
    "samsum/bertscore_precision": 0.6306706219911575,
    "samsum/bertscore_recall": 0.7116748541593552,
    "samsum/bertscore_f1": 0.6681777884562811,
    "xsum/rouge1": 0.07577425197474222,
    "xsum/rouge2": 0.015173697537833916,
    "xsum/rougeL": 0.06791431460162027,
    "xsum/rougeLsum": 0.06226845688932118,
    "xsum/bertscore_precision": 0.685419961810112,
    "xsum/bertscore_recall": 0.6998258779446284,
    "xsum/bertscore_f1": 0.6894047458966573,
    "gigaword/rouge1": 0.01673533809218624,
    "gigaword/rouge2": 0.002688706117978111,
    "gigaword/rougeL": 0.016448820149720444,
    "gigaword/rougeLsum": 0.015425704010653109,
    "gigaword/bertscore_precision": 0.5675247123837471,
    "gigaword/bertscore_recall": 0.6473609921336174,
    "gigaword/bertscore_f1": 0.6041367189586162,
    "cnndm/rouge1": 0.12628644770575717,
    "cnndm/rouge2": 0.04433825429137848,
    "cnndm/rougeL": 0.09875398462560836,
    "cnndm/rougeLsum": 0.10678645281701828,
    "cnndm/bertscore_precision": 0.6603378802537918,
    "cnndm/bertscore_recall": 0.7300801773866018,
    "cnndm/bertscore_f1": 0.6927102555831274,
    "eval_agg/avg_all_rougef": 0.052643335804179334,
    "eval_agg/avg_all_bertf": 0.6636073772236705,
    "eval_agg/avg_all": 0.3581253565139249,
    "num_rl_rollout": 4,
    "lm_epoch": 0,
    "rl_epoch": 0,
    "step": 400,
    "total_data_token": 668982,
    "total_rl_token": 849448,
    "total_lm_token": 625452,
    "total_token": 1474900,
    "completed_steps": 400,
    "tune_objective": 0.784820971275843
}