| { | |
| "best_metric": 0.02781366929411888, | |
| "best_model_checkpoint": "./teapotllm/checkpoint-372", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 5580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.0, | |
| "epoch": 1.0, | |
| "mean_em": 0.13636363636363635, | |
| "mean_similarity": 0.7151702378848286, | |
| "mean_word_count_diff": 4.0606060606060606, | |
| "similarity_boolean": 0.6859854146838188, | |
| "similarity_extraction": 0.6776353690773249, | |
| "similarity_qa": 0.7660255491733551, | |
| "similarity_summarization": 0.7428409457206726, | |
| "similarity_unanswerable": 0.7166690183803439, | |
| "word_count_diff_boolean": 1.9, | |
| "word_count_diff_extraction": 3.1, | |
| "word_count_diff_qa": 4.7, | |
| "word_count_diff_summarization": 19.25, | |
| "word_count_diff_unanswerable": 2.9375 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.1633770614862442, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 0.0597, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.02993270382285118, | |
| "eval_runtime": 4.3173, | |
| "eval_samples_per_second": 15.287, | |
| "eval_steps_per_second": 2.085, | |
| "step": 186 | |
| }, | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.2, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.03125, | |
| "epoch": 2.0, | |
| "mean_em": 0.16666666666666666, | |
| "mean_similarity": 0.6823387349193747, | |
| "mean_word_count_diff": 4.393939393939394, | |
| "similarity_boolean": 0.6429030410945415, | |
| "similarity_extraction": 0.5662426881492137, | |
| "similarity_qa": 0.8177247762680053, | |
| "similarity_summarization": 0.6609117537736893, | |
| "similarity_unanswerable": 0.6913126385770738, | |
| "word_count_diff_boolean": 2.0, | |
| "word_count_diff_extraction": 4.5, | |
| "word_count_diff_qa": 4.4, | |
| "word_count_diff_summarization": 18.25, | |
| "word_count_diff_unanswerable": 3.375 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.11337712407112122, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.0236, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.02781366929411888, | |
| "eval_runtime": 4.3427, | |
| "eval_samples_per_second": 15.198, | |
| "eval_steps_per_second": 2.072, | |
| "step": 372 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.03125, | |
| "epoch": 3.0, | |
| "mean_em": 0.19696969696969696, | |
| "mean_similarity": 0.7136092173556486, | |
| "mean_word_count_diff": 4.0606060606060606, | |
| "similarity_boolean": 0.7808200724422931, | |
| "similarity_extraction": 0.5972317367792129, | |
| "similarity_qa": 0.8595586240291595, | |
| "similarity_summarization": 0.624700665473938, | |
| "similarity_unanswerable": 0.6944781672209501, | |
| "word_count_diff_boolean": 1.2, | |
| "word_count_diff_extraction": 5.3, | |
| "word_count_diff_qa": 3.5, | |
| "word_count_diff_summarization": 18.0, | |
| "word_count_diff_unanswerable": 3.0 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3930008113384247, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.015, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.02964785322546959, | |
| "eval_runtime": 4.3399, | |
| "eval_samples_per_second": 15.208, | |
| "eval_steps_per_second": 2.074, | |
| "step": 558 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.3, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.09375, | |
| "epoch": 4.0, | |
| "mean_em": 0.18181818181818182, | |
| "mean_similarity": 0.7020406423305924, | |
| "mean_word_count_diff": 4.287878787878788, | |
| "similarity_boolean": 0.7741398263722659, | |
| "similarity_extraction": 0.5548319160938263, | |
| "similarity_qa": 0.8464474320411682, | |
| "similarity_summarization": 0.6931064277887344, | |
| "similarity_unanswerable": 0.681502029299736, | |
| "word_count_diff_boolean": 1.4, | |
| "word_count_diff_extraction": 4.6, | |
| "word_count_diff_qa": 3.8, | |
| "word_count_diff_summarization": 18.75, | |
| "word_count_diff_unanswerable": 3.4375 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2295440137386322, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.0102, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.03305710479617119, | |
| "eval_runtime": 4.326, | |
| "eval_samples_per_second": 15.257, | |
| "eval_steps_per_second": 2.08, | |
| "step": 744 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.125, | |
| "epoch": 5.0, | |
| "mean_em": 0.21212121212121213, | |
| "mean_similarity": 0.7231389704075727, | |
| "mean_word_count_diff": 4.363636363636363, | |
| "similarity_boolean": 0.7543415576219559, | |
| "similarity_extraction": 0.5749588221311569, | |
| "similarity_qa": 0.7916842460632324, | |
| "similarity_summarization": 0.8001345098018646, | |
| "similarity_unanswerable": 0.7286496171727777, | |
| "word_count_diff_boolean": 1.6, | |
| "word_count_diff_extraction": 5.9, | |
| "word_count_diff_qa": 6.0, | |
| "word_count_diff_summarization": 15.75, | |
| "word_count_diff_unanswerable": 2.8125 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.16206224262714386, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0073, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.03383969888091087, | |
| "eval_runtime": 4.3253, | |
| "eval_samples_per_second": 15.259, | |
| "eval_steps_per_second": 2.081, | |
| "step": 930 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.09375, | |
| "epoch": 6.0, | |
| "mean_em": 0.18181818181818182, | |
| "mean_similarity": 0.7190481367887873, | |
| "mean_word_count_diff": 3.712121212121212, | |
| "similarity_boolean": 0.7263102680444717, | |
| "similarity_extraction": 0.5906288996338844, | |
| "similarity_qa": 0.8479191601276398, | |
| "similarity_summarization": 0.7078308612108231, | |
| "similarity_unanswerable": 0.7180396970361471, | |
| "word_count_diff_boolean": 1.6, | |
| "word_count_diff_extraction": 4.1, | |
| "word_count_diff_qa": 3.7, | |
| "word_count_diff_summarization": 18.5, | |
| "word_count_diff_unanswerable": 2.40625 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.11092964559793472, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0063, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.036639757454395294, | |
| "eval_runtime": 4.3411, | |
| "eval_samples_per_second": 15.204, | |
| "eval_steps_per_second": 2.073, | |
| "step": 1116 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.09375, | |
| "epoch": 7.0, | |
| "mean_em": 0.19696969696969696, | |
| "mean_similarity": 0.7222502547683138, | |
| "mean_word_count_diff": 3.4696969696969697, | |
| "similarity_boolean": 0.7063896596431732, | |
| "similarity_extraction": 0.613977988064289, | |
| "similarity_qa": 0.8624586880207061, | |
| "similarity_summarization": 0.7022114843130112, | |
| "similarity_unanswerable": 0.7197314850054681, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.8, | |
| "word_count_diff_qa": 4.0, | |
| "word_count_diff_summarization": 18.25, | |
| "word_count_diff_unanswerable": 2.4375 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.0025342688895761967, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 0.0051, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.03565499931573868, | |
| "eval_runtime": 4.3456, | |
| "eval_samples_per_second": 15.188, | |
| "eval_steps_per_second": 2.071, | |
| "step": 1302 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.09375, | |
| "epoch": 8.0, | |
| "mean_em": 0.22727272727272727, | |
| "mean_similarity": 0.7595055665482174, | |
| "mean_word_count_diff": 3.6363636363636362, | |
| "similarity_boolean": 0.827492555975914, | |
| "similarity_extraction": 0.6385330930352211, | |
| "similarity_qa": 0.8492773473262787, | |
| "similarity_summarization": 0.7314918637275696, | |
| "similarity_unanswerable": 0.7515115616843104, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.0, | |
| "word_count_diff_qa": 4.7, | |
| "word_count_diff_summarization": 17.75, | |
| "word_count_diff_unanswerable": 2.875 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.015746578574180603, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.0059, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.03911704570055008, | |
| "eval_runtime": 4.3337, | |
| "eval_samples_per_second": 15.229, | |
| "eval_steps_per_second": 2.077, | |
| "step": 1488 | |
| }, | |
| { | |
| "em_boolean": 0.8, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.1875, | |
| "epoch": 9.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7606699131993633, | |
| "mean_word_count_diff": 3.621212121212121, | |
| "similarity_boolean": 0.8291448082774877, | |
| "similarity_extraction": 0.7344626411795616, | |
| "similarity_qa": 0.8531238198280334, | |
| "similarity_summarization": 0.6869091093540192, | |
| "similarity_unanswerable": 0.7277895356528461, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 2.5, | |
| "word_count_diff_qa": 4.3, | |
| "word_count_diff_summarization": 18.25, | |
| "word_count_diff_unanswerable": 2.53125 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.2743629217147827, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0032, | |
| "step": 1674 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.04249922186136246, | |
| "eval_runtime": 4.3316, | |
| "eval_samples_per_second": 15.237, | |
| "eval_steps_per_second": 2.078, | |
| "step": 1674 | |
| }, | |
| { | |
| "em_boolean": 0.9, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.1875, | |
| "epoch": 10.0, | |
| "mean_em": 0.3181818181818182, | |
| "mean_similarity": 0.7545153351895737, | |
| "mean_word_count_diff": 3.5606060606060606, | |
| "similarity_boolean": 0.9258624315261841, | |
| "similarity_extraction": 0.6548894591629505, | |
| "similarity_qa": 0.7920802533626556, | |
| "similarity_summarization": 0.7080870717763901, | |
| "similarity_unanswerable": 0.7261669498402625, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.5, | |
| "word_count_diff_qa": 4.3, | |
| "word_count_diff_summarization": 17.75, | |
| "word_count_diff_unanswerable": 2.6875 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.19264960289001465, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0023, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.044133350253105164, | |
| "eval_runtime": 4.332, | |
| "eval_samples_per_second": 15.236, | |
| "eval_steps_per_second": 2.078, | |
| "step": 1860 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 11.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7554914088863315, | |
| "mean_word_count_diff": 3.4545454545454546, | |
| "similarity_boolean": 0.8498284339904785, | |
| "similarity_extraction": 0.6388141810894012, | |
| "similarity_qa": 0.8062415808439255, | |
| "similarity_summarization": 0.7252775132656097, | |
| "similarity_unanswerable": 0.7503900304436684, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.4, | |
| "word_count_diff_qa": 5.3, | |
| "word_count_diff_summarization": 15.25, | |
| "word_count_diff_unanswerable": 2.5 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.026405496522784233, | |
| "learning_rate": 3.1666666666666666e-05, | |
| "loss": 0.0018, | |
| "step": 2046 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.04769740626215935, | |
| "eval_runtime": 4.3231, | |
| "eval_samples_per_second": 15.267, | |
| "eval_steps_per_second": 2.082, | |
| "step": 2046 | |
| }, | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.09375, | |
| "epoch": 12.0, | |
| "mean_em": 0.18181818181818182, | |
| "mean_similarity": 0.70488141804482, | |
| "mean_word_count_diff": 3.712121212121212, | |
| "similarity_boolean": 0.6591130249202252, | |
| "similarity_extraction": 0.7003911420702934, | |
| "similarity_qa": 0.7918749868869781, | |
| "similarity_summarization": 0.6370653212070465, | |
| "similarity_unanswerable": 0.7018787739798427, | |
| "word_count_diff_boolean": 1.6, | |
| "word_count_diff_extraction": 3.2, | |
| "word_count_diff_qa": 3.5, | |
| "word_count_diff_summarization": 18.75, | |
| "word_count_diff_unanswerable": 2.71875 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.009582425467669964, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0021, | |
| "step": 2232 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.04564819857478142, | |
| "eval_runtime": 4.3388, | |
| "eval_samples_per_second": 15.211, | |
| "eval_steps_per_second": 2.074, | |
| "step": 2232 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 13.0, | |
| "mean_em": 0.2727272727272727, | |
| "mean_similarity": 0.732733571506811, | |
| "mean_word_count_diff": 3.5303030303030303, | |
| "similarity_boolean": 0.7299912668764591, | |
| "similarity_extraction": 0.6099361270666123, | |
| "similarity_qa": 0.8338733792304993, | |
| "similarity_summarization": 0.6718393042683601, | |
| "similarity_unanswerable": 0.7479703365825117, | |
| "word_count_diff_boolean": 1.4, | |
| "word_count_diff_extraction": 4.2, | |
| "word_count_diff_qa": 3.3, | |
| "word_count_diff_summarization": 16.75, | |
| "word_count_diff_unanswerable": 2.40625 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.015427525155246258, | |
| "learning_rate": 2.8333333333333335e-05, | |
| "loss": 0.006, | |
| "step": 2418 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.0508425235748291, | |
| "eval_runtime": 4.3235, | |
| "eval_samples_per_second": 15.265, | |
| "eval_steps_per_second": 2.082, | |
| "step": 2418 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.1875, | |
| "epoch": 14.0, | |
| "mean_em": 0.2727272727272727, | |
| "mean_similarity": 0.7405690112800309, | |
| "mean_word_count_diff": 3.4242424242424243, | |
| "similarity_boolean": 0.8725614547729492, | |
| "similarity_extraction": 0.6553262293338775, | |
| "similarity_qa": 0.7748158514499665, | |
| "similarity_summarization": 0.7374982982873917, | |
| "similarity_unanswerable": 0.7156414436176419, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.3, | |
| "word_count_diff_qa": 5.1, | |
| "word_count_diff_summarization": 18.0, | |
| "word_count_diff_unanswerable": 2.1875 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.02267725020647049, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.0014, | |
| "step": 2604 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.0475737489759922, | |
| "eval_runtime": 4.3517, | |
| "eval_samples_per_second": 15.166, | |
| "eval_steps_per_second": 2.068, | |
| "step": 2604 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.6, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 15.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7692180305267825, | |
| "mean_word_count_diff": 3.1666666666666665, | |
| "similarity_boolean": 0.7965274572372436, | |
| "similarity_extraction": 0.7734156735241413, | |
| "similarity_qa": 0.8109998881816864, | |
| "similarity_summarization": 0.6901126950979233, | |
| "similarity_unanswerable": 0.7562034076545388, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 1.5, | |
| "word_count_diff_qa": 3.8, | |
| "word_count_diff_summarization": 18.5, | |
| "word_count_diff_unanswerable": 2.5625 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.04372488334774971, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0014, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.04670372232794762, | |
| "eval_runtime": 4.3278, | |
| "eval_samples_per_second": 15.25, | |
| "eval_steps_per_second": 2.08, | |
| "step": 2790 | |
| }, | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.7, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 16.0, | |
| "mean_em": 0.2727272727272727, | |
| "mean_similarity": 0.7683560192923654, | |
| "mean_word_count_diff": 3.5757575757575757, | |
| "similarity_boolean": 0.6552163552492857, | |
| "similarity_extraction": 0.7891500160098076, | |
| "similarity_qa": 0.8301358222961426, | |
| "similarity_summarization": 0.7006907612085342, | |
| "similarity_unanswerable": 0.7863660091534257, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 1.9, | |
| "word_count_diff_qa": 4.9, | |
| "word_count_diff_summarization": 21.25, | |
| "word_count_diff_unanswerable": 2.0625 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.017858723178505898, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.0013, | |
| "step": 2976 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.0506870336830616, | |
| "eval_runtime": 4.3168, | |
| "eval_samples_per_second": 15.289, | |
| "eval_steps_per_second": 2.085, | |
| "step": 2976 | |
| }, | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.3, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 17.0, | |
| "mean_em": 0.21212121212121213, | |
| "mean_similarity": 0.7196075666808721, | |
| "mean_word_count_diff": 3.803030303030303, | |
| "similarity_boolean": 0.6327047817409038, | |
| "similarity_extraction": 0.6252625986933709, | |
| "similarity_qa": 0.8158903121948242, | |
| "similarity_summarization": 0.6718292534351349, | |
| "similarity_unanswerable": 0.7521314206533134, | |
| "word_count_diff_boolean": 1.6, | |
| "word_count_diff_extraction": 3.6, | |
| "word_count_diff_qa": 3.9, | |
| "word_count_diff_summarization": 18.25, | |
| "word_count_diff_unanswerable": 2.71875 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.0030904498416930437, | |
| "learning_rate": 2.1666666666666667e-05, | |
| "loss": 0.0009, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.05454666167497635, | |
| "eval_runtime": 4.3302, | |
| "eval_samples_per_second": 15.242, | |
| "eval_steps_per_second": 2.078, | |
| "step": 3162 | |
| }, | |
| { | |
| "em_boolean": 0.4, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.1875, | |
| "epoch": 18.0, | |
| "mean_em": 0.22727272727272727, | |
| "mean_similarity": 0.7074560875524626, | |
| "mean_word_count_diff": 3.6818181818181817, | |
| "similarity_boolean": 0.6091547535732389, | |
| "similarity_extraction": 0.6446325138211251, | |
| "similarity_qa": 0.8525063633918762, | |
| "similarity_summarization": 0.6083656400442123, | |
| "similarity_unanswerable": 0.7248657159507275, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 2.6, | |
| "word_count_diff_qa": 4.0, | |
| "word_count_diff_summarization": 18.75, | |
| "word_count_diff_unanswerable": 2.65625 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.0558372437953949, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0008, | |
| "step": 3348 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.054993122816085815, | |
| "eval_runtime": 4.3426, | |
| "eval_samples_per_second": 15.198, | |
| "eval_steps_per_second": 2.073, | |
| "step": 3348 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.6, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 19.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7609088637612083, | |
| "mean_word_count_diff": 3.196969696969697, | |
| "similarity_boolean": 0.7781090468168259, | |
| "similarity_extraction": 0.7332883447408676, | |
| "similarity_qa": 0.7839098423719406, | |
| "similarity_summarization": 0.6715002506971359, | |
| "similarity_unanswerable": 0.7681534895673394, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 1.9, | |
| "word_count_diff_qa": 4.7, | |
| "word_count_diff_summarization": 18.25, | |
| "word_count_diff_unanswerable": 2.25 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.2534376084804535, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 0.001, | |
| "step": 3534 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.05550788715481758, | |
| "eval_runtime": 4.3349, | |
| "eval_samples_per_second": 15.225, | |
| "eval_steps_per_second": 2.076, | |
| "step": 3534 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 20.0, | |
| "mean_em": 0.25757575757575757, | |
| "mean_similarity": 0.7586507165070736, | |
| "mean_word_count_diff": 3.5454545454545454, | |
| "similarity_boolean": 0.6818089708685875, | |
| "similarity_extraction": 0.6717077270150185, | |
| "similarity_qa": 0.8592052400112152, | |
| "similarity_summarization": 0.6685295552015305, | |
| "similarity_unanswerable": 0.7896753028035164, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 2.1, | |
| "word_count_diff_qa": 4.4, | |
| "word_count_diff_summarization": 18.0, | |
| "word_count_diff_unanswerable": 2.5 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.011831851676106453, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0008, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.053514618426561356, | |
| "eval_runtime": 4.3467, | |
| "eval_samples_per_second": 15.184, | |
| "eval_steps_per_second": 2.071, | |
| "step": 3720 | |
| }, | |
| { | |
| "em_boolean": 0.8, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 21.0, | |
| "mean_em": 0.30303030303030304, | |
| "mean_similarity": 0.7693727000644712, | |
| "mean_word_count_diff": 3.212121212121212, | |
| "similarity_boolean": 0.8541430443525314, | |
| "similarity_extraction": 0.7111731946468354, | |
| "similarity_qa": 0.8433568000793457, | |
| "similarity_summarization": 0.538779329508543, | |
| "similarity_unanswerable": 0.7667734529823065, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 1.8, | |
| "word_count_diff_qa": 4.5, | |
| "word_count_diff_summarization": 17.0, | |
| "word_count_diff_unanswerable": 2.53125 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.021487198770046234, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0007, | |
| "step": 3906 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.05830617994070053, | |
| "eval_runtime": 4.3417, | |
| "eval_samples_per_second": 15.201, | |
| "eval_steps_per_second": 2.073, | |
| "step": 3906 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.3, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 22.0, | |
| "mean_em": 0.22727272727272727, | |
| "mean_similarity": 0.7141596485267986, | |
| "mean_word_count_diff": 3.5303030303030303, | |
| "similarity_boolean": 0.7537968754768372, | |
| "similarity_extraction": 0.5005216524004936, | |
| "similarity_qa": 0.8255742013454437, | |
| "similarity_summarization": 0.6753722429275513, | |
| "similarity_unanswerable": 0.7385662668384612, | |
| "word_count_diff_boolean": 1.2, | |
| "word_count_diff_extraction": 3.1, | |
| "word_count_diff_qa": 3.6, | |
| "word_count_diff_summarization": 18.75, | |
| "word_count_diff_unanswerable": 2.46875 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.001079858047887683, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0009, | |
| "step": 4092 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.0558372400701046, | |
| "eval_runtime": 4.3409, | |
| "eval_samples_per_second": 15.204, | |
| "eval_steps_per_second": 2.073, | |
| "step": 4092 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 23.0, | |
| "mean_em": 0.24242424242424243, | |
| "mean_similarity": 0.7432461308710503, | |
| "mean_word_count_diff": 3.272727272727273, | |
| "similarity_boolean": 0.7965274572372436, | |
| "similarity_extraction": 0.6360977619886399, | |
| "similarity_qa": 0.8315042436122895, | |
| "similarity_summarization": 0.7279509454965591, | |
| "similarity_unanswerable": 0.7344108195975423, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 3.4, | |
| "word_count_diff_qa": 4.2, | |
| "word_count_diff_summarization": 15.25, | |
| "word_count_diff_unanswerable": 2.46875 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.1313343644142151, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 0.0005, | |
| "step": 4278 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.05678420513868332, | |
| "eval_runtime": 4.3484, | |
| "eval_samples_per_second": 15.178, | |
| "eval_steps_per_second": 2.07, | |
| "step": 4278 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 24.0, | |
| "mean_em": 0.25757575757575757, | |
| "mean_similarity": 0.733292786009384, | |
| "mean_word_count_diff": 3.3333333333333335, | |
| "similarity_boolean": 0.6607738882303238, | |
| "similarity_extraction": 0.6927396818995476, | |
| "similarity_qa": 0.8479106187820434, | |
| "similarity_summarization": 0.6271217167377472, | |
| "similarity_unanswerable": 0.7460810975171626, | |
| "word_count_diff_boolean": 1.4, | |
| "word_count_diff_extraction": 2.5, | |
| "word_count_diff_qa": 4.7, | |
| "word_count_diff_summarization": 17.75, | |
| "word_count_diff_unanswerable": 1.96875 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.00035786593798547983, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0004, | |
| "step": 4464 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.058857183903455734, | |
| "eval_runtime": 4.3529, | |
| "eval_samples_per_second": 15.162, | |
| "eval_steps_per_second": 2.068, | |
| "step": 4464 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 25.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7524795749535164, | |
| "mean_word_count_diff": 3.6515151515151514, | |
| "similarity_boolean": 0.7810821939259768, | |
| "similarity_extraction": 0.6710324764251709, | |
| "similarity_qa": 0.820537006855011, | |
| "similarity_summarization": 0.6545907258987427, | |
| "similarity_unanswerable": 0.75996163347736, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 3.1, | |
| "word_count_diff_qa": 3.8, | |
| "word_count_diff_summarization": 17.75, | |
| "word_count_diff_unanswerable": 2.625 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.2973511815071106, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0006, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.05786564573645592, | |
| "eval_runtime": 4.3502, | |
| "eval_samples_per_second": 15.172, | |
| "eval_steps_per_second": 2.069, | |
| "step": 4650 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.6, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.25, | |
| "epoch": 26.0, | |
| "mean_em": 0.2878787878787879, | |
| "mean_similarity": 0.7584869705817916, | |
| "mean_word_count_diff": 3.0303030303030303, | |
| "similarity_boolean": 0.7514585584402085, | |
| "similarity_extraction": 0.7340315699577331, | |
| "similarity_qa": 0.8186799943447113, | |
| "similarity_summarization": 0.7270446717739105, | |
| "similarity_unanswerable": 0.7534456294961274, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 2.4, | |
| "word_count_diff_qa": 4.2, | |
| "word_count_diff_summarization": 17.0, | |
| "word_count_diff_unanswerable": 2.0625 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.03420831263065338, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0005, | |
| "step": 4836 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.05816643312573433, | |
| "eval_runtime": 4.3476, | |
| "eval_samples_per_second": 15.181, | |
| "eval_steps_per_second": 2.07, | |
| "step": 4836 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 27.0, | |
| "mean_em": 0.25757575757575757, | |
| "mean_similarity": 0.7683728897210323, | |
| "mean_word_count_diff": 3.227272727272727, | |
| "similarity_boolean": 0.823177945613861, | |
| "similarity_extraction": 0.674461716413498, | |
| "similarity_qa": 0.8566352903842926, | |
| "similarity_summarization": 0.6613511592149734, | |
| "similarity_unanswerable": 0.7663892675191164, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 2.8, | |
| "word_count_diff_qa": 4.6, | |
| "word_count_diff_summarization": 17.25, | |
| "word_count_diff_unanswerable": 2.1875 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 0.20164477825164795, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0006, | |
| "step": 5022 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.05884123966097832, | |
| "eval_runtime": 4.3217, | |
| "eval_samples_per_second": 15.272, | |
| "eval_steps_per_second": 2.083, | |
| "step": 5022 | |
| }, | |
| { | |
| "em_boolean": 0.7, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.21875, | |
| "epoch": 28.0, | |
| "mean_em": 0.30303030303030304, | |
| "mean_similarity": 0.7534916117561586, | |
| "mean_word_count_diff": 3.378787878787879, | |
| "similarity_boolean": 0.8725614547729492, | |
| "similarity_extraction": 0.6645052798092366, | |
| "similarity_qa": 0.8722499251365662, | |
| "similarity_summarization": 0.6813686788082123, | |
| "similarity_unanswerable": 0.7159939082339406, | |
| "word_count_diff_boolean": 0.0, | |
| "word_count_diff_extraction": 4.4, | |
| "word_count_diff_qa": 3.3, | |
| "word_count_diff_summarization": 15.75, | |
| "word_count_diff_unanswerable": 2.59375 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.09254211187362671, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0005, | |
| "step": 5208 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.05950010567903519, | |
| "eval_runtime": 4.3472, | |
| "eval_samples_per_second": 15.182, | |
| "eval_steps_per_second": 2.07, | |
| "step": 5208 | |
| }, | |
| { | |
| "em_boolean": 0.3, | |
| "em_extraction": 0.5, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.15625, | |
| "epoch": 29.0, | |
| "mean_em": 0.19696969696969696, | |
| "mean_similarity": 0.7206367963755672, | |
| "mean_word_count_diff": 3.257575757575758, | |
| "similarity_boolean": 0.5599547553807497, | |
| "similarity_extraction": 0.6752738565206527, | |
| "similarity_qa": 0.826910001039505, | |
| "similarity_summarization": 0.6975099295377731, | |
| "similarity_unanswerable": 0.7547063347883523, | |
| "word_count_diff_boolean": 2.1, | |
| "word_count_diff_extraction": 3.4, | |
| "word_count_diff_qa": 3.3, | |
| "word_count_diff_summarization": 15.5, | |
| "word_count_diff_unanswerable": 2.03125 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.002610220806673169, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.0003, | |
| "step": 5394 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.05941936746239662, | |
| "eval_runtime": 4.3317, | |
| "eval_samples_per_second": 15.236, | |
| "eval_steps_per_second": 2.078, | |
| "step": 5394 | |
| }, | |
| { | |
| "em_boolean": 0.6, | |
| "em_extraction": 0.4, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.1875, | |
| "epoch": 30.0, | |
| "mean_em": 0.24242424242424243, | |
| "mean_similarity": 0.7168677732574217, | |
| "mean_word_count_diff": 4.393939393939394, | |
| "similarity_boolean": 0.7321859009563922, | |
| "similarity_extraction": 0.5674642950296402, | |
| "similarity_qa": 0.8223672747612, | |
| "similarity_summarization": 0.6349779516458511, | |
| "similarity_unanswerable": 0.7360370787791908, | |
| "word_count_diff_boolean": 1.7, | |
| "word_count_diff_extraction": 5.9, | |
| "word_count_diff_qa": 4.9, | |
| "word_count_diff_summarization": 18.0, | |
| "word_count_diff_unanswerable": 2.90625 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.08226452767848969, | |
| "learning_rate": 0.0, | |
| "loss": 0.0004, | |
| "step": 5580 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.05968955159187317, | |
| "eval_runtime": 4.3488, | |
| "eval_samples_per_second": 15.177, | |
| "eval_steps_per_second": 2.07, | |
| "step": 5580 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5580, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.572121033146368e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |