| { |
| "best_global_step": 18252, |
| "best_metric": 0.8571, |
| "best_model_checkpoint": "models/NED/SPACCC_human_only_tfidf_hybrid_long_v2_addheaders/Llama-3.2-1B-Instruct/checkpoint-18252", |
| "epoch": 50.0, |
| "eval_steps": 500, |
| "global_step": 304200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.1268050391018822, |
| "epoch": 1.0, |
| "grad_norm": 71.5, |
| "learning_rate": 1.999671268902038e-05, |
| "loss": 1.1856, |
| "mean_token_accuracy": 0.7717264709329229, |
| "num_tokens": 23190253.0, |
| "step": 6084 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_entropy": 2.1406238079071045, |
| "eval_loss": 0.4971363842487335, |
| "eval_mean_token_accuracy": 0.9041083097457886, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 23190253.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.1758, |
| "eval_samples_per_second": 79.649, |
| "eval_steps_per_second": 28.446, |
| "step": 6084 |
| }, |
| { |
| "entropy": 1.9585949286358049, |
| "epoch": 2.0, |
| "grad_norm": 32.5, |
| "learning_rate": 2.9690823318896277e-05, |
| "loss": 0.654, |
| "mean_token_accuracy": 0.8564344449832124, |
| "num_tokens": 46380506.0, |
| "step": 12168 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_entropy": 1.99363112449646, |
| "eval_loss": 0.4503121078014374, |
| "eval_mean_token_accuracy": 0.901867413520813, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 46380506.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1565, |
| "eval_samples_per_second": 89.474, |
| "eval_steps_per_second": 31.955, |
| "step": 12168 |
| }, |
| { |
| "entropy": 1.6538763878221656, |
| "epoch": 3.0, |
| "grad_norm": 163.0, |
| "learning_rate": 2.9072266617865347e-05, |
| "loss": 0.3304, |
| "mean_token_accuracy": 0.9195070968381204, |
| "num_tokens": 69570759.0, |
| "step": 18252 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_entropy": 1.6258357524871827, |
| "eval_loss": 0.527018129825592, |
| "eval_mean_token_accuracy": 0.9253968238830567, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 69570759.0, |
| "eval_recall": 0.8571, |
| "eval_runtime": 0.1557, |
| "eval_samples_per_second": 89.896, |
| "eval_steps_per_second": 32.106, |
| "step": 18252 |
| }, |
| { |
| "entropy": 1.458226298067148, |
| "epoch": 4.0, |
| "grad_norm": 0.09423828125, |
| "learning_rate": 2.845370991683442e-05, |
| "loss": 0.1547, |
| "mean_token_accuracy": 0.9605004129383926, |
| "num_tokens": 92761012.0, |
| "step": 24336 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_entropy": 1.5267343997955323, |
| "eval_loss": 0.43075770139694214, |
| "eval_mean_token_accuracy": 0.9333333253860474, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 92761012.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.1616, |
| "eval_samples_per_second": 86.629, |
| "eval_steps_per_second": 30.939, |
| "step": 24336 |
| }, |
| { |
| "entropy": 1.3115615775560094, |
| "epoch": 5.0, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 2.7835153215803492e-05, |
| "loss": 0.0645, |
| "mean_token_accuracy": 0.9830160907994439, |
| "num_tokens": 115951265.0, |
| "step": 30420 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_entropy": 1.3103554964065551, |
| "eval_loss": 0.6741493940353394, |
| "eval_mean_token_accuracy": 0.8914098978042603, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 115951265.0, |
| "eval_recall": 0.6429, |
| "eval_runtime": 0.1548, |
| "eval_samples_per_second": 90.451, |
| "eval_steps_per_second": 32.304, |
| "step": 30420 |
| }, |
| { |
| "entropy": 1.2145140489255966, |
| "epoch": 6.0, |
| "grad_norm": 0.051513671875, |
| "learning_rate": 2.7216596514772566e-05, |
| "loss": 0.0227, |
| "mean_token_accuracy": 0.9940000502178341, |
| "num_tokens": 139141518.0, |
| "step": 36504 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_entropy": 1.287993311882019, |
| "eval_loss": 0.6404770612716675, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 139141518.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1536, |
| "eval_samples_per_second": 91.127, |
| "eval_steps_per_second": 32.545, |
| "step": 36504 |
| }, |
| { |
| "entropy": 1.1663444105885987, |
| "epoch": 7.0, |
| "grad_norm": 0.177734375, |
| "learning_rate": 2.659803981374164e-05, |
| "loss": 0.0072, |
| "mean_token_accuracy": 0.9983230459012145, |
| "num_tokens": 162331771.0, |
| "step": 42588 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_entropy": 1.220553469657898, |
| "eval_loss": 0.7181684374809265, |
| "eval_mean_token_accuracy": 0.9009337067604065, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 162331771.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1554, |
| "eval_samples_per_second": 90.075, |
| "eval_steps_per_second": 32.17, |
| "step": 42588 |
| }, |
| { |
| "entropy": 1.1479155327304267, |
| "epoch": 8.0, |
| "grad_norm": 0.0177001953125, |
| "learning_rate": 2.597948311271071e-05, |
| "loss": 0.0018, |
| "mean_token_accuracy": 0.9995507592730896, |
| "num_tokens": 185522024.0, |
| "step": 48672 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_entropy": 1.2225802183151244, |
| "eval_loss": 0.8296155333518982, |
| "eval_mean_token_accuracy": 0.9009337067604065, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 185522024.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1539, |
| "eval_samples_per_second": 90.954, |
| "eval_steps_per_second": 32.484, |
| "step": 48672 |
| }, |
| { |
| "entropy": 1.1291984924158716, |
| "epoch": 9.0, |
| "grad_norm": 0.04833984375, |
| "learning_rate": 2.5360926411679782e-05, |
| "loss": 0.0006, |
| "mean_token_accuracy": 0.9999462363752858, |
| "num_tokens": 208712277.0, |
| "step": 54756 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_entropy": 1.1875993251800536, |
| "eval_loss": 0.8487787842750549, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 208712277.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1586, |
| "eval_samples_per_second": 88.285, |
| "eval_steps_per_second": 31.53, |
| "step": 54756 |
| }, |
| { |
| "entropy": 1.1119468484776809, |
| "epoch": 10.0, |
| "grad_norm": 0.05029296875, |
| "learning_rate": 2.4742369710648856e-05, |
| "loss": 0.0003, |
| "mean_token_accuracy": 0.9999655756920596, |
| "num_tokens": 231902530.0, |
| "step": 60840 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_entropy": 1.1920786857604981, |
| "eval_loss": 0.8497948050498962, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 231902530.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1574, |
| "eval_samples_per_second": 88.919, |
| "eval_steps_per_second": 31.757, |
| "step": 60840 |
| }, |
| { |
| "entropy": 1.1129700567262255, |
| "epoch": 11.0, |
| "grad_norm": 0.0830078125, |
| "learning_rate": 2.4123813009617927e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 255092783.0, |
| "step": 66924 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_entropy": 1.1868632793426515, |
| "eval_loss": 0.8353831171989441, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 255092783.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1588, |
| "eval_samples_per_second": 88.164, |
| "eval_steps_per_second": 31.487, |
| "step": 66924 |
| }, |
| { |
| "entropy": 1.1077099534769266, |
| "epoch": 12.0, |
| "grad_norm": 0.07568359375, |
| "learning_rate": 2.3505256308586998e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 0.9999945211437954, |
| "num_tokens": 278283036.0, |
| "step": 73008 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_entropy": 1.1834761381149292, |
| "eval_loss": 0.8413147330284119, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 278283036.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1558, |
| "eval_samples_per_second": 89.841, |
| "eval_steps_per_second": 32.086, |
| "step": 73008 |
| }, |
| { |
| "entropy": 1.1058313136701126, |
| "epoch": 13.0, |
| "grad_norm": 0.0159912109375, |
| "learning_rate": 2.2886699607556072e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 301473289.0, |
| "step": 79092 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_entropy": 1.1844575166702271, |
| "eval_loss": 0.8453251123428345, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 301473289.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1604, |
| "eval_samples_per_second": 87.297, |
| "eval_steps_per_second": 31.177, |
| "step": 79092 |
| }, |
| { |
| "entropy": 1.106135592432809, |
| "epoch": 14.0, |
| "grad_norm": 0.11328125, |
| "learning_rate": 2.2268142906525143e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 324663542.0, |
| "step": 85176 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_entropy": 1.1829971313476562, |
| "eval_loss": 0.8412962555885315, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 324663542.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1597, |
| "eval_samples_per_second": 87.671, |
| "eval_steps_per_second": 31.311, |
| "step": 85176 |
| }, |
| { |
| "entropy": 1.1052913846525214, |
| "epoch": 15.0, |
| "grad_norm": 0.06689453125, |
| "learning_rate": 2.1649586205494213e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 347853795.0, |
| "step": 91260 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_entropy": 1.183160948753357, |
| "eval_loss": 0.8468051552772522, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 347853795.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1548, |
| "eval_samples_per_second": 90.433, |
| "eval_steps_per_second": 32.298, |
| "step": 91260 |
| }, |
| { |
| "entropy": 1.1050384549973906, |
| "epoch": 16.0, |
| "grad_norm": 0.061279296875, |
| "learning_rate": 2.1031029504463287e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 371044048.0, |
| "step": 97344 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_entropy": 1.183035182952881, |
| "eval_loss": 0.8468660712242126, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 371044048.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1566, |
| "eval_samples_per_second": 89.398, |
| "eval_steps_per_second": 31.928, |
| "step": 97344 |
| }, |
| { |
| "entropy": 1.1046691402685165, |
| "epoch": 17.0, |
| "grad_norm": 0.0390625, |
| "learning_rate": 2.0412472803432358e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 394234301.0, |
| "step": 103428 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_entropy": 1.1834327936172486, |
| "eval_loss": 0.8357403874397278, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 394234301.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1667, |
| "eval_samples_per_second": 83.986, |
| "eval_steps_per_second": 29.995, |
| "step": 103428 |
| }, |
| { |
| "entropy": 1.1045704223289057, |
| "epoch": 18.0, |
| "grad_norm": 0.0034637451171875, |
| "learning_rate": 1.979391610240143e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 417424554.0, |
| "step": 109512 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_entropy": 1.1848695039749146, |
| "eval_loss": 0.8536882400512695, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 417424554.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1648, |
| "eval_samples_per_second": 84.942, |
| "eval_steps_per_second": 30.337, |
| "step": 109512 |
| }, |
| { |
| "entropy": 1.1047596051495456, |
| "epoch": 19.0, |
| "grad_norm": 0.10498046875, |
| "learning_rate": 1.9175359401370503e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 440614807.0, |
| "step": 115596 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_entropy": 1.1839102029800415, |
| "eval_loss": 0.8556452989578247, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 440614807.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1644, |
| "eval_samples_per_second": 85.182, |
| "eval_steps_per_second": 30.422, |
| "step": 115596 |
| }, |
| { |
| "entropy": 1.1042878528436024, |
| "epoch": 20.0, |
| "grad_norm": 0.07373046875, |
| "learning_rate": 1.8556802700339577e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 463805060.0, |
| "step": 121680 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_entropy": 1.183977437019348, |
| "eval_loss": 0.8611962199211121, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 463805060.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1597, |
| "eval_samples_per_second": 87.654, |
| "eval_steps_per_second": 31.305, |
| "step": 121680 |
| }, |
| { |
| "entropy": 1.104654125815391, |
| "epoch": 21.0, |
| "grad_norm": 0.0478515625, |
| "learning_rate": 1.793824599930865e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 486995313.0, |
| "step": 127764 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_entropy": 1.1869411706924438, |
| "eval_loss": 0.8467808961868286, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 486995313.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1555, |
| "eval_samples_per_second": 90.033, |
| "eval_steps_per_second": 32.155, |
| "step": 127764 |
| }, |
| { |
| "entropy": 1.1043770736785095, |
| "epoch": 22.0, |
| "grad_norm": 0.040771484375, |
| "learning_rate": 1.7319689298277722e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 510185566.0, |
| "step": 133848 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_entropy": 1.1831506013870239, |
| "eval_loss": 0.8473494648933411, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 510185566.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.153, |
| "eval_samples_per_second": 91.486, |
| "eval_steps_per_second": 32.674, |
| "step": 133848 |
| }, |
| { |
| "entropy": 1.1042385918384003, |
| "epoch": 23.0, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.6701132597246793e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 533375819.0, |
| "step": 139932 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_entropy": 1.1850621223449707, |
| "eval_loss": 0.8552234768867493, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 533375819.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1571, |
| "eval_samples_per_second": 89.113, |
| "eval_steps_per_second": 31.826, |
| "step": 139932 |
| }, |
| { |
| "entropy": 1.104157416445107, |
| "epoch": 24.0, |
| "grad_norm": 0.2021484375, |
| "learning_rate": 1.6082575896215867e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 556566072.0, |
| "step": 146016 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_entropy": 1.186206865310669, |
| "eval_loss": 0.8547914624214172, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 556566072.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1564, |
| "eval_samples_per_second": 89.51, |
| "eval_steps_per_second": 31.968, |
| "step": 146016 |
| }, |
| { |
| "entropy": 1.1040764231673201, |
| "epoch": 25.0, |
| "grad_norm": 0.0257568359375, |
| "learning_rate": 1.5464019195184938e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 579756325.0, |
| "step": 152100 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_entropy": 1.1851461410522461, |
| "eval_loss": 0.8527544736862183, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 579756325.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1568, |
| "eval_samples_per_second": 89.303, |
| "eval_steps_per_second": 31.894, |
| "step": 152100 |
| }, |
| { |
| "entropy": 1.1041492314165473, |
| "epoch": 26.0, |
| "grad_norm": 0.06494140625, |
| "learning_rate": 1.4845462494154008e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 602946578.0, |
| "step": 158184 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_entropy": 1.1824380159378052, |
| "eval_loss": 0.8571881055831909, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 602946578.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1542, |
| "eval_samples_per_second": 90.8, |
| "eval_steps_per_second": 32.428, |
| "step": 158184 |
| }, |
| { |
| "entropy": 1.1045205590463953, |
| "epoch": 27.0, |
| "grad_norm": 0.05517578125, |
| "learning_rate": 1.4226905793123081e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 626136831.0, |
| "step": 164268 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_entropy": 1.1850265502929687, |
| "eval_loss": 0.8606659173965454, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 626136831.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1562, |
| "eval_samples_per_second": 89.64, |
| "eval_steps_per_second": 32.014, |
| "step": 164268 |
| }, |
| { |
| "entropy": 1.1041291631968848, |
| "epoch": 28.0, |
| "grad_norm": 0.130859375, |
| "learning_rate": 1.3608349092092153e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 649327084.0, |
| "step": 170352 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_entropy": 1.186257529258728, |
| "eval_loss": 0.854324996471405, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 649327084.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1578, |
| "eval_samples_per_second": 88.707, |
| "eval_steps_per_second": 31.681, |
| "step": 170352 |
| }, |
| { |
| "entropy": 1.1043956205885634, |
| "epoch": 29.0, |
| "grad_norm": 0.004913330078125, |
| "learning_rate": 1.2989792391061224e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 672517337.0, |
| "step": 176436 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_entropy": 1.185477328300476, |
| "eval_loss": 0.8509793877601624, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 672517337.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1565, |
| "eval_samples_per_second": 89.43, |
| "eval_steps_per_second": 31.939, |
| "step": 176436 |
| }, |
| { |
| "entropy": 1.1042519219508695, |
| "epoch": 30.0, |
| "grad_norm": 0.12158203125, |
| "learning_rate": 1.2371235690030298e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 695707590.0, |
| "step": 182520 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_entropy": 1.1846540451049805, |
| "eval_loss": 0.8473905920982361, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 695707590.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.155, |
| "eval_samples_per_second": 90.317, |
| "eval_steps_per_second": 32.256, |
| "step": 182520 |
| }, |
| { |
| "entropy": 1.104193890843244, |
| "epoch": 31.0, |
| "grad_norm": 0.028564453125, |
| "learning_rate": 1.175267898899937e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 718897843.0, |
| "step": 188604 |
| }, |
| { |
| "epoch": 31.0, |
| "eval_entropy": 1.1850831031799316, |
| "eval_loss": 0.8546524047851562, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 718897843.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.165, |
| "eval_samples_per_second": 84.858, |
| "eval_steps_per_second": 30.306, |
| "step": 188604 |
| }, |
| { |
| "entropy": 1.1044594693799148, |
| "epoch": 32.0, |
| "grad_norm": 0.01507568359375, |
| "learning_rate": 1.1134122287968443e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 742088096.0, |
| "step": 194688 |
| }, |
| { |
| "epoch": 32.0, |
| "eval_entropy": 1.1852279663085938, |
| "eval_loss": 0.8612449765205383, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 742088096.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1576, |
| "eval_samples_per_second": 88.852, |
| "eval_steps_per_second": 31.733, |
| "step": 194688 |
| }, |
| { |
| "entropy": 1.1042602037537028, |
| "epoch": 33.0, |
| "grad_norm": 0.049072265625, |
| "learning_rate": 1.0515565586937514e-05, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 765278349.0, |
| "step": 200772 |
| }, |
| { |
| "epoch": 33.0, |
| "eval_entropy": 1.1850273609161377, |
| "eval_loss": 0.854258120059967, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 765278349.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1569, |
| "eval_samples_per_second": 89.235, |
| "eval_steps_per_second": 31.869, |
| "step": 200772 |
| }, |
| { |
| "entropy": 1.1045278703633459, |
| "epoch": 34.0, |
| "grad_norm": 0.1767578125, |
| "learning_rate": 9.897008885906586e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 788468602.0, |
| "step": 206856 |
| }, |
| { |
| "epoch": 34.0, |
| "eval_entropy": 1.1850582599639892, |
| "eval_loss": 0.850601077079773, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 788468602.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1584, |
| "eval_samples_per_second": 88.362, |
| "eval_steps_per_second": 31.558, |
| "step": 206856 |
| }, |
| { |
| "entropy": 1.104611723027991, |
| "epoch": 35.0, |
| "grad_norm": 0.033203125, |
| "learning_rate": 9.278452184875659e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 811658855.0, |
| "step": 212940 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_entropy": 1.1844811916351319, |
| "eval_loss": 0.8560983538627625, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 811658855.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1556, |
| "eval_samples_per_second": 89.96, |
| "eval_steps_per_second": 32.129, |
| "step": 212940 |
| }, |
| { |
| "entropy": 1.104373964606579, |
| "epoch": 36.0, |
| "grad_norm": 0.0027313232421875, |
| "learning_rate": 8.65989548384473e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 834849108.0, |
| "step": 219024 |
| }, |
| { |
| "epoch": 36.0, |
| "eval_entropy": 1.1842174053192138, |
| "eval_loss": 0.8440762758255005, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 834849108.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1546, |
| "eval_samples_per_second": 90.57, |
| "eval_steps_per_second": 32.346, |
| "step": 219024 |
| }, |
| { |
| "entropy": 1.1041848019186717, |
| "epoch": 37.0, |
| "grad_norm": 0.06689453125, |
| "learning_rate": 8.041338782813804e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 858039361.0, |
| "step": 225108 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_entropy": 1.1856298685073852, |
| "eval_loss": 0.8591736555099487, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 858039361.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1682, |
| "eval_samples_per_second": 83.241, |
| "eval_steps_per_second": 29.729, |
| "step": 225108 |
| }, |
| { |
| "entropy": 1.1047284195467801, |
| "epoch": 38.0, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 7.4227820817828744e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 881229614.0, |
| "step": 231192 |
| }, |
| { |
| "epoch": 38.0, |
| "eval_entropy": 1.1842705011367798, |
| "eval_loss": 0.8513706922531128, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 881229614.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1567, |
| "eval_samples_per_second": 89.326, |
| "eval_steps_per_second": 31.902, |
| "step": 231192 |
| }, |
| { |
| "entropy": 1.1047490643071627, |
| "epoch": 39.0, |
| "grad_norm": 0.025634765625, |
| "learning_rate": 6.804225380751948e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 904419867.0, |
| "step": 237276 |
| }, |
| { |
| "epoch": 39.0, |
| "eval_entropy": 1.1818478107452393, |
| "eval_loss": 0.8533371090888977, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 904419867.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1577, |
| "eval_samples_per_second": 88.773, |
| "eval_steps_per_second": 31.705, |
| "step": 237276 |
| }, |
| { |
| "entropy": 1.1046844204051334, |
| "epoch": 40.0, |
| "grad_norm": 0.03466796875, |
| "learning_rate": 6.185668679721019e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 927610120.0, |
| "step": 243360 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_entropy": 1.1807388305664062, |
| "eval_loss": 0.8424216508865356, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 927610120.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.157, |
| "eval_samples_per_second": 89.195, |
| "eval_steps_per_second": 31.855, |
| "step": 243360 |
| }, |
| { |
| "entropy": 1.1044822435563995, |
| "epoch": 41.0, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 5.567111978690091e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 950800373.0, |
| "step": 249444 |
| }, |
| { |
| "epoch": 41.0, |
| "eval_entropy": 1.1833556652069093, |
| "eval_loss": 0.8545311093330383, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 950800373.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1548, |
| "eval_samples_per_second": 90.464, |
| "eval_steps_per_second": 32.309, |
| "step": 249444 |
| }, |
| { |
| "entropy": 1.1045338336101571, |
| "epoch": 42.0, |
| "grad_norm": 0.028564453125, |
| "learning_rate": 4.948555277659164e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 973990626.0, |
| "step": 255528 |
| }, |
| { |
| "epoch": 42.0, |
| "eval_entropy": 1.185540795326233, |
| "eval_loss": 0.848736584186554, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 973990626.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1564, |
| "eval_samples_per_second": 89.51, |
| "eval_steps_per_second": 31.968, |
| "step": 255528 |
| }, |
| { |
| "entropy": 1.104217090569853, |
| "epoch": 43.0, |
| "grad_norm": 0.01708984375, |
| "learning_rate": 4.329998576628236e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 997180879.0, |
| "step": 261612 |
| }, |
| { |
| "epoch": 43.0, |
| "eval_entropy": 1.1844725608825684, |
| "eval_loss": 0.8541809320449829, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 997180879.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1566, |
| "eval_samples_per_second": 89.418, |
| "eval_steps_per_second": 31.935, |
| "step": 261612 |
| }, |
| { |
| "entropy": 1.1048875654276697, |
| "epoch": 44.0, |
| "grad_norm": 0.068359375, |
| "learning_rate": 3.711441875597308e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1020371132.0, |
| "step": 267696 |
| }, |
| { |
| "epoch": 44.0, |
| "eval_entropy": 1.1849848747253418, |
| "eval_loss": 0.8501148223876953, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1020371132.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.156, |
| "eval_samples_per_second": 89.746, |
| "eval_steps_per_second": 32.052, |
| "step": 267696 |
| }, |
| { |
| "entropy": 1.1046531631180834, |
| "epoch": 45.0, |
| "grad_norm": 0.1015625, |
| "learning_rate": 3.09288517456638e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1043561385.0, |
| "step": 273780 |
| }, |
| { |
| "epoch": 45.0, |
| "eval_entropy": 1.183105206489563, |
| "eval_loss": 0.8626060485839844, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1043561385.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1555, |
| "eval_samples_per_second": 90.048, |
| "eval_steps_per_second": 32.16, |
| "step": 273780 |
| }, |
| { |
| "entropy": 1.104986509314342, |
| "epoch": 46.0, |
| "grad_norm": 0.0206298828125, |
| "learning_rate": 2.474328473535452e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1066751638.0, |
| "step": 279864 |
| }, |
| { |
| "epoch": 46.0, |
| "eval_entropy": 1.1854262113571168, |
| "eval_loss": 0.848030686378479, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1066751638.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1552, |
| "eval_samples_per_second": 90.181, |
| "eval_steps_per_second": 32.207, |
| "step": 279864 |
| }, |
| { |
| "entropy": 1.1049189729509503, |
| "epoch": 47.0, |
| "grad_norm": 0.0830078125, |
| "learning_rate": 1.8557717725045243e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1089941891.0, |
| "step": 285948 |
| }, |
| { |
| "epoch": 47.0, |
| "eval_entropy": 1.1844752788543702, |
| "eval_loss": 0.8457481265068054, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1089941891.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1547, |
| "eval_samples_per_second": 90.504, |
| "eval_steps_per_second": 32.323, |
| "step": 285948 |
| }, |
| { |
| "entropy": 1.1046031564865983, |
| "epoch": 48.0, |
| "grad_norm": 0.0223388671875, |
| "learning_rate": 1.2372150714735964e-06, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1113132144.0, |
| "step": 292032 |
| }, |
| { |
| "epoch": 48.0, |
| "eval_entropy": 1.1853480339050293, |
| "eval_loss": 0.8546704649925232, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1113132144.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1565, |
| "eval_samples_per_second": 89.473, |
| "eval_steps_per_second": 31.955, |
| "step": 292032 |
| }, |
| { |
| "entropy": 1.104954168171102, |
| "epoch": 49.0, |
| "grad_norm": 0.0157470703125, |
| "learning_rate": 6.186583704426686e-07, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1136322397.0, |
| "step": 298116 |
| }, |
| { |
| "epoch": 49.0, |
| "eval_entropy": 1.1826888084411622, |
| "eval_loss": 0.8540498614311218, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1136322397.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1583, |
| "eval_samples_per_second": 88.419, |
| "eval_steps_per_second": 31.578, |
| "step": 298116 |
| }, |
| { |
| "entropy": 1.104515764203454, |
| "epoch": 50.0, |
| "grad_norm": 0.126953125, |
| "learning_rate": 1.0166941174078366e-10, |
| "loss": 0.0002, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 1159512650.0, |
| "step": 304200 |
| }, |
| { |
| "epoch": 50.0, |
| "eval_entropy": 1.1829030990600586, |
| "eval_loss": 0.8473049998283386, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 1159512650.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.1545, |
| "eval_samples_per_second": 90.586, |
| "eval_steps_per_second": 32.352, |
| "step": 304200 |
| } |
| ], |
| "logging_steps": 0, |
| "max_steps": 304200, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 0, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.770251437965312e+18, |
| "train_batch_size": 3, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
