| { |
| "best_global_step": 18252, |
| "best_metric": 0.8571, |
| "best_model_checkpoint": "models/NED/SPACCC_human_only_tfidf_hybrid_long_v2_addheaders/Llama-3.1-8B-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": 1.6452295544859141, |
| "epoch": 1.0, |
| "grad_norm": 36.5, |
| "learning_rate": 1.999671268902038e-05, |
| "loss": 0.8315, |
| "mean_token_accuracy": 0.8217970482187001, |
| "num_tokens": 23190253.0, |
| "step": 6084 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_entropy": 1.7185479640960692, |
| "eval_loss": 0.4542328417301178, |
| "eval_mean_token_accuracy": 0.8898225903511048, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 23190253.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.3763, |
| "eval_samples_per_second": 37.199, |
| "eval_steps_per_second": 13.286, |
| "step": 6084 |
| }, |
| { |
| "entropy": 1.6400107836190374, |
| "epoch": 2.0, |
| "grad_norm": 24.0, |
| "learning_rate": 2.9690823318896277e-05, |
| "loss": 0.5539, |
| "mean_token_accuracy": 0.869478269118294, |
| "num_tokens": 46380506.0, |
| "step": 12168 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_entropy": 1.826171350479126, |
| "eval_loss": 0.3982131779193878, |
| "eval_mean_token_accuracy": 0.9231559276580811, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 46380506.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3572, |
| "eval_samples_per_second": 39.19, |
| "eval_steps_per_second": 13.996, |
| "step": 12168 |
| }, |
| { |
| "entropy": 1.369841670310082, |
| "epoch": 3.0, |
| "grad_norm": 48.5, |
| "learning_rate": 2.9072266617865347e-05, |
| "loss": 0.2993, |
| "mean_token_accuracy": 0.9245726638164354, |
| "num_tokens": 69570759.0, |
| "step": 18252 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_entropy": 1.2688756227493285, |
| "eval_loss": 0.4362812042236328, |
| "eval_mean_token_accuracy": 0.9365079402923584, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 69570759.0, |
| "eval_recall": 0.8571, |
| "eval_runtime": 0.3611, |
| "eval_samples_per_second": 38.773, |
| "eval_steps_per_second": 13.848, |
| "step": 18252 |
| }, |
| { |
| "entropy": 1.019543299896007, |
| "epoch": 4.0, |
| "grad_norm": 0.63671875, |
| "learning_rate": 2.845370991683442e-05, |
| "loss": 0.1426, |
| "mean_token_accuracy": 0.9636940412482958, |
| "num_tokens": 92761012.0, |
| "step": 24336 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_entropy": 1.04158456325531, |
| "eval_loss": 0.5508492588996887, |
| "eval_mean_token_accuracy": 0.9152194261550903, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 92761012.0, |
| "eval_recall": 0.7143, |
| "eval_runtime": 0.3625, |
| "eval_samples_per_second": 38.617, |
| "eval_steps_per_second": 13.792, |
| "step": 24336 |
| }, |
| { |
| "entropy": 0.8108051640179343, |
| "epoch": 5.0, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 2.7835153215803492e-05, |
| "loss": 0.0607, |
| "mean_token_accuracy": 0.9852782511283974, |
| "num_tokens": 115951265.0, |
| "step": 30420 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_entropy": 0.8801020622253418, |
| "eval_loss": 0.6922857165336609, |
| "eval_mean_token_accuracy": 0.9152194261550903, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 115951265.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3885, |
| "eval_samples_per_second": 36.034, |
| "eval_steps_per_second": 12.869, |
| "step": 30420 |
| }, |
| { |
| "entropy": 0.718129879927886, |
| "epoch": 6.0, |
| "grad_norm": 0.0277099609375, |
| "learning_rate": 2.7216596514772566e-05, |
| "loss": 0.0218, |
| "mean_token_accuracy": 0.9949554541466034, |
| "num_tokens": 139141518.0, |
| "step": 36504 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_entropy": 0.7900081038475036, |
| "eval_loss": 0.6174029111862183, |
| "eval_mean_token_accuracy": 0.9247432351112366, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 139141518.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3581, |
| "eval_samples_per_second": 39.098, |
| "eval_steps_per_second": 13.964, |
| "step": 36504 |
| }, |
| { |
| "entropy": 0.707064643982286, |
| "epoch": 7.0, |
| "grad_norm": 0.00396728515625, |
| "learning_rate": 2.659803981374164e-05, |
| "loss": 0.0071, |
| "mean_token_accuracy": 0.9984249214566436, |
| "num_tokens": 162331771.0, |
| "step": 42588 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_entropy": 0.8332192063331604, |
| "eval_loss": 0.6586180925369263, |
| "eval_mean_token_accuracy": 0.9476190447807312, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 162331771.0, |
| "eval_recall": 0.8571, |
| "eval_runtime": 0.3996, |
| "eval_samples_per_second": 35.034, |
| "eval_steps_per_second": 12.512, |
| "step": 42588 |
| }, |
| { |
| "entropy": 0.7012911128574575, |
| "epoch": 8.0, |
| "grad_norm": 6.628036499023438e-05, |
| "learning_rate": 2.597948311271071e-05, |
| "loss": 0.0021, |
| "mean_token_accuracy": 0.9994771537578553, |
| "num_tokens": 185522024.0, |
| "step": 48672 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_entropy": 0.8370911121368408, |
| "eval_loss": 0.695911705493927, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 185522024.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.4123, |
| "eval_samples_per_second": 33.959, |
| "eval_steps_per_second": 12.128, |
| "step": 48672 |
| }, |
| { |
| "entropy": 0.7100966135621619, |
| "epoch": 9.0, |
| "grad_norm": 0.0034332275390625, |
| "learning_rate": 2.5360926411679782e-05, |
| "loss": 0.0006, |
| "mean_token_accuracy": 0.9998389675889965, |
| "num_tokens": 208712277.0, |
| "step": 54756 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_entropy": 0.848201584815979, |
| "eval_loss": 0.6835612654685974, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 208712277.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.4367, |
| "eval_samples_per_second": 32.06, |
| "eval_steps_per_second": 11.45, |
| "step": 54756 |
| }, |
| { |
| "entropy": 0.7072051375869228, |
| "epoch": 10.0, |
| "grad_norm": 0.01507568359375, |
| "learning_rate": 2.4742369710648856e-05, |
| "loss": 0.0001, |
| "mean_token_accuracy": 0.9999864275501716, |
| "num_tokens": 231902530.0, |
| "step": 60840 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_entropy": 0.8428204655647278, |
| "eval_loss": 0.6963585615158081, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 231902530.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3772, |
| "eval_samples_per_second": 37.119, |
| "eval_steps_per_second": 13.257, |
| "step": 60840 |
| }, |
| { |
| "entropy": 0.7021450536628054, |
| "epoch": 11.0, |
| "grad_norm": 0.04638671875, |
| "learning_rate": 2.4123813009617927e-05, |
| "loss": 0.0001, |
| "mean_token_accuracy": 0.9999853084543517, |
| "num_tokens": 255092783.0, |
| "step": 66924 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_entropy": 0.8535857558250427, |
| "eval_loss": 0.703579306602478, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 255092783.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3599, |
| "eval_samples_per_second": 38.9, |
| "eval_steps_per_second": 13.893, |
| "step": 66924 |
| }, |
| { |
| "entropy": 0.7062483392484403, |
| "epoch": 12.0, |
| "grad_norm": 0.0003795623779296875, |
| "learning_rate": 2.3505256308586998e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 278283036.0, |
| "step": 73008 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_entropy": 0.84964017868042, |
| "eval_loss": 0.7106609344482422, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 278283036.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3587, |
| "eval_samples_per_second": 39.032, |
| "eval_steps_per_second": 13.94, |
| "step": 73008 |
| }, |
| { |
| "entropy": 0.7034359549041801, |
| "epoch": 13.0, |
| "grad_norm": 0.004058837890625, |
| "learning_rate": 2.2886699607556072e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 0.9999969561953517, |
| "num_tokens": 301473289.0, |
| "step": 79092 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_entropy": 0.846993887424469, |
| "eval_loss": 0.7097910642623901, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 301473289.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3594, |
| "eval_samples_per_second": 38.956, |
| "eval_steps_per_second": 13.913, |
| "step": 79092 |
| }, |
| { |
| "entropy": 0.7021650211110387, |
| "epoch": 14.0, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 2.2268142906525143e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 324663542.0, |
| "step": 85176 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_entropy": 0.8468610405921936, |
| "eval_loss": 0.7046443223953247, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 324663542.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3627, |
| "eval_samples_per_second": 38.595, |
| "eval_steps_per_second": 13.784, |
| "step": 85176 |
| }, |
| { |
| "entropy": 0.7020127680115916, |
| "epoch": 15.0, |
| "grad_norm": 0.01336669921875, |
| "learning_rate": 2.1649586205494213e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 347853795.0, |
| "step": 91260 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_entropy": 0.8471502661705017, |
| "eval_loss": 0.7057820558547974, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 347853795.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3597, |
| "eval_samples_per_second": 38.921, |
| "eval_steps_per_second": 13.9, |
| "step": 91260 |
| }, |
| { |
| "entropy": 0.7020189780925623, |
| "epoch": 16.0, |
| "grad_norm": 0.002655029296875, |
| "learning_rate": 2.1031029504463287e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 23190253.0, |
| "step": 97344 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_entropy": 0.8467319965362549, |
| "eval_loss": 0.7018277049064636, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 23190253.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3922, |
| "eval_samples_per_second": 35.698, |
| "eval_steps_per_second": 12.749, |
| "step": 97344 |
| }, |
| { |
| "entropy": 0.7017412878901139, |
| "epoch": 17.0, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 2.0412472803432358e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 46380506.0, |
| "step": 103428 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_entropy": 0.8470642566680908, |
| "eval_loss": 0.7061834335327148, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 46380506.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3627, |
| "eval_samples_per_second": 38.601, |
| "eval_steps_per_second": 13.786, |
| "step": 103428 |
| }, |
| { |
| "entropy": 0.7014865135280495, |
| "epoch": 18.0, |
| "grad_norm": 0.0004730224609375, |
| "learning_rate": 1.979391610240143e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 69570759.0, |
| "step": 109512 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_entropy": 0.8469521641731262, |
| "eval_loss": 0.705422043800354, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 69570759.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3685, |
| "eval_samples_per_second": 37.988, |
| "eval_steps_per_second": 13.567, |
| "step": 109512 |
| }, |
| { |
| "entropy": 0.7014951153384628, |
| "epoch": 19.0, |
| "grad_norm": 0.0012359619140625, |
| "learning_rate": 1.9175359401370503e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 92761012.0, |
| "step": 115596 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_entropy": 0.8468135356903076, |
| "eval_loss": 0.7019368410110474, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 92761012.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3664, |
| "eval_samples_per_second": 38.213, |
| "eval_steps_per_second": 13.648, |
| "step": 115596 |
| }, |
| { |
| "entropy": 0.7013456000359257, |
| "epoch": 20.0, |
| "grad_norm": 0.0032958984375, |
| "learning_rate": 1.8556802700339577e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 115951265.0, |
| "step": 121680 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_entropy": 0.8471927762031555, |
| "eval_loss": 0.7029876708984375, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 115951265.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3699, |
| "eval_samples_per_second": 37.852, |
| "eval_steps_per_second": 13.519, |
| "step": 121680 |
| }, |
| { |
| "entropy": 0.701600033552297, |
| "epoch": 21.0, |
| "grad_norm": 0.007232666015625, |
| "learning_rate": 1.793824599930865e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 139141518.0, |
| "step": 127764 |
| }, |
| { |
| "epoch": 21.0, |
| "eval_entropy": 0.8469080567359925, |
| "eval_loss": 0.6987797617912292, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 139141518.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3683, |
| "eval_samples_per_second": 38.011, |
| "eval_steps_per_second": 13.575, |
| "step": 127764 |
| }, |
| { |
| "entropy": 0.7013353118568875, |
| "epoch": 22.0, |
| "grad_norm": 0.005035400390625, |
| "learning_rate": 1.7319689298277722e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 162331771.0, |
| "step": 133848 |
| }, |
| { |
| "epoch": 22.0, |
| "eval_entropy": 0.8477846264839173, |
| "eval_loss": 0.7038993835449219, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 162331771.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3637, |
| "eval_samples_per_second": 38.496, |
| "eval_steps_per_second": 13.749, |
| "step": 133848 |
| }, |
| { |
| "entropy": 0.701190030512098, |
| "epoch": 23.0, |
| "grad_norm": 0.0034942626953125, |
| "learning_rate": 1.6701132597246793e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 185522024.0, |
| "step": 139932 |
| }, |
| { |
| "epoch": 23.0, |
| "eval_entropy": 0.8471533417701721, |
| "eval_loss": 0.7041516900062561, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 185522024.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3647, |
| "eval_samples_per_second": 38.392, |
| "eval_steps_per_second": 13.711, |
| "step": 139932 |
| }, |
| { |
| "entropy": 0.7012511798518022, |
| "epoch": 24.0, |
| "grad_norm": 0.033447265625, |
| "learning_rate": 1.6082575896215867e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 208712277.0, |
| "step": 146016 |
| }, |
| { |
| "epoch": 24.0, |
| "eval_entropy": 0.8475937247276306, |
| "eval_loss": 0.7037870287895203, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 208712277.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3629, |
| "eval_samples_per_second": 38.578, |
| "eval_steps_per_second": 13.778, |
| "step": 146016 |
| }, |
| { |
| "entropy": 0.7011014092799318, |
| "epoch": 25.0, |
| "grad_norm": 0.0004138946533203125, |
| "learning_rate": 1.5464019195184938e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 231902530.0, |
| "step": 152100 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_entropy": 0.8474008202552795, |
| "eval_loss": 0.7015256881713867, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 231902530.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3667, |
| "eval_samples_per_second": 38.176, |
| "eval_steps_per_second": 13.634, |
| "step": 152100 |
| }, |
| { |
| "entropy": 0.7010043011849684, |
| "epoch": 26.0, |
| "grad_norm": 0.005462646484375, |
| "learning_rate": 1.4845462494154008e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 255092783.0, |
| "step": 158184 |
| }, |
| { |
| "epoch": 26.0, |
| "eval_entropy": 0.8452415347099305, |
| "eval_loss": 0.7029375433921814, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 255092783.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3822, |
| "eval_samples_per_second": 36.63, |
| "eval_steps_per_second": 13.082, |
| "step": 158184 |
| }, |
| { |
| "entropy": 0.7014232551303057, |
| "epoch": 27.0, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.4226905793123081e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 278283036.0, |
| "step": 164268 |
| }, |
| { |
| "epoch": 27.0, |
| "eval_entropy": 0.8463044404983521, |
| "eval_loss": 0.7052632570266724, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 278283036.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.388, |
| "eval_samples_per_second": 36.085, |
| "eval_steps_per_second": 12.888, |
| "step": 164268 |
| }, |
| { |
| "entropy": 0.701196315226235, |
| "epoch": 28.0, |
| "grad_norm": 0.00665283203125, |
| "learning_rate": 1.3608349092092153e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 301473289.0, |
| "step": 170352 |
| }, |
| { |
| "epoch": 28.0, |
| "eval_entropy": 0.8478581547737122, |
| "eval_loss": 0.7016162872314453, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 301473289.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3628, |
| "eval_samples_per_second": 38.588, |
| "eval_steps_per_second": 13.781, |
| "step": 170352 |
| }, |
| { |
| "entropy": 0.7013136386910526, |
| "epoch": 29.0, |
| "grad_norm": 0.002960205078125, |
| "learning_rate": 1.2989792391061224e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 324663542.0, |
| "step": 176436 |
| }, |
| { |
| "epoch": 29.0, |
| "eval_entropy": 0.848319697380066, |
| "eval_loss": 0.7054678797721863, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 324663542.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.4297, |
| "eval_samples_per_second": 32.584, |
| "eval_steps_per_second": 11.637, |
| "step": 176436 |
| }, |
| { |
| "entropy": 0.7013322007080037, |
| "epoch": 30.0, |
| "grad_norm": 0.023193359375, |
| "learning_rate": 1.2371235690030298e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 347853795.0, |
| "step": 182520 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_entropy": 0.8468625545501709, |
| "eval_loss": 0.7053791284561157, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 347853795.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3794, |
| "eval_samples_per_second": 36.904, |
| "eval_steps_per_second": 13.18, |
| "step": 182520 |
| }, |
| { |
| "entropy": 0.7009657870618079, |
| "epoch": 31.0, |
| "grad_norm": 0.00531005859375, |
| "learning_rate": 1.175267898899937e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 23190253.0, |
| "step": 188604 |
| }, |
| { |
| "epoch": 31.0, |
| "eval_entropy": 0.8471902132034301, |
| "eval_loss": 0.7000441551208496, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 23190253.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3469, |
| "eval_samples_per_second": 40.36, |
| "eval_steps_per_second": 14.414, |
| "step": 188604 |
| }, |
| { |
| "entropy": 0.7013605682964312, |
| "epoch": 32.0, |
| "grad_norm": 0.004486083984375, |
| "learning_rate": 1.1134122287968443e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 46380506.0, |
| "step": 194688 |
| }, |
| { |
| "epoch": 32.0, |
| "eval_entropy": 0.8475266933441162, |
| "eval_loss": 0.7029227018356323, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 46380506.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3352, |
| "eval_samples_per_second": 41.764, |
| "eval_steps_per_second": 14.916, |
| "step": 194688 |
| }, |
| { |
| "entropy": 0.7009823948223288, |
| "epoch": 33.0, |
| "grad_norm": 0.00072479248046875, |
| "learning_rate": 1.0515565586937514e-05, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 69570759.0, |
| "step": 200772 |
| }, |
| { |
| "epoch": 33.0, |
| "eval_entropy": 0.8483369827270508, |
| "eval_loss": 0.6992912292480469, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 69570759.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3344, |
| "eval_samples_per_second": 41.86, |
| "eval_steps_per_second": 14.95, |
| "step": 200772 |
| }, |
| { |
| "entropy": 0.7011937351406287, |
| "epoch": 34.0, |
| "grad_norm": 0.0191650390625, |
| "learning_rate": 9.897008885906586e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 92761012.0, |
| "step": 206856 |
| }, |
| { |
| "epoch": 34.0, |
| "eval_entropy": 0.8469909429550171, |
| "eval_loss": 0.7026051878929138, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 92761012.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3375, |
| "eval_samples_per_second": 41.484, |
| "eval_steps_per_second": 14.816, |
| "step": 206856 |
| }, |
| { |
| "entropy": 0.7014018389580674, |
| "epoch": 35.0, |
| "grad_norm": 0.0016021728515625, |
| "learning_rate": 9.278452184875659e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 115951265.0, |
| "step": 212940 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_entropy": 0.8470338106155395, |
| "eval_loss": 0.701595664024353, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 115951265.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3346, |
| "eval_samples_per_second": 41.838, |
| "eval_steps_per_second": 14.942, |
| "step": 212940 |
| }, |
| { |
| "entropy": 0.7011160486862112, |
| "epoch": 36.0, |
| "grad_norm": 0.00010251998901367188, |
| "learning_rate": 8.65989548384473e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 139141518.0, |
| "step": 219024 |
| }, |
| { |
| "epoch": 36.0, |
| "eval_entropy": 0.8465482711791992, |
| "eval_loss": 0.7029346823692322, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 139141518.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3364, |
| "eval_samples_per_second": 41.619, |
| "eval_steps_per_second": 14.864, |
| "step": 219024 |
| }, |
| { |
| "entropy": 0.7011719772899566, |
| "epoch": 37.0, |
| "grad_norm": 0.0274658203125, |
| "learning_rate": 8.041338782813804e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 162331771.0, |
| "step": 225108 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_entropy": 0.8483202934265137, |
| "eval_loss": 0.7015349268913269, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 162331771.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3421, |
| "eval_samples_per_second": 40.92, |
| "eval_steps_per_second": 14.614, |
| "step": 225108 |
| }, |
| { |
| "entropy": 0.7014031262087399, |
| "epoch": 38.0, |
| "grad_norm": 0.00131988525390625, |
| "learning_rate": 7.4227820817828744e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 185522024.0, |
| "step": 231192 |
| }, |
| { |
| "epoch": 38.0, |
| "eval_entropy": 0.8476453304290772, |
| "eval_loss": 0.7029452323913574, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 185522024.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.365, |
| "eval_samples_per_second": 38.357, |
| "eval_steps_per_second": 13.699, |
| "step": 231192 |
| }, |
| { |
| "entropy": 0.7013709572664928, |
| "epoch": 39.0, |
| "grad_norm": 0.00299072265625, |
| "learning_rate": 6.804225380751948e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 208712277.0, |
| "step": 237276 |
| }, |
| { |
| "epoch": 39.0, |
| "eval_entropy": 0.8475553870201111, |
| "eval_loss": 0.7111210227012634, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 208712277.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3606, |
| "eval_samples_per_second": 38.824, |
| "eval_steps_per_second": 13.866, |
| "step": 237276 |
| }, |
| { |
| "entropy": 0.701420904293333, |
| "epoch": 40.0, |
| "grad_norm": 0.0274658203125, |
| "learning_rate": 6.185668679721019e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 231902530.0, |
| "step": 243360 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_entropy": 0.8471273303031921, |
| "eval_loss": 0.7025880813598633, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 231902530.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3628, |
| "eval_samples_per_second": 38.594, |
| "eval_steps_per_second": 13.783, |
| "step": 243360 |
| }, |
| { |
| "entropy": 0.7012593725219829, |
| "epoch": 41.0, |
| "grad_norm": 0.00136566162109375, |
| "learning_rate": 5.567111978690091e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 255092783.0, |
| "step": 249444 |
| }, |
| { |
| "epoch": 41.0, |
| "eval_entropy": 0.8475102543830871, |
| "eval_loss": 0.701640784740448, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 255092783.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3585, |
| "eval_samples_per_second": 39.047, |
| "eval_steps_per_second": 13.945, |
| "step": 249444 |
| }, |
| { |
| "entropy": 0.7012932761612102, |
| "epoch": 42.0, |
| "grad_norm": 0.004669189453125, |
| "learning_rate": 4.948555277659164e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 278283036.0, |
| "step": 255528 |
| }, |
| { |
| "epoch": 42.0, |
| "eval_entropy": 0.8476259231567382, |
| "eval_loss": 0.7012701034545898, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 278283036.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3591, |
| "eval_samples_per_second": 38.982, |
| "eval_steps_per_second": 13.922, |
| "step": 255528 |
| }, |
| { |
| "entropy": 0.7010069375638505, |
| "epoch": 43.0, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 4.329998576628236e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 301473289.0, |
| "step": 261612 |
| }, |
| { |
| "epoch": 43.0, |
| "eval_entropy": 0.8467528820037842, |
| "eval_loss": 0.701405942440033, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 301473289.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3595, |
| "eval_samples_per_second": 38.939, |
| "eval_steps_per_second": 13.907, |
| "step": 261612 |
| }, |
| { |
| "entropy": 0.701388671474971, |
| "epoch": 44.0, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 3.711441875597308e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 324663542.0, |
| "step": 267696 |
| }, |
| { |
| "epoch": 44.0, |
| "eval_entropy": 0.847226333618164, |
| "eval_loss": 0.7051442265510559, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 324663542.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3574, |
| "eval_samples_per_second": 39.168, |
| "eval_steps_per_second": 13.989, |
| "step": 267696 |
| }, |
| { |
| "entropy": 0.7011981701121998, |
| "epoch": 45.0, |
| "grad_norm": 0.0189208984375, |
| "learning_rate": 3.09288517456638e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 347853795.0, |
| "step": 273780 |
| }, |
| { |
| "epoch": 45.0, |
| "eval_entropy": 0.8469288349151611, |
| "eval_loss": 0.699046790599823, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 347853795.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3618, |
| "eval_samples_per_second": 38.7, |
| "eval_steps_per_second": 13.821, |
| "step": 273780 |
| }, |
| { |
| "entropy": 0.7014733728004239, |
| "epoch": 46.0, |
| "grad_norm": 0.007354736328125, |
| "learning_rate": 2.474328473535452e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 23190253.0, |
| "step": 279864 |
| }, |
| { |
| "epoch": 46.0, |
| "eval_entropy": 0.8493005037307739, |
| "eval_loss": 0.7080731391906738, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 23190253.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.7195, |
| "eval_samples_per_second": 19.457, |
| "eval_steps_per_second": 6.949, |
| "step": 279864 |
| }, |
| { |
| "entropy": 0.7013301944983468, |
| "epoch": 47.0, |
| "grad_norm": 0.039306640625, |
| "learning_rate": 1.8557717725045243e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 46380506.0, |
| "step": 285948 |
| }, |
| { |
| "epoch": 47.0, |
| "eval_entropy": 0.8469571113586426, |
| "eval_loss": 0.7055781483650208, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 46380506.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3604, |
| "eval_samples_per_second": 38.845, |
| "eval_steps_per_second": 13.873, |
| "step": 285948 |
| }, |
| { |
| "entropy": 0.7012139727887474, |
| "epoch": 48.0, |
| "grad_norm": 0.004791259765625, |
| "learning_rate": 1.2372150714735964e-06, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 69570759.0, |
| "step": 292032 |
| }, |
| { |
| "epoch": 48.0, |
| "eval_entropy": 0.8478953242301941, |
| "eval_loss": 0.7026681900024414, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 69570759.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.3695, |
| "eval_samples_per_second": 37.885, |
| "eval_steps_per_second": 13.53, |
| "step": 292032 |
| }, |
| { |
| "entropy": 0.7014247281409501, |
| "epoch": 49.0, |
| "grad_norm": 0.00099945068359375, |
| "learning_rate": 6.186583704426686e-07, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 92761012.0, |
| "step": 298116 |
| }, |
| { |
| "epoch": 49.0, |
| "eval_entropy": 0.8473766565322876, |
| "eval_loss": 0.7014277577400208, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 92761012.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.4396, |
| "eval_samples_per_second": 31.845, |
| "eval_steps_per_second": 11.373, |
| "step": 298116 |
| }, |
| { |
| "entropy": 0.7012487682443477, |
| "epoch": 50.0, |
| "grad_norm": 0.01092529296875, |
| "learning_rate": 1.0166941174078366e-10, |
| "loss": 0.0, |
| "mean_token_accuracy": 1.0, |
| "num_tokens": 115951265.0, |
| "step": 304200 |
| }, |
| { |
| "epoch": 50.0, |
| "eval_entropy": 0.8481771588325501, |
| "eval_loss": 0.7044681906700134, |
| "eval_mean_token_accuracy": 0.9358543395996094, |
| "eval_num_gold": 14, |
| "eval_num_guess": 14, |
| "eval_num_tokens": 115951265.0, |
| "eval_recall": 0.7857, |
| "eval_runtime": 0.9148, |
| "eval_samples_per_second": 15.304, |
| "eval_steps_per_second": 5.466, |
| "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": 5.221235748535009e+19, |
| "train_batch_size": 3, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
