{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.8, "eval_steps": 500, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "completion_length": 179.5625, "epoch": 0.004, "grad_norm": 2.046875, "kl": 0.0, "learning_rate": 2.0000000000000003e-06, "loss": 0.0, "reward": 0.9819051176309586, "reward_std": 0.2500194739550352, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.05803571408614516, "rewards/no_repetition_reward_func": 0.946190819144249, "rewards/verse_reward_func": 0.09375, "step": 1 }, { "completion_length": 183.1875, "epoch": 0.008, "grad_norm": 1.1875, "kl": 0.0, "learning_rate": 4.000000000000001e-06, "loss": -0.0, "reward": 1.0611461400985718, "reward_std": 0.16349453944712877, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0052083334885537624, "rewards/no_repetition_reward_func": 0.962187796831131, "rewards/verse_reward_func": 0.09375, "step": 2 }, { "completion_length": 177.3125, "epoch": 0.012, "grad_norm": 3.34375, "kl": 0.0009225639951182529, "learning_rate": 6e-06, "loss": 0.0, "reward": 0.927303671836853, "reward_std": 0.20073290541768074, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.958553671836853, "rewards/verse_reward_func": 0.03125, "step": 3 }, { "completion_length": 164.375, "epoch": 0.016, "grad_norm": 1.3515625, "kl": 0.0010245055600535125, "learning_rate": 8.000000000000001e-06, "loss": 0.0, "reward": 0.9359359890222549, "reward_std": 0.16188888170290738, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9671860039234161, "rewards/verse_reward_func": 0.03125, "step": 4 }, { "completion_length": 181.9375, "epoch": 0.02, "grad_norm": 1.2109375, "kl": 0.000991553533822298, "learning_rate": 1e-05, "loss": 0.0, "reward": 1.0427572429180145, "reward_std": 0.14885016926564276, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.010416666977107525, "rewards/no_repetition_reward_func": 0.9698406159877777, "rewards/verse_reward_func": 0.0625, "step": 5 }, { "completion_length": 199.3125, "epoch": 0.024, "grad_norm": 0.78125, "kl": 0.0008464615239063278, "learning_rate": 1.2e-05, "loss": 0.0, "reward": 0.897506594657898, "reward_std": 0.0972918642219156, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.897506594657898, "rewards/verse_reward_func": 0.0, "step": 6 }, { "completion_length": 179.0625, "epoch": 0.028, "grad_norm": 0.95703125, "kl": 0.0009047269122675061, "learning_rate": 1.4000000000000001e-05, "loss": 0.0, "reward": 0.9670905768871307, "reward_std": 0.11869156261673197, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9358405917882919, "rewards/verse_reward_func": 0.03125, "step": 7 }, { "completion_length": 200.0, "epoch": 0.032, "grad_norm": 0.82421875, "kl": 0.0009274971816921607, "learning_rate": 1.6000000000000003e-05, "loss": 0.0, "reward": 1.0096274018287659, "reward_std": 0.1000128339510411, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.003289473708719015, "rewards/no_repetition_reward_func": 0.9438379555940628, "rewards/verse_reward_func": 0.0625, "step": 8 }, { "completion_length": 191.4375, "epoch": 0.036, "grad_norm": 0.9609375, "kl": 0.0009898433345369995, "learning_rate": 1.8e-05, "loss": 0.0, "reward": 1.0010679364204407, "reward_std": 0.0678855258738622, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9698179215192795, "rewards/verse_reward_func": 0.03125, "step": 9 }, { "completion_length": 179.625, "epoch": 0.04, "grad_norm": 1.296875, "kl": 0.0010007202363340184, "learning_rate": 2e-05, "loss": 0.0, "reward": 1.006456971168518, "reward_std": 0.1597397131845355, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9439570009708405, "rewards/verse_reward_func": 0.0625, "step": 10 }, { "completion_length": 183.375, "epoch": 0.044, "grad_norm": 1.75, "kl": 0.0008183487225323915, "learning_rate": 2.2000000000000003e-05, "loss": 0.0, "reward": 0.9860096573829651, "reward_std": 0.23296335770282894, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9235096573829651, "rewards/verse_reward_func": 0.125, "step": 11 }, { "completion_length": 188.8125, "epoch": 0.048, "grad_norm": 2.109375, "kl": 0.0009898586868075654, "learning_rate": 2.4e-05, "loss": 0.0, "reward": 0.9310034066438675, "reward_std": 0.2854597745463252, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9310034215450287, "rewards/verse_reward_func": 0.0625, "step": 12 }, { "completion_length": 182.0, "epoch": 0.052, "grad_norm": 0.9453125, "kl": 0.000811553152743727, "learning_rate": 2.6000000000000002e-05, "loss": 0.0, "reward": 1.0062111169099808, "reward_std": 0.17000664526131004, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9437110871076584, "rewards/verse_reward_func": 0.0625, "step": 13 }, { "completion_length": 178.5, "epoch": 0.056, "grad_norm": 1.3359375, "kl": 0.0009504656482022256, "learning_rate": 2.8000000000000003e-05, "loss": 0.0, "reward": 0.9895833879709244, "reward_std": 0.07891088706674054, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9583334028720856, "rewards/verse_reward_func": 0.03125, "step": 14 }, { "completion_length": 192.9375, "epoch": 0.06, "grad_norm": 0.98828125, "kl": 0.0009447222109884024, "learning_rate": 3e-05, "loss": 0.0, "reward": 0.8987747207283974, "reward_std": 0.2582971692318097, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.125, "rewards/no_repetition_reward_func": 0.9612747132778168, "rewards/verse_reward_func": 0.0625, "step": 15 }, { "completion_length": 186.9375, "epoch": 0.064, "grad_norm": 1.0, "kl": 0.0009807306778384373, "learning_rate": 3.2000000000000005e-05, "loss": 0.0, "reward": 1.0321685820817947, "reward_std": 0.12863060203380883, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9696685969829559, "rewards/verse_reward_func": 0.0625, "step": 16 }, { "completion_length": 155.5625, "epoch": 0.068, "grad_norm": 1.21875, "kl": 0.0010486226383363828, "learning_rate": 3.4000000000000007e-05, "loss": 0.0, "reward": 0.9713962525129318, "reward_std": 0.24358075205236673, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.125, "rewards/no_repetition_reward_func": 0.9713962525129318, "rewards/verse_reward_func": 0.125, "step": 17 }, { "completion_length": 187.5625, "epoch": 0.072, "grad_norm": 0.91015625, "kl": 0.0009779602842172608, "learning_rate": 3.6e-05, "loss": 0.0, "reward": 1.0452248454093933, "reward_std": 0.216551274061203, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004464285913854837, "rewards/no_repetition_reward_func": 0.9470105767250061, "rewards/verse_reward_func": 0.09375, "step": 18 }, { "completion_length": 185.5625, "epoch": 0.076, "grad_norm": 1.2890625, "kl": 0.001001911296043545, "learning_rate": 3.8e-05, "loss": 0.0, "reward": 0.9728774726390839, "reward_std": 0.09631168603664264, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9416275024414062, "rewards/verse_reward_func": 0.03125, "step": 19 }, { "completion_length": 200.0, "epoch": 0.08, "grad_norm": 0.8125, "kl": 0.0009419274720130488, "learning_rate": 4e-05, "loss": 0.0, "reward": 0.9773482233285904, "reward_std": 0.07271566009148955, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004464285913854837, "rewards/no_repetition_reward_func": 0.9416339844465256, "rewards/verse_reward_func": 0.03125, "step": 20 }, { "completion_length": 196.3125, "epoch": 0.084, "grad_norm": 0.90625, "kl": 0.0009593678259989247, "learning_rate": 4.2e-05, "loss": 0.0, "reward": 0.9732275754213333, "reward_std": 0.10389877262059599, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9419775754213333, "rewards/verse_reward_func": 0.03125, "step": 21 }, { "completion_length": 186.625, "epoch": 0.088, "grad_norm": 1.546875, "kl": 0.0011393697932362556, "learning_rate": 4.4000000000000006e-05, "loss": 0.0, "reward": 1.0079743266105652, "reward_std": 0.07848087884485722, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.007708333432674408, "rewards/no_repetition_reward_func": 0.9690160155296326, "rewards/verse_reward_func": 0.03125, "step": 22 }, { "completion_length": 199.0625, "epoch": 0.092, "grad_norm": 0.9609375, "kl": 0.0011462626862339675, "learning_rate": 4.600000000000001e-05, "loss": 0.0, "reward": 0.9748821258544922, "reward_std": 0.020194193988572806, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.007352941203862429, "rewards/no_repetition_reward_func": 0.9675291776657104, "rewards/verse_reward_func": 0.0, "step": 23 }, { "completion_length": 175.0, "epoch": 0.096, "grad_norm": 1.5859375, "kl": 0.0012009456986561418, "learning_rate": 4.8e-05, "loss": 0.0, "reward": 0.9423503875732422, "reward_std": 0.1669474468799308, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.05859375, "rewards/no_repetition_reward_func": 0.9696941375732422, "rewards/verse_reward_func": 0.03125, "step": 24 }, { "completion_length": 164.0, "epoch": 0.1, "grad_norm": 1.1171875, "kl": 0.0012017716944683343, "learning_rate": 5e-05, "loss": 0.0, "reward": 0.998960480093956, "reward_std": 0.05968155374284834, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.967710480093956, "rewards/verse_reward_func": 0.03125, "step": 25 }, { "completion_length": 186.1875, "epoch": 0.104, "grad_norm": 1.25, "kl": 0.001386465271934867, "learning_rate": 4.999756310023261e-05, "loss": 0.0001, "reward": 1.024965062737465, "reward_std": 0.12524886144092306, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9624650627374649, "rewards/verse_reward_func": 0.0625, "step": 26 }, { "completion_length": 199.6875, "epoch": 0.108, "grad_norm": 0.890625, "kl": 0.0012058749562129378, "learning_rate": 4.999025287600886e-05, "loss": 0.0, "reward": 0.9964274168014526, "reward_std": 0.26108699198812246, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9651773869991302, "rewards/verse_reward_func": 0.09375, "step": 27 }, { "completion_length": 179.75, "epoch": 0.112, "grad_norm": 1.0703125, "kl": 0.00140558643033728, "learning_rate": 4.997807075247146e-05, "loss": 0.0001, "reward": 0.9315525591373444, "reward_std": 0.1886110061313957, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9628025740385056, "rewards/verse_reward_func": 0.03125, "step": 28 }, { "completion_length": 163.5, "epoch": 0.116, "grad_norm": 1.6171875, "kl": 0.00141397793777287, "learning_rate": 4.996101910454953e-05, "loss": 0.0001, "reward": 0.9228900671005249, "reward_std": 0.08080775220878422, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9541400671005249, "rewards/verse_reward_func": 0.03125, "step": 29 }, { "completion_length": 185.0, "epoch": 0.12, "grad_norm": 0.921875, "kl": 0.0018486627377569675, "learning_rate": 4.993910125649561e-05, "loss": 0.0001, "reward": 0.9309101402759552, "reward_std": 0.07874627423007041, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9309101402759552, "rewards/verse_reward_func": 0.0, "step": 30 }, { "completion_length": 181.6875, "epoch": 0.124, "grad_norm": 1.265625, "kl": 0.0024124052142724395, "learning_rate": 4.991232148123761e-05, "loss": 0.0001, "reward": 0.9530736654996872, "reward_std": 0.2654994917102158, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9530736953020096, "rewards/verse_reward_func": 0.0625, "step": 31 }, { "completion_length": 193.25, "epoch": 0.128, "grad_norm": 1.0234375, "kl": 0.002259129600133747, "learning_rate": 4.988068499954578e-05, "loss": 0.0001, "reward": 1.0645287036895752, "reward_std": 0.22907709190621972, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9395287036895752, "rewards/verse_reward_func": 0.125, "step": 32 }, { "completion_length": 200.0, "epoch": 0.132, "grad_norm": 0.875, "kl": 0.002345684450119734, "learning_rate": 4.984419797901491e-05, "loss": 0.0001, "reward": 0.9817457795143127, "reward_std": 0.05011130444472656, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004166666883975267, "rewards/no_repetition_reward_func": 0.9463291019201279, "rewards/verse_reward_func": 0.03125, "step": 33 }, { "completion_length": 171.875, "epoch": 0.136, "grad_norm": 1.1484375, "kl": 0.0031502785277552903, "learning_rate": 4.980286753286195e-05, "loss": 0.0001, "reward": 1.0629110634326935, "reward_std": 0.1909176445333287, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9691610336303711, "rewards/verse_reward_func": 0.09375, "step": 34 }, { "completion_length": 192.5, "epoch": 0.14, "grad_norm": 1.0390625, "kl": 0.0024105910561047494, "learning_rate": 4.975670171853926e-05, "loss": 0.0001, "reward": 1.0318033248186111, "reward_std": 0.12928975140675902, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9693033397197723, "rewards/verse_reward_func": 0.0625, "step": 35 }, { "completion_length": 168.8125, "epoch": 0.144, "grad_norm": 1.4765625, "kl": 0.003135324106551707, "learning_rate": 4.9705709536163824e-05, "loss": 0.0001, "reward": 0.9991441071033478, "reward_std": 0.07299526745919138, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.967894122004509, "rewards/verse_reward_func": 0.03125, "step": 36 }, { "completion_length": 176.75, "epoch": 0.148, "grad_norm": 1.0, "kl": 0.003276034549344331, "learning_rate": 4.964990092676263e-05, "loss": 0.0001, "reward": 1.008391559123993, "reward_std": 0.07070710451807827, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.01131535996682942, "rewards/no_repetition_reward_func": 0.9658262133598328, "rewards/verse_reward_func": 0.03125, "step": 37 }, { "completion_length": 186.25, "epoch": 0.152, "grad_norm": 1.0390625, "kl": 0.0030353819020092487, "learning_rate": 4.9589286770334654e-05, "loss": 0.0001, "reward": 0.972317710518837, "reward_std": 0.016954098246060312, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004166666883975267, "rewards/no_repetition_reward_func": 0.9681510329246521, "rewards/verse_reward_func": 0.0, "step": 38 }, { "completion_length": 200.0, "epoch": 0.156, "grad_norm": 0.8828125, "kl": 0.0033272147993557155, "learning_rate": 4.952387888372979e-05, "loss": 0.0001, "reward": 1.0669118165969849, "reward_std": 0.1348436245461926, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9419118165969849, "rewards/verse_reward_func": 0.125, "step": 39 }, { "completion_length": 173.625, "epoch": 0.16, "grad_norm": 1.84375, "kl": 0.003484633460175246, "learning_rate": 4.9453690018345144e-05, "loss": 0.0001, "reward": 0.999244287610054, "reward_std": 0.18777411000337452, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9679943025112152, "rewards/verse_reward_func": 0.09375, "step": 40 }, { "completion_length": 179.6875, "epoch": 0.164, "grad_norm": 1.140625, "kl": 0.004894184530712664, "learning_rate": 4.937873385763908e-05, "loss": 0.0002, "reward": 1.0016028583049774, "reward_std": 0.07005300477612764, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9703528732061386, "rewards/verse_reward_func": 0.03125, "step": 41 }, { "completion_length": 190.625, "epoch": 0.168, "grad_norm": 1.03125, "kl": 0.0029558211099356413, "learning_rate": 4.929902501446366e-05, "loss": 0.0001, "reward": 1.0353728234767914, "reward_std": 0.1293391860090196, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004166666883975267, "rewards/no_repetition_reward_func": 0.9687061756849289, "rewards/verse_reward_func": 0.0625, "step": 42 }, { "completion_length": 187.6875, "epoch": 0.172, "grad_norm": 1.1484375, "kl": 0.0038812385755591094, "learning_rate": 4.9214579028215776e-05, "loss": 0.0002, "reward": 0.9657279551029205, "reward_std": 0.015553490375168622, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9657279700040817, "rewards/verse_reward_func": 0.0, "step": 43 }, { "completion_length": 190.25, "epoch": 0.176, "grad_norm": 0.83203125, "kl": 0.0032730000093579292, "learning_rate": 4.912541236180779e-05, "loss": 0.0001, "reward": 0.8466661870479584, "reward_std": 0.20375583856366575, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9091661721467972, "rewards/verse_reward_func": 0.0, "step": 44 }, { "completion_length": 189.4375, "epoch": 0.18, "grad_norm": 1.03125, "kl": 0.004002011672127992, "learning_rate": 4.9031542398457974e-05, "loss": 0.0002, "reward": 1.0241023004055023, "reward_std": 0.2693894528783858, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.05833333311602473, "rewards/no_repetition_reward_func": 0.9574356377124786, "rewards/verse_reward_func": 0.125, "step": 45 }, { "completion_length": 200.0, "epoch": 0.184, "grad_norm": 1.109375, "kl": 0.0036516414838843048, "learning_rate": 4.893298743830168e-05, "loss": 0.0001, "reward": 1.031081184744835, "reward_std": 0.12839890411123633, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9685812145471573, "rewards/verse_reward_func": 0.0625, "step": 46 }, { "completion_length": 184.0625, "epoch": 0.188, "grad_norm": 1.28125, "kl": 0.004063341184519231, "learning_rate": 4.882976669482367e-05, "loss": 0.0002, "reward": 1.010192185640335, "reward_std": 0.06739407801069319, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008928571827709675, "rewards/no_repetition_reward_func": 0.9700136184692383, "rewards/verse_reward_func": 0.03125, "step": 47 }, { "completion_length": 198.125, "epoch": 0.192, "grad_norm": 0.83203125, "kl": 0.003968458739109337, "learning_rate": 4.8721900291112415e-05, "loss": 0.0002, "reward": 0.9995202720165253, "reward_std": 0.07142159587237984, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0024999999441206455, "rewards/no_repetition_reward_func": 0.9657702594995499, "rewards/verse_reward_func": 0.03125, "step": 48 }, { "completion_length": 186.8125, "epoch": 0.196, "grad_norm": 0.875, "kl": 0.004230177321005613, "learning_rate": 4.860940925593703e-05, "loss": 0.0002, "reward": 0.9467289745807648, "reward_std": 0.05237383279018104, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9467289596796036, "rewards/verse_reward_func": 0.0, "step": 49 }, { "completion_length": 172.25, "epoch": 0.2, "grad_norm": 1.0546875, "kl": 0.007191646727733314, "learning_rate": 4.849231551964771e-05, "loss": 0.0003, "reward": 1.1267552971839905, "reward_std": 0.3819158934056759, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9705053418874741, "rewards/verse_reward_func": 0.21875, "step": 50 }, { "completion_length": 166.9375, "epoch": 0.204, "grad_norm": 1.40625, "kl": 0.006566129217389971, "learning_rate": 4.837064190990036e-05, "loss": 0.0003, "reward": 0.8921804428100586, "reward_std": 0.1621550468262285, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.055921051651239395, "rewards/no_repetition_reward_func": 0.9481014907360077, "rewards/verse_reward_func": 0.0, "step": 51 }, { "completion_length": 199.3125, "epoch": 0.208, "grad_norm": 0.8359375, "kl": 0.006002974580042064, "learning_rate": 4.8244412147206284e-05, "loss": 0.0002, "reward": 1.02969092130661, "reward_std": 0.12929325131699443, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9671909362077713, "rewards/verse_reward_func": 0.0625, "step": 52 }, { "completion_length": 158.75, "epoch": 0.212, "grad_norm": 1.75, "kl": 0.006504544056952, "learning_rate": 4.8113650840307834e-05, "loss": 0.0003, "reward": 0.9824613779783249, "reward_std": 0.16255538212135434, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.058823529398068786, "rewards/no_repetition_reward_func": 0.9162849336862564, "rewards/verse_reward_func": 0.125, "step": 53 }, { "completion_length": 177.375, "epoch": 0.216, "grad_norm": 1.171875, "kl": 0.005495292483828962, "learning_rate": 4.797838348138086e-05, "loss": 0.0002, "reward": 1.1409614980220795, "reward_std": 0.25640933960676193, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9534614980220795, "rewards/verse_reward_func": 0.1875, "step": 54 }, { "completion_length": 200.0, "epoch": 0.22, "grad_norm": 0.9375, "kl": 0.005827408516779542, "learning_rate": 4.783863644106502e-05, "loss": 0.0002, "reward": 0.9815671294927597, "reward_std": 0.08528327068779618, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9503171741962433, "rewards/verse_reward_func": 0.03125, "step": 55 }, { "completion_length": 161.75, "epoch": 0.224, "grad_norm": 2.875, "kl": 0.0055718234507367015, "learning_rate": 4.769443696332272e-05, "loss": 0.0002, "reward": 0.9660014510154724, "reward_std": 0.19446178257931024, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9660014659166336, "rewards/verse_reward_func": 0.0625, "step": 56 }, { "completion_length": 168.1875, "epoch": 0.228, "grad_norm": 1.5625, "kl": 0.007422439637593925, "learning_rate": 4.754581316012785e-05, "loss": 0.0003, "reward": 0.8729029893875122, "reward_std": 0.31897079292684793, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.125, "rewards/no_repetition_reward_func": 0.966652974486351, "rewards/verse_reward_func": 0.03125, "step": 57 }, { "completion_length": 174.875, "epoch": 0.232, "grad_norm": 1.0625, "kl": 0.005080131231807172, "learning_rate": 4.7392794005985326e-05, "loss": 0.0002, "reward": 0.9698505848646164, "reward_std": 0.0073038917616941035, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9698505699634552, "rewards/verse_reward_func": 0.0, "step": 58 }, { "completion_length": 167.9375, "epoch": 0.236, "grad_norm": 1.65625, "kl": 0.005669898469932377, "learning_rate": 4.723540933228244e-05, "loss": 0.0002, "reward": 0.99271559715271, "reward_std": 0.08150344673776999, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9614655822515488, "rewards/verse_reward_func": 0.03125, "step": 59 }, { "completion_length": 191.0625, "epoch": 0.24, "grad_norm": 0.9375, "kl": 0.007179695880040526, "learning_rate": 4.707368982147318e-05, "loss": 0.0003, "reward": 1.0425692796707153, "reward_std": 0.15895756683312356, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9488192647695541, "rewards/verse_reward_func": 0.09375, "step": 60 }, { "completion_length": 178.8125, "epoch": 0.244, "grad_norm": 1.234375, "kl": 0.007374793640337884, "learning_rate": 4.690766700109659e-05, "loss": 0.0003, "reward": 1.0702029168605804, "reward_std": 0.14750042068772018, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004166666883975267, "rewards/no_repetition_reward_func": 0.972286269068718, "rewards/verse_reward_func": 0.09375, "step": 61 }, { "completion_length": 177.5625, "epoch": 0.248, "grad_norm": 1.1328125, "kl": 0.009947373066097498, "learning_rate": 4.6737373237630476e-05, "loss": 0.0004, "reward": 1.010836973786354, "reward_std": 0.18967258161865175, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9170869737863541, "rewards/verse_reward_func": 0.09375, "step": 62 }, { "completion_length": 181.75, "epoch": 0.252, "grad_norm": 0.921875, "kl": 0.00658240367192775, "learning_rate": 4.656284173018144e-05, "loss": 0.0003, "reward": 1.0603050589561462, "reward_std": 0.28348227217793465, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9353050738573074, "rewards/verse_reward_func": 0.125, "step": 63 }, { "completion_length": 178.125, "epoch": 0.256, "grad_norm": 1.4921875, "kl": 0.009526908048428595, "learning_rate": 4.638410650401267e-05, "loss": 0.0004, "reward": 1.0653773248195648, "reward_std": 0.1409751852042973, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.971627339720726, "rewards/verse_reward_func": 0.09375, "step": 64 }, { "completion_length": 193.0625, "epoch": 0.26, "grad_norm": 0.921875, "kl": 0.007291483459994197, "learning_rate": 4.620120240391065e-05, "loss": 0.0003, "reward": 0.992166668176651, "reward_std": 0.08354807726573199, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9609166830778122, "rewards/verse_reward_func": 0.03125, "step": 65 }, { "completion_length": 163.1875, "epoch": 0.264, "grad_norm": 2.4375, "kl": 0.010100604966282845, "learning_rate": 4.601416508739211e-05, "loss": 0.0004, "reward": 0.9094432145357132, "reward_std": 0.13521920214407146, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.940693199634552, "rewards/verse_reward_func": 0.03125, "step": 66 }, { "completion_length": 186.25, "epoch": 0.268, "grad_norm": 0.94140625, "kl": 0.006481092656031251, "learning_rate": 4.5823031017752485e-05, "loss": 0.0003, "reward": 0.9819363504648209, "reward_std": 0.09091703849844635, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9506863355636597, "rewards/verse_reward_func": 0.03125, "step": 67 }, { "completion_length": 175.8125, "epoch": 0.272, "grad_norm": 2.109375, "kl": 0.008945175679400563, "learning_rate": 4.562783745695738e-05, "loss": 0.0004, "reward": 0.9387106597423553, "reward_std": 0.16390067897737026, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9699606597423553, "rewards/verse_reward_func": 0.03125, "step": 68 }, { "completion_length": 200.0, "epoch": 0.276, "grad_norm": 0.83984375, "kl": 0.0076279257191345096, "learning_rate": 4.542862245837821e-05, "loss": 0.0003, "reward": 0.9555844515562057, "reward_std": 0.1185589594533667, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9243344515562057, "rewards/verse_reward_func": 0.03125, "step": 69 }, { "completion_length": 190.375, "epoch": 0.28, "grad_norm": 1.0390625, "kl": 0.010221289936453104, "learning_rate": 4.522542485937369e-05, "loss": 0.0004, "reward": 1.0356830060482025, "reward_std": 0.14343186398036778, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0052083334885537624, "rewards/no_repetition_reward_func": 0.9679746478796005, "rewards/verse_reward_func": 0.0625, "step": 70 }, { "completion_length": 183.6875, "epoch": 0.284, "grad_norm": 1.0703125, "kl": 0.009065072517842054, "learning_rate": 4.5018284273718336e-05, "loss": 0.0004, "reward": 0.9543251842260361, "reward_std": 0.019520931062288582, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9543252140283585, "rewards/verse_reward_func": 0.0, "step": 71 }, { "completion_length": 178.125, "epoch": 0.288, "grad_norm": 1.1875, "kl": 0.009216391015797853, "learning_rate": 4.480724108387977e-05, "loss": 0.0004, "reward": 0.9448710381984711, "reward_std": 0.046969235059805214, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9448710232973099, "rewards/verse_reward_func": 0.0, "step": 72 }, { "completion_length": 195.4375, "epoch": 0.292, "grad_norm": 0.91796875, "kl": 0.007251435425132513, "learning_rate": 4.4592336433146e-05, "loss": 0.0003, "reward": 1.004251092672348, "reward_std": 0.07232979987747967, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.003289473708719015, "rewards/no_repetition_reward_func": 0.9697116166353226, "rewards/verse_reward_func": 0.03125, "step": 73 }, { "completion_length": 200.0, "epoch": 0.296, "grad_norm": 1.0546875, "kl": 0.008636260172352195, "learning_rate": 4.4373612217604496e-05, "loss": 0.0003, "reward": 1.030202716588974, "reward_std": 0.13229261268861592, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9677027016878128, "rewards/verse_reward_func": 0.0625, "step": 74 }, { "completion_length": 188.8125, "epoch": 0.3, "grad_norm": 1.1640625, "kl": 0.008647361653856933, "learning_rate": 4.415111107797445e-05, "loss": 0.0003, "reward": 1.0327503681182861, "reward_std": 0.12821446859743446, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9702503681182861, "rewards/verse_reward_func": 0.0625, "step": 75 }, { "completion_length": 190.25, "epoch": 0.304, "grad_norm": 0.94140625, "kl": 0.010642669745720923, "learning_rate": 4.3924876391293915e-05, "loss": 0.0004, "reward": 0.9804545342922211, "reward_std": 0.02781949588097632, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.011174242943525314, "rewards/no_repetition_reward_func": 0.969280332326889, "rewards/verse_reward_func": 0.0, "step": 76 }, { "completion_length": 177.5, "epoch": 0.308, "grad_norm": 1.6640625, "kl": 0.009210785734467208, "learning_rate": 4.36949522624633e-05, "loss": 0.0004, "reward": 0.906844437122345, "reward_std": 0.12991982704261318, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.969344437122345, "rewards/verse_reward_func": 0.0, "step": 77 }, { "completion_length": 169.9375, "epoch": 0.312, "grad_norm": 1.4296875, "kl": 0.0085853380151093, "learning_rate": 4.3461383515647106e-05, "loss": 0.0003, "reward": 1.0604548156261444, "reward_std": 0.1915177572518587, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.966704785823822, "rewards/verse_reward_func": 0.09375, "step": 78 }, { "completion_length": 161.625, "epoch": 0.316, "grad_norm": 1.40625, "kl": 0.008558343281038105, "learning_rate": 4.3224215685535294e-05, "loss": 0.0003, "reward": 0.9392005652189255, "reward_std": 0.19322428456507623, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9704505503177643, "rewards/verse_reward_func": 0.03125, "step": 79 }, { "completion_length": 159.0625, "epoch": 0.32, "grad_norm": 2.21875, "kl": 0.009619904682040215, "learning_rate": 4.2983495008466276e-05, "loss": 0.0004, "reward": 0.9338277578353882, "reward_std": 0.17408786993473768, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.059027777751907706, "rewards/no_repetition_reward_func": 0.9616055637598038, "rewards/verse_reward_func": 0.03125, "step": 80 }, { "completion_length": 171.875, "epoch": 0.324, "grad_norm": 1.0703125, "kl": 0.008021688903681934, "learning_rate": 4.273926841341302e-05, "loss": 0.0003, "reward": 1.0092779248952866, "reward_std": 0.10741436225362122, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9467779695987701, "rewards/verse_reward_func": 0.0625, "step": 81 }, { "completion_length": 190.3125, "epoch": 0.328, "grad_norm": 0.953125, "kl": 0.008311162469908595, "learning_rate": 4.249158351283414e-05, "loss": 0.0003, "reward": 1.0095551759004593, "reward_std": 0.08130261686164886, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0069444444961845875, "rewards/no_repetition_reward_func": 0.9713607430458069, "rewards/verse_reward_func": 0.03125, "step": 82 }, { "completion_length": 191.5, "epoch": 0.332, "grad_norm": 1.265625, "kl": 0.00910053146071732, "learning_rate": 4.224048859339175e-05, "loss": 0.0004, "reward": 1.019190862774849, "reward_std": 0.14867421332746744, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9566908329725266, "rewards/verse_reward_func": 0.0625, "step": 83 }, { "completion_length": 163.5625, "epoch": 0.336, "grad_norm": 1.5390625, "kl": 0.010091160424053669, "learning_rate": 4.198603260653792e-05, "loss": 0.0004, "reward": 0.968821257352829, "reward_std": 0.19755066523794085, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.968821257352829, "rewards/verse_reward_func": 0.0625, "step": 84 }, { "completion_length": 189.875, "epoch": 0.34, "grad_norm": 1.1796875, "kl": 0.011119180475361645, "learning_rate": 4.172826515897146e-05, "loss": 0.0004, "reward": 1.0337166488170624, "reward_std": 0.1272729904158041, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.97121661901474, "rewards/verse_reward_func": 0.0625, "step": 85 }, { "completion_length": 174.5, "epoch": 0.344, "grad_norm": 1.2734375, "kl": 0.00940567790530622, "learning_rate": 4.146723650296701e-05, "loss": 0.0004, "reward": 0.9978608191013336, "reward_std": 0.07362514256965369, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9666108191013336, "rewards/verse_reward_func": 0.03125, "step": 86 }, { "completion_length": 173.0625, "epoch": 0.348, "grad_norm": 1.2578125, "kl": 0.010684633627533913, "learning_rate": 4.1202997526578276e-05, "loss": 0.0004, "reward": 0.9645528197288513, "reward_std": 0.2310585738159716, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9645527899265289, "rewards/verse_reward_func": 0.0625, "step": 87 }, { "completion_length": 199.9375, "epoch": 0.352, "grad_norm": 1.0234375, "kl": 0.009741432731971145, "learning_rate": 4.093559974371725e-05, "loss": 0.0004, "reward": 1.0686962455511093, "reward_std": 0.20054577942937613, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0052083334885537624, "rewards/no_repetition_reward_func": 0.9697379469871521, "rewards/verse_reward_func": 0.09375, "step": 88 }, { "completion_length": 200.0, "epoch": 0.356, "grad_norm": 0.97265625, "kl": 0.010457327123731375, "learning_rate": 4.066509528411152e-05, "loss": 0.0004, "reward": 1.063912272453308, "reward_std": 0.1371896315831691, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.00390625, "rewards/no_repetition_reward_func": 0.9662560373544693, "rewards/verse_reward_func": 0.09375, "step": 89 }, { "completion_length": 186.25, "epoch": 0.36, "grad_norm": 0.95703125, "kl": 0.0091336437035352, "learning_rate": 4.039153688314145e-05, "loss": 0.0004, "reward": 0.9959292709827423, "reward_std": 0.06627507868688554, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9646792411804199, "rewards/verse_reward_func": 0.03125, "step": 90 }, { "completion_length": 197.5625, "epoch": 0.364, "grad_norm": 0.90625, "kl": 0.0125860923435539, "learning_rate": 4.011497787155938e-05, "loss": 0.0005, "reward": 1.0568312108516693, "reward_std": 0.1421456339303404, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9630811959505081, "rewards/verse_reward_func": 0.09375, "step": 91 }, { "completion_length": 177.4375, "epoch": 0.368, "grad_norm": 6.8125, "kl": 0.011468705954030156, "learning_rate": 3.983547216509254e-05, "loss": 0.0005, "reward": 1.0019132494926453, "reward_std": 0.22998703457415104, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9706632345914841, "rewards/verse_reward_func": 0.09375, "step": 92 }, { "completion_length": 191.625, "epoch": 0.372, "grad_norm": 1.140625, "kl": 0.011732830666005611, "learning_rate": 3.955307425393224e-05, "loss": 0.0005, "reward": 0.9975786060094833, "reward_std": 0.07108220970258117, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9663286209106445, "rewards/verse_reward_func": 0.03125, "step": 93 }, { "completion_length": 184.5, "epoch": 0.376, "grad_norm": 1.40625, "kl": 0.01064238091930747, "learning_rate": 3.92678391921108e-05, "loss": 0.0004, "reward": 1.0643680691719055, "reward_std": 0.19065000605769455, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9706180691719055, "rewards/verse_reward_func": 0.09375, "step": 94 }, { "completion_length": 190.25, "epoch": 0.38, "grad_norm": 1.0625, "kl": 0.008287778589874506, "learning_rate": 3.897982258676867e-05, "loss": 0.0003, "reward": 1.0316542387008667, "reward_std": 0.1355991712771356, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9691542387008667, "rewards/verse_reward_func": 0.0625, "step": 95 }, { "completion_length": 179.625, "epoch": 0.384, "grad_norm": 1.7265625, "kl": 0.01421071938239038, "learning_rate": 3.868908058731376e-05, "loss": 0.0006, "reward": 0.9379890263080597, "reward_std": 0.06504874012898654, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9692390412092209, "rewards/verse_reward_func": 0.03125, "step": 96 }, { "completion_length": 200.0, "epoch": 0.388, "grad_norm": 0.98046875, "kl": 0.013900347286835313, "learning_rate": 3.8395669874474915e-05, "loss": 0.0006, "reward": 1.0974175035953522, "reward_std": 0.15276630711741745, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0027173913549631834, "rewards/no_repetition_reward_func": 0.9697001129388809, "rewards/verse_reward_func": 0.125, "step": 97 }, { "completion_length": 151.125, "epoch": 0.392, "grad_norm": 2.171875, "kl": 0.016044694697484374, "learning_rate": 3.8099647649251986e-05, "loss": 0.0006, "reward": 0.9290718883275986, "reward_std": 0.18168721569236368, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9603218883275986, "rewards/verse_reward_func": 0.03125, "step": 98 }, { "completion_length": 187.4375, "epoch": 0.396, "grad_norm": 1.1484375, "kl": 0.019918689038604498, "learning_rate": 3.780107162176429e-05, "loss": 0.0008, "reward": 1.0208609402179718, "reward_std": 0.15501247788779438, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0052083334885537624, "rewards/no_repetition_reward_func": 0.9531526267528534, "rewards/verse_reward_func": 0.0625, "step": 99 }, { "completion_length": 169.875, "epoch": 0.4, "grad_norm": 21.625, "kl": 0.02106661314610392, "learning_rate": 3.7500000000000003e-05, "loss": 0.0008, "reward": 0.9671216309070587, "reward_std": 0.2393743423745036, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.059210526291280985, "rewards/no_repetition_reward_func": 0.9638321995735168, "rewards/verse_reward_func": 0.0625, "step": 100 }, { "completion_length": 187.625, "epoch": 0.404, "grad_norm": 1.046875, "kl": 0.012206163723021746, "learning_rate": 3.719649147846832e-05, "loss": 0.0005, "reward": 1.0463195890188217, "reward_std": 0.1548451292910613, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9525695890188217, "rewards/verse_reward_func": 0.09375, "step": 101 }, { "completion_length": 172.25, "epoch": 0.408, "grad_norm": 1.5234375, "kl": 0.011775880353525281, "learning_rate": 3.689060522675689e-05, "loss": 0.0005, "reward": 0.9707594662904739, "reward_std": 0.008258502697572112, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9707594811916351, "rewards/verse_reward_func": 0.0, "step": 102 }, { "completion_length": 179.5, "epoch": 0.412, "grad_norm": 1.015625, "kl": 0.01282069692388177, "learning_rate": 3.6582400877996546e-05, "loss": 0.0005, "reward": 1.0097861289978027, "reward_std": 0.07917119329795241, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0064484127797186375, "rewards/no_repetition_reward_func": 0.9720877408981323, "rewards/verse_reward_func": 0.03125, "step": 103 }, { "completion_length": 176.75, "epoch": 0.416, "grad_norm": 1.6015625, "kl": 0.013389652129262686, "learning_rate": 3.627193851723577e-05, "loss": 0.0005, "reward": 1.058149516582489, "reward_std": 0.19120246334932745, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.964399516582489, "rewards/verse_reward_func": 0.09375, "step": 104 }, { "completion_length": 193.5, "epoch": 0.42, "grad_norm": 1.0703125, "kl": 0.012109125847928226, "learning_rate": 3.5959278669726935e-05, "loss": 0.0005, "reward": 0.9079252034425735, "reward_std": 0.12893440295010805, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9704252034425735, "rewards/verse_reward_func": 0.0, "step": 105 }, { "completion_length": 191.25, "epoch": 0.424, "grad_norm": 0.9296875, "kl": 0.011083262274041772, "learning_rate": 3.564448228912682e-05, "loss": 0.0004, "reward": 1.0317822843790054, "reward_std": 0.08086646965239197, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.969282329082489, "rewards/verse_reward_func": 0.0625, "step": 106 }, { "completion_length": 199.5, "epoch": 0.428, "grad_norm": 0.92578125, "kl": 0.011120409704744816, "learning_rate": 3.532761074561355e-05, "loss": 0.0004, "reward": 0.9677194654941559, "reward_std": 0.010984104126691818, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9677194505929947, "rewards/verse_reward_func": 0.0, "step": 107 }, { "completion_length": 187.3125, "epoch": 0.432, "grad_norm": 1.7578125, "kl": 0.00994372135028243, "learning_rate": 3.5008725813922386e-05, "loss": 0.0004, "reward": 0.900108203291893, "reward_std": 0.1357139606261626, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9626082181930542, "rewards/verse_reward_func": 0.0, "step": 108 }, { "completion_length": 195.5625, "epoch": 0.436, "grad_norm": 0.93359375, "kl": 0.011622026795521379, "learning_rate": 3.4687889661302576e-05, "loss": 0.0005, "reward": 1.0201395899057388, "reward_std": 0.1402094402583316, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9576396048069, "rewards/verse_reward_func": 0.0625, "step": 109 }, { "completion_length": 194.4375, "epoch": 0.44, "grad_norm": 0.92578125, "kl": 0.01124755502678454, "learning_rate": 3.436516483539781e-05, "loss": 0.0004, "reward": 1.0611599385738373, "reward_std": 0.18473112024366856, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9674099236726761, "rewards/verse_reward_func": 0.09375, "step": 110 }, { "completion_length": 165.9375, "epoch": 0.444, "grad_norm": 1.4375, "kl": 0.011312742251902819, "learning_rate": 3.4040614252052305e-05, "loss": 0.0005, "reward": 1.0056527256965637, "reward_std": 0.2893778228899464, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0580357126891613, "rewards/no_repetition_reward_func": 0.9699384570121765, "rewards/verse_reward_func": 0.09375, "step": 111 }, { "completion_length": 189.4375, "epoch": 0.448, "grad_norm": 1.40625, "kl": 0.011389761231839657, "learning_rate": 3.3714301183045385e-05, "loss": 0.0005, "reward": 0.978916808962822, "reward_std": 0.10797232540789992, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.947666808962822, "rewards/verse_reward_func": 0.03125, "step": 112 }, { "completion_length": 188.75, "epoch": 0.452, "grad_norm": 1.078125, "kl": 0.01380365202203393, "learning_rate": 3.338628924375638e-05, "loss": 0.0006, "reward": 1.1212629675865173, "reward_std": 0.20934525597840548, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9650130420923233, "rewards/verse_reward_func": 0.15625, "step": 113 }, { "completion_length": 185.5625, "epoch": 0.456, "grad_norm": 1.234375, "kl": 0.01276180287823081, "learning_rate": 3.305664238076278e-05, "loss": 0.0005, "reward": 1.0015516877174377, "reward_std": 0.07264067209325731, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9703017175197601, "rewards/verse_reward_func": 0.03125, "step": 114 }, { "completion_length": 186.125, "epoch": 0.46, "grad_norm": 1.0234375, "kl": 0.01292411470785737, "learning_rate": 3.272542485937369e-05, "loss": 0.0005, "reward": 0.971293568611145, "reward_std": 0.008211376029066741, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9712935835123062, "rewards/verse_reward_func": 0.0, "step": 115 }, { "completion_length": 185.5625, "epoch": 0.464, "grad_norm": 0.9921875, "kl": 0.010555712506175041, "learning_rate": 3.239270125110117e-05, "loss": 0.0004, "reward": 0.9998664557933807, "reward_std": 0.06927325238939375, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9686164557933807, "rewards/verse_reward_func": 0.03125, "step": 116 }, { "completion_length": 200.0, "epoch": 0.468, "grad_norm": 0.8984375, "kl": 0.010709483409300447, "learning_rate": 3.205853642107192e-05, "loss": 0.0004, "reward": 1.0861131846904755, "reward_std": 0.1826954057905823, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008333333767950535, "rewards/no_repetition_reward_func": 0.9527798146009445, "rewards/verse_reward_func": 0.125, "step": 117 }, { "completion_length": 195.75, "epoch": 0.472, "grad_norm": 1.0390625, "kl": 0.011763982591219246, "learning_rate": 3.172299551538164e-05, "loss": 0.0005, "reward": 1.0221904069185257, "reward_std": 0.13792194984853268, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9596904069185257, "rewards/verse_reward_func": 0.0625, "step": 118 }, { "completion_length": 177.1875, "epoch": 0.476, "grad_norm": 0.9375, "kl": 0.016657143598422408, "learning_rate": 3.138614394839476e-05, "loss": 0.0007, "reward": 1.1469204425811768, "reward_std": 0.21332770731532946, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9594204723834991, "rewards/verse_reward_func": 0.1875, "step": 119 }, { "completion_length": 198.9375, "epoch": 0.48, "grad_norm": 0.8515625, "kl": 0.011479404289275408, "learning_rate": 3.104804738999169e-05, "loss": 0.0005, "reward": 0.9883647859096527, "reward_std": 0.08675403287634254, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9571147710084915, "rewards/verse_reward_func": 0.03125, "step": 120 }, { "completion_length": 193.125, "epoch": 0.484, "grad_norm": 0.91796875, "kl": 0.012866375269368291, "learning_rate": 3.0708771752766394e-05, "loss": 0.0005, "reward": 1.0705659836530685, "reward_std": 0.08083172226906754, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008806818397715688, "rewards/no_repetition_reward_func": 0.9680091291666031, "rewards/verse_reward_func": 0.09375, "step": 121 }, { "completion_length": 170.25, "epoch": 0.488, "grad_norm": 2.9375, "kl": 0.012929918710142374, "learning_rate": 3.0368383179176585e-05, "loss": 0.0005, "reward": 0.9417696744203568, "reward_std": 0.33038245933130383, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.125, "rewards/no_repetition_reward_func": 0.9730196893215179, "rewards/verse_reward_func": 0.09375, "step": 122 }, { "completion_length": 190.75, "epoch": 0.492, "grad_norm": 0.95703125, "kl": 0.011973332846537232, "learning_rate": 3.002694802864912e-05, "loss": 0.0005, "reward": 1.00479157269001, "reward_std": 0.0681580058299005, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9735415726900101, "rewards/verse_reward_func": 0.03125, "step": 123 }, { "completion_length": 182.625, "epoch": 0.496, "grad_norm": 1.578125, "kl": 0.012948470888659358, "learning_rate": 2.9684532864643122e-05, "loss": 0.0005, "reward": 0.9993221163749695, "reward_std": 0.06764203141210601, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9680720865726471, "rewards/verse_reward_func": 0.03125, "step": 124 }, { "completion_length": 166.125, "epoch": 0.5, "grad_norm": 1.421875, "kl": 0.013671528082340956, "learning_rate": 2.9341204441673266e-05, "loss": 0.0005, "reward": 1.000758871436119, "reward_std": 0.06821087165735662, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9695088416337967, "rewards/verse_reward_func": 0.03125, "step": 125 }, { "completion_length": 192.1875, "epoch": 0.504, "grad_norm": 1.0, "kl": 0.010890441946685314, "learning_rate": 2.8997029692295874e-05, "loss": 0.0004, "reward": 1.099923998117447, "reward_std": 0.1977190513862297, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.00390625, "rewards/no_repetition_reward_func": 0.9710177630186081, "rewards/verse_reward_func": 0.125, "step": 126 }, { "completion_length": 184.0, "epoch": 0.508, "grad_norm": 1.0625, "kl": 0.01217685965821147, "learning_rate": 2.8652075714060295e-05, "loss": 0.0005, "reward": 1.0954584777355194, "reward_std": 0.14531454246025532, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9704584330320358, "rewards/verse_reward_func": 0.125, "step": 127 }, { "completion_length": 199.1875, "epoch": 0.512, "grad_norm": 1.1015625, "kl": 0.015260245883837342, "learning_rate": 2.8306409756428064e-05, "loss": 0.0006, "reward": 1.0273628383874893, "reward_std": 0.12785551248816773, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9648628383874893, "rewards/verse_reward_func": 0.0625, "step": 128 }, { "completion_length": 168.25, "epoch": 0.516, "grad_norm": 2.015625, "kl": 0.014295702800154686, "learning_rate": 2.7960099207662532e-05, "loss": 0.0006, "reward": 0.9671203643083572, "reward_std": 0.20635199826210737, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.059027777751907706, "rewards/no_repetition_reward_func": 0.9636481553316116, "rewards/verse_reward_func": 0.0625, "step": 129 }, { "completion_length": 183.3125, "epoch": 0.52, "grad_norm": 0.921875, "kl": 0.010565208503976464, "learning_rate": 2.761321158169134e-05, "loss": 0.0004, "reward": 0.9713365137577057, "reward_std": 0.0052234133472666144, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9713364988565445, "rewards/verse_reward_func": 0.0, "step": 130 }, { "completion_length": 186.75, "epoch": 0.524, "grad_norm": 0.87890625, "kl": 0.009015375981107354, "learning_rate": 2.726581450494451e-05, "loss": 0.0004, "reward": 0.9892549365758896, "reward_std": 0.0885618986794725, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9580049067735672, "rewards/verse_reward_func": 0.03125, "step": 131 }, { "completion_length": 171.9375, "epoch": 0.528, "grad_norm": 1.4296875, "kl": 0.014596091117709875, "learning_rate": 2.6917975703170466e-05, "loss": 0.0006, "reward": 1.0221654921770096, "reward_std": 0.11091072077397257, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.02083333395421505, "rewards/no_repetition_reward_func": 0.9700821489095688, "rewards/verse_reward_func": 0.03125, "step": 132 }, { "completion_length": 184.6875, "epoch": 0.532, "grad_norm": 1.2578125, "kl": 0.010223059449344873, "learning_rate": 2.656976298823284e-05, "loss": 0.0004, "reward": 1.0608211010694504, "reward_std": 0.13878573069814593, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9670711010694504, "rewards/verse_reward_func": 0.09375, "step": 133 }, { "completion_length": 175.5625, "epoch": 0.536, "grad_norm": 1.265625, "kl": 0.014912077924236655, "learning_rate": 2.6221244244890336e-05, "loss": 0.0006, "reward": 0.899513378739357, "reward_std": 0.12862814648542553, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9620133638381958, "rewards/verse_reward_func": 0.0, "step": 134 }, { "completion_length": 161.0, "epoch": 0.54, "grad_norm": 1.4296875, "kl": 0.011082642246037722, "learning_rate": 2.587248741756253e-05, "loss": 0.0004, "reward": 1.0980724096298218, "reward_std": 0.12940445519052446, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.00390625, "rewards/no_repetition_reward_func": 0.9691661596298218, "rewards/verse_reward_func": 0.125, "step": 135 }, { "completion_length": 191.125, "epoch": 0.544, "grad_norm": 1.140625, "kl": 0.013581054750829935, "learning_rate": 2.5523560497083926e-05, "loss": 0.0005, "reward": 1.0009849965572357, "reward_std": 0.06628520804224536, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9697349965572357, "rewards/verse_reward_func": 0.03125, "step": 136 }, { "completion_length": 165.8125, "epoch": 0.548, "grad_norm": 1.46875, "kl": 0.014468337409198284, "learning_rate": 2.517453150744904e-05, "loss": 0.0006, "reward": 0.9878086149692535, "reward_std": 0.1958834775723517, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.05937499995343387, "rewards/no_repetition_reward_func": 0.9534336477518082, "rewards/verse_reward_func": 0.09375, "step": 137 }, { "completion_length": 191.0, "epoch": 0.552, "grad_norm": 1.0625, "kl": 0.017647058703005314, "learning_rate": 2.4825468492550964e-05, "loss": 0.0007, "reward": 1.0782214999198914, "reward_std": 0.14270076889079064, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.01458333432674408, "rewards/no_repetition_reward_func": 0.9698881953954697, "rewards/verse_reward_func": 0.09375, "step": 138 }, { "completion_length": 171.0, "epoch": 0.556, "grad_norm": 1.0390625, "kl": 0.011567339533939958, "learning_rate": 2.447643950291608e-05, "loss": 0.0005, "reward": 1.0274786055088043, "reward_std": 0.1633491888642311, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008484163088724017, "rewards/no_repetition_reward_func": 0.9564944505691528, "rewards/verse_reward_func": 0.0625, "step": 139 }, { "completion_length": 175.9375, "epoch": 0.56, "grad_norm": 1.1875, "kl": 0.012567983940243721, "learning_rate": 2.4127512582437485e-05, "loss": 0.0005, "reward": 0.9732318967580795, "reward_std": 0.006703993887640536, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9732318818569183, "rewards/verse_reward_func": 0.0, "step": 140 }, { "completion_length": 168.5625, "epoch": 0.564, "grad_norm": 1.359375, "kl": 0.014170312788337469, "learning_rate": 2.377875575510967e-05, "loss": 0.0006, "reward": 0.9996987283229828, "reward_std": 0.07236416323576123, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9684487730264664, "rewards/verse_reward_func": 0.03125, "step": 141 }, { "completion_length": 185.3125, "epoch": 0.568, "grad_norm": 0.9765625, "kl": 0.012509799096733332, "learning_rate": 2.3430237011767167e-05, "loss": 0.0005, "reward": 1.000573217868805, "reward_std": 0.07302216510288417, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9693232327699661, "rewards/verse_reward_func": 0.03125, "step": 142 }, { "completion_length": 172.0625, "epoch": 0.572, "grad_norm": 1.09375, "kl": 0.01766343624331057, "learning_rate": 2.3082024296829536e-05, "loss": 0.0007, "reward": 1.0561566203832626, "reward_std": 0.18419526610523462, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0036764706019312143, "rewards/no_repetition_reward_func": 0.9587301462888718, "rewards/verse_reward_func": 0.09375, "step": 143 }, { "completion_length": 192.3125, "epoch": 0.576, "grad_norm": 0.9609375, "kl": 0.012620434630662203, "learning_rate": 2.2734185495055503e-05, "loss": 0.0005, "reward": 1.0322028547525406, "reward_std": 0.12665497395209968, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9697028696537018, "rewards/verse_reward_func": 0.0625, "step": 144 }, { "completion_length": 195.375, "epoch": 0.58, "grad_norm": 0.984375, "kl": 0.01540983165614307, "learning_rate": 2.238678841830867e-05, "loss": 0.0006, "reward": 1.0079124569892883, "reward_std": 0.08523962972685695, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008680555736646056, "rewards/no_repetition_reward_func": 0.9679818898439407, "rewards/verse_reward_func": 0.03125, "step": 145 }, { "completion_length": 188.75, "epoch": 0.584, "grad_norm": 1.203125, "kl": 0.022039128933101892, "learning_rate": 2.2039900792337474e-05, "loss": 0.0009, "reward": 1.0038699060678482, "reward_std": 0.06854990951251239, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.972619891166687, "rewards/verse_reward_func": 0.03125, "step": 146 }, { "completion_length": 192.5625, "epoch": 0.588, "grad_norm": 1.0078125, "kl": 0.010647273156791925, "learning_rate": 2.1693590243571938e-05, "loss": 0.0004, "reward": 1.0933727622032166, "reward_std": 0.20574917271733284, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9683727622032166, "rewards/verse_reward_func": 0.125, "step": 147 }, { "completion_length": 170.9375, "epoch": 0.592, "grad_norm": 1.140625, "kl": 0.00910773070063442, "learning_rate": 2.1347924285939714e-05, "loss": 0.0004, "reward": 1.065127283334732, "reward_std": 0.18485231965314597, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9713772684335709, "rewards/verse_reward_func": 0.09375, "step": 148 }, { "completion_length": 188.0625, "epoch": 0.596, "grad_norm": 1.1640625, "kl": 0.012752162292599678, "learning_rate": 2.1002970307704132e-05, "loss": 0.0005, "reward": 1.0038089752197266, "reward_std": 0.06542766433267388, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.972559005022049, "rewards/verse_reward_func": 0.03125, "step": 149 }, { "completion_length": 184.875, "epoch": 0.6, "grad_norm": 1.1328125, "kl": 0.010926142102107406, "learning_rate": 2.0658795558326743e-05, "loss": 0.0004, "reward": 1.0259553045034409, "reward_std": 0.1254618337843567, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.963455319404602, "rewards/verse_reward_func": 0.0625, "step": 150 }, { "completion_length": 179.5625, "epoch": 0.604, "grad_norm": 1.375, "kl": 0.013023809995502234, "learning_rate": 2.031546713535688e-05, "loss": 0.0005, "reward": 1.0287172943353653, "reward_std": 0.326163575053215, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9662172794342041, "rewards/verse_reward_func": 0.125, "step": 151 }, { "completion_length": 180.75, "epoch": 0.608, "grad_norm": 1.0546875, "kl": 0.013435242464765906, "learning_rate": 1.9973051971350888e-05, "loss": 0.0005, "reward": 1.0376519858837128, "reward_std": 0.12582509743515402, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0024999999441206455, "rewards/no_repetition_reward_func": 0.9726519733667374, "rewards/verse_reward_func": 0.0625, "step": 152 }, { "completion_length": 195.0, "epoch": 0.612, "grad_norm": 0.9296875, "kl": 0.012024191906675696, "learning_rate": 1.963161682082342e-05, "loss": 0.0005, "reward": 1.0774504840373993, "reward_std": 0.12127143458928913, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9524504691362381, "rewards/verse_reward_func": 0.125, "step": 153 }, { "completion_length": 191.375, "epoch": 0.616, "grad_norm": 0.96484375, "kl": 0.019440859090536833, "learning_rate": 1.9291228247233605e-05, "loss": 0.0008, "reward": 1.0334815829992294, "reward_std": 0.13698444701731205, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.00390625, "rewards/no_repetition_reward_func": 0.967075303196907, "rewards/verse_reward_func": 0.0625, "step": 154 }, { "completion_length": 189.75, "epoch": 0.62, "grad_norm": 1.046875, "kl": 0.019582425709813833, "learning_rate": 1.895195261000831e-05, "loss": 0.0008, "reward": 1.0371443033218384, "reward_std": 0.13878979021683335, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004166666883975267, "rewards/no_repetition_reward_func": 0.9704776406288147, "rewards/verse_reward_func": 0.0625, "step": 155 }, { "completion_length": 196.5, "epoch": 0.624, "grad_norm": 0.98046875, "kl": 0.01221572607755661, "learning_rate": 1.8613856051605243e-05, "loss": 0.0005, "reward": 0.934912696480751, "reward_std": 0.19153758394531906, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9661626815795898, "rewards/verse_reward_func": 0.03125, "step": 156 }, { "completion_length": 190.4375, "epoch": 0.628, "grad_norm": 1.109375, "kl": 0.01733283861540258, "learning_rate": 1.827700448461836e-05, "loss": 0.0007, "reward": 1.0045756101608276, "reward_std": 0.09838507196400315, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.01666666753590107, "rewards/no_repetition_reward_func": 0.9566589593887329, "rewards/verse_reward_func": 0.03125, "step": 157 }, { "completion_length": 189.6875, "epoch": 0.632, "grad_norm": 1.0234375, "kl": 0.011717171175405383, "learning_rate": 1.7941463578928086e-05, "loss": 0.0005, "reward": 1.081895649433136, "reward_std": 0.16829337610397488, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9568956643342972, "rewards/verse_reward_func": 0.125, "step": 158 }, { "completion_length": 193.1875, "epoch": 0.636, "grad_norm": 0.875, "kl": 0.010208276333287358, "learning_rate": 1.7607298748898842e-05, "loss": 0.0004, "reward": 1.0351229459047318, "reward_std": 0.08973796060308814, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.004807692486792803, "rewards/no_repetition_reward_func": 0.9678152203559875, "rewards/verse_reward_func": 0.0625, "step": 159 }, { "completion_length": 185.125, "epoch": 0.64, "grad_norm": 1.1875, "kl": 0.01279965229332447, "learning_rate": 1.7274575140626318e-05, "loss": 0.0005, "reward": 1.017332836985588, "reward_std": 0.1597475241869688, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9548328071832657, "rewards/verse_reward_func": 0.0625, "step": 160 }, { "completion_length": 177.0, "epoch": 0.644, "grad_norm": 1.0234375, "kl": 0.015851975651457906, "learning_rate": 1.6943357619237226e-05, "loss": 0.0006, "reward": 1.0032832622528076, "reward_std": 0.06546501937555149, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9720332771539688, "rewards/verse_reward_func": 0.03125, "step": 161 }, { "completion_length": 183.1875, "epoch": 0.648, "grad_norm": 1.015625, "kl": 0.013623830629512668, "learning_rate": 1.6613710756243626e-05, "loss": 0.0005, "reward": 1.0095181614160538, "reward_std": 0.25556074921041727, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.050260416232049465, "rewards/no_repetition_reward_func": 0.9660285711288452, "rewards/verse_reward_func": 0.09375, "step": 162 }, { "completion_length": 167.0, "epoch": 0.652, "grad_norm": 1.9609375, "kl": 0.013558724895119667, "learning_rate": 1.6285698816954624e-05, "loss": 0.0005, "reward": 0.9124084264039993, "reward_std": 0.1921169520355761, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9436584264039993, "rewards/verse_reward_func": 0.03125, "step": 163 }, { "completion_length": 200.0, "epoch": 0.656, "grad_norm": 0.87890625, "kl": 0.010941914282739162, "learning_rate": 1.5959385747947698e-05, "loss": 0.0004, "reward": 1.0318201780319214, "reward_std": 0.12901913665700704, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9693201929330826, "rewards/verse_reward_func": 0.0625, "step": 164 }, { "completion_length": 174.625, "epoch": 0.66, "grad_norm": 1.265625, "kl": 0.015559795778244734, "learning_rate": 1.56348351646022e-05, "loss": 0.0006, "reward": 0.9770224392414093, "reward_std": 0.017092078109271824, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0052083334885537624, "rewards/no_repetition_reward_func": 0.9718141108751297, "rewards/verse_reward_func": 0.0, "step": 165 }, { "completion_length": 200.0, "epoch": 0.664, "grad_norm": 0.9140625, "kl": 0.01205143821425736, "learning_rate": 1.5312110338697426e-05, "loss": 0.0005, "reward": 1.0031522065401077, "reward_std": 0.0669463153462857, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9719022214412689, "rewards/verse_reward_func": 0.03125, "step": 166 }, { "completion_length": 167.875, "epoch": 0.668, "grad_norm": 1.2109375, "kl": 0.012573536252602935, "learning_rate": 1.4991274186077632e-05, "loss": 0.0005, "reward": 1.0363154411315918, "reward_std": 0.12498233921360224, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9738154411315918, "rewards/verse_reward_func": 0.0625, "step": 167 }, { "completion_length": 165.3125, "epoch": 0.672, "grad_norm": 3.9375, "kl": 0.01248105801641941, "learning_rate": 1.467238925438646e-05, "loss": 0.0005, "reward": 0.902620941400528, "reward_std": 0.33662330894730985, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.125, "rewards/no_repetition_reward_func": 0.9651209115982056, "rewards/verse_reward_func": 0.0625, "step": 168 }, { "completion_length": 189.8125, "epoch": 0.676, "grad_norm": 1.171875, "kl": 0.014783380553126335, "learning_rate": 1.4355517710873184e-05, "loss": 0.0006, "reward": 0.9686231762170792, "reward_std": 0.007990248850546777, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9686231911182404, "rewards/verse_reward_func": 0.0, "step": 169 }, { "completion_length": 176.75, "epoch": 0.68, "grad_norm": 1.984375, "kl": 0.00982649892102927, "learning_rate": 1.4040721330273062e-05, "loss": 0.0004, "reward": 1.0318966805934906, "reward_std": 0.19692795420996845, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9693966805934906, "rewards/verse_reward_func": 0.125, "step": 170 }, { "completion_length": 171.625, "epoch": 0.684, "grad_norm": 1.109375, "kl": 0.01388387638144195, "learning_rate": 1.3728061482764238e-05, "loss": 0.0006, "reward": 1.1590849459171295, "reward_std": 0.20052161253988743, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9715849459171295, "rewards/verse_reward_func": 0.1875, "step": 171 }, { "completion_length": 188.6875, "epoch": 0.688, "grad_norm": 1.0390625, "kl": 0.013264609966427088, "learning_rate": 1.3417599122003464e-05, "loss": 0.0005, "reward": 0.9750119894742966, "reward_std": 0.020695629762485623, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0060068650636821985, "rewards/no_repetition_reward_func": 0.9690051376819611, "rewards/verse_reward_func": 0.0, "step": 172 }, { "completion_length": 175.6875, "epoch": 0.692, "grad_norm": 0.9609375, "kl": 0.012299251044169068, "learning_rate": 1.3109394773243117e-05, "loss": 0.0005, "reward": 1.027367189526558, "reward_std": 0.1334714696276933, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9648671746253967, "rewards/verse_reward_func": 0.0625, "step": 173 }, { "completion_length": 196.25, "epoch": 0.696, "grad_norm": 0.9296875, "kl": 0.012535229092463851, "learning_rate": 1.280350852153168e-05, "loss": 0.0005, "reward": 0.965914785861969, "reward_std": 0.009997655171900988, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9659148007631302, "rewards/verse_reward_func": 0.0, "step": 174 }, { "completion_length": 189.9375, "epoch": 0.7, "grad_norm": 1.09375, "kl": 0.011037069605663419, "learning_rate": 1.2500000000000006e-05, "loss": 0.0004, "reward": 1.0548015385866165, "reward_std": 0.1706442846916616, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.005681818351149559, "rewards/no_repetition_reward_func": 0.9553697109222412, "rewards/verse_reward_func": 0.09375, "step": 175 }, { "completion_length": 191.0625, "epoch": 0.704, "grad_norm": 1.0, "kl": 0.011079935124143958, "learning_rate": 1.2198928378235716e-05, "loss": 0.0004, "reward": 1.0105179399251938, "reward_std": 0.0821902226889506, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0078125, "rewards/no_repetition_reward_func": 0.971455454826355, "rewards/verse_reward_func": 0.03125, "step": 176 }, { "completion_length": 197.8125, "epoch": 0.708, "grad_norm": 0.8828125, "kl": 0.009368012193590403, "learning_rate": 1.1900352350748026e-05, "loss": 0.0004, "reward": 0.9249662607908249, "reward_std": 0.21413116506300867, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9562162756919861, "rewards/verse_reward_func": 0.03125, "step": 177 }, { "completion_length": 193.25, "epoch": 0.712, "grad_norm": 1.03125, "kl": 0.011819357518106699, "learning_rate": 1.1604330125525079e-05, "loss": 0.0005, "reward": 0.964430496096611, "reward_std": 0.19670187239535153, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9644304662942886, "rewards/verse_reward_func": 0.0625, "step": 178 }, { "completion_length": 188.5, "epoch": 0.716, "grad_norm": 1.421875, "kl": 0.011817316291853786, "learning_rate": 1.1310919412686247e-05, "loss": 0.0005, "reward": 1.1032432317733765, "reward_std": 0.2703648883616552, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.052483974024653435, "rewards/no_repetition_reward_func": 0.9682272225618362, "rewards/verse_reward_func": 0.1875, "step": 179 }, { "completion_length": 168.625, "epoch": 0.72, "grad_norm": 1.125, "kl": 0.015864971559494734, "learning_rate": 1.1020177413231334e-05, "loss": 0.0006, "reward": 1.0682349503040314, "reward_std": 0.13318812719080597, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.003289473708719015, "rewards/no_repetition_reward_func": 0.9711955040693283, "rewards/verse_reward_func": 0.09375, "step": 180 }, { "completion_length": 181.5, "epoch": 0.724, "grad_norm": 1.140625, "kl": 0.014111926779150963, "learning_rate": 1.0732160807889211e-05, "loss": 0.0006, "reward": 0.972553163766861, "reward_std": 0.0047605925356037915, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9725531786680222, "rewards/verse_reward_func": 0.0, "step": 181 }, { "completion_length": 200.0, "epoch": 0.728, "grad_norm": 0.921875, "kl": 0.010279940441250801, "learning_rate": 1.0446925746067768e-05, "loss": 0.0004, "reward": 1.0327564477920532, "reward_std": 0.13015011709649116, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9702564477920532, "rewards/verse_reward_func": 0.0625, "step": 182 }, { "completion_length": 176.0, "epoch": 0.732, "grad_norm": 0.8515625, "kl": 0.011694031534716487, "learning_rate": 1.0164527834907467e-05, "loss": 0.0005, "reward": 1.0148691684007645, "reward_std": 0.1078255403554067, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.008275691885501146, "rewards/no_repetition_reward_func": 0.9440935105085373, "rewards/verse_reward_func": 0.0625, "step": 183 }, { "completion_length": 187.6875, "epoch": 0.736, "grad_norm": 1.1796875, "kl": 0.011029373039491475, "learning_rate": 9.88502212844063e-06, "loss": 0.0004, "reward": 1.063307762145996, "reward_std": 0.18685762688983232, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9695577621459961, "rewards/verse_reward_func": 0.09375, "step": 184 }, { "completion_length": 186.5625, "epoch": 0.74, "grad_norm": 1.046875, "kl": 0.015813767444342375, "learning_rate": 9.608463116858542e-06, "loss": 0.0006, "reward": 1.038159966468811, "reward_std": 0.1321913282154128, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0036764706019312143, "rewards/no_repetition_reward_func": 0.9719835072755814, "rewards/verse_reward_func": 0.0625, "step": 185 }, { "completion_length": 200.0, "epoch": 0.744, "grad_norm": 0.9296875, "kl": 0.010945732123218477, "learning_rate": 9.334904715888495e-06, "loss": 0.0004, "reward": 1.002086028456688, "reward_std": 0.06929445278365165, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9708360135555267, "rewards/verse_reward_func": 0.03125, "step": 186 }, { "completion_length": 194.5, "epoch": 0.748, "grad_norm": 0.9921875, "kl": 0.012873378116637468, "learning_rate": 9.064400256282757e-06, "loss": 0.0005, "reward": 0.9956127107143402, "reward_std": 0.07935338048264384, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9643627256155014, "rewards/verse_reward_func": 0.03125, "step": 187 }, { "completion_length": 194.8125, "epoch": 0.752, "grad_norm": 0.8359375, "kl": 0.0095058165024966, "learning_rate": 8.797002473421728e-06, "loss": 0.0004, "reward": 1.001351773738861, "reward_std": 0.18411000445485115, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9388517886400223, "rewards/verse_reward_func": 0.0625, "step": 188 }, { "completion_length": 188.375, "epoch": 0.756, "grad_norm": 1.828125, "kl": 0.013147451216354966, "learning_rate": 8.532763497032987e-06, "loss": 0.0005, "reward": 0.9736805409193039, "reward_std": 0.24951145879458636, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.9736805409193039, "rewards/verse_reward_func": 0.0625, "step": 189 }, { "completion_length": 183.0625, "epoch": 0.76, "grad_norm": 0.9921875, "kl": 0.01549130748026073, "learning_rate": 8.271734841028553e-06, "loss": 0.0006, "reward": 1.0413794666528702, "reward_std": 0.14600584097206593, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.010273692896589637, "rewards/no_repetition_reward_func": 0.9686057716608047, "rewards/verse_reward_func": 0.0625, "step": 190 }, { "completion_length": 176.4375, "epoch": 0.764, "grad_norm": 1.296875, "kl": 0.01476406422443688, "learning_rate": 8.013967393462094e-06, "loss": 0.0006, "reward": 1.004815697669983, "reward_std": 0.06899422314018011, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9735656976699829, "rewards/verse_reward_func": 0.03125, "step": 191 }, { "completion_length": 184.625, "epoch": 0.768, "grad_norm": 1.2734375, "kl": 0.013548333663493395, "learning_rate": 7.759511406608255e-06, "loss": 0.0005, "reward": 0.9655581712722778, "reward_std": 0.014207061380147934, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9655582010746002, "rewards/verse_reward_func": 0.0, "step": 192 }, { "completion_length": 182.1875, "epoch": 0.772, "grad_norm": 1.1328125, "kl": 0.010045995935797691, "learning_rate": 7.508416487165862e-06, "loss": 0.0004, "reward": 0.9854150861501694, "reward_std": 0.09294597333064303, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9541650712490082, "rewards/verse_reward_func": 0.03125, "step": 193 }, { "completion_length": 184.3125, "epoch": 0.776, "grad_norm": 1.234375, "kl": 0.010615430539473891, "learning_rate": 7.260731586586983e-06, "loss": 0.0004, "reward": 1.0294946879148483, "reward_std": 0.13081860658712685, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9669946581125259, "rewards/verse_reward_func": 0.0625, "step": 194 }, { "completion_length": 199.75, "epoch": 0.78, "grad_norm": 1.0078125, "kl": 0.010976087069138885, "learning_rate": 7.016504991533726e-06, "loss": 0.0004, "reward": 1.027074545621872, "reward_std": 0.1214328232454136, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9645745754241943, "rewards/verse_reward_func": 0.0625, "step": 195 }, { "completion_length": 178.75, "epoch": 0.784, "grad_norm": 2.421875, "kl": 0.014907070668414235, "learning_rate": 6.775784314464717e-06, "loss": 0.0006, "reward": 0.9615938365459442, "reward_std": 0.24925772845745087, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.0625, "rewards/no_repetition_reward_func": 0.961593821644783, "rewards/verse_reward_func": 0.0625, "step": 196 }, { "completion_length": 199.0625, "epoch": 0.788, "grad_norm": 0.953125, "kl": 0.013197696302086115, "learning_rate": 6.538616484352902e-06, "loss": 0.0005, "reward": 1.0271732360124588, "reward_std": 0.1338005670113489, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9646732360124588, "rewards/verse_reward_func": 0.0625, "step": 197 }, { "completion_length": 188.875, "epoch": 0.792, "grad_norm": 1.0546875, "kl": 0.009762313449755311, "learning_rate": 6.305047737536707e-06, "loss": 0.0004, "reward": 1.0017031729221344, "reward_std": 0.06734973879065365, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9704531729221344, "rewards/verse_reward_func": 0.03125, "step": 198 }, { "completion_length": 177.625, "epoch": 0.796, "grad_norm": 0.8359375, "kl": 0.011908727698028088, "learning_rate": 6.075123608706093e-06, "loss": 0.0005, "reward": 0.97906094789505, "reward_std": 0.10488309944048524, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": 0.0, "rewards/no_repetition_reward_func": 0.9478109329938889, "rewards/verse_reward_func": 0.03125, "step": 199 }, { "completion_length": 188.5625, "epoch": 0.8, "grad_norm": 1.7109375, "kl": 0.010567213874310255, "learning_rate": 5.848888922025553e-06, "loss": 0.0004, "reward": 0.9393830448389053, "reward_std": 0.0688461393583566, "rewards/check_divine_comedy_plagiarism": 0.0, "rewards/endecasillabo_reward_func": -0.059523809468373656, "rewards/no_repetition_reward_func": 0.9676568508148193, "rewards/verse_reward_func": 0.03125, "step": 200 } ], "logging_steps": 1, "max_steps": 250, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }