{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.4, "eval_steps": 500, "global_step": 100, "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 } ], "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 }