| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 27, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.037037037037037035, | |
| "grad_norm": 27.55777176285727, | |
| "learning_rate": 1.6666666666666665e-07, | |
| "logits/chosen": -1.3260200023651123, | |
| "logits/rejected": -1.3555822372436523, | |
| "logps/chosen": -620.6215209960938, | |
| "logps/rejected": -538.2889404296875, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 25.11934361945706, | |
| "learning_rate": 3.333333333333333e-07, | |
| "logits/chosen": -1.3862979412078857, | |
| "logits/rejected": -1.418839931488037, | |
| "logps/chosen": -548.8743286132812, | |
| "logps/rejected": -502.6883544921875, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 26.735838110076433, | |
| "learning_rate": 5e-07, | |
| "logits/chosen": -1.3722721338272095, | |
| "logits/rejected": -1.4019339084625244, | |
| "logps/chosen": -593.66552734375, | |
| "logps/rejected": -518.8505249023438, | |
| "loss": 0.6926, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": 0.0003613852895796299, | |
| "rewards/margins": 0.002076806966215372, | |
| "rewards/rejected": -0.001715421793051064, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 27.831815566219696, | |
| "learning_rate": 4.978612153434526e-07, | |
| "logits/chosen": -1.3279492855072021, | |
| "logits/rejected": -1.4584531784057617, | |
| "logps/chosen": -616.8908081054688, | |
| "logps/rejected": -478.5179443359375, | |
| "loss": 0.6906, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 0.002412938978523016, | |
| "rewards/margins": 0.00437101349234581, | |
| "rewards/rejected": -0.001958074513822794, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.18518518518518517, | |
| "grad_norm": 26.45387550515481, | |
| "learning_rate": 4.91481456572267e-07, | |
| "logits/chosen": -1.2913222312927246, | |
| "logits/rejected": -1.4004031419754028, | |
| "logps/chosen": -596.9254150390625, | |
| "logps/rejected": -491.03643798828125, | |
| "loss": 0.6846, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.012376622296869755, | |
| "rewards/margins": 0.019729193300008774, | |
| "rewards/rejected": -0.007352571468800306, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 26.113162913083638, | |
| "learning_rate": 4.809698831278217e-07, | |
| "logits/chosen": -1.2476236820220947, | |
| "logits/rejected": -1.3365733623504639, | |
| "logps/chosen": -576.7242431640625, | |
| "logps/rejected": -499.41351318359375, | |
| "loss": 0.6689, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 0.014585882425308228, | |
| "rewards/margins": 0.039108239114284515, | |
| "rewards/rejected": -0.024522356688976288, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.25925925925925924, | |
| "grad_norm": 24.212448317206544, | |
| "learning_rate": 4.6650635094610966e-07, | |
| "logits/chosen": -1.3422677516937256, | |
| "logits/rejected": -1.4066834449768066, | |
| "logps/chosen": -550.8783569335938, | |
| "logps/rejected": -444.896484375, | |
| "loss": 0.6611, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 0.01903172954916954, | |
| "rewards/margins": 0.05026581883430481, | |
| "rewards/rejected": -0.031234093010425568, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 25.92207003651225, | |
| "learning_rate": 4.483383350728088e-07, | |
| "logits/chosen": -1.155773401260376, | |
| "logits/rejected": -1.2405778169631958, | |
| "logps/chosen": -660.4547119140625, | |
| "logps/rejected": -549.8304443359375, | |
| "loss": 0.6308, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.04862784221768379, | |
| "rewards/margins": 0.11470220237970352, | |
| "rewards/rejected": -0.06607436388731003, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 22.804718326627064, | |
| "learning_rate": 4.2677669529663686e-07, | |
| "logits/chosen": -1.1948490142822266, | |
| "logits/rejected": -1.2298305034637451, | |
| "logps/chosen": -661.9083251953125, | |
| "logps/rejected": -571.0025634765625, | |
| "loss": 0.6201, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.05401439964771271, | |
| "rewards/margins": 0.20858871936798096, | |
| "rewards/rejected": -0.15457431972026825, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 21.49943323053994, | |
| "learning_rate": 4.0219035725218013e-07, | |
| "logits/chosen": -1.2277066707611084, | |
| "logits/rejected": -1.2695496082305908, | |
| "logps/chosen": -664.5496826171875, | |
| "logps/rejected": -551.2160034179688, | |
| "loss": 0.5948, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.05593748763203621, | |
| "rewards/margins": 0.2475208044052124, | |
| "rewards/rejected": -0.1915833204984665, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4074074074074074, | |
| "grad_norm": 21.083207323643638, | |
| "learning_rate": 3.75e-07, | |
| "logits/chosen": -1.2142881155014038, | |
| "logits/rejected": -1.2608280181884766, | |
| "logps/chosen": -493.3670654296875, | |
| "logps/rejected": -460.8230285644531, | |
| "loss": 0.5937, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.02144962176680565, | |
| "rewards/margins": 0.19565340876579285, | |
| "rewards/rejected": -0.2171030193567276, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 22.30950608958288, | |
| "learning_rate": 3.4567085809127245e-07, | |
| "logits/chosen": -1.1867148876190186, | |
| "logits/rejected": -1.204742431640625, | |
| "logps/chosen": -518.848876953125, | |
| "logps/rejected": -477.93121337890625, | |
| "loss": 0.5571, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -0.030759546905755997, | |
| "rewards/margins": 0.27591627836227417, | |
| "rewards/rejected": -0.3066757917404175, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.48148148148148145, | |
| "grad_norm": 21.647385072039942, | |
| "learning_rate": 3.147047612756302e-07, | |
| "logits/chosen": -1.1367645263671875, | |
| "logits/rejected": -1.1599252223968506, | |
| "logps/chosen": -575.5039672851562, | |
| "logps/rejected": -499.92059326171875, | |
| "loss": 0.5172, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.07214581221342087, | |
| "rewards/margins": 0.3617976903915405, | |
| "rewards/rejected": -0.4339434802532196, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5185185185185185, | |
| "grad_norm": 21.79316246052631, | |
| "learning_rate": 2.826315480550129e-07, | |
| "logits/chosen": -1.077150821685791, | |
| "logits/rejected": -1.1288504600524902, | |
| "logps/chosen": -556.2161865234375, | |
| "logps/rejected": -495.59539794921875, | |
| "loss": 0.4981, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -0.12262667715549469, | |
| "rewards/margins": 0.3966625928878784, | |
| "rewards/rejected": -0.5192892551422119, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 21.27092910668923, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": -1.135519027709961, | |
| "logits/rejected": -1.1375882625579834, | |
| "logps/chosen": -600.9283447265625, | |
| "logps/rejected": -557.6646118164062, | |
| "loss": 0.4817, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.17058339715003967, | |
| "rewards/margins": 0.44923293590545654, | |
| "rewards/rejected": -0.6198163628578186, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 20.418108861546624, | |
| "learning_rate": 2.1736845194498716e-07, | |
| "logits/chosen": -1.065222144126892, | |
| "logits/rejected": -1.1098757982254028, | |
| "logps/chosen": -516.3076782226562, | |
| "logps/rejected": -414.37371826171875, | |
| "loss": 0.4632, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -0.3122747242450714, | |
| "rewards/margins": 0.4216526746749878, | |
| "rewards/rejected": -0.7339274287223816, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6296296296296297, | |
| "grad_norm": 19.79277850701313, | |
| "learning_rate": 1.8529523872436977e-07, | |
| "logits/chosen": -1.017812967300415, | |
| "logits/rejected": -1.0660429000854492, | |
| "logps/chosen": -720.6757202148438, | |
| "logps/rejected": -684.0345458984375, | |
| "loss": 0.4552, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.32912153005599976, | |
| "rewards/margins": 0.5706872940063477, | |
| "rewards/rejected": -0.8998088240623474, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 20.42225137733936, | |
| "learning_rate": 1.5432914190872756e-07, | |
| "logits/chosen": -1.0442301034927368, | |
| "logits/rejected": -1.0349968671798706, | |
| "logps/chosen": -747.7850952148438, | |
| "logps/rejected": -685.4484252929688, | |
| "loss": 0.4673, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -0.3863644599914551, | |
| "rewards/margins": 0.5358256101608276, | |
| "rewards/rejected": -0.9221900701522827, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7037037037037037, | |
| "grad_norm": 22.47255209023482, | |
| "learning_rate": 1.2500000000000005e-07, | |
| "logits/chosen": -1.1190571784973145, | |
| "logits/rejected": -1.1354550123214722, | |
| "logps/chosen": -685.4273681640625, | |
| "logps/rejected": -646.656005859375, | |
| "loss": 0.4169, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.4319349527359009, | |
| "rewards/margins": 0.7433109879493713, | |
| "rewards/rejected": -1.175246000289917, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 20.30453136903234, | |
| "learning_rate": 9.780964274781983e-08, | |
| "logits/chosen": -0.9758723974227905, | |
| "logits/rejected": -0.9760102033615112, | |
| "logps/chosen": -701.876220703125, | |
| "logps/rejected": -603.935546875, | |
| "loss": 0.4295, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -0.4580194056034088, | |
| "rewards/margins": 0.6639784574508667, | |
| "rewards/rejected": -1.1219978332519531, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 19.920032346717264, | |
| "learning_rate": 7.322330470336313e-08, | |
| "logits/chosen": -1.0556690692901611, | |
| "logits/rejected": -1.038096308708191, | |
| "logps/chosen": -790.4100952148438, | |
| "logps/rejected": -738.6947021484375, | |
| "loss": 0.3989, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.5596134662628174, | |
| "rewards/margins": 0.6681773662567139, | |
| "rewards/rejected": -1.2277909517288208, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8148148148148148, | |
| "grad_norm": 19.811761984641322, | |
| "learning_rate": 5.166166492719124e-08, | |
| "logits/chosen": -1.0022271871566772, | |
| "logits/rejected": -0.9993726015090942, | |
| "logps/chosen": -648.9810180664062, | |
| "logps/rejected": -578.921142578125, | |
| "loss": 0.3949, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.5851563215255737, | |
| "rewards/margins": 0.786234974861145, | |
| "rewards/rejected": -1.3713912963867188, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.8518518518518519, | |
| "grad_norm": 19.77551914709964, | |
| "learning_rate": 3.349364905389032e-08, | |
| "logits/chosen": -0.924078106880188, | |
| "logits/rejected": -0.9447305202484131, | |
| "logps/chosen": -727.5433959960938, | |
| "logps/rejected": -727.569580078125, | |
| "loss": 0.3777, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.3620569705963135, | |
| "rewards/margins": 1.0265687704086304, | |
| "rewards/rejected": -1.3886257410049438, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 20.99090983878074, | |
| "learning_rate": 1.9030116872178314e-08, | |
| "logits/chosen": -0.9339395761489868, | |
| "logits/rejected": -1.002177357673645, | |
| "logps/chosen": -772.19921875, | |
| "logps/rejected": -641.7393798828125, | |
| "loss": 0.3809, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.4156669080257416, | |
| "rewards/margins": 0.8427646160125732, | |
| "rewards/rejected": -1.2584315538406372, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9259259259259259, | |
| "grad_norm": 19.60455577981017, | |
| "learning_rate": 8.518543427732949e-09, | |
| "logits/chosen": -1.0230871438980103, | |
| "logits/rejected": -0.9638053178787231, | |
| "logps/chosen": -676.6509399414062, | |
| "logps/rejected": -698.9414672851562, | |
| "loss": 0.3781, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.5820339918136597, | |
| "rewards/margins": 0.7440013885498047, | |
| "rewards/rejected": -1.326035499572754, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9629629629629629, | |
| "grad_norm": 20.184672646004614, | |
| "learning_rate": 2.1387846565474044e-09, | |
| "logits/chosen": -0.9462152123451233, | |
| "logits/rejected": -0.9009606242179871, | |
| "logps/chosen": -672.621337890625, | |
| "logps/rejected": -633.5792236328125, | |
| "loss": 0.3642, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -0.6688261032104492, | |
| "rewards/margins": 0.8014542460441589, | |
| "rewards/rejected": -1.470280408859253, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 19.209616333813695, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -1.027219533920288, | |
| "logits/rejected": -1.0129032135009766, | |
| "logps/chosen": -630.52880859375, | |
| "logps/rejected": -564.9932250976562, | |
| "loss": 0.3937, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -0.5863191485404968, | |
| "rewards/margins": 0.763579249382019, | |
| "rewards/rejected": -1.349898338317871, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 27, | |
| "total_flos": 0.0, | |
| "train_loss": 0.525856320504789, | |
| "train_runtime": 769.8946, | |
| "train_samples_per_second": 2.244, | |
| "train_steps_per_second": 0.035 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 27, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |