| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 1000, | |
| "global_step": 169, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005917159763313609, | |
| "grad_norm": 86.11190361804692, | |
| "learning_rate": 2.941176470588235e-08, | |
| "logits/chosen": -0.1997361034154892, | |
| "logits/rejected": -0.19101263582706451, | |
| "logps/chosen": -210.513671875, | |
| "logps/rejected": -157.554931640625, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": 78.94990602073459, | |
| "learning_rate": 2.941176470588235e-07, | |
| "logits/chosen": -0.13639792799949646, | |
| "logits/rejected": -0.1734662801027298, | |
| "logps/chosen": -218.1241912841797, | |
| "logps/rejected": -238.053466796875, | |
| "loss": 0.6903, | |
| "rewards/accuracies": 0.37037035822868347, | |
| "rewards/chosen": -0.008327378891408443, | |
| "rewards/margins": -0.011080991476774216, | |
| "rewards/rejected": 0.002753612119704485, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "grad_norm": 76.16607821763226, | |
| "learning_rate": 4.901315789473684e-07, | |
| "logits/chosen": -0.1947246491909027, | |
| "logits/rejected": -0.24646611511707306, | |
| "logps/chosen": -177.68190002441406, | |
| "logps/rejected": -174.81893920898438, | |
| "loss": 0.6672, | |
| "rewards/accuracies": 0.5833333730697632, | |
| "rewards/chosen": -0.04245181754231453, | |
| "rewards/margins": 0.05021858215332031, | |
| "rewards/rejected": -0.09267039597034454, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17751479289940827, | |
| "grad_norm": 109.70241370606963, | |
| "learning_rate": 4.5723684210526313e-07, | |
| "logits/chosen": -0.06654810905456543, | |
| "logits/rejected": -0.10471577942371368, | |
| "logps/chosen": -205.10757446289062, | |
| "logps/rejected": -221.33865356445312, | |
| "loss": 0.6287, | |
| "rewards/accuracies": 0.6333333253860474, | |
| "rewards/chosen": -0.1406942903995514, | |
| "rewards/margins": 0.19435003399848938, | |
| "rewards/rejected": -0.33504432439804077, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 104.51396601202018, | |
| "learning_rate": 4.243421052631579e-07, | |
| "logits/chosen": -0.28040310740470886, | |
| "logits/rejected": -0.22528812289237976, | |
| "logps/chosen": -218.5230712890625, | |
| "logps/rejected": -223.2051239013672, | |
| "loss": 0.5973, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.35167446732521057, | |
| "rewards/margins": 0.22981619834899902, | |
| "rewards/rejected": -0.5814906358718872, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2958579881656805, | |
| "grad_norm": 77.4636654508096, | |
| "learning_rate": 3.914473684210526e-07, | |
| "logits/chosen": -0.09673222154378891, | |
| "logits/rejected": -0.046957988291978836, | |
| "logps/chosen": -151.72000122070312, | |
| "logps/rejected": -199.17935180664062, | |
| "loss": 0.5671, | |
| "rewards/accuracies": 0.7833333611488342, | |
| "rewards/chosen": -0.3736603260040283, | |
| "rewards/margins": 0.3569501042366028, | |
| "rewards/rejected": -0.7306104898452759, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "grad_norm": 64.26194560632132, | |
| "learning_rate": 3.5855263157894734e-07, | |
| "logits/chosen": -0.08753165602684021, | |
| "logits/rejected": -0.07491908967494965, | |
| "logps/chosen": -178.92002868652344, | |
| "logps/rejected": -237.271240234375, | |
| "loss": 0.5653, | |
| "rewards/accuracies": 0.6666666865348816, | |
| "rewards/chosen": -0.44847816228866577, | |
| "rewards/margins": 0.366828054189682, | |
| "rewards/rejected": -0.8153061866760254, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.41420118343195267, | |
| "grad_norm": 75.47610499926685, | |
| "learning_rate": 3.2565789473684206e-07, | |
| "logits/chosen": -0.25761300325393677, | |
| "logits/rejected": -0.25849616527557373, | |
| "logps/chosen": -191.11752319335938, | |
| "logps/rejected": -177.6903839111328, | |
| "loss": 0.5494, | |
| "rewards/accuracies": 0.6500000357627869, | |
| "rewards/chosen": -0.5434780120849609, | |
| "rewards/margins": 0.3972131609916687, | |
| "rewards/rejected": -0.9406911730766296, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "grad_norm": 68.31555898880357, | |
| "learning_rate": 2.9276315789473684e-07, | |
| "logits/chosen": -0.06761662662029266, | |
| "logits/rejected": -0.08591805398464203, | |
| "logps/chosen": -151.31748962402344, | |
| "logps/rejected": -214.83871459960938, | |
| "loss": 0.5368, | |
| "rewards/accuracies": 0.7666666507720947, | |
| "rewards/chosen": -0.34307175874710083, | |
| "rewards/margins": 0.5734245181083679, | |
| "rewards/rejected": -0.9164963960647583, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5325443786982249, | |
| "grad_norm": 89.70054927345474, | |
| "learning_rate": 2.5986842105263156e-07, | |
| "logits/chosen": 0.28517499566078186, | |
| "logits/rejected": 0.20042014122009277, | |
| "logps/chosen": -219.8415985107422, | |
| "logps/rejected": -200.08901977539062, | |
| "loss": 0.5263, | |
| "rewards/accuracies": 0.7166666388511658, | |
| "rewards/chosen": -0.5583446621894836, | |
| "rewards/margins": 0.48568302392959595, | |
| "rewards/rejected": -1.0440276861190796, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "grad_norm": 70.5257802097204, | |
| "learning_rate": 2.2697368421052633e-07, | |
| "logits/chosen": -0.08118040859699249, | |
| "logits/rejected": -0.05347698926925659, | |
| "logps/chosen": -132.26165771484375, | |
| "logps/rejected": -201.5222625732422, | |
| "loss": 0.5441, | |
| "rewards/accuracies": 0.6833332777023315, | |
| "rewards/chosen": -0.5124568939208984, | |
| "rewards/margins": 0.4847620129585266, | |
| "rewards/rejected": -0.997218906879425, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.650887573964497, | |
| "grad_norm": 88.85554517295625, | |
| "learning_rate": 1.9407894736842102e-07, | |
| "logits/chosen": 0.028139472007751465, | |
| "logits/rejected": -0.018909335136413574, | |
| "logps/chosen": -192.33306884765625, | |
| "logps/rejected": -235.2200469970703, | |
| "loss": 0.5436, | |
| "rewards/accuracies": 0.7666667103767395, | |
| "rewards/chosen": -0.45699724555015564, | |
| "rewards/margins": 0.7322259545326233, | |
| "rewards/rejected": -1.1892231702804565, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "grad_norm": 64.03669024190756, | |
| "learning_rate": 1.611842105263158e-07, | |
| "logits/chosen": 0.00658189645037055, | |
| "logits/rejected": -0.047699280083179474, | |
| "logps/chosen": -223.26205444335938, | |
| "logps/rejected": -258.1011657714844, | |
| "loss": 0.547, | |
| "rewards/accuracies": 0.7333333492279053, | |
| "rewards/chosen": -0.7406572103500366, | |
| "rewards/margins": 0.8165454864501953, | |
| "rewards/rejected": -1.557202935218811, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 120.82289691922682, | |
| "learning_rate": 1.2828947368421054e-07, | |
| "logits/chosen": -0.1693594604730606, | |
| "logits/rejected": -0.184026300907135, | |
| "logps/chosen": -212.06900024414062, | |
| "logps/rejected": -218.6104278564453, | |
| "loss": 0.5153, | |
| "rewards/accuracies": 0.7500000596046448, | |
| "rewards/chosen": -0.5004691481590271, | |
| "rewards/margins": 0.5659549832344055, | |
| "rewards/rejected": -1.0664241313934326, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8284023668639053, | |
| "grad_norm": 75.4403830508224, | |
| "learning_rate": 9.539473684210526e-08, | |
| "logits/chosen": -0.0047453404404222965, | |
| "logits/rejected": 0.06134886294603348, | |
| "logps/chosen": -215.7347412109375, | |
| "logps/rejected": -216.7386016845703, | |
| "loss": 0.5365, | |
| "rewards/accuracies": 0.7500000596046448, | |
| "rewards/chosen": -0.47300204634666443, | |
| "rewards/margins": 0.6861640214920044, | |
| "rewards/rejected": -1.1591660976409912, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8875739644970414, | |
| "grad_norm": 403.81954432259465, | |
| "learning_rate": 6.25e-08, | |
| "logits/chosen": 0.07171961665153503, | |
| "logits/rejected": 0.06573580205440521, | |
| "logps/chosen": -168.96322631835938, | |
| "logps/rejected": -237.2980194091797, | |
| "loss": 0.5128, | |
| "rewards/accuracies": 0.783333420753479, | |
| "rewards/chosen": -0.5706448554992676, | |
| "rewards/margins": 0.8216232061386108, | |
| "rewards/rejected": -1.3922679424285889, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "grad_norm": 77.59593509813354, | |
| "learning_rate": 2.9605263157894734e-08, | |
| "logits/chosen": 0.030731897801160812, | |
| "logits/rejected": 0.0011664718622341752, | |
| "logps/chosen": -192.11680603027344, | |
| "logps/rejected": -223.5975799560547, | |
| "loss": 0.5015, | |
| "rewards/accuracies": 0.7833333611488342, | |
| "rewards/chosen": -0.5610911250114441, | |
| "rewards/margins": 0.7299409508705139, | |
| "rewards/rejected": -1.291032075881958, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 169, | |
| "total_flos": 0.0, | |
| "train_loss": 0.5670995542988975, | |
| "train_runtime": 1780.4141, | |
| "train_samples_per_second": 4.532, | |
| "train_steps_per_second": 0.095 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 169, | |
| "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": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |