| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.15808974886784685, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013174145738987238, | |
| "grad_norm": 0.39343583583831787, | |
| "learning_rate": 6.25e-08, | |
| "logits/chosen": 10.071717262268066, | |
| "logits/rejected": 10.610974311828613, | |
| "logps/chosen": -121.14067077636719, | |
| "logps/ref_chosen": -121.14067077636719, | |
| "logps/ref_rejected": -137.65684509277344, | |
| "logps/rejected": -137.65684509277344, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.026348291477974475, | |
| "grad_norm": 0.7521647214889526, | |
| "learning_rate": 1.25e-07, | |
| "logits/chosen": 10.222262382507324, | |
| "logits/rejected": 10.754176139831543, | |
| "logps/chosen": -116.48068237304688, | |
| "logps/ref_chosen": -116.48068237304688, | |
| "logps/ref_rejected": -130.27796936035156, | |
| "logps/rejected": -130.27796936035156, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03952243721696171, | |
| "grad_norm": 0.5880491137504578, | |
| "learning_rate": 1.875e-07, | |
| "logits/chosen": 10.01984977722168, | |
| "logits/rejected": 10.62405776977539, | |
| "logps/chosen": -123.36822509765625, | |
| "logps/ref_chosen": -122.6683349609375, | |
| "logps/ref_rejected": -132.69850158691406, | |
| "logps/rejected": -133.207275390625, | |
| "loss": 0.6941, | |
| "rewards/accuracies": 0.4453125, | |
| "rewards/chosen": -0.006998830940574408, | |
| "rewards/margins": -0.0019110905705019832, | |
| "rewards/rejected": -0.005087739787995815, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05269658295594895, | |
| "grad_norm": 0.4605408310890198, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": 9.99990463256836, | |
| "logits/rejected": 10.736846923828125, | |
| "logps/chosen": -123.02133178710938, | |
| "logps/ref_chosen": -122.59739685058594, | |
| "logps/ref_rejected": -129.70767211914062, | |
| "logps/rejected": -129.98374938964844, | |
| "loss": 0.6939, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.004239337984472513, | |
| "rewards/margins": -0.0014785109087824821, | |
| "rewards/rejected": -0.0027608266100287437, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06587072869493618, | |
| "grad_norm": 0.469856321811676, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": 10.075482368469238, | |
| "logits/rejected": 10.892666816711426, | |
| "logps/chosen": -117.554931640625, | |
| "logps/ref_chosen": -117.5941162109375, | |
| "logps/ref_rejected": -132.1708984375, | |
| "logps/rejected": -132.1227264404297, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.546875, | |
| "rewards/chosen": 0.0003917121794074774, | |
| "rewards/margins": -9.010493522509933e-05, | |
| "rewards/rejected": 0.0004818170564249158, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07904487443392343, | |
| "grad_norm": 0.5541566610336304, | |
| "learning_rate": 3.75e-07, | |
| "logits/chosen": 10.660999298095703, | |
| "logits/rejected": 11.290507316589355, | |
| "logps/chosen": -127.00320434570312, | |
| "logps/ref_chosen": -126.12411499023438, | |
| "logps/ref_rejected": -136.9976043701172, | |
| "logps/rejected": -137.75950622558594, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.00879070907831192, | |
| "rewards/margins": -0.0011717069428414106, | |
| "rewards/rejected": -0.007619001902639866, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09221902017291066, | |
| "grad_norm": 0.4422788619995117, | |
| "learning_rate": 4.375e-07, | |
| "logits/chosen": 9.958097457885742, | |
| "logits/rejected": 10.642163276672363, | |
| "logps/chosen": -115.61244201660156, | |
| "logps/ref_chosen": -115.08863830566406, | |
| "logps/ref_rejected": -125.91255187988281, | |
| "logps/rejected": -126.15577697753906, | |
| "loss": 0.6946, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.005238103214651346, | |
| "rewards/margins": -0.0028058765456080437, | |
| "rewards/rejected": -0.002432226436212659, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1053931659118979, | |
| "grad_norm": 0.5495327115058899, | |
| "learning_rate": 5e-07, | |
| "logits/chosen": 10.345503807067871, | |
| "logits/rejected": 10.97708797454834, | |
| "logps/chosen": -121.48406982421875, | |
| "logps/ref_chosen": -121.4114761352539, | |
| "logps/ref_rejected": -134.62770080566406, | |
| "logps/rejected": -134.7586212158203, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.0007259202538989484, | |
| "rewards/margins": 0.0005832896567881107, | |
| "rewards/rejected": -0.0013092098524793983, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11856731165088513, | |
| "grad_norm": 0.43922048807144165, | |
| "learning_rate": 4.997252228714278e-07, | |
| "logits/chosen": 10.159126281738281, | |
| "logits/rejected": 11.002123832702637, | |
| "logps/chosen": -122.51399230957031, | |
| "logps/ref_chosen": -121.59207153320312, | |
| "logps/ref_rejected": -134.70025634765625, | |
| "logps/rejected": -135.55740356445312, | |
| "loss": 0.6935, | |
| "rewards/accuracies": 0.4453125, | |
| "rewards/chosen": -0.00921926274895668, | |
| "rewards/margins": -0.0006479143630713224, | |
| "rewards/rejected": -0.008571348153054714, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13174145738987236, | |
| "grad_norm": 0.6643485426902771, | |
| "learning_rate": 4.989014955054745e-07, | |
| "logits/chosen": 9.9464111328125, | |
| "logits/rejected": 10.739057540893555, | |
| "logps/chosen": -117.91310119628906, | |
| "logps/ref_chosen": -117.16349029541016, | |
| "logps/ref_rejected": -129.98167419433594, | |
| "logps/rejected": -130.65924072265625, | |
| "loss": 0.6935, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.007496046833693981, | |
| "rewards/margins": -0.0007203805143944919, | |
| "rewards/rejected": -0.00677566509693861, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14491560312885962, | |
| "grad_norm": 0.4432956576347351, | |
| "learning_rate": 4.975306286336627e-07, | |
| "logits/chosen": 10.095781326293945, | |
| "logits/rejected": 11.001167297363281, | |
| "logps/chosen": -123.26506042480469, | |
| "logps/ref_chosen": -121.95927429199219, | |
| "logps/ref_rejected": -136.18655395507812, | |
| "logps/rejected": -137.39865112304688, | |
| "loss": 0.6936, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.01305788941681385, | |
| "rewards/margins": -0.0009369202307425439, | |
| "rewards/rejected": -0.012120969593524933, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15808974886784685, | |
| "grad_norm": 0.3955974280834198, | |
| "learning_rate": 4.956156357188939e-07, | |
| "logits/chosen": 10.129995346069336, | |
| "logits/rejected": 10.673677444458008, | |
| "logps/chosen": -126.05357360839844, | |
| "logps/ref_chosen": -124.12315368652344, | |
| "logps/ref_rejected": -134.275390625, | |
| "logps/rejected": -136.24940490722656, | |
| "loss": 0.693, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.01930420845746994, | |
| "rewards/margins": 0.00043603626545518637, | |
| "rewards/rejected": -0.01974024437367916, | |
| "step": 12 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 12, | |
| "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": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |