| { | |
| "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": 0.48070791363716125, | |
| "learning_rate": 6.25e-08, | |
| "logits/chosen": 10.32492733001709, | |
| "logits/rejected": 10.282785415649414, | |
| "logps/chosen": -176.09544372558594, | |
| "logps/ref_chosen": -176.09544372558594, | |
| "logps/ref_rejected": -181.75552368164062, | |
| "logps/rejected": -181.75552368164062, | |
| "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": 0.4185059070587158, | |
| "learning_rate": 1.25e-07, | |
| "logits/chosen": 10.530074119567871, | |
| "logits/rejected": 10.672085762023926, | |
| "logps/chosen": -173.2084503173828, | |
| "logps/ref_chosen": -173.2084503173828, | |
| "logps/ref_rejected": -187.02206420898438, | |
| "logps/rejected": -187.02206420898438, | |
| "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": 0.3848375082015991, | |
| "learning_rate": 1.875e-07, | |
| "logits/chosen": 10.340475082397461, | |
| "logits/rejected": 10.486526489257812, | |
| "logps/chosen": -169.6623992919922, | |
| "logps/ref_chosen": -169.080810546875, | |
| "logps/ref_rejected": -182.52792358398438, | |
| "logps/rejected": -182.80642700195312, | |
| "loss": 0.6947, | |
| "rewards/accuracies": 0.3828125, | |
| "rewards/chosen": -0.005815833806991577, | |
| "rewards/margins": -0.0030306153930723667, | |
| "rewards/rejected": -0.002785218646749854, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05269658295594895, | |
| "grad_norm": 0.7261964678764343, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": 10.3652925491333, | |
| "logits/rejected": 10.412069320678711, | |
| "logps/chosen": -177.03428649902344, | |
| "logps/ref_chosen": -176.78369140625, | |
| "logps/ref_rejected": -180.0931396484375, | |
| "logps/rejected": -180.21463012695312, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.0025058696046471596, | |
| "rewards/margins": -0.0012909012148156762, | |
| "rewards/rejected": -0.0012149682734161615, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06587072869493618, | |
| "grad_norm": 0.3572319746017456, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": 10.757452011108398, | |
| "logits/rejected": 10.791389465332031, | |
| "logps/chosen": -169.12631225585938, | |
| "logps/ref_chosen": -168.522216796875, | |
| "logps/ref_rejected": -176.08982849121094, | |
| "logps/rejected": -176.59901428222656, | |
| "loss": 0.6936, | |
| "rewards/accuracies": 0.4609375, | |
| "rewards/chosen": -0.0060411859303712845, | |
| "rewards/margins": -0.0009493756806477904, | |
| "rewards/rejected": -0.0050918105989694595, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07904487443392343, | |
| "grad_norm": 0.4071093201637268, | |
| "learning_rate": 3.75e-07, | |
| "logits/chosen": 11.003137588500977, | |
| "logits/rejected": 11.056475639343262, | |
| "logps/chosen": -174.662353515625, | |
| "logps/ref_chosen": -174.5843048095703, | |
| "logps/ref_rejected": -181.65040588378906, | |
| "logps/rejected": -181.72401428222656, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.0007804610067978501, | |
| "rewards/margins": -4.43047538283281e-05, | |
| "rewards/rejected": -0.0007361561874859035, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09221902017291066, | |
| "grad_norm": 0.7951124906539917, | |
| "learning_rate": 4.375e-07, | |
| "logits/chosen": 10.003807067871094, | |
| "logits/rejected": 10.067156791687012, | |
| "logps/chosen": -171.4049530029297, | |
| "logps/ref_chosen": -170.4031219482422, | |
| "logps/ref_rejected": -181.00917053222656, | |
| "logps/rejected": -181.8280792236328, | |
| "loss": 0.6941, | |
| "rewards/accuracies": 0.4453125, | |
| "rewards/chosen": -0.010018287226557732, | |
| "rewards/margins": -0.0018291514134034514, | |
| "rewards/rejected": -0.008189136162400246, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1053931659118979, | |
| "grad_norm": 0.5591660737991333, | |
| "learning_rate": 5e-07, | |
| "logits/chosen": 10.874868392944336, | |
| "logits/rejected": 10.79828929901123, | |
| "logps/chosen": -179.36700439453125, | |
| "logps/ref_chosen": -178.47369384765625, | |
| "logps/ref_rejected": -184.52243041992188, | |
| "logps/rejected": -185.44284057617188, | |
| "loss": 0.693, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.008933190256357193, | |
| "rewards/margins": 0.0002710025873966515, | |
| "rewards/rejected": -0.009204192087054253, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11856731165088513, | |
| "grad_norm": 0.45744186639785767, | |
| "learning_rate": 4.997252228714278e-07, | |
| "logits/chosen": 10.472944259643555, | |
| "logits/rejected": 10.640350341796875, | |
| "logps/chosen": -175.36553955078125, | |
| "logps/ref_chosen": -174.38418579101562, | |
| "logps/ref_rejected": -182.0985565185547, | |
| "logps/rejected": -183.1609649658203, | |
| "loss": 0.6928, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.009813489392399788, | |
| "rewards/margins": 0.0008107352769002318, | |
| "rewards/rejected": -0.010624224320054054, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13174145738987236, | |
| "grad_norm": 0.3784727156162262, | |
| "learning_rate": 4.989014955054745e-07, | |
| "logits/chosen": 10.471797943115234, | |
| "logits/rejected": 10.57437515258789, | |
| "logps/chosen": -173.78768920898438, | |
| "logps/ref_chosen": -172.48837280273438, | |
| "logps/ref_rejected": -179.625, | |
| "logps/rejected": -180.91595458984375, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.012993087992072105, | |
| "rewards/margins": -8.335959864780307e-05, | |
| "rewards/rejected": -0.012909728102385998, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14491560312885962, | |
| "grad_norm": 0.4041476547718048, | |
| "learning_rate": 4.975306286336627e-07, | |
| "logits/chosen": 10.539186477661133, | |
| "logits/rejected": 10.477883338928223, | |
| "logps/chosen": -172.62416076660156, | |
| "logps/ref_chosen": -169.97216796875, | |
| "logps/ref_rejected": -180.00955200195312, | |
| "logps/rejected": -182.72525024414062, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.546875, | |
| "rewards/chosen": -0.026519589126110077, | |
| "rewards/margins": 0.0006374535150825977, | |
| "rewards/rejected": -0.027157040312886238, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15808974886784685, | |
| "grad_norm": 0.5964677333831787, | |
| "learning_rate": 4.956156357188939e-07, | |
| "logits/chosen": 10.28010082244873, | |
| "logits/rejected": 10.401304244995117, | |
| "logps/chosen": -178.43714904785156, | |
| "logps/ref_chosen": -175.83639526367188, | |
| "logps/ref_rejected": -181.73045349121094, | |
| "logps/rejected": -184.20223999023438, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.026007818058133125, | |
| "rewards/margins": -0.0012897354317829013, | |
| "rewards/rejected": -0.024718083441257477, | |
| "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 | |
| } | |