| { | |
| "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": 0.3937181234359741, | |
| "learning_rate": 6.25e-08, | |
| "logits/chosen": 10.038352012634277, | |
| "logits/rejected": 10.592904090881348, | |
| "logps/chosen": -132.56948852539062, | |
| "logps/ref_chosen": -132.56948852539062, | |
| "logps/ref_rejected": -148.778076171875, | |
| "logps/rejected": -148.778076171875, | |
| "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": 0.7554967999458313, | |
| "learning_rate": 1.25e-07, | |
| "logits/chosen": 10.320584297180176, | |
| "logits/rejected": 10.721946716308594, | |
| "logps/chosen": -132.46385192871094, | |
| "logps/ref_chosen": -132.46385192871094, | |
| "logps/ref_rejected": -142.69049072265625, | |
| "logps/rejected": -142.69049072265625, | |
| "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": 0.5059730410575867, | |
| "learning_rate": 1.875e-07, | |
| "logits/chosen": 9.891550064086914, | |
| "logits/rejected": 10.499095916748047, | |
| "logps/chosen": -136.52415466308594, | |
| "logps/ref_chosen": -135.79869079589844, | |
| "logps/ref_rejected": -145.69479370117188, | |
| "logps/rejected": -146.24124145507812, | |
| "loss": 0.6941, | |
| "rewards/accuracies": 0.3984375, | |
| "rewards/chosen": -0.007254539057612419, | |
| "rewards/margins": -0.0017902727704495192, | |
| "rewards/rejected": -0.005464266519993544, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05269658295594895, | |
| "grad_norm": 0.6780113577842712, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": 10.122194290161133, | |
| "logits/rejected": 10.64792251586914, | |
| "logps/chosen": -137.41903686523438, | |
| "logps/ref_chosen": -136.85159301757812, | |
| "logps/ref_rejected": -142.64901733398438, | |
| "logps/rejected": -142.94912719726562, | |
| "loss": 0.6945, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.00567453121766448, | |
| "rewards/margins": -0.0026732482947409153, | |
| "rewards/rejected": -0.0030012831557542086, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06587072869493618, | |
| "grad_norm": 0.5103796124458313, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": 10.327558517456055, | |
| "logits/rejected": 11.079102516174316, | |
| "logps/chosen": -130.6144561767578, | |
| "logps/ref_chosen": -130.2961883544922, | |
| "logps/ref_rejected": -144.1498565673828, | |
| "logps/rejected": -144.40802001953125, | |
| "loss": 0.6935, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.0031829699873924255, | |
| "rewards/margins": -0.000601244973950088, | |
| "rewards/rejected": -0.0025817248970270157, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07904487443392343, | |
| "grad_norm": 0.7068921327590942, | |
| "learning_rate": 3.75e-07, | |
| "logits/chosen": 10.794892311096191, | |
| "logits/rejected": 11.326902389526367, | |
| "logps/chosen": -138.65377807617188, | |
| "logps/ref_chosen": -138.59861755371094, | |
| "logps/ref_rejected": -151.63775634765625, | |
| "logps/rejected": -151.48056030273438, | |
| "loss": 0.6942, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.0005516067612916231, | |
| "rewards/margins": -0.002123636193573475, | |
| "rewards/rejected": 0.0015720291994512081, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09221902017291066, | |
| "grad_norm": 0.4639431834220886, | |
| "learning_rate": 4.375e-07, | |
| "logits/chosen": 9.949363708496094, | |
| "logits/rejected": 10.44249153137207, | |
| "logps/chosen": -130.56231689453125, | |
| "logps/ref_chosen": -129.5895538330078, | |
| "logps/ref_rejected": -139.67816162109375, | |
| "logps/rejected": -140.58245849609375, | |
| "loss": 0.6935, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.009727789089083672, | |
| "rewards/margins": -0.000684772094246, | |
| "rewards/rejected": -0.009043017402291298, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1053931659118979, | |
| "grad_norm": 0.5582533478736877, | |
| "learning_rate": 5e-07, | |
| "logits/chosen": 10.655470848083496, | |
| "logits/rejected": 11.168412208557129, | |
| "logps/chosen": -134.7186737060547, | |
| "logps/ref_chosen": -134.24497985839844, | |
| "logps/ref_rejected": -147.19009399414062, | |
| "logps/rejected": -147.70596313476562, | |
| "loss": 0.693, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.0047368621453642845, | |
| "rewards/margins": 0.0004215884255245328, | |
| "rewards/rejected": -0.0051584504544734955, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11856731165088513, | |
| "grad_norm": 0.559306263923645, | |
| "learning_rate": 4.997252228714278e-07, | |
| "logits/chosen": 10.525918006896973, | |
| "logits/rejected": 10.821675300598145, | |
| "logps/chosen": -142.31239318847656, | |
| "logps/ref_chosen": -141.59860229492188, | |
| "logps/ref_rejected": -149.33889770507812, | |
| "logps/rejected": -149.89822387695312, | |
| "loss": 0.6939, | |
| "rewards/accuracies": 0.4140625, | |
| "rewards/chosen": -0.007138015236705542, | |
| "rewards/margins": -0.001544784870930016, | |
| "rewards/rejected": -0.005593230482190847, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13174145738987236, | |
| "grad_norm": 0.6123551726341248, | |
| "learning_rate": 4.989014955054745e-07, | |
| "logits/chosen": 10.101277351379395, | |
| "logits/rejected": 10.55870246887207, | |
| "logps/chosen": -133.06532287597656, | |
| "logps/ref_chosen": -132.13449096679688, | |
| "logps/ref_rejected": -142.75973510742188, | |
| "logps/rejected": -143.90049743652344, | |
| "loss": 0.6921, | |
| "rewards/accuracies": 0.5390625, | |
| "rewards/chosen": -0.00930829904973507, | |
| "rewards/margins": 0.002099357545375824, | |
| "rewards/rejected": -0.011407657526433468, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14491560312885962, | |
| "grad_norm": 0.6729086637496948, | |
| "learning_rate": 4.975306286336627e-07, | |
| "logits/chosen": 10.170294761657715, | |
| "logits/rejected": 10.829442024230957, | |
| "logps/chosen": -137.3282012939453, | |
| "logps/ref_chosen": -135.85043334960938, | |
| "logps/ref_rejected": -152.14999389648438, | |
| "logps/rejected": -153.4501953125, | |
| "loss": 0.6941, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.014777721837162971, | |
| "rewards/margins": -0.0017757893074303865, | |
| "rewards/rejected": -0.013001934625208378, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15808974886784685, | |
| "grad_norm": 0.3486435115337372, | |
| "learning_rate": 4.956156357188939e-07, | |
| "logits/chosen": 10.261972427368164, | |
| "logits/rejected": 10.63994312286377, | |
| "logps/chosen": -141.65684509277344, | |
| "logps/ref_chosen": -139.63204956054688, | |
| "logps/ref_rejected": -148.03170776367188, | |
| "logps/rejected": -150.0493621826172, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.484375, | |
| "rewards/chosen": -0.02024805173277855, | |
| "rewards/margins": -7.15737696737051e-05, | |
| "rewards/rejected": -0.0201764777302742, | |
| "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 | |
| } | |