| { | |
| "best_metric": 0.25283440947532654, | |
| "best_model_checkpoint": "tuna6/mistral-saiga-journal-finetune6/checkpoint-1000", | |
| "epoch": 0.06540008502011052, | |
| "eval_steps": 40, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6344897150993347, | |
| "learning_rate": 0.00028828828828828825, | |
| "loss": 0.8513, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 0.6471168398857117, | |
| "eval_runtime": 596.3818, | |
| "eval_samples_per_second": 12.821, | |
| "eval_steps_per_second": 1.603, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.4552938938140869, | |
| "learning_rate": 0.00027627627627627627, | |
| "loss": 0.5674, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.5346186757087708, | |
| "eval_runtime": 595.5393, | |
| "eval_samples_per_second": 12.839, | |
| "eval_steps_per_second": 1.605, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.505496621131897, | |
| "learning_rate": 0.00026426426426426423, | |
| "loss": 0.4685, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.45588552951812744, | |
| "eval_runtime": 595.3281, | |
| "eval_samples_per_second": 12.843, | |
| "eval_steps_per_second": 1.606, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.49343597888946533, | |
| "learning_rate": 0.00025225225225225225, | |
| "loss": 0.4164, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.4150841534137726, | |
| "eval_runtime": 595.6775, | |
| "eval_samples_per_second": 12.836, | |
| "eval_steps_per_second": 1.605, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7332279682159424, | |
| "learning_rate": 0.00024024024024024023, | |
| "loss": 0.3707, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 0.38310083746910095, | |
| "eval_runtime": 595.6271, | |
| "eval_samples_per_second": 12.837, | |
| "eval_steps_per_second": 1.605, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.30062803626060486, | |
| "learning_rate": 0.0002282282282282282, | |
| "loss": 0.3497, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.3645000457763672, | |
| "eval_runtime": 596.1295, | |
| "eval_samples_per_second": 12.826, | |
| "eval_steps_per_second": 1.604, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.3287501633167267, | |
| "learning_rate": 0.0002162162162162162, | |
| "loss": 0.296, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.3538116216659546, | |
| "eval_runtime": 597.2837, | |
| "eval_samples_per_second": 12.801, | |
| "eval_steps_per_second": 1.601, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5396926999092102, | |
| "learning_rate": 0.00020420420420420418, | |
| "loss": 0.2976, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.3392348885536194, | |
| "eval_runtime": 596.5398, | |
| "eval_samples_per_second": 12.817, | |
| "eval_steps_per_second": 1.603, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.1267449855804443, | |
| "learning_rate": 0.00019219219219219217, | |
| "loss": 0.3018, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.3298206925392151, | |
| "eval_runtime": 596.9832, | |
| "eval_samples_per_second": 12.808, | |
| "eval_steps_per_second": 1.601, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.375235378742218, | |
| "learning_rate": 0.00018018018018018016, | |
| "loss": 0.3019, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.3226415514945984, | |
| "eval_runtime": 597.4868, | |
| "eval_samples_per_second": 12.797, | |
| "eval_steps_per_second": 1.6, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4212028980255127, | |
| "learning_rate": 0.00016816816816816817, | |
| "loss": 0.323, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.31242895126342773, | |
| "eval_runtime": 596.6515, | |
| "eval_samples_per_second": 12.815, | |
| "eval_steps_per_second": 1.602, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.3362777531147003, | |
| "learning_rate": 0.00015615615615615616, | |
| "loss": 0.299, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.3068313002586365, | |
| "eval_runtime": 596.6534, | |
| "eval_samples_per_second": 12.815, | |
| "eval_steps_per_second": 1.602, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.29624176025390625, | |
| "learning_rate": 0.00014414414414414412, | |
| "loss": 0.2671, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.30176782608032227, | |
| "eval_runtime": 596.3445, | |
| "eval_samples_per_second": 12.821, | |
| "eval_steps_per_second": 1.603, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3254912197589874, | |
| "learning_rate": 0.00013213213213213211, | |
| "loss": 0.2799, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.2952657639980316, | |
| "eval_runtime": 596.6573, | |
| "eval_samples_per_second": 12.815, | |
| "eval_steps_per_second": 1.602, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3114880919456482, | |
| "learning_rate": 0.00012012012012012012, | |
| "loss": 0.2674, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.29019972681999207, | |
| "eval_runtime": 596.4197, | |
| "eval_samples_per_second": 12.82, | |
| "eval_steps_per_second": 1.603, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.2812901437282562, | |
| "learning_rate": 0.0001081081081081081, | |
| "loss": 0.2395, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.2831648588180542, | |
| "eval_runtime": 596.2022, | |
| "eval_samples_per_second": 12.825, | |
| "eval_steps_per_second": 1.603, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.553597092628479, | |
| "learning_rate": 9.609609609609608e-05, | |
| "loss": 0.2436, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.27831366658210754, | |
| "eval_runtime": 595.5956, | |
| "eval_samples_per_second": 12.838, | |
| "eval_steps_per_second": 1.605, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.46639448404312134, | |
| "learning_rate": 8.408408408408409e-05, | |
| "loss": 0.2537, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.27261102199554443, | |
| "eval_runtime": 595.7096, | |
| "eval_samples_per_second": 12.835, | |
| "eval_steps_per_second": 1.605, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.3072221279144287, | |
| "learning_rate": 7.207207207207206e-05, | |
| "loss": 0.238, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.2690950334072113, | |
| "eval_runtime": 595.7471, | |
| "eval_samples_per_second": 12.834, | |
| "eval_steps_per_second": 1.605, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.3728131055831909, | |
| "learning_rate": 6.006006006006006e-05, | |
| "loss": 0.2583, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.2644895017147064, | |
| "eval_runtime": 595.5546, | |
| "eval_samples_per_second": 12.838, | |
| "eval_steps_per_second": 1.605, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4398050010204315, | |
| "learning_rate": 4.804804804804804e-05, | |
| "loss": 0.2274, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.26183322072029114, | |
| "eval_runtime": 595.4684, | |
| "eval_samples_per_second": 12.84, | |
| "eval_steps_per_second": 1.605, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.37743285298347473, | |
| "learning_rate": 3.603603603603603e-05, | |
| "loss": 0.2279, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.258406400680542, | |
| "eval_runtime": 595.383, | |
| "eval_samples_per_second": 12.842, | |
| "eval_steps_per_second": 1.606, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.25567543506622314, | |
| "learning_rate": 2.402402402402402e-05, | |
| "loss": 0.2497, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.25550028681755066, | |
| "eval_runtime": 595.2117, | |
| "eval_samples_per_second": 12.846, | |
| "eval_steps_per_second": 1.606, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.379567414522171, | |
| "learning_rate": 1.201201201201201e-05, | |
| "loss": 0.2599, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.25368040800094604, | |
| "eval_runtime": 595.2349, | |
| "eval_samples_per_second": 12.845, | |
| "eval_steps_per_second": 1.606, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3067691922187805, | |
| "learning_rate": 0.0, | |
| "loss": 0.2332, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.25283440947532654, | |
| "eval_runtime": 595.1198, | |
| "eval_samples_per_second": 12.848, | |
| "eval_steps_per_second": 1.606, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 40, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 40, | |
| "total_flos": 3.1676545818624e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |