| { | |
| "best_metric": 0.055875860154628754, | |
| "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-250", | |
| "epoch": 0.12876641771825909, | |
| "eval_steps": 50, | |
| "global_step": 250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025753283543651817, | |
| "grad_norm": 14.230540018977528, | |
| "learning_rate": 2.9411764705882355e-06, | |
| "loss": 1.0954, | |
| "num_input_tokens_seen": 49920, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0051506567087303634, | |
| "grad_norm": 12.725175128208063, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.9793, | |
| "num_input_tokens_seen": 99840, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007725985063095545, | |
| "grad_norm": 11.908760913870685, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 1.0964, | |
| "num_input_tokens_seen": 149760, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010301313417460727, | |
| "grad_norm": 7.5714179545602835, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.7079, | |
| "num_input_tokens_seen": 199680, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012876641771825908, | |
| "grad_norm": 2.273110551179123, | |
| "learning_rate": 1.4705882352941177e-05, | |
| "loss": 0.4213, | |
| "num_input_tokens_seen": 249600, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01545197012619109, | |
| "grad_norm": 1.4511084067844011, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.3359, | |
| "num_input_tokens_seen": 299520, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018027298480556272, | |
| "grad_norm": 1.7462649262033438, | |
| "learning_rate": 2.058823529411765e-05, | |
| "loss": 0.3463, | |
| "num_input_tokens_seen": 349440, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.020602626834921454, | |
| "grad_norm": 1.2143595820366577, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.28, | |
| "num_input_tokens_seen": 399360, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.023177955189286635, | |
| "grad_norm": 0.8468719125045373, | |
| "learning_rate": 2.647058823529412e-05, | |
| "loss": 0.3442, | |
| "num_input_tokens_seen": 449280, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.025753283543651816, | |
| "grad_norm": 0.9726867932660042, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.3441, | |
| "num_input_tokens_seen": 499200, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025753283543651816, | |
| "eval_loss": 0.33834776282310486, | |
| "eval_runtime": 47.4467, | |
| "eval_samples_per_second": 1.265, | |
| "eval_steps_per_second": 0.316, | |
| "num_input_tokens_seen": 499200, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.028328611898016998, | |
| "grad_norm": 1.4016556961184263, | |
| "learning_rate": 3.235294117647059e-05, | |
| "loss": 0.3182, | |
| "num_input_tokens_seen": 549120, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.03090394025238218, | |
| "grad_norm": 0.6437613769459606, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.3294, | |
| "num_input_tokens_seen": 599040, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03347926860674736, | |
| "grad_norm": 0.7389008951321312, | |
| "learning_rate": 3.8235294117647055e-05, | |
| "loss": 0.3097, | |
| "num_input_tokens_seen": 648960, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.036054596961112545, | |
| "grad_norm": 0.771553860801019, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.3008, | |
| "num_input_tokens_seen": 698880, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.03862992531547772, | |
| "grad_norm": 0.6965369148334918, | |
| "learning_rate": 4.411764705882353e-05, | |
| "loss": 0.3278, | |
| "num_input_tokens_seen": 748800, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.04120525366984291, | |
| "grad_norm": 0.912943461315541, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 0.3074, | |
| "num_input_tokens_seen": 798720, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.043780582024208085, | |
| "grad_norm": 0.8407481737577445, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3423, | |
| "num_input_tokens_seen": 848640, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.04635591037857327, | |
| "grad_norm": 0.9112879058417015, | |
| "learning_rate": 5.294117647058824e-05, | |
| "loss": 0.3008, | |
| "num_input_tokens_seen": 898560, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04893123873293845, | |
| "grad_norm": 2.391489040464162, | |
| "learning_rate": 5.588235294117647e-05, | |
| "loss": 0.2815, | |
| "num_input_tokens_seen": 948480, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.05150656708730363, | |
| "grad_norm": 2.155211791607199, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.2274, | |
| "num_input_tokens_seen": 998400, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05150656708730363, | |
| "eval_loss": 0.18663176894187927, | |
| "eval_runtime": 18.9199, | |
| "eval_samples_per_second": 3.171, | |
| "eval_steps_per_second": 0.793, | |
| "num_input_tokens_seen": 998400, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05408189544166881, | |
| "grad_norm": 2.2181531422996716, | |
| "learning_rate": 6.176470588235295e-05, | |
| "loss": 0.168, | |
| "num_input_tokens_seen": 1048320, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.056657223796033995, | |
| "grad_norm": 3.1829920225573236, | |
| "learning_rate": 6.470588235294118e-05, | |
| "loss": 0.0709, | |
| "num_input_tokens_seen": 1098240, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05923255215039917, | |
| "grad_norm": 4.337350477588576, | |
| "learning_rate": 6.764705882352942e-05, | |
| "loss": 0.1609, | |
| "num_input_tokens_seen": 1148160, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.06180788050476436, | |
| "grad_norm": 2.1010046045637365, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.0354, | |
| "num_input_tokens_seen": 1198080, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.06438320885912954, | |
| "grad_norm": 2.232308844812103, | |
| "learning_rate": 7.352941176470589e-05, | |
| "loss": 0.1133, | |
| "num_input_tokens_seen": 1248000, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.06695853721349472, | |
| "grad_norm": 5.641631090993415, | |
| "learning_rate": 7.647058823529411e-05, | |
| "loss": 0.0867, | |
| "num_input_tokens_seen": 1297920, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0695338655678599, | |
| "grad_norm": 1.5031437609685787, | |
| "learning_rate": 7.941176470588235e-05, | |
| "loss": 0.1352, | |
| "num_input_tokens_seen": 1347840, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.07210919392222509, | |
| "grad_norm": 3.2992644431188465, | |
| "learning_rate": 8.23529411764706e-05, | |
| "loss": 0.101, | |
| "num_input_tokens_seen": 1397760, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.07468452227659027, | |
| "grad_norm": 3.494236832758233, | |
| "learning_rate": 8.529411764705883e-05, | |
| "loss": 0.0334, | |
| "num_input_tokens_seen": 1447680, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.07725985063095545, | |
| "grad_norm": 0.0602113869322109, | |
| "learning_rate": 8.823529411764706e-05, | |
| "loss": 0.0667, | |
| "num_input_tokens_seen": 1497600, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07725985063095545, | |
| "eval_loss": 0.09665286540985107, | |
| "eval_runtime": 19.2745, | |
| "eval_samples_per_second": 3.113, | |
| "eval_steps_per_second": 0.778, | |
| "num_input_tokens_seen": 1497600, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07983517898532062, | |
| "grad_norm": 6.096163706748617, | |
| "learning_rate": 9.11764705882353e-05, | |
| "loss": 0.06, | |
| "num_input_tokens_seen": 1547520, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.08241050733968582, | |
| "grad_norm": 4.278069142242893, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.0265, | |
| "num_input_tokens_seen": 1597440, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.08498583569405099, | |
| "grad_norm": 1.7183097652953412, | |
| "learning_rate": 9.705882352941177e-05, | |
| "loss": 0.1743, | |
| "num_input_tokens_seen": 1647360, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.08756116404841617, | |
| "grad_norm": 4.95401899568707, | |
| "learning_rate": 0.0001, | |
| "loss": 0.051, | |
| "num_input_tokens_seen": 1697280, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.09013649240278135, | |
| "grad_norm": 0.34558354886099124, | |
| "learning_rate": 9.999940874631277e-05, | |
| "loss": 0.0584, | |
| "num_input_tokens_seen": 1747200, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.09271182075714654, | |
| "grad_norm": 3.6326401692458146, | |
| "learning_rate": 9.999763499923432e-05, | |
| "loss": 0.0704, | |
| "num_input_tokens_seen": 1797120, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.09528714911151172, | |
| "grad_norm": 0.4999720825488852, | |
| "learning_rate": 9.999467880071402e-05, | |
| "loss": 0.0278, | |
| "num_input_tokens_seen": 1847040, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.0978624774658769, | |
| "grad_norm": 0.5211964505880501, | |
| "learning_rate": 9.999054022066641e-05, | |
| "loss": 0.0862, | |
| "num_input_tokens_seen": 1896960, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.10043780582024209, | |
| "grad_norm": 0.8767038751832389, | |
| "learning_rate": 9.998521935696953e-05, | |
| "loss": 0.0565, | |
| "num_input_tokens_seen": 1946880, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.10301313417460727, | |
| "grad_norm": 0.1848827361202722, | |
| "learning_rate": 9.997871633546257e-05, | |
| "loss": 0.0459, | |
| "num_input_tokens_seen": 1996800, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10301313417460727, | |
| "eval_loss": 0.09957947582006454, | |
| "eval_runtime": 19.2651, | |
| "eval_samples_per_second": 3.114, | |
| "eval_steps_per_second": 0.779, | |
| "num_input_tokens_seen": 1996800, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10558846252897244, | |
| "grad_norm": 1.0969393658164421, | |
| "learning_rate": 9.997103130994296e-05, | |
| "loss": 0.0539, | |
| "num_input_tokens_seen": 2046720, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.10816379088333762, | |
| "grad_norm": 2.885869194934028, | |
| "learning_rate": 9.996216446216267e-05, | |
| "loss": 0.0654, | |
| "num_input_tokens_seen": 2096640, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.11073911923770281, | |
| "grad_norm": 0.5225257245731217, | |
| "learning_rate": 9.995211600182397e-05, | |
| "loss": 0.0316, | |
| "num_input_tokens_seen": 2146560, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.11331444759206799, | |
| "grad_norm": 2.1553510734212797, | |
| "learning_rate": 9.994088616657444e-05, | |
| "loss": 0.1169, | |
| "num_input_tokens_seen": 2196480, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.11588977594643317, | |
| "grad_norm": 1.1133884703723633, | |
| "learning_rate": 9.992847522200133e-05, | |
| "loss": 0.0382, | |
| "num_input_tokens_seen": 2246400, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.11846510430079835, | |
| "grad_norm": 0.8875243341616034, | |
| "learning_rate": 9.99148834616253e-05, | |
| "loss": 0.0406, | |
| "num_input_tokens_seen": 2296320, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.12104043265516354, | |
| "grad_norm": 1.81283533812695, | |
| "learning_rate": 9.990011120689351e-05, | |
| "loss": 0.0182, | |
| "num_input_tokens_seen": 2346240, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.12361576100952872, | |
| "grad_norm": 3.873083258671571, | |
| "learning_rate": 9.988415880717194e-05, | |
| "loss": 0.0881, | |
| "num_input_tokens_seen": 2396160, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1261910893638939, | |
| "grad_norm": 3.427761103620865, | |
| "learning_rate": 9.986702663973722e-05, | |
| "loss": 0.0565, | |
| "num_input_tokens_seen": 2446080, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.12876641771825909, | |
| "grad_norm": 1.531943599765959, | |
| "learning_rate": 9.98487151097676e-05, | |
| "loss": 0.0805, | |
| "num_input_tokens_seen": 2496000, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.12876641771825909, | |
| "eval_loss": 0.055875860154628754, | |
| "eval_runtime": 19.5106, | |
| "eval_samples_per_second": 3.075, | |
| "eval_steps_per_second": 0.769, | |
| "num_input_tokens_seen": 2496000, | |
| "step": 250 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 3400, | |
| "num_input_tokens_seen": 2496000, | |
| "num_train_epochs": 2, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 164644726243328.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |