| { | |
| "best_global_step": 275, | |
| "best_metric": 0.20948709547519684, | |
| "best_model_checkpoint": "./llava_dora_weather_model/checkpoint-250", | |
| "epoch": 3.0, | |
| "eval_steps": 25, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10282776349614396, | |
| "grad_norm": 0.5152251720428467, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0854, | |
| "mean_token_accuracy": 0.5449784517288208, | |
| "num_tokens": 645254.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20565552699228792, | |
| "grad_norm": 0.8606380820274353, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4846, | |
| "mean_token_accuracy": 0.6233440130949021, | |
| "num_tokens": 1290816.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2570694087403599, | |
| "eval_loss": 0.7531503438949585, | |
| "eval_mean_token_accuracy": 0.7884166170389224, | |
| "eval_num_tokens": 1613469.0, | |
| "eval_runtime": 262.0989, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.30848329048843187, | |
| "grad_norm": 0.7467573881149292, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8003, | |
| "mean_token_accuracy": 0.7741554424166679, | |
| "num_tokens": 1936067.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.41131105398457585, | |
| "grad_norm": 0.35523706674575806, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4382, | |
| "mean_token_accuracy": 0.8741505913436413, | |
| "num_tokens": 2581570.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5141388174807198, | |
| "grad_norm": 0.3796336352825165, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3519, | |
| "mean_token_accuracy": 0.8941524222493171, | |
| "num_tokens": 3226740.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5141388174807198, | |
| "eval_loss": 0.33243268728256226, | |
| "eval_mean_token_accuracy": 0.9004789113998413, | |
| "eval_num_tokens": 3226740.0, | |
| "eval_runtime": 262.0578, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6169665809768637, | |
| "grad_norm": 0.37023383378982544, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3303, | |
| "mean_token_accuracy": 0.8990803204476834, | |
| "num_tokens": 3872000.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7197943444730077, | |
| "grad_norm": 0.4363132417201996, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2998, | |
| "mean_token_accuracy": 0.9075073637068272, | |
| "num_tokens": 4517650.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7712082262210797, | |
| "eval_loss": 0.28144383430480957, | |
| "eval_mean_token_accuracy": 0.9120607966031784, | |
| "eval_num_tokens": 4840076.0, | |
| "eval_runtime": 262.1457, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8226221079691517, | |
| "grad_norm": 0.37328270077705383, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2725, | |
| "mean_token_accuracy": 0.9144053891301155, | |
| "num_tokens": 5162524.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9254498714652957, | |
| "grad_norm": 0.37761208415031433, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2699, | |
| "mean_token_accuracy": 0.914168405532837, | |
| "num_tokens": 5808159.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0205655526992288, | |
| "grad_norm": 0.4116646647453308, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2577, | |
| "mean_token_accuracy": 0.9184043850447681, | |
| "num_tokens": 6405024.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0205655526992288, | |
| "eval_loss": 0.26103895902633667, | |
| "eval_mean_token_accuracy": 0.9167074130131648, | |
| "eval_num_tokens": 6405024.0, | |
| "eval_runtime": 262.2039, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1233933161953726, | |
| "grad_norm": 0.47321799397468567, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2496, | |
| "mean_token_accuracy": 0.9199652373790741, | |
| "num_tokens": 7050233.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2262210796915167, | |
| "grad_norm": 0.43690064549446106, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2435, | |
| "mean_token_accuracy": 0.9212526880204678, | |
| "num_tokens": 7695653.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.2776349614395888, | |
| "eval_loss": 0.23867399990558624, | |
| "eval_mean_token_accuracy": 0.9220352160624968, | |
| "eval_num_tokens": 8018413.0, | |
| "eval_runtime": 262.0533, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.3290488431876606, | |
| "grad_norm": 0.4848271608352661, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2325, | |
| "mean_token_accuracy": 0.9234701178967952, | |
| "num_tokens": 8341317.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4318766066838047, | |
| "grad_norm": 0.471851110458374, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2214, | |
| "mean_token_accuracy": 0.9266947895288468, | |
| "num_tokens": 8986431.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5347043701799485, | |
| "grad_norm": 0.4235078692436218, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2204, | |
| "mean_token_accuracy": 0.9253557249903679, | |
| "num_tokens": 9631882.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.5347043701799485, | |
| "eval_loss": 0.22957810759544373, | |
| "eval_mean_token_accuracy": 0.9235361132866297, | |
| "eval_num_tokens": 9631882.0, | |
| "eval_runtime": 262.1591, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6375321336760926, | |
| "grad_norm": 0.4376716613769531, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2201, | |
| "mean_token_accuracy": 0.9248783186078071, | |
| "num_tokens": 10277533.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7403598971722365, | |
| "grad_norm": 0.43904566764831543, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2287, | |
| "mean_token_accuracy": 0.9224192254245281, | |
| "num_tokens": 10922895.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.7917737789203085, | |
| "eval_loss": 0.22500084340572357, | |
| "eval_mean_token_accuracy": 0.9243928493597569, | |
| "eval_num_tokens": 11245730.0, | |
| "eval_runtime": 262.5112, | |
| "eval_samples_per_second": 0.743, | |
| "eval_steps_per_second": 0.743, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.8431876606683804, | |
| "grad_norm": 0.39153727889060974, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2142, | |
| "mean_token_accuracy": 0.9261229902505874, | |
| "num_tokens": 11568708.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9460154241645244, | |
| "grad_norm": 0.44200599193573, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2144, | |
| "mean_token_accuracy": 0.9287430047988892, | |
| "num_tokens": 12213156.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0411311053984575, | |
| "grad_norm": 0.40645667910575867, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2113, | |
| "mean_token_accuracy": 0.9269847950419864, | |
| "num_tokens": 12809961.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0411311053984575, | |
| "eval_loss": 0.21981394290924072, | |
| "eval_mean_token_accuracy": 0.9264772112552936, | |
| "eval_num_tokens": 12809961.0, | |
| "eval_runtime": 262.3645, | |
| "eval_samples_per_second": 0.743, | |
| "eval_steps_per_second": 0.743, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1439588688946016, | |
| "grad_norm": 0.4270681142807007, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1988, | |
| "mean_token_accuracy": 0.9318850792944431, | |
| "num_tokens": 13455171.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2467866323907453, | |
| "grad_norm": 0.40930086374282837, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2014, | |
| "mean_token_accuracy": 0.9315063305199146, | |
| "num_tokens": 14099963.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.2982005141388173, | |
| "eval_loss": 0.21635691821575165, | |
| "eval_mean_token_accuracy": 0.9275793613531651, | |
| "eval_num_tokens": 14422581.0, | |
| "eval_runtime": 262.4739, | |
| "eval_samples_per_second": 0.743, | |
| "eval_steps_per_second": 0.743, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.3496143958868894, | |
| "grad_norm": 0.4065852165222168, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1935, | |
| "mean_token_accuracy": 0.9315552815794945, | |
| "num_tokens": 14745386.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4524421593830334, | |
| "grad_norm": 0.4231654405593872, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2007, | |
| "mean_token_accuracy": 0.9310491763055324, | |
| "num_tokens": 15390859.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5552699228791775, | |
| "grad_norm": 0.4411866366863251, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2009, | |
| "mean_token_accuracy": 0.9314465291798115, | |
| "num_tokens": 16035959.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.5552699228791775, | |
| "eval_loss": 0.2139820158481598, | |
| "eval_mean_token_accuracy": 0.9278741191595029, | |
| "eval_num_tokens": 16035959.0, | |
| "eval_runtime": 262.5501, | |
| "eval_samples_per_second": 0.743, | |
| "eval_steps_per_second": 0.743, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.658097686375321, | |
| "grad_norm": 0.4447077512741089, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1968, | |
| "mean_token_accuracy": 0.9322382904589176, | |
| "num_tokens": 16681649.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7609254498714653, | |
| "grad_norm": 0.3979589641094208, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2035, | |
| "mean_token_accuracy": 0.9300756581127644, | |
| "num_tokens": 17327083.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8123393316195373, | |
| "eval_loss": 0.20948709547519684, | |
| "eval_mean_token_accuracy": 0.9293834570126656, | |
| "eval_num_tokens": 17650109.0, | |
| "eval_runtime": 262.8881, | |
| "eval_samples_per_second": 0.742, | |
| "eval_steps_per_second": 0.742, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.8637532133676094, | |
| "grad_norm": 0.41979947686195374, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1886, | |
| "mean_token_accuracy": 0.9338090866804123, | |
| "num_tokens": 17972810.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.966580976863753, | |
| "grad_norm": 0.4241204857826233, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1917, | |
| "mean_token_accuracy": 0.9328915029764175, | |
| "num_tokens": 18618053.0, | |
| "step": 290 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.470276663138264e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |