| { | |
| "best_metric": 1.7786831855773926, | |
| "best_model_checkpoint": "$../logs/03-08-2024T09-55-11-detailclip-b32/checkpoint-116", | |
| "epoch": 1.5130434782608697, | |
| "eval_steps": 58, | |
| "global_step": 174, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05217391304347826, | |
| "grad_norm": 2.7094666957855225, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.7099, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10434782608695652, | |
| "grad_norm": 2.6887893676757812, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.6948, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1565217391304348, | |
| "grad_norm": 2.56785249710083, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7276, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 2.413269281387329, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5937, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 2.0822973251342773, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4702, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3130434782608696, | |
| "grad_norm": 1.9383469820022583, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.4579, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3652173913043478, | |
| "grad_norm": 1.6925252676010132, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.4028, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 1.5693070888519287, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.3466, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.46956521739130436, | |
| "grad_norm": 1.8624712228775024, | |
| "learning_rate": 9.999673735634258e-05, | |
| "loss": 0.31, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5043478260869565, | |
| "eval_loss": 1.88290536403656, | |
| "eval_runtime": 371.0087, | |
| "eval_samples_per_second": 6.021, | |
| "eval_steps_per_second": 0.049, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5043478260869565, | |
| "neg_increments": 0.40031376906933136, | |
| "positive_increments": 0.6000771337979885, | |
| "rate_increments": 0.60599292530573, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 1.565119743347168, | |
| "learning_rate": 9.997960964140947e-05, | |
| "loss": 0.289, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5739130434782609, | |
| "grad_norm": 1.695056438446045, | |
| "learning_rate": 9.994780621691156e-05, | |
| "loss": 0.2486, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "grad_norm": 1.3525909185409546, | |
| "learning_rate": 9.990133642141359e-05, | |
| "loss": 0.2476, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6782608695652174, | |
| "grad_norm": 1.7046144008636475, | |
| "learning_rate": 9.984021390002458e-05, | |
| "loss": 0.2677, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.7304347826086957, | |
| "grad_norm": 1.199271559715271, | |
| "learning_rate": 9.976445660039118e-05, | |
| "loss": 0.2425, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 1.1520024538040161, | |
| "learning_rate": 9.967408676742751e-05, | |
| "loss": 0.2444, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 1.023667335510254, | |
| "learning_rate": 9.956913093678348e-05, | |
| "loss": 0.2127, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.8869565217391304, | |
| "grad_norm": 1.418879747390747, | |
| "learning_rate": 9.944961992705288e-05, | |
| "loss": 0.2375, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.9391304347826087, | |
| "grad_norm": 1.6150606870651245, | |
| "learning_rate": 9.931558883072403e-05, | |
| "loss": 0.2214, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.991304347826087, | |
| "grad_norm": 1.3231348991394043, | |
| "learning_rate": 9.916707700387547e-05, | |
| "loss": 0.2283, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.008695652173913, | |
| "eval_loss": 1.7786831855773926, | |
| "eval_runtime": 352.6788, | |
| "eval_samples_per_second": 6.334, | |
| "eval_steps_per_second": 0.051, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.008695652173913, | |
| "neg_increments": 0.38078110707552626, | |
| "positive_increments": 0.6381213280123575, | |
| "rate_increments": 0.6405809892973509, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 1.1350363492965698, | |
| "learning_rate": 9.900412805461967e-05, | |
| "loss": 0.2019, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0956521739130434, | |
| "grad_norm": 1.305219054222107, | |
| "learning_rate": 9.882678983029819e-05, | |
| "loss": 0.1942, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.1478260869565218, | |
| "grad_norm": 0.8753030300140381, | |
| "learning_rate": 9.863511440343206e-05, | |
| "loss": 0.19, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.1253970861434937, | |
| "learning_rate": 9.842915805643155e-05, | |
| "loss": 0.1917, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.2521739130434781, | |
| "grad_norm": 1.4084516763687134, | |
| "learning_rate": 9.820898126506979e-05, | |
| "loss": 0.1898, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 1.483275294303894, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.2025, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3565217391304347, | |
| "grad_norm": 1.4686787128448486, | |
| "learning_rate": 9.772622911139622e-05, | |
| "loss": 0.1918, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.4086956521739131, | |
| "grad_norm": 1.3151410818099976, | |
| "learning_rate": 9.746379550150009e-05, | |
| "loss": 0.222, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.4608695652173913, | |
| "grad_norm": 1.1024744510650635, | |
| "learning_rate": 9.71874249104506e-05, | |
| "loss": 0.179, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.5130434782608697, | |
| "grad_norm": 1.1871354579925537, | |
| "learning_rate": 9.68971984900326e-05, | |
| "loss": 0.1768, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 1.5130434782608697, | |
| "eval_loss": 1.8319995403289795, | |
| "eval_runtime": 375.6242, | |
| "eval_samples_per_second": 5.947, | |
| "eval_steps_per_second": 0.048, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 1.5130434782608697, | |
| "neg_increments": 0.36799247711332744, | |
| "positive_increments": 0.6630822809526313, | |
| "rate_increments": 0.6599125175329515, | |
| "step": 174 | |
| } | |
| ], | |
| "logging_steps": 6, | |
| "max_steps": 1150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 58, | |
| "total_flos": 3942368157607110.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |