| { | |
| "best_metric": 0.9536784291267395, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-9000", | |
| "epoch": 1.658069270449521, | |
| "eval_steps": 500, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 20.04043960571289, | |
| "learning_rate": 4.8848563006632274e-08, | |
| "loss": 0.3758, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 1.4184927940368652, | |
| "eval_runtime": 74.3562, | |
| "eval_samples_per_second": 16.233, | |
| "eval_steps_per_second": 2.031, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 74.25025939941406, | |
| "learning_rate": 4.7697126013264556e-08, | |
| "loss": 0.4103, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 1.3500770330429077, | |
| "eval_runtime": 75.7013, | |
| "eval_samples_per_second": 15.944, | |
| "eval_steps_per_second": 1.995, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 0.5102410912513733, | |
| "learning_rate": 4.6545689019896826e-08, | |
| "loss": 0.433, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 1.2885024547576904, | |
| "eval_runtime": 75.197, | |
| "eval_samples_per_second": 16.051, | |
| "eval_steps_per_second": 2.008, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 0.1667678952217102, | |
| "learning_rate": 4.539425202652911e-08, | |
| "loss": 0.3424, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 1.239119052886963, | |
| "eval_runtime": 75.387, | |
| "eval_samples_per_second": 16.011, | |
| "eval_steps_per_second": 2.003, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 343.8175048828125, | |
| "learning_rate": 4.4242815033161385e-08, | |
| "loss": 0.3645, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 1.1902339458465576, | |
| "eval_runtime": 75.9899, | |
| "eval_samples_per_second": 15.884, | |
| "eval_steps_per_second": 1.987, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 305.5626525878906, | |
| "learning_rate": 4.309137803979366e-08, | |
| "loss": 0.3172, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 1.1506118774414062, | |
| "eval_runtime": 74.8655, | |
| "eval_samples_per_second": 16.122, | |
| "eval_steps_per_second": 2.017, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 65.59048461914062, | |
| "learning_rate": 4.193994104642594e-08, | |
| "loss": 0.2751, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 1.1169357299804688, | |
| "eval_runtime": 76.2936, | |
| "eval_samples_per_second": 15.82, | |
| "eval_steps_per_second": 1.979, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 18.106201171875, | |
| "learning_rate": 4.078850405305821e-08, | |
| "loss": 0.2919, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 1.0920671224594116, | |
| "eval_runtime": 74.9101, | |
| "eval_samples_per_second": 16.113, | |
| "eval_steps_per_second": 2.016, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 506.40570068359375, | |
| "learning_rate": 3.9637067059690496e-08, | |
| "loss": 0.2583, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 1.0721209049224854, | |
| "eval_runtime": 75.8053, | |
| "eval_samples_per_second": 15.922, | |
| "eval_steps_per_second": 1.992, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 1.1508910655975342, | |
| "learning_rate": 3.848563006632277e-08, | |
| "loss": 0.2679, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 1.0519349575042725, | |
| "eval_runtime": 75.226, | |
| "eval_samples_per_second": 16.045, | |
| "eval_steps_per_second": 2.007, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 0.1393290013074875, | |
| "learning_rate": 3.733419307295505e-08, | |
| "loss": 0.2472, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 1.0355629920959473, | |
| "eval_runtime": 74.6535, | |
| "eval_samples_per_second": 16.168, | |
| "eval_steps_per_second": 2.023, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "grad_norm": 200.34664916992188, | |
| "learning_rate": 3.6182756079587324e-08, | |
| "loss": 0.26, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "eval_loss": 1.0176944732666016, | |
| "eval_runtime": 74.4374, | |
| "eval_samples_per_second": 16.215, | |
| "eval_steps_per_second": 2.029, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "grad_norm": 431.1804504394531, | |
| "learning_rate": 3.50313190862196e-08, | |
| "loss": 0.2153, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "eval_loss": 1.0044586658477783, | |
| "eval_runtime": 74.3247, | |
| "eval_samples_per_second": 16.24, | |
| "eval_steps_per_second": 2.032, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "grad_norm": 0.6204941272735596, | |
| "learning_rate": 3.387988209285188e-08, | |
| "loss": 0.1791, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "eval_loss": 0.9926707148551941, | |
| "eval_runtime": 75.7317, | |
| "eval_samples_per_second": 15.938, | |
| "eval_steps_per_second": 1.994, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "grad_norm": 52.27273178100586, | |
| "learning_rate": 3.272844509948415e-08, | |
| "loss": 0.2082, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "eval_loss": 0.9803809523582458, | |
| "eval_runtime": 74.9936, | |
| "eval_samples_per_second": 16.095, | |
| "eval_steps_per_second": 2.014, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "grad_norm": 0.0976635068655014, | |
| "learning_rate": 3.1577008106116435e-08, | |
| "loss": 0.196, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "eval_loss": 0.9711655378341675, | |
| "eval_runtime": 74.3537, | |
| "eval_samples_per_second": 16.233, | |
| "eval_steps_per_second": 2.031, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "grad_norm": 46.64162826538086, | |
| "learning_rate": 3.042557111274871e-08, | |
| "loss": 0.1946, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "eval_loss": 0.9621181488037109, | |
| "eval_runtime": 75.7205, | |
| "eval_samples_per_second": 15.94, | |
| "eval_steps_per_second": 1.994, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "grad_norm": 0.05690698325634003, | |
| "learning_rate": 2.927413411938099e-08, | |
| "loss": 0.2422, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "eval_loss": 0.9536784291267395, | |
| "eval_runtime": 75.1745, | |
| "eval_samples_per_second": 16.056, | |
| "eval_steps_per_second": 2.009, | |
| "step": 9000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 21712, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3238544678042520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |