| { | |
| "best_metric": 1.9574228525161743, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-9000", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 16284, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 0.030529862269759178, | |
| "learning_rate": 4.8464750675509704e-05, | |
| "loss": 0.7076, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 2.07871675491333, | |
| "eval_runtime": 73.4415, | |
| "eval_samples_per_second": 16.435, | |
| "eval_steps_per_second": 2.056, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 0.02040918357670307, | |
| "learning_rate": 4.6929501351019406e-05, | |
| "loss": 0.7045, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 2.0783257484436035, | |
| "eval_runtime": 74.6654, | |
| "eval_samples_per_second": 16.165, | |
| "eval_steps_per_second": 2.022, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 0.8010361194610596, | |
| "learning_rate": 4.539425202652911e-05, | |
| "loss": 0.7106, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 2.0473792552948, | |
| "eval_runtime": 75.4141, | |
| "eval_samples_per_second": 16.005, | |
| "eval_steps_per_second": 2.002, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 1.1971429586410522, | |
| "learning_rate": 4.3859002702038817e-05, | |
| "loss": 0.704, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 2.281594753265381, | |
| "eval_runtime": 76.3969, | |
| "eval_samples_per_second": 15.799, | |
| "eval_steps_per_second": 1.977, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 1.5381091833114624, | |
| "learning_rate": 4.232375337754851e-05, | |
| "loss": 0.6996, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 2.045022964477539, | |
| "eval_runtime": 76.3329, | |
| "eval_samples_per_second": 15.812, | |
| "eval_steps_per_second": 1.978, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 5.1257524490356445, | |
| "learning_rate": 4.078850405305822e-05, | |
| "loss": 0.6739, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 2.0969045162200928, | |
| "eval_runtime": 76.2547, | |
| "eval_samples_per_second": 15.829, | |
| "eval_steps_per_second": 1.98, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 6.156252861022949, | |
| "learning_rate": 3.925325472856792e-05, | |
| "loss": 0.6687, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 2.0315287113189697, | |
| "eval_runtime": 76.4108, | |
| "eval_samples_per_second": 15.796, | |
| "eval_steps_per_second": 1.976, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 0.053296659141778946, | |
| "learning_rate": 3.7718005404077624e-05, | |
| "loss": 0.8387, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 2.3810112476348877, | |
| "eval_runtime": 74.9737, | |
| "eval_samples_per_second": 16.099, | |
| "eval_steps_per_second": 2.014, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 0.1077703982591629, | |
| "learning_rate": 3.6182756079587326e-05, | |
| "loss": 0.6988, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 2.0409226417541504, | |
| "eval_runtime": 76.7206, | |
| "eval_samples_per_second": 15.732, | |
| "eval_steps_per_second": 1.968, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 4.246299743652344, | |
| "learning_rate": 3.4647506755097034e-05, | |
| "loss": 0.6808, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 1.9745088815689087, | |
| "eval_runtime": 74.928, | |
| "eval_samples_per_second": 16.109, | |
| "eval_steps_per_second": 2.015, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 2.2185893058776855, | |
| "learning_rate": 3.311225743060673e-05, | |
| "loss": 0.6584, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 1.9648858308792114, | |
| "eval_runtime": 76.0743, | |
| "eval_samples_per_second": 15.866, | |
| "eval_steps_per_second": 1.985, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "grad_norm": 6.992915153503418, | |
| "learning_rate": 3.157700810611644e-05, | |
| "loss": 0.6567, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "eval_loss": 2.0568628311157227, | |
| "eval_runtime": 76.0502, | |
| "eval_samples_per_second": 15.871, | |
| "eval_steps_per_second": 1.986, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "grad_norm": 13.946557998657227, | |
| "learning_rate": 3.004175878162614e-05, | |
| "loss": 0.6542, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "eval_loss": 1.9788861274719238, | |
| "eval_runtime": 75.9196, | |
| "eval_samples_per_second": 15.898, | |
| "eval_steps_per_second": 1.989, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "grad_norm": 2.426790237426758, | |
| "learning_rate": 2.8506509457135838e-05, | |
| "loss": 0.6536, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "eval_loss": 1.9676496982574463, | |
| "eval_runtime": 75.155, | |
| "eval_samples_per_second": 16.06, | |
| "eval_steps_per_second": 2.009, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "grad_norm": 1.7985050678253174, | |
| "learning_rate": 2.6971260132645544e-05, | |
| "loss": 0.637, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "eval_loss": 2.003986358642578, | |
| "eval_runtime": 74.9913, | |
| "eval_samples_per_second": 16.095, | |
| "eval_steps_per_second": 2.014, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "grad_norm": 22.398109436035156, | |
| "learning_rate": 2.543601080815525e-05, | |
| "loss": 0.6409, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "eval_loss": 1.9904581308364868, | |
| "eval_runtime": 74.9317, | |
| "eval_samples_per_second": 16.108, | |
| "eval_steps_per_second": 2.015, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "grad_norm": 4.255030155181885, | |
| "learning_rate": 2.390076148366495e-05, | |
| "loss": 0.6127, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "eval_loss": 2.2638871669769287, | |
| "eval_runtime": 74.9634, | |
| "eval_samples_per_second": 16.101, | |
| "eval_steps_per_second": 2.014, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "grad_norm": 11.839683532714844, | |
| "learning_rate": 2.2365512159174652e-05, | |
| "loss": 0.6564, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "eval_loss": 1.9574228525161743, | |
| "eval_runtime": 74.8273, | |
| "eval_samples_per_second": 16.13, | |
| "eval_steps_per_second": 2.018, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "grad_norm": 7.12531042098999, | |
| "learning_rate": 2.0830262834684354e-05, | |
| "loss": 0.6813, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "eval_loss": 2.036475896835327, | |
| "eval_runtime": 75.9457, | |
| "eval_samples_per_second": 15.893, | |
| "eval_steps_per_second": 1.988, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "grad_norm": 2.8198225498199463, | |
| "learning_rate": 1.9295013510194056e-05, | |
| "loss": 0.6421, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "eval_loss": 2.2431833744049072, | |
| "eval_runtime": 75.2519, | |
| "eval_samples_per_second": 16.039, | |
| "eval_steps_per_second": 2.007, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "grad_norm": 23.107166290283203, | |
| "learning_rate": 1.7759764185703758e-05, | |
| "loss": 0.6531, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "eval_loss": 1.9985452890396118, | |
| "eval_runtime": 74.9704, | |
| "eval_samples_per_second": 16.1, | |
| "eval_steps_per_second": 2.014, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "grad_norm": 4.0393805503845215, | |
| "learning_rate": 1.622451486121346e-05, | |
| "loss": 0.6388, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "eval_loss": 1.9725395441055298, | |
| "eval_runtime": 74.9174, | |
| "eval_samples_per_second": 16.111, | |
| "eval_steps_per_second": 2.016, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 21.68863868713379, | |
| "learning_rate": 1.4689265536723165e-05, | |
| "loss": 0.6394, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "eval_loss": 2.0170812606811523, | |
| "eval_runtime": 75.0053, | |
| "eval_samples_per_second": 16.092, | |
| "eval_steps_per_second": 2.013, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "grad_norm": 1.8485018014907837, | |
| "learning_rate": 1.3154016212232867e-05, | |
| "loss": 0.6151, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "eval_loss": 1.9586845636367798, | |
| "eval_runtime": 74.7694, | |
| "eval_samples_per_second": 16.143, | |
| "eval_steps_per_second": 2.02, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "grad_norm": 14.168105125427246, | |
| "learning_rate": 1.161876688774257e-05, | |
| "loss": 0.5947, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "eval_loss": 2.0683774948120117, | |
| "eval_runtime": 74.9327, | |
| "eval_samples_per_second": 16.108, | |
| "eval_steps_per_second": 2.015, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "grad_norm": 1.8051279783248901, | |
| "learning_rate": 1.0083517563252274e-05, | |
| "loss": 0.6166, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "eval_loss": 2.05798602104187, | |
| "eval_runtime": 74.9282, | |
| "eval_samples_per_second": 16.109, | |
| "eval_steps_per_second": 2.015, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "grad_norm": 23.26294708251953, | |
| "learning_rate": 8.548268238761976e-06, | |
| "loss": 0.6382, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "eval_loss": 2.020942211151123, | |
| "eval_runtime": 74.9346, | |
| "eval_samples_per_second": 16.107, | |
| "eval_steps_per_second": 2.015, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "grad_norm": 14.594226837158203, | |
| "learning_rate": 7.0130189142716784e-06, | |
| "loss": 0.5949, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "eval_loss": 2.0775811672210693, | |
| "eval_runtime": 74.871, | |
| "eval_samples_per_second": 16.121, | |
| "eval_steps_per_second": 2.017, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "grad_norm": 6.04954719543457, | |
| "learning_rate": 5.47776958978138e-06, | |
| "loss": 0.6282, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "eval_loss": 2.072066307067871, | |
| "eval_runtime": 74.9104, | |
| "eval_samples_per_second": 16.113, | |
| "eval_steps_per_second": 2.016, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "grad_norm": 1.6737792491912842, | |
| "learning_rate": 3.942520265291084e-06, | |
| "loss": 0.6057, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "eval_loss": 2.0812695026397705, | |
| "eval_runtime": 74.9464, | |
| "eval_samples_per_second": 16.105, | |
| "eval_steps_per_second": 2.015, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "grad_norm": 13.746774673461914, | |
| "learning_rate": 2.407270940800786e-06, | |
| "loss": 0.6229, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "eval_loss": 2.0467605590820312, | |
| "eval_runtime": 74.8136, | |
| "eval_samples_per_second": 16.133, | |
| "eval_steps_per_second": 2.018, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "grad_norm": 11.62869644165039, | |
| "learning_rate": 8.720216163104889e-07, | |
| "loss": 0.612, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "eval_loss": 2.0670080184936523, | |
| "eval_runtime": 74.9975, | |
| "eval_samples_per_second": 16.094, | |
| "eval_steps_per_second": 2.013, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 16284, | |
| "total_flos": 5859024508620360.0, | |
| "train_loss": 0.6559209774413702, | |
| "train_runtime": 12369.044, | |
| "train_samples_per_second": 2.633, | |
| "train_steps_per_second": 1.317 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16284, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5859024508620360.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |