| { | |
| "best_metric": 0.9258912801742554, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-16000", | |
| "epoch": 2.9476787030213707, | |
| "eval_steps": 500, | |
| "global_step": 16000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 19.808820724487305, | |
| "learning_rate": 4.84647506755097e-08, | |
| "loss": 0.3758, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 1.418845295906067, | |
| "eval_runtime": 75.2377, | |
| "eval_samples_per_second": 16.042, | |
| "eval_steps_per_second": 2.007, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 76.37991333007812, | |
| "learning_rate": 4.69295013510194e-08, | |
| "loss": 0.4108, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 1.3513683080673218, | |
| "eval_runtime": 74.9686, | |
| "eval_samples_per_second": 16.1, | |
| "eval_steps_per_second": 2.014, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 0.5795987844467163, | |
| "learning_rate": 4.539425202652911e-08, | |
| "loss": 0.4335, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 1.2895005941390991, | |
| "eval_runtime": 76.713, | |
| "eval_samples_per_second": 15.734, | |
| "eval_steps_per_second": 1.968, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 0.1463615745306015, | |
| "learning_rate": 4.385900270203881e-08, | |
| "loss": 0.3436, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 1.2414860725402832, | |
| "eval_runtime": 76.9749, | |
| "eval_samples_per_second": 15.68, | |
| "eval_steps_per_second": 1.962, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 323.8576354980469, | |
| "learning_rate": 4.232375337754851e-08, | |
| "loss": 0.366, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 1.1949334144592285, | |
| "eval_runtime": 77.0381, | |
| "eval_samples_per_second": 15.668, | |
| "eval_steps_per_second": 1.96, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 339.28533935546875, | |
| "learning_rate": 4.078850405305821e-08, | |
| "loss": 0.3191, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 1.1560280323028564, | |
| "eval_runtime": 77.0026, | |
| "eval_samples_per_second": 15.675, | |
| "eval_steps_per_second": 1.961, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 60.27859878540039, | |
| "learning_rate": 3.925325472856792e-08, | |
| "loss": 0.2779, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 1.1233855485916138, | |
| "eval_runtime": 76.9006, | |
| "eval_samples_per_second": 15.696, | |
| "eval_steps_per_second": 1.964, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 18.268455505371094, | |
| "learning_rate": 3.7718005404077616e-08, | |
| "loss": 0.2942, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 1.0993207693099976, | |
| "eval_runtime": 77.2834, | |
| "eval_samples_per_second": 15.618, | |
| "eval_steps_per_second": 1.954, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 505.22723388671875, | |
| "learning_rate": 3.6182756079587324e-08, | |
| "loss": 0.2615, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 1.080470085144043, | |
| "eval_runtime": 77.1214, | |
| "eval_samples_per_second": 15.651, | |
| "eval_steps_per_second": 1.958, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 1.2971173524856567, | |
| "learning_rate": 3.4647506755097025e-08, | |
| "loss": 0.2715, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 1.0614728927612305, | |
| "eval_runtime": 77.2982, | |
| "eval_samples_per_second": 15.615, | |
| "eval_steps_per_second": 1.953, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 0.17611829936504364, | |
| "learning_rate": 3.3112257430606727e-08, | |
| "loss": 0.2509, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 1.0461839437484741, | |
| "eval_runtime": 75.3494, | |
| "eval_samples_per_second": 16.019, | |
| "eval_steps_per_second": 2.004, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "grad_norm": 228.2562713623047, | |
| "learning_rate": 3.1577008106116435e-08, | |
| "loss": 0.2653, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "eval_loss": 1.0305675268173218, | |
| "eval_runtime": 77.0542, | |
| "eval_samples_per_second": 15.664, | |
| "eval_steps_per_second": 1.96, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "grad_norm": 432.4784240722656, | |
| "learning_rate": 3.0041758781626136e-08, | |
| "loss": 0.2199, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "eval_loss": 1.0182703733444214, | |
| "eval_runtime": 76.296, | |
| "eval_samples_per_second": 15.82, | |
| "eval_steps_per_second": 1.979, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "grad_norm": 0.6385967135429382, | |
| "learning_rate": 2.8506509457135834e-08, | |
| "loss": 0.1844, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "eval_loss": 1.0078601837158203, | |
| "eval_runtime": 76.5441, | |
| "eval_samples_per_second": 15.769, | |
| "eval_steps_per_second": 1.973, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "grad_norm": 57.47743606567383, | |
| "learning_rate": 2.697126013264554e-08, | |
| "loss": 0.2143, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "eval_loss": 0.9969784617424011, | |
| "eval_runtime": 75.6186, | |
| "eval_samples_per_second": 15.962, | |
| "eval_steps_per_second": 1.997, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "grad_norm": 0.11704438179731369, | |
| "learning_rate": 2.5436010808155244e-08, | |
| "loss": 0.2011, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "eval_loss": 0.9886976480484009, | |
| "eval_runtime": 75.238, | |
| "eval_samples_per_second": 16.042, | |
| "eval_steps_per_second": 2.007, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "grad_norm": 39.71356964111328, | |
| "learning_rate": 2.3900761483664948e-08, | |
| "loss": 0.1995, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "eval_loss": 0.980699360370636, | |
| "eval_runtime": 75.7533, | |
| "eval_samples_per_second": 15.933, | |
| "eval_steps_per_second": 1.993, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "grad_norm": 0.09355564415454865, | |
| "learning_rate": 2.236551215917465e-08, | |
| "loss": 0.2493, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "eval_loss": 0.9733120203018188, | |
| "eval_runtime": 75.1219, | |
| "eval_samples_per_second": 16.067, | |
| "eval_steps_per_second": 2.01, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "grad_norm": 0.12566468119621277, | |
| "learning_rate": 2.083026283468435e-08, | |
| "loss": 0.2192, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "eval_loss": 0.966866672039032, | |
| "eval_runtime": 76.5812, | |
| "eval_samples_per_second": 15.761, | |
| "eval_steps_per_second": 1.972, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "grad_norm": 1.9423623085021973, | |
| "learning_rate": 1.9295013510194056e-08, | |
| "loss": 0.1882, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "eval_loss": 0.9616385698318481, | |
| "eval_runtime": 75.5454, | |
| "eval_samples_per_second": 15.977, | |
| "eval_steps_per_second": 1.999, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "grad_norm": 0.12048076093196869, | |
| "learning_rate": 1.7759764185703757e-08, | |
| "loss": 0.2193, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "eval_loss": 0.9550592303276062, | |
| "eval_runtime": 75.1134, | |
| "eval_samples_per_second": 16.069, | |
| "eval_steps_per_second": 2.01, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "grad_norm": 0.19933176040649414, | |
| "learning_rate": 1.622451486121346e-08, | |
| "loss": 0.2148, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "eval_loss": 0.9494845271110535, | |
| "eval_runtime": 76.2803, | |
| "eval_samples_per_second": 15.823, | |
| "eval_steps_per_second": 1.98, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 13.599303245544434, | |
| "learning_rate": 1.4689265536723163e-08, | |
| "loss": 0.1975, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "eval_loss": 0.9449298977851868, | |
| "eval_runtime": 75.6422, | |
| "eval_samples_per_second": 15.957, | |
| "eval_steps_per_second": 1.996, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "grad_norm": 0.8174373507499695, | |
| "learning_rate": 1.3154016212232866e-08, | |
| "loss": 0.1791, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "eval_loss": 0.940927267074585, | |
| "eval_runtime": 75.2379, | |
| "eval_samples_per_second": 16.042, | |
| "eval_steps_per_second": 2.007, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "grad_norm": 705.062744140625, | |
| "learning_rate": 1.161876688774257e-08, | |
| "loss": 0.2057, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "eval_loss": 0.9382032155990601, | |
| "eval_runtime": 76.1075, | |
| "eval_samples_per_second": 15.859, | |
| "eval_steps_per_second": 1.984, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "grad_norm": 0.7270774841308594, | |
| "learning_rate": 1.0083517563252273e-08, | |
| "loss": 0.2037, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "eval_loss": 0.9352426528930664, | |
| "eval_runtime": 75.5821, | |
| "eval_samples_per_second": 15.969, | |
| "eval_steps_per_second": 1.998, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "grad_norm": 2.034397602081299, | |
| "learning_rate": 8.548268238761974e-09, | |
| "loss": 0.2011, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "eval_loss": 0.9317955374717712, | |
| "eval_runtime": 75.3697, | |
| "eval_samples_per_second": 16.014, | |
| "eval_steps_per_second": 2.003, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "grad_norm": 299.313232421875, | |
| "learning_rate": 7.013018914271678e-09, | |
| "loss": 0.1551, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "eval_loss": 0.9295068979263306, | |
| "eval_runtime": 76.109, | |
| "eval_samples_per_second": 15.859, | |
| "eval_steps_per_second": 1.984, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "grad_norm": 0.04151364788413048, | |
| "learning_rate": 5.47776958978138e-09, | |
| "loss": 0.1565, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "eval_loss": 0.9278931617736816, | |
| "eval_runtime": 75.8114, | |
| "eval_samples_per_second": 15.921, | |
| "eval_steps_per_second": 1.992, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "grad_norm": 4.239564895629883, | |
| "learning_rate": 3.942520265291083e-09, | |
| "loss": 0.1963, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "eval_loss": 0.9268263578414917, | |
| "eval_runtime": 75.5159, | |
| "eval_samples_per_second": 15.983, | |
| "eval_steps_per_second": 2.0, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "grad_norm": 0.001815026975236833, | |
| "learning_rate": 2.407270940800786e-09, | |
| "loss": 0.1823, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "eval_loss": 0.9261995553970337, | |
| "eval_runtime": 76.2027, | |
| "eval_samples_per_second": 15.839, | |
| "eval_steps_per_second": 1.982, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "grad_norm": 221.31707763671875, | |
| "learning_rate": 8.720216163104888e-10, | |
| "loss": 0.1854, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "eval_loss": 0.9258912801742554, | |
| "eval_runtime": 75.783, | |
| "eval_samples_per_second": 15.927, | |
| "eval_steps_per_second": 1.993, | |
| "step": 16000 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5756853136862520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |