| { | |
| "best_metric": 0.541618287563324, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-11500", | |
| "epoch": 1.2265358361774745, | |
| "eval_steps": 500, | |
| "global_step": 11500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05332764505119454, | |
| "grad_norm": 414.62841796875, | |
| "learning_rate": 4.911120591581342e-07, | |
| "loss": 0.3812, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05332764505119454, | |
| "eval_loss": 1.1163015365600586, | |
| "eval_runtime": 61.8962, | |
| "eval_samples_per_second": 15.946, | |
| "eval_steps_per_second": 2.003, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10665529010238908, | |
| "grad_norm": 34.08418655395508, | |
| "learning_rate": 4.822241183162685e-07, | |
| "loss": 0.2683, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10665529010238908, | |
| "eval_loss": 0.9684447646141052, | |
| "eval_runtime": 61.821, | |
| "eval_samples_per_second": 15.965, | |
| "eval_steps_per_second": 2.006, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1599829351535836, | |
| "grad_norm": 350.4574890136719, | |
| "learning_rate": 4.733361774744027e-07, | |
| "loss": 0.2119, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1599829351535836, | |
| "eval_loss": 0.9099885821342468, | |
| "eval_runtime": 61.9606, | |
| "eval_samples_per_second": 15.929, | |
| "eval_steps_per_second": 2.001, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.21331058020477817, | |
| "grad_norm": 0.01787523180246353, | |
| "learning_rate": 4.6444823663253695e-07, | |
| "loss": 0.1889, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21331058020477817, | |
| "eval_loss": 0.8620074987411499, | |
| "eval_runtime": 63.2824, | |
| "eval_samples_per_second": 15.597, | |
| "eval_steps_per_second": 1.959, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2666382252559727, | |
| "grad_norm": 118.66907501220703, | |
| "learning_rate": 4.5556029579067116e-07, | |
| "loss": 0.2071, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2666382252559727, | |
| "eval_loss": 0.791786253452301, | |
| "eval_runtime": 63.7932, | |
| "eval_samples_per_second": 15.472, | |
| "eval_steps_per_second": 1.944, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3199658703071672, | |
| "grad_norm": 7.21087323185543e-09, | |
| "learning_rate": 4.4667235494880547e-07, | |
| "loss": 0.1588, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3199658703071672, | |
| "eval_loss": 0.7657254338264465, | |
| "eval_runtime": 62.4738, | |
| "eval_samples_per_second": 15.799, | |
| "eval_steps_per_second": 1.985, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.37329351535836175, | |
| "grad_norm": 0.0014836661284789443, | |
| "learning_rate": 4.377844141069397e-07, | |
| "loss": 0.1718, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.37329351535836175, | |
| "eval_loss": 0.7610095739364624, | |
| "eval_runtime": 63.9118, | |
| "eval_samples_per_second": 15.443, | |
| "eval_steps_per_second": 1.94, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.42662116040955633, | |
| "grad_norm": 3.2739710807800293, | |
| "learning_rate": 4.2889647326507393e-07, | |
| "loss": 0.1113, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42662116040955633, | |
| "eval_loss": 0.7458393573760986, | |
| "eval_runtime": 62.1803, | |
| "eval_samples_per_second": 15.873, | |
| "eval_steps_per_second": 1.994, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.47994880546075086, | |
| "grad_norm": 1.788787841796875, | |
| "learning_rate": 4.2000853242320814e-07, | |
| "loss": 0.1313, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47994880546075086, | |
| "eval_loss": 0.7168156504631042, | |
| "eval_runtime": 62.1477, | |
| "eval_samples_per_second": 15.882, | |
| "eval_steps_per_second": 1.995, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5332764505119454, | |
| "grad_norm": 4.031916250823997e-05, | |
| "learning_rate": 4.1112059158134245e-07, | |
| "loss": 0.1649, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5332764505119454, | |
| "eval_loss": 0.7019046545028687, | |
| "eval_runtime": 62.2092, | |
| "eval_samples_per_second": 15.866, | |
| "eval_steps_per_second": 1.993, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.58660409556314, | |
| "grad_norm": 3.883213139488362e-06, | |
| "learning_rate": 4.0223265073947665e-07, | |
| "loss": 0.1245, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.58660409556314, | |
| "eval_loss": 0.6812178492546082, | |
| "eval_runtime": 62.492, | |
| "eval_samples_per_second": 15.794, | |
| "eval_steps_per_second": 1.984, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6399317406143344, | |
| "grad_norm": 80.17021942138672, | |
| "learning_rate": 3.933447098976109e-07, | |
| "loss": 0.1286, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6399317406143344, | |
| "eval_loss": 0.6501584649085999, | |
| "eval_runtime": 62.5236, | |
| "eval_samples_per_second": 15.786, | |
| "eval_steps_per_second": 1.983, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.693259385665529, | |
| "grad_norm": 0.0001418297761119902, | |
| "learning_rate": 3.8445676905574517e-07, | |
| "loss": 0.1076, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.693259385665529, | |
| "eval_loss": 0.6153982281684875, | |
| "eval_runtime": 62.2927, | |
| "eval_samples_per_second": 15.845, | |
| "eval_steps_per_second": 1.991, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7465870307167235, | |
| "grad_norm": 5.804526495012396e-07, | |
| "learning_rate": 3.755688282138794e-07, | |
| "loss": 0.1477, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7465870307167235, | |
| "eval_loss": 0.6118016839027405, | |
| "eval_runtime": 63.1588, | |
| "eval_samples_per_second": 15.627, | |
| "eval_steps_per_second": 1.963, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7999146757679181, | |
| "grad_norm": 2.66489315032959, | |
| "learning_rate": 3.6668088737201363e-07, | |
| "loss": 0.1315, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7999146757679181, | |
| "eval_loss": 0.601563036441803, | |
| "eval_runtime": 62.9163, | |
| "eval_samples_per_second": 15.688, | |
| "eval_steps_per_second": 1.971, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8532423208191127, | |
| "grad_norm": 851.2547607421875, | |
| "learning_rate": 3.5779294653014783e-07, | |
| "loss": 0.1413, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8532423208191127, | |
| "eval_loss": 0.584888756275177, | |
| "eval_runtime": 62.6735, | |
| "eval_samples_per_second": 15.748, | |
| "eval_steps_per_second": 1.979, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9065699658703071, | |
| "grad_norm": 6.041275628376752e-05, | |
| "learning_rate": 3.4890500568828214e-07, | |
| "loss": 0.124, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9065699658703071, | |
| "eval_loss": 0.5766209363937378, | |
| "eval_runtime": 63.3493, | |
| "eval_samples_per_second": 15.58, | |
| "eval_steps_per_second": 1.957, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9598976109215017, | |
| "grad_norm": 3.0901598930358887, | |
| "learning_rate": 3.4001706484641635e-07, | |
| "loss": 0.1215, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9598976109215017, | |
| "eval_loss": 0.5558739304542542, | |
| "eval_runtime": 63.0862, | |
| "eval_samples_per_second": 15.645, | |
| "eval_steps_per_second": 1.966, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.0132252559726962, | |
| "grad_norm": 2509.93212890625, | |
| "learning_rate": 3.311291240045506e-07, | |
| "loss": 0.131, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0132252559726962, | |
| "eval_loss": 0.563326358795166, | |
| "eval_runtime": 62.3607, | |
| "eval_samples_per_second": 15.827, | |
| "eval_steps_per_second": 1.988, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0665529010238908, | |
| "grad_norm": 32.05735778808594, | |
| "learning_rate": 3.2224118316268486e-07, | |
| "loss": 0.0348, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0665529010238908, | |
| "eval_loss": 0.553093671798706, | |
| "eval_runtime": 62.9895, | |
| "eval_samples_per_second": 15.669, | |
| "eval_steps_per_second": 1.969, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.1198805460750854, | |
| "grad_norm": 0.004324838053435087, | |
| "learning_rate": 3.133532423208191e-07, | |
| "loss": 0.0687, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.1198805460750854, | |
| "eval_loss": 0.5494681000709534, | |
| "eval_runtime": 62.3211, | |
| "eval_samples_per_second": 15.837, | |
| "eval_steps_per_second": 1.99, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.17320819112628, | |
| "grad_norm": 0.09391242265701294, | |
| "learning_rate": 3.044653014789533e-07, | |
| "loss": 0.0749, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.17320819112628, | |
| "eval_loss": 0.5473881363868713, | |
| "eval_runtime": 62.3454, | |
| "eval_samples_per_second": 15.831, | |
| "eval_steps_per_second": 1.989, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2265358361774745, | |
| "grad_norm": 1.9322777752850406e-11, | |
| "learning_rate": 2.955773606370876e-07, | |
| "loss": 0.1083, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2265358361774745, | |
| "eval_loss": 0.541618287563324, | |
| "eval_runtime": 63.1868, | |
| "eval_samples_per_second": 15.62, | |
| "eval_steps_per_second": 1.962, | |
| "step": 11500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 28128, | |
| "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": 4137940556192520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |