| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.7573065757751465, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.56, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7243107769423559, | |
| "eval_f1": 0.6401043033324587, | |
| "eval_loss": 0.4985913038253784, | |
| "eval_precision": 0.6583725987676694, | |
| "eval_recall": 0.6324331696672122, | |
| "eval_runtime": 5.1414, | |
| "eval_samples_per_second": 77.605, | |
| "eval_steps_per_second": 9.725, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.935351848602295, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4895, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_f1": 0.7272049226179378, | |
| "eval_loss": 0.4648020267486572, | |
| "eval_precision": 0.7208157524613221, | |
| "eval_recall": 0.7569558101472995, | |
| "eval_runtime": 5.0899, | |
| "eval_samples_per_second": 78.39, | |
| "eval_steps_per_second": 9.823, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.3584516048431396, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.407, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8014506303154594, | |
| "eval_loss": 0.3817030191421509, | |
| "eval_precision": 0.8, | |
| "eval_recall": 0.8029641753046008, | |
| "eval_runtime": 5.0712, | |
| "eval_samples_per_second": 78.68, | |
| "eval_steps_per_second": 9.86, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.330716848373413, | |
| "learning_rate": 4e-05, | |
| "loss": 0.347, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.816408656658218, | |
| "eval_loss": 0.3600384294986725, | |
| "eval_precision": 0.8270654903728508, | |
| "eval_recall": 0.8078741589379888, | |
| "eval_runtime": 5.1423, | |
| "eval_samples_per_second": 77.592, | |
| "eval_steps_per_second": 9.723, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.4344160556793213, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3094, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.801956204379562, | |
| "eval_loss": 0.36615419387817383, | |
| "eval_precision": 0.7932330827067668, | |
| "eval_recall": 0.8144208037825059, | |
| "eval_runtime": 5.0766, | |
| "eval_samples_per_second": 78.595, | |
| "eval_steps_per_second": 9.849, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 5.476891994476318, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2924, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8176959060086304, | |
| "eval_loss": 0.367689311504364, | |
| "eval_precision": 0.8074396108253187, | |
| "eval_recall": 0.8332878705219131, | |
| "eval_runtime": 5.0721, | |
| "eval_samples_per_second": 78.665, | |
| "eval_steps_per_second": 9.858, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.243276596069336, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2723, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8271953405017921, | |
| "eval_loss": 0.33599093556404114, | |
| "eval_precision": 0.8228172499116295, | |
| "eval_recall": 0.8321513002364066, | |
| "eval_runtime": 5.09, | |
| "eval_samples_per_second": 78.388, | |
| "eval_steps_per_second": 9.823, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.912153959274292, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2611, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8219298245614035, | |
| "eval_loss": 0.32096049189567566, | |
| "eval_precision": 0.8273273273273273, | |
| "eval_recall": 0.8171485724677214, | |
| "eval_runtime": 5.0461, | |
| "eval_samples_per_second": 79.071, | |
| "eval_steps_per_second": 9.909, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 7.307469844818115, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2584, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8270277816557308, | |
| "eval_loss": 0.34841448068618774, | |
| "eval_precision": 0.8187875321704318, | |
| "eval_recall": 0.8378796144753592, | |
| "eval_runtime": 5.0698, | |
| "eval_samples_per_second": 78.701, | |
| "eval_steps_per_second": 9.862, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.1979804039001465, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2432, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8289650949173301, | |
| "eval_loss": 0.3248250186443329, | |
| "eval_precision": 0.8325081997648369, | |
| "eval_recall": 0.82569558101473, | |
| "eval_runtime": 5.0514, | |
| "eval_samples_per_second": 78.988, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.695533812046051, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2387, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8306964902709584, | |
| "eval_loss": 0.31659966707229614, | |
| "eval_precision": 0.8306964902709584, | |
| "eval_recall": 0.8306964902709584, | |
| "eval_runtime": 5.0901, | |
| "eval_samples_per_second": 78.388, | |
| "eval_steps_per_second": 9.823, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 12.895674705505371, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2273, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8255172205802521, | |
| "eval_loss": 0.3223665952682495, | |
| "eval_precision": 0.8239495798319327, | |
| "eval_recall": 0.8271503909801782, | |
| "eval_runtime": 5.0669, | |
| "eval_samples_per_second": 78.747, | |
| "eval_steps_per_second": 9.868, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.45775964856147766, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2259, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8342105263157895, | |
| "eval_loss": 0.3087916076183319, | |
| "eval_precision": 0.8398085585585586, | |
| "eval_recall": 0.82924168030551, | |
| "eval_runtime": 5.0958, | |
| "eval_samples_per_second": 78.3, | |
| "eval_steps_per_second": 9.812, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 12.224174499511719, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2268, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8375505157126486, | |
| "eval_loss": 0.3109127879142761, | |
| "eval_precision": 0.8359243697478991, | |
| "eval_recall": 0.8392434988179669, | |
| "eval_runtime": 5.0692, | |
| "eval_samples_per_second": 78.711, | |
| "eval_steps_per_second": 9.864, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.779151201248169, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2226, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8484099018899409, | |
| "eval_loss": 0.31573131680488586, | |
| "eval_precision": 0.8430645161290322, | |
| "eval_recall": 0.8545644662665939, | |
| "eval_runtime": 5.1042, | |
| "eval_samples_per_second": 78.171, | |
| "eval_steps_per_second": 9.796, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.361120700836182, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2214, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.3111570477485657, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0563, | |
| "eval_samples_per_second": 78.912, | |
| "eval_steps_per_second": 9.889, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.201449871063232, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2148, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8367430441898527, | |
| "eval_loss": 0.3018483519554138, | |
| "eval_precision": 0.8367430441898527, | |
| "eval_recall": 0.8367430441898527, | |
| "eval_runtime": 5.1148, | |
| "eval_samples_per_second": 78.008, | |
| "eval_steps_per_second": 9.775, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.3120012283325195, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2088, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8372817261706151, | |
| "eval_loss": 0.31123366951942444, | |
| "eval_precision": 0.8308913308913308, | |
| "eval_recall": 0.8449718130569195, | |
| "eval_runtime": 5.0776, | |
| "eval_samples_per_second": 78.58, | |
| "eval_steps_per_second": 9.847, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.43890380859375, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2116, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8315338681464504, | |
| "eval_loss": 0.302460640668869, | |
| "eval_precision": 0.8299369747899159, | |
| "eval_recall": 0.8331969448990726, | |
| "eval_runtime": 5.0766, | |
| "eval_samples_per_second": 78.596, | |
| "eval_steps_per_second": 9.849, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.140255928039551, | |
| "learning_rate": 0.0, | |
| "loss": 0.2099, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8375505157126486, | |
| "eval_loss": 0.30324476957321167, | |
| "eval_precision": 0.8359243697478991, | |
| "eval_recall": 0.8392434988179669, | |
| "eval_runtime": 5.0544, | |
| "eval_samples_per_second": 78.942, | |
| "eval_steps_per_second": 9.892, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7635661149264000.0, | |
| "train_loss": 0.28240973284987153, | |
| "train_runtime": 1940.7402, | |
| "train_samples_per_second": 37.491, | |
| "train_steps_per_second": 1.257 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7635661149264000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |