| { | |
| "best_metric": 0.11086247861385345, | |
| "best_model_checkpoint": "output_pipe/prom_300_notata/origin/checkpoint-1000", | |
| "epoch": 4.0, | |
| "eval_steps": 200, | |
| "global_step": 2656, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15060240963855423, | |
| "grad_norm": 22.862253189086914, | |
| "learning_rate": 2.9470452801227934e-05, | |
| "loss": 0.3165, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.30120481927710846, | |
| "grad_norm": 15.997178077697754, | |
| "learning_rate": 2.831926323867997e-05, | |
| "loss": 0.174, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.30120481927710846, | |
| "eval_accuracy": 0.9440361786319955, | |
| "eval_f1": 0.9440350340667568, | |
| "eval_loss": 0.13831785321235657, | |
| "eval_matthews_correlation": 0.8884885605641855, | |
| "eval_precision": 0.9443041112978641, | |
| "eval_recall": 0.9441844573233038, | |
| "eval_runtime": 1.5317, | |
| "eval_samples_per_second": 3464.713, | |
| "eval_steps_per_second": 54.187, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45180722891566266, | |
| "grad_norm": 9.916468620300293, | |
| "learning_rate": 2.7168073676132003e-05, | |
| "loss": 0.1727, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 8.223430633544922, | |
| "learning_rate": 2.6016884113584036e-05, | |
| "loss": 0.1306, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "eval_accuracy": 0.9470510646316186, | |
| "eval_f1": 0.9469116079894582, | |
| "eval_loss": 0.1320875585079193, | |
| "eval_matthews_correlation": 0.8970111970995432, | |
| "eval_precision": 0.9503975774077025, | |
| "eval_recall": 0.946621567301194, | |
| "eval_runtime": 1.5336, | |
| "eval_samples_per_second": 3460.585, | |
| "eval_steps_per_second": 54.123, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7530120481927711, | |
| "grad_norm": 8.465655326843262, | |
| "learning_rate": 2.4865694551036073e-05, | |
| "loss": 0.1449, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9036144578313253, | |
| "grad_norm": 8.037919998168945, | |
| "learning_rate": 2.3714504988488106e-05, | |
| "loss": 0.1268, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9036144578313253, | |
| "eval_accuracy": 0.9502543810062182, | |
| "eval_f1": 0.9502384353255368, | |
| "eval_loss": 0.1223142221570015, | |
| "eval_matthews_correlation": 0.9019972975694879, | |
| "eval_precision": 0.9514557669338246, | |
| "eval_recall": 0.9505419934872148, | |
| "eval_runtime": 1.5394, | |
| "eval_samples_per_second": 3447.515, | |
| "eval_steps_per_second": 53.918, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0542168674698795, | |
| "grad_norm": 5.252899646759033, | |
| "learning_rate": 2.256331542594014e-05, | |
| "loss": 0.0967, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "grad_norm": 2.8755605220794678, | |
| "learning_rate": 2.1412125863392172e-05, | |
| "loss": 0.0793, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2048192771084336, | |
| "eval_accuracy": 0.9534576973808178, | |
| "eval_f1": 0.9534544978603143, | |
| "eval_loss": 0.12600094079971313, | |
| "eval_matthews_correlation": 0.9075729094229542, | |
| "eval_precision": 0.9539269039717451, | |
| "eval_recall": 0.9536460489075334, | |
| "eval_runtime": 1.546, | |
| "eval_samples_per_second": 3432.697, | |
| "eval_steps_per_second": 53.686, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.355421686746988, | |
| "grad_norm": 15.925186157226562, | |
| "learning_rate": 2.026093630084421e-05, | |
| "loss": 0.0632, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.5060240963855422, | |
| "grad_norm": 5.883889198303223, | |
| "learning_rate": 1.9109746738296238e-05, | |
| "loss": 0.0664, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5060240963855422, | |
| "eval_accuracy": 0.9634445072545694, | |
| "eval_f1": 0.9634296411014179, | |
| "eval_loss": 0.11086247861385345, | |
| "eval_matthews_correlation": 0.9270444420166869, | |
| "eval_precision": 0.9637054919711341, | |
| "eval_recall": 0.9633390224799887, | |
| "eval_runtime": 1.5519, | |
| "eval_samples_per_second": 3419.769, | |
| "eval_steps_per_second": 53.484, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6566265060240963, | |
| "grad_norm": 9.317992210388184, | |
| "learning_rate": 1.7958557175748274e-05, | |
| "loss": 0.0651, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "grad_norm": 7.534521102905273, | |
| "learning_rate": 1.6807367613200308e-05, | |
| "loss": 0.0644, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8072289156626506, | |
| "eval_accuracy": 0.9570378745053703, | |
| "eval_f1": 0.9570372017874856, | |
| "eval_loss": 0.14527226984500885, | |
| "eval_matthews_correlation": 0.9144670396513508, | |
| "eval_precision": 0.9572853548448508, | |
| "eval_recall": 0.9571816906821948, | |
| "eval_runtime": 1.5434, | |
| "eval_samples_per_second": 3438.416, | |
| "eval_steps_per_second": 53.776, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9578313253012047, | |
| "grad_norm": 7.254329681396484, | |
| "learning_rate": 1.565617805065234e-05, | |
| "loss": 0.0635, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.108433734939759, | |
| "grad_norm": 6.914297580718994, | |
| "learning_rate": 1.4504988488104375e-05, | |
| "loss": 0.0374, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.108433734939759, | |
| "eval_accuracy": 0.9628792161296401, | |
| "eval_f1": 0.9628401830979684, | |
| "eval_loss": 0.18788941204547882, | |
| "eval_matthews_correlation": 0.9266187165348982, | |
| "eval_precision": 0.9639729425818305, | |
| "eval_recall": 0.9626467230259594, | |
| "eval_runtime": 1.5439, | |
| "eval_samples_per_second": 3437.375, | |
| "eval_steps_per_second": 53.76, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.2590361445783134, | |
| "grad_norm": 0.7644485235214233, | |
| "learning_rate": 1.3353798925556408e-05, | |
| "loss": 0.0279, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "grad_norm": 8.384675979614258, | |
| "learning_rate": 1.2202609363008443e-05, | |
| "loss": 0.0164, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4096385542168672, | |
| "eval_accuracy": 0.9657056717542868, | |
| "eval_f1": 0.9656756314944064, | |
| "eval_loss": 0.16530963778495789, | |
| "eval_matthews_correlation": 0.9320662060488338, | |
| "eval_precision": 0.9665640323608415, | |
| "eval_recall": 0.9655027778622123, | |
| "eval_runtime": 1.5424, | |
| "eval_samples_per_second": 3440.766, | |
| "eval_steps_per_second": 53.813, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.5602409638554215, | |
| "grad_norm": 18.41074562072754, | |
| "learning_rate": 1.1051419800460476e-05, | |
| "loss": 0.0179, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.710843373493976, | |
| "grad_norm": 0.21532420814037323, | |
| "learning_rate": 9.90023023791251e-06, | |
| "loss": 0.0251, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.710843373493976, | |
| "eval_accuracy": 0.9591106086301112, | |
| "eval_f1": 0.95910930195127, | |
| "eval_loss": 0.22635453939437866, | |
| "eval_matthews_correlation": 0.9187108433086283, | |
| "eval_precision": 0.9594386335644487, | |
| "eval_recall": 0.9592722248152282, | |
| "eval_runtime": 1.5425, | |
| "eval_samples_per_second": 3440.442, | |
| "eval_steps_per_second": 53.808, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.86144578313253, | |
| "grad_norm": 5.3087263107299805, | |
| "learning_rate": 8.749040675364544e-06, | |
| "loss": 0.0224, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "grad_norm": 0.07524475455284119, | |
| "learning_rate": 7.597851112816577e-06, | |
| "loss": 0.0179, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0120481927710845, | |
| "eval_accuracy": 0.9692858488788393, | |
| "eval_f1": 0.9692716697192839, | |
| "eval_loss": 0.17637819051742554, | |
| "eval_matthews_correlation": 0.9387831838919006, | |
| "eval_precision": 0.9696159708142937, | |
| "eval_recall": 0.9691673202840899, | |
| "eval_runtime": 1.5414, | |
| "eval_samples_per_second": 3442.9, | |
| "eval_steps_per_second": 53.846, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.1626506024096384, | |
| "grad_norm": 0.002716244664043188, | |
| "learning_rate": 6.446661550268611e-06, | |
| "loss": 0.0032, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.3132530120481927, | |
| "grad_norm": 0.0046457210555672646, | |
| "learning_rate": 5.295471987720644e-06, | |
| "loss": 0.0037, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.3132530120481927, | |
| "eval_accuracy": 0.9677784058790277, | |
| "eval_f1": 0.9677761908263713, | |
| "eval_loss": 0.21041572093963623, | |
| "eval_matthews_correlation": 0.9355606983847059, | |
| "eval_precision": 0.9677642707707517, | |
| "eval_recall": 0.9677964281666167, | |
| "eval_runtime": 1.5409, | |
| "eval_samples_per_second": 3444.045, | |
| "eval_steps_per_second": 53.864, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.463855421686747, | |
| "grad_norm": 7.352254390716553, | |
| "learning_rate": 4.144282425172678e-06, | |
| "loss": 0.0021, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "grad_norm": 0.1295173019170761, | |
| "learning_rate": 2.993092862624712e-06, | |
| "loss": 0.0014, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.6144578313253013, | |
| "eval_accuracy": 0.9683436970039571, | |
| "eval_f1": 0.9683377061336159, | |
| "eval_loss": 0.22055356204509735, | |
| "eval_matthews_correlation": 0.936697700228757, | |
| "eval_precision": 0.9683925624578065, | |
| "eval_recall": 0.9683051418503684, | |
| "eval_runtime": 1.5412, | |
| "eval_samples_per_second": 3443.434, | |
| "eval_steps_per_second": 53.854, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.765060240963855, | |
| "grad_norm": 0.0010891439160332084, | |
| "learning_rate": 1.8419033000767458e-06, | |
| "loss": 0.0035, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.9156626506024095, | |
| "grad_norm": 0.010914706625044346, | |
| "learning_rate": 6.907137375287798e-07, | |
| "loss": 0.0026, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.9156626506024095, | |
| "eval_accuracy": 0.9690974185038628, | |
| "eval_f1": 0.9690902179184598, | |
| "eval_loss": 0.21992520987987518, | |
| "eval_matthews_correlation": 0.938226978614608, | |
| "eval_precision": 0.9691836691199455, | |
| "eval_recall": 0.9690433199920572, | |
| "eval_runtime": 1.5378, | |
| "eval_samples_per_second": 3450.927, | |
| "eval_steps_per_second": 53.972, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 2656, | |
| "total_flos": 8308929553956864.0, | |
| "train_loss": 0.06583031266366682, | |
| "train_runtime": 193.8482, | |
| "train_samples_per_second": 875.984, | |
| "train_steps_per_second": 13.701 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2656, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8308929553956864.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |