adapters-mistral-bnb8-QLORA-super_glue-multirc
/
trainer_state-mistral-bnb8-QLORA-super_glue-multirc-sequence_classification.json
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.792, | |
| "eval_steps": 1, | |
| "global_step": 14, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 426.2940979003906, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.2927, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "eval_exact_match": 0.0, | |
| "eval_f1_a": 0.0, | |
| "eval_f1_m": 0.35077572820670067, | |
| "eval_loss": 6.642009735107422, | |
| "eval_runtime": 84.411, | |
| "eval_samples_per_second": 2.962, | |
| "eval_steps_per_second": 0.746, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 358.92913818359375, | |
| "learning_rate": 5e-05, | |
| "loss": 6.6477, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "eval_exact_match": 0.16326530612244897, | |
| "eval_f1_a": 0.6279863481228669, | |
| "eval_f1_m": 0.5070784590892632, | |
| "eval_loss": 2.5748679637908936, | |
| "eval_runtime": 84.5004, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 202.6668243408203, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 1.5318, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "eval_exact_match": 0.2653061224489796, | |
| "eval_f1_a": 0.6779661016949152, | |
| "eval_f1_m": 0.6622228112024031, | |
| "eval_loss": 1.5085991621017456, | |
| "eval_runtime": 84.4987, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 89.18372344970703, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.021, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "eval_exact_match": 0.0, | |
| "eval_f1_a": 0.038834951456310676, | |
| "eval_f1_m": 0.36717514511632166, | |
| "eval_loss": 3.572444438934326, | |
| "eval_runtime": 84.4787, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 249.71437072753906, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 3.3149, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_exact_match": 0.04081632653061224, | |
| "eval_f1_a": 0.05769230769230769, | |
| "eval_f1_m": 0.3837689254856121, | |
| "eval_loss": 2.9223897457122803, | |
| "eval_runtime": 84.4654, | |
| "eval_samples_per_second": 2.96, | |
| "eval_steps_per_second": 0.746, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 234.4349365234375, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.0938, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "eval_exact_match": 0.24489795918367346, | |
| "eval_f1_a": 0.6127167630057804, | |
| "eval_f1_m": 0.6370587195917328, | |
| "eval_loss": 1.037977933883667, | |
| "eval_runtime": 84.4697, | |
| "eval_samples_per_second": 2.96, | |
| "eval_steps_per_second": 0.746, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 44.01259994506836, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 1.3183, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "eval_exact_match": 0.2653061224489796, | |
| "eval_f1_a": 0.6926406926406927, | |
| "eval_f1_m": 0.6772793532997615, | |
| "eval_loss": 0.9939583539962769, | |
| "eval_runtime": 84.4865, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 80.42048645019531, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.9766, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "eval_exact_match": 0.32653061224489793, | |
| "eval_f1_a": 0.709090909090909, | |
| "eval_f1_m": 0.7124046588332301, | |
| "eval_loss": 0.9605630040168762, | |
| "eval_runtime": 84.508, | |
| "eval_samples_per_second": 2.958, | |
| "eval_steps_per_second": 0.745, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 50.82283401489258, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.5821, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "eval_exact_match": 0.3469387755102041, | |
| "eval_f1_a": 0.694300518134715, | |
| "eval_f1_m": 0.6903760638454516, | |
| "eval_loss": 0.932707667350769, | |
| "eval_runtime": 84.4808, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 47.22996139526367, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6158, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_exact_match": 0.3673469387755102, | |
| "eval_f1_a": 0.6989247311827956, | |
| "eval_f1_m": 0.7010895000690919, | |
| "eval_loss": 0.9111232757568359, | |
| "eval_runtime": 84.4887, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 70.53364562988281, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.5635, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "eval_exact_match": 0.3469387755102041, | |
| "eval_f1_a": 0.7040816326530612, | |
| "eval_f1_m": 0.7049636304738346, | |
| "eval_loss": 0.8203014135360718, | |
| "eval_runtime": 84.4658, | |
| "eval_samples_per_second": 2.96, | |
| "eval_steps_per_second": 0.746, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 24.0622501373291, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.5367, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "eval_exact_match": 0.2857142857142857, | |
| "eval_f1_a": 0.7115384615384615, | |
| "eval_f1_m": 0.7037444414995435, | |
| "eval_loss": 0.8254165649414062, | |
| "eval_runtime": 84.4794, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 49.5471076965332, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.5805, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "eval_exact_match": 0.3877551020408163, | |
| "eval_f1_a": 0.7317073170731707, | |
| "eval_f1_m": 0.7391583473216126, | |
| "eval_loss": 0.7916401624679565, | |
| "eval_runtime": 84.4743, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 32.582847595214844, | |
| "learning_rate": 0.0, | |
| "loss": 0.4737, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "eval_exact_match": 0.3469387755102041, | |
| "eval_f1_a": 0.6903553299492386, | |
| "eval_f1_m": 0.6909562093235563, | |
| "eval_loss": 0.7645593881607056, | |
| "eval_runtime": 84.4968, | |
| "eval_samples_per_second": 2.959, | |
| "eval_steps_per_second": 0.746, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "step": 14, | |
| "total_flos": 4.3944687722496e+16, | |
| "train_loss": 2.0392316984278813, | |
| "train_runtime": 3611.0931, | |
| "train_samples_per_second": 0.554, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 14, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 4.3944687722496e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |