| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 648, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.0, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.0, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.6820276497695853, | |
| "precision": 0.6540404040404041, | |
| "recall": 0.7125171939477304, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.43557936284183824, | |
| "precision": 0.5235513385293121, | |
| "recall": 0.3729181752353367, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.5835814547001276, | |
| "precision": 0.5468314069350338, | |
| "recall": 0.6256269949840402, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8638885700072705, | |
| "precision": 0.8318349299926308, | |
| "recall": 0.8985115020297699, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.8838224593742419, | |
| "precision": 0.8730234786775275, | |
| "recall": 0.8948919449901768, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.7876854942657897, | |
| "eval_loss": 0.5461820363998413, | |
| "eval_macro avg": { | |
| "f1-score": 0.4926999280990091, | |
| "precision": 0.4898973654535584, | |
| "recall": 0.500637973026722, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0655, | |
| "eval_samples_per_second": 15.991, | |
| "eval_steps_per_second": 2.172, | |
| "eval_weighted avg": { | |
| "f1-score": 0.7743853744001297, | |
| "precision": 0.7661527277359783, | |
| "recall": 0.7876854942657897, | |
| "support": 30257.0 | |
| }, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.46397188049209137, | |
| "precision": 0.48175182481751827, | |
| "recall": 0.44745762711864406, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.6402439024390244, | |
| "precision": 0.6104651162790697, | |
| "recall": 0.6730769230769231, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7515299877600979, | |
| "precision": 0.6769570011025359, | |
| "recall": 0.844566712517194, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5573419078242231, | |
| "precision": 0.5500705218617772, | |
| "recall": 0.5648081100651702, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7704582651391162, | |
| "precision": 0.6987012987012987, | |
| "recall": 0.8586411308709531, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8736837792436597, | |
| "precision": 0.9002026342451874, | |
| "recall": 0.8486826394969355, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9114385688236163, | |
| "precision": 0.9145485115221046, | |
| "recall": 0.9083497053045186, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8256932280133523, | |
| "eval_loss": 0.4454804062843323, | |
| "eval_macro avg": { | |
| "f1-score": 0.7098097559602613, | |
| "precision": 0.6903852726470703, | |
| "recall": 0.7350832640643341, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0817, | |
| "eval_samples_per_second": 15.94, | |
| "eval_steps_per_second": 2.165, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8274557458920446, | |
| "precision": 0.8315444607100787, | |
| "recall": 0.8256932280133523, | |
| "support": 30257.0 | |
| }, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5551724137931033, | |
| "precision": 0.5649122807017544, | |
| "recall": 0.5457627118644067, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7108433734939757, | |
| "precision": 0.6704545454545454, | |
| "recall": 0.7564102564102564, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7860892388451444, | |
| "precision": 0.7515683814303639, | |
| "recall": 0.8239339752407153, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5878616796047988, | |
| "precision": 0.573296627666896, | |
| "recall": 0.6031860970311369, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7989093387866394, | |
| "precision": 0.7961956521739131, | |
| "recall": 0.801641586867305, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8824656529006482, | |
| "precision": 0.9044793581815143, | |
| "recall": 0.8614980498288626, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9169814976266588, | |
| "precision": 0.9044525128989107, | |
| "recall": 0.9298624754420433, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8402683676504611, | |
| "eval_loss": 0.41143855452537537, | |
| "eval_macro avg": { | |
| "f1-score": 0.7483318850072812, | |
| "precision": 0.7379084797868425, | |
| "recall": 0.7603278789549609, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0654, | |
| "eval_samples_per_second": 15.991, | |
| "eval_steps_per_second": 2.172, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8412916372002002, | |
| "precision": 0.8430827906594975, | |
| "recall": 0.8402683676504611, | |
| "support": 30257.0 | |
| }, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5518394648829431, | |
| "precision": 0.5445544554455446, | |
| "recall": 0.559322033898305, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7048192771084337, | |
| "precision": 0.6647727272727273, | |
| "recall": 0.75, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.792225201072386, | |
| "precision": 0.7725490196078432, | |
| "recall": 0.81292984869326, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.577999737153371, | |
| "precision": 0.6344489324870167, | |
| "recall": 0.5307748008689356, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.800089746466233, | |
| "precision": 0.7875441696113075, | |
| "recall": 0.8130414956680346, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8919449901768174, | |
| "precision": 0.8642131979695431, | |
| "recall": 0.9215155615696887, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9180246175312703, | |
| "precision": 0.9316273895013655, | |
| "recall": 0.9048133595284872, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8475063621641273, | |
| "eval_loss": 0.476072758436203, | |
| "eval_macro avg": { | |
| "f1-score": 0.7481347191987793, | |
| "precision": 0.7428156988421926, | |
| "recall": 0.7560567286038158, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0796, | |
| "eval_samples_per_second": 15.946, | |
| "eval_steps_per_second": 2.166, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8443975745657389, | |
| "precision": 0.8435295888400192, | |
| "recall": 0.8475063621641273, | |
| "support": 30257.0 | |
| }, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5932203389830508, | |
| "precision": 0.5932203389830508, | |
| "recall": 0.5932203389830508, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7398119122257053, | |
| "precision": 0.7239263803680982, | |
| "recall": 0.7564102564102564, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7774172615184944, | |
| "precision": 0.7358722358722358, | |
| "recall": 0.8239339752407153, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5827523883294604, | |
| "precision": 0.6264224257563141, | |
| "recall": 0.5447743181269611, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.79782711384034, | |
| "precision": 0.8275355218030377, | |
| "recall": 0.7701778385772914, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8862478777589133, | |
| "precision": 0.8792698213726104, | |
| "recall": 0.8933375786038367, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9135231488172665, | |
| "precision": 0.8963789354259242, | |
| "recall": 0.931335952848723, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8441682916349935, | |
| "eval_loss": 0.526358962059021, | |
| "eval_macro avg": { | |
| "f1-score": 0.7558285773533188, | |
| "precision": 0.7546608085116101, | |
| "recall": 0.7590271798272621, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.076, | |
| "eval_samples_per_second": 15.958, | |
| "eval_steps_per_second": 2.167, | |
| "eval_weighted avg": { | |
| "f1-score": 0.841232400970915, | |
| "precision": 0.8396195745969204, | |
| "recall": 0.8441682916349935, | |
| "support": 30257.0 | |
| }, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5809682804674458, | |
| "precision": 0.5723684210526315, | |
| "recall": 0.5898305084745763, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7402597402597402, | |
| "precision": 0.75, | |
| "recall": 0.7307692307692307, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7956298200514138, | |
| "precision": 0.7466827503015682, | |
| "recall": 0.8514442916093535, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.580742605412209, | |
| "precision": 0.6067859021567596, | |
| "recall": 0.556842867487328, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.798828125, | |
| "precision": 0.859695218076721, | |
| "recall": 0.7460100319197447, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8905630763194036, | |
| "precision": 0.8781922303049063, | |
| "recall": 0.9032874313460161, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9183922046285018, | |
| "precision": 0.9110681488641856, | |
| "recall": 0.9258349705304518, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8468453580989523, | |
| "eval_loss": 0.5279645919799805, | |
| "eval_macro avg": { | |
| "f1-score": 0.7579119788769592, | |
| "precision": 0.7606846672509676, | |
| "recall": 0.7577170474481001, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0684, | |
| "eval_samples_per_second": 15.981, | |
| "eval_steps_per_second": 2.17, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8447801270668736, | |
| "precision": 0.8439473640584135, | |
| "recall": 0.8468453580989523, | |
| "support": 30257.0 | |
| }, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "grad_norm": 3.7896041870117188, | |
| "learning_rate": 1.7530864197530865e-05, | |
| "loss": 0.4007, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5654450261780105, | |
| "precision": 0.5827338129496403, | |
| "recall": 0.5491525423728814, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7454545454545455, | |
| "precision": 0.7068965517241379, | |
| "recall": 0.7884615384615384, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.8080155138978669, | |
| "precision": 0.7621951219512195, | |
| "recall": 0.859697386519945, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5279187817258884, | |
| "precision": 0.6115702479338843, | |
| "recall": 0.4643977793869177, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.8239801667793555, | |
| "precision": 0.8146167557932263, | |
| "recall": 0.833561331509348, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8874376008707728, | |
| "precision": 0.8396306818181818, | |
| "recall": 0.9410172729443604, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.916984006092917, | |
| "precision": 0.9490278507619548, | |
| "recall": 0.8870333988212181, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8432428859437485, | |
| "eval_loss": 0.673554003238678, | |
| "eval_macro avg": { | |
| "f1-score": 0.7536050915713366, | |
| "precision": 0.7523815747046064, | |
| "recall": 0.7604744642880299, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0724, | |
| "eval_samples_per_second": 15.969, | |
| "eval_steps_per_second": 2.169, | |
| "eval_weighted avg": { | |
| "f1-score": 0.8377716313785262, | |
| "precision": 0.8383472270757442, | |
| "recall": 0.8432428859437485, | |
| "support": 30257.0 | |
| }, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_B-Claim": { | |
| "f1-score": 0.5714285714285714, | |
| "precision": 0.554140127388535, | |
| "recall": 0.5898305084745763, | |
| "support": 295.0 | |
| }, | |
| "eval_B-MajorClaim": { | |
| "f1-score": 0.7103825136612022, | |
| "precision": 0.6190476190476191, | |
| "recall": 0.8333333333333334, | |
| "support": 156.0 | |
| }, | |
| "eval_B-Premise": { | |
| "f1-score": 0.7976111479761113, | |
| "precision": 0.7705128205128206, | |
| "recall": 0.8266850068775791, | |
| "support": 727.0 | |
| }, | |
| "eval_I-Claim": { | |
| "f1-score": 0.5769132977370963, | |
| "precision": 0.6094547408004297, | |
| "recall": 0.5476707699734492, | |
| "support": 4143.0 | |
| }, | |
| "eval_I-MajorClaim": { | |
| "f1-score": 0.7987654320987654, | |
| "precision": 0.7277840269966255, | |
| "recall": 0.8850889192886456, | |
| "support": 2193.0 | |
| }, | |
| "eval_I-Premise": { | |
| "f1-score": 0.8899450117831893, | |
| "precision": 0.8784213382957277, | |
| "recall": 0.9017750537292049, | |
| "support": 12563.0 | |
| }, | |
| "eval_O": { | |
| "f1-score": 0.9160536128187039, | |
| "precision": 0.9404096834264432, | |
| "recall": 0.8929273084479371, | |
| "support": 10180.0 | |
| }, | |
| "eval_accuracy": 0.8439038900089235, | |
| "eval_loss": 0.6914934515953064, | |
| "eval_macro avg": { | |
| "f1-score": 0.7515856553576629, | |
| "precision": 0.7285386223526001, | |
| "recall": 0.7824729857321036, | |
| "support": 30257.0 | |
| }, | |
| "eval_runtime": 5.0936, | |
| "eval_samples_per_second": 15.902, | |
| "eval_steps_per_second": 2.16, | |
| "eval_weighted avg": { | |
| "f1-score": 0.843008353502502, | |
| "precision": 0.8444388009152946, | |
| "recall": 0.8439038900089235, | |
| "support": 30257.0 | |
| }, | |
| "step": 648 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4050, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1146754059192000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |