{ "best_metric": null, "best_model_checkpoint": null, "epoch": 17.0, "eval_steps": 500, "global_step": 1377, "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": 0.32421052631578945, "precision": 0.42777777777777776, "recall": 0.26101694915254237, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.192090395480226, "precision": 0.8095238095238095, "recall": 0.10897435897435898, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8566265060240965, "precision": 0.7620578778135049, "recall": 0.9779917469050894, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.5001390820584144, "precision": 0.5900886117492615, "recall": 0.43398503499879315, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.776470588235294, "precision": 0.6649333766655834, "recall": 0.93296853625171, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8861458374413376, "precision": 0.8781460059402845, "recall": 0.8942927644670858, "support": 12563.0 }, "eval_O": { "f1-score": 0.9987647926046939, "precision": 0.9977708781147998, "recall": 0.9997606892150607, "support": 12536.0 }, "eval_accuracy": 0.8713396498328887, "eval_loss": 0.3092103600502014, "eval_macro avg": { "f1-score": 0.647778246879979, "precision": 0.732899762512146, "recall": 0.6584271542806628, "support": 32613.0 }, "eval_runtime": 4.965, "eval_samples_per_second": 16.314, "eval_steps_per_second": 2.215, "eval_weighted avg": { "f1-score": 0.8639627599323451, "precision": 0.8662078389010613, "recall": 0.8713396498328887, "support": 32613.0 }, "step": 81 }, { "epoch": 2.0, "eval_B-Claim": { "f1-score": 0.6186291739894552, "precision": 0.6423357664233577, "recall": 0.5966101694915255, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8113879003558718, "precision": 0.912, "recall": 0.7307692307692307, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8871181938911022, "precision": 0.8575096277278562, "recall": 0.9188445667125172, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.656253149249219, "precision": 0.5633217993079584, "recall": 0.7859039343470915, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8368159203980099, "precision": 0.9206349206349206, "recall": 0.7669858641130871, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8831027253668763, "precision": 0.933020288827855, "recall": 0.8382551938231314, "support": 12563.0 }, "eval_O": { "f1-score": 0.9998006141085457, "precision": 0.9996013077107089, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.8880814399165977, "eval_loss": 0.2581149935722351, "eval_macro avg": { "f1-score": 0.8133010967655829, "precision": 0.8326319586618081, "recall": 0.8053384227509405, "support": 32613.0 }, "eval_runtime": 5.0296, "eval_samples_per_second": 16.105, "eval_steps_per_second": 2.187, "eval_weighted avg": { "f1-score": 0.8933836566327042, "precision": 0.9064023864086126, "recall": 0.8880814399165977, "support": 32613.0 }, "step": 162 }, { "epoch": 3.0, "eval_B-Claim": { "f1-score": 0.6576271186440678, "precision": 0.6576271186440678, "recall": 0.6576271186440678, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8673139158576051, "precision": 0.8758169934640523, "recall": 0.8589743589743589, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8838487972508591, "precision": 0.8832417582417582, "recall": 0.8844566712517193, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6468460321049062, "precision": 0.6083351052519669, "recall": 0.6905623944001931, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8695855771482088, "precision": 0.8936477382098171, "recall": 0.8467852257181943, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8926256077795786, "precision": 0.9090533960551291, "recall": 0.8767810236408501, "support": 12563.0 }, "eval_O": { "f1-score": 0.9998803589232302, "precision": 0.9997607464710104, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.8965749854352559, "eval_loss": 0.2399493306875229, "eval_macro avg": { "f1-score": 0.8311039153869222, "precision": 0.8324975509054003, "recall": 0.8307409703756262, "support": 32613.0 }, "eval_runtime": 5.0024, "eval_samples_per_second": 16.192, "eval_steps_per_second": 2.199, "eval_weighted avg": { "f1-score": 0.8986385496019461, "precision": 0.9016737059631544, "recall": 0.8965749854352559, "support": 32613.0 }, "step": 243 }, { "epoch": 4.0, "eval_B-Claim": { "f1-score": 0.6106346483704975, "precision": 0.6180555555555556, "recall": 0.6033898305084746, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8498583569405098, "precision": 0.7614213197969543, "recall": 0.9615384615384616, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8746478873239436, "precision": 0.8961038961038961, "recall": 0.8541953232462174, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.5964912280701754, "precision": 0.6109845608706657, "recall": 0.5826695631185131, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8631961259079903, "precision": 0.7741585233441911, "recall": 0.9753761969904241, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8933694290678411, "precision": 0.9071885770556376, "recall": 0.8799649765183475, "support": 12563.0 }, "eval_O": { "f1-score": 0.9999601132782897, "precision": 1.0, "recall": 0.9999202297383536, "support": 12536.0 }, "eval_accuracy": 0.8920369177935179, "eval_loss": 0.2815629839897156, "eval_macro avg": { "f1-score": 0.8125939698513209, "precision": 0.7954160618181286, "recall": 0.8367220830941131, "support": 32613.0 }, "eval_runtime": 4.9474, "eval_samples_per_second": 16.372, "eval_steps_per_second": 2.223, "eval_weighted avg": { "f1-score": 0.8914155863760895, "precision": 0.892730641712682, "recall": 0.8920369177935179, "support": 32613.0 }, "step": 324 }, { "epoch": 5.0, "eval_B-Claim": { "f1-score": 0.6796116504854369, "precision": 0.6501547987616099, "recall": 0.711864406779661, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8746177370030581, "precision": 0.8362573099415205, "recall": 0.9166666666666666, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8836879432624114, "precision": 0.9121522693997072, "recall": 0.8569463548830811, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6557152635181382, "precision": 0.6216742375081116, "recall": 0.6937002172338885, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8859746610357857, "precision": 0.8642671292281006, "recall": 0.9088007295941632, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8956745077255495, "precision": 0.9180177168644492, "recall": 0.874393058982727, "support": 12563.0 }, "eval_O": { "f1-score": 0.9998006141085457, "precision": 0.9996013077107089, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9003771502161715, "eval_loss": 0.2959797978401184, "eval_macro avg": { "f1-score": 0.839297482448418, "precision": 0.8288749670591724, "recall": 0.8517673477343124, "support": 32613.0 }, "eval_runtime": 4.9768, "eval_samples_per_second": 16.275, "eval_steps_per_second": 2.21, "eval_weighted avg": { "f1-score": 0.9022413468876027, "precision": 0.9051721478141229, "recall": 0.9003771502161715, "support": 32613.0 }, "step": 405 }, { "epoch": 6.0, "eval_B-Claim": { "f1-score": 0.6655518394648828, "precision": 0.6567656765676567, "recall": 0.6745762711864407, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8807339449541285, "precision": 0.8421052631578947, "recall": 0.9230769230769231, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8812849162011174, "precision": 0.8950354609929078, "recall": 0.8679504814305364, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.634203994799669, "precision": 0.6213524779990737, "recall": 0.647598358677287, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8926454445664106, "precision": 0.8607112616426758, "recall": 0.9270405836753306, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8911586596689544, "precision": 0.9041533546325878, "recall": 0.8785321977234737, "support": 12563.0 }, "eval_O": { "f1-score": 0.9995614559661922, "precision": 0.9991232964055152, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.8972802256768774, "eval_loss": 0.33718833327293396, "eval_macro avg": { "f1-score": 0.8350200365173363, "precision": 0.8256066844854731, "recall": 0.8455392593957132, "support": 32613.0 }, "eval_runtime": 4.9878, "eval_samples_per_second": 16.24, "eval_steps_per_second": 2.205, "eval_weighted avg": { "f1-score": 0.8979742337698349, "precision": 0.8990738793396261, "recall": 0.8972802256768774, "support": 32613.0 }, "step": 486 }, { "epoch": 6.17, "grad_norm": 10.722956657409668, "learning_rate": 1.7530864197530865e-05, "loss": 0.2434, "step": 500 }, { "epoch": 7.0, "eval_B-Claim": { "f1-score": 0.6608695652173914, "precision": 0.6785714285714286, "recall": 0.6440677966101694, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.875, "precision": 0.8536585365853658, "recall": 0.8974358974358975, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8905608755129958, "precision": 0.8857142857142857, "recall": 0.8954607977991746, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6306948332909138, "precision": 0.6670255720053836, "recall": 0.5981173062997828, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.890302764370338, "precision": 0.8579281183932347, "recall": 0.9252165982672138, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9037468972853708, "precision": 0.8947573724449992, "recall": 0.9129188888004458, "support": 12563.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9043326280930917, "eval_loss": 0.43479764461517334, "eval_macro avg": { "f1-score": 0.8358821336681442, "precision": 0.8339507591020997, "recall": 0.8390310407446692, "support": 32613.0 }, "eval_runtime": 4.9479, "eval_samples_per_second": 16.371, "eval_steps_per_second": 2.223, "eval_weighted avg": { "f1-score": 0.9025256555797259, "precision": 0.9014510706568353, "recall": 0.9043326280930917, "support": 32613.0 }, "step": 567 }, { "epoch": 8.0, "eval_B-Claim": { "f1-score": 0.6611295681063122, "precision": 0.6482084690553745, "recall": 0.6745762711864407, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8925081433224756, "precision": 0.9072847682119205, "recall": 0.8782051282051282, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8810511756569848, "precision": 0.885952712100139, "recall": 0.8762035763411279, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6376404494382022, "precision": 0.6189502385821404, "recall": 0.6574945691527878, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8858131487889275, "precision": 0.896358543417367, "recall": 0.8755129958960328, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8946945982823662, "precision": 0.9022258195062728, "recall": 0.8872880681365916, "support": 12563.0 }, "eval_O": { "f1-score": 0.9999202361011406, "precision": 0.9998404849258254, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.8984147425873118, "eval_loss": 0.3937377631664276, "eval_macro avg": { "f1-score": 0.8361081885280585, "precision": 0.836974433685577, "recall": 0.8356115155597299, "support": 32613.0 }, "eval_runtime": 5.0071, "eval_samples_per_second": 16.177, "eval_steps_per_second": 2.197, "eval_weighted avg": { "f1-score": 0.8994624730308206, "precision": 0.9007308115646135, "recall": 0.8984147425873118, "support": 32613.0 }, "step": 648 }, { "epoch": 9.0, "eval_B-Claim": { "f1-score": 0.6821192052980132, "precision": 0.6666666666666666, "recall": 0.6983050847457627, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8510638297872339, "precision": 0.9523809523809523, "recall": 0.7692307692307693, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8959891230455472, "precision": 0.885752688172043, "recall": 0.90646492434663, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6547967670141736, "precision": 0.636094674556213, "recall": 0.6746319092445088, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8507386653082017, "precision": 0.9636468551644547, "recall": 0.7615139078887369, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9060476867203537, "precision": 0.8987391338397682, "recall": 0.9134760805540078, "support": 12563.0 }, "eval_O": { "f1-score": 0.9999202361011406, "precision": 0.9998404849258254, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9033820868978628, "eval_loss": 0.4608707129955292, "eval_macro avg": { "f1-score": 0.8343822161820948, "precision": 0.8575887793865605, "recall": 0.8176603822872022, "support": 32613.0 }, "eval_runtime": 4.9809, "eval_samples_per_second": 16.262, "eval_steps_per_second": 2.208, "eval_weighted avg": { "f1-score": 0.9039813964448579, "precision": 0.9064685293306464, "recall": 0.9033820868978628, "support": 32613.0 }, "step": 729 }, { "epoch": 10.0, "eval_B-Claim": { "f1-score": 0.6863354037267081, "precision": 0.6332378223495702, "recall": 0.7491525423728813, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8976897689768977, "precision": 0.9251700680272109, "recall": 0.8717948717948718, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8780141843971632, "precision": 0.9062957540263543, "recall": 0.8514442916093535, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6615143369175628, "precision": 0.6171368861024034, "recall": 0.7127685252232682, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9019424292066465, "precision": 0.9264423076923077, "recall": 0.8787049703602371, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8934422897101273, "precision": 0.9130796077779625, "recall": 0.8746318554485394, "support": 12563.0 }, "eval_O": { "f1-score": 0.9999601132782897, "precision": 1.0, "recall": 0.9999202297383536, "support": 12536.0 }, "eval_accuracy": 0.9008370895041854, "eval_loss": 0.49426138401031494, "eval_macro avg": { "f1-score": 0.8455569323161993, "precision": 0.8459089208536871, "recall": 0.8483453266496436, "support": 32613.0 }, "eval_runtime": 4.962, "eval_samples_per_second": 16.324, "eval_steps_per_second": 2.217, "eval_weighted avg": { "f1-score": 0.9032978864532892, "precision": 0.9071693470023781, "recall": 0.9008370895041854, "support": 32613.0 }, "step": 810 }, { "epoch": 11.0, "eval_B-Claim": { "f1-score": 0.6654991243432573, "precision": 0.6884057971014492, "recall": 0.6440677966101694, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8888888888888888, "precision": 0.9361702127659575, "recall": 0.8461538461538461, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8911290322580646, "precision": 0.871222076215506, "recall": 0.9119669876203577, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6376628304034401, "precision": 0.6697662061636557, "recall": 0.6084962587496983, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.8884068450228971, "precision": 0.9422290388548057, "recall": 0.8404012767897857, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.904972804972805, "precision": 0.8838885937618578, "recall": 0.9270874791053092, "support": 12563.0 }, "eval_O": { "f1-score": 0.9999202361011406, "precision": 0.9998404849258254, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9055284702419281, "eval_loss": 0.5303362607955933, "eval_macro avg": { "f1-score": 0.8394971088557848, "precision": 0.8559317728270083, "recall": 0.8254533778613096, "support": 32613.0 }, "eval_runtime": 4.9749, "eval_samples_per_second": 16.282, "eval_steps_per_second": 2.211, "eval_weighted avg": { "f1-score": 0.9038459038093231, "precision": 0.9033803424876604, "recall": 0.9055284702419281, "support": 32613.0 }, "step": 891 }, { "epoch": 12.0, "eval_B-Claim": { "f1-score": 0.6374045801526718, "precision": 0.7292576419213974, "recall": 0.5661016949152542, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.9114754098360656, "precision": 0.9328859060402684, "recall": 0.8910256410256411, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8913612565445026, "precision": 0.850187265917603, "recall": 0.936726272352132, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.5983549421441516, "precision": 0.7082508250825083, "recall": 0.5179821385469466, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9206057824690226, "precision": 0.9265588914549654, "recall": 0.9147286821705426, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9017204628501827, "precision": 0.8640309285870595, "recall": 0.9428480458489215, "support": 12563.0 }, "eval_O": { "f1-score": 0.9997606319317004, "precision": 1.0, "recall": 0.9995213784301212, "support": 12536.0 }, "eval_accuracy": 0.9049765430963113, "eval_loss": 0.5691992044448853, "eval_macro avg": { "f1-score": 0.8372404379897568, "precision": 0.8587387798576859, "recall": 0.8241334076127942, "support": 32613.0 }, "eval_runtime": 4.9632, "eval_samples_per_second": 16.32, "eval_steps_per_second": 2.216, "eval_weighted avg": { "f1-score": 0.8995624631750451, "precision": 0.8995123024912762, "recall": 0.9049765430963113, "support": 32613.0 }, "step": 972 }, { "epoch": 12.35, "grad_norm": 8.504526138305664, "learning_rate": 1.506172839506173e-05, "loss": 0.0319, "step": 1000 }, { "epoch": 13.0, "eval_B-Claim": { "f1-score": 0.6779661016949152, "precision": 0.6779661016949152, "recall": 0.6779661016949152, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.9096774193548387, "precision": 0.9155844155844156, "recall": 0.9038461538461539, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8873626373626373, "precision": 0.8861454046639232, "recall": 0.8885832187070152, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.663892288861689, "precision": 0.6734541842562701, "recall": 0.6545981173062998, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9122568973315242, "precision": 0.9048900852400179, "recall": 0.9197446420428637, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9055930186433956, "precision": 0.9025855934213648, "recall": 0.9086205524158243, "support": 12563.0 }, "eval_O": { "f1-score": 0.9998404340194671, "precision": 1.0, "recall": 0.9996809189534142, "support": 12536.0 }, "eval_accuracy": 0.9095452733572502, "eval_loss": 0.5995727777481079, "eval_macro avg": { "f1-score": 0.8509412567526382, "precision": 0.8515179692658439, "recall": 0.8504342435666409, "support": 32613.0 }, "eval_runtime": 4.967, "eval_samples_per_second": 16.308, "eval_steps_per_second": 2.215, "eval_weighted avg": { "f1-score": 0.9091182110651892, "precision": 0.908741493590885, "recall": 0.9095452733572502, "support": 32613.0 }, "step": 1053 }, { "epoch": 14.0, "eval_B-Claim": { "f1-score": 0.6543438077634012, "precision": 0.7195121951219512, "recall": 0.6, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8918032786885245, "precision": 0.912751677852349, "recall": 0.8717948717948718, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.895364238410596, "precision": 0.8633461047254151, "recall": 0.9298486932599724, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.612272977435333, "precision": 0.712, "recall": 0.5370504465363264, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9033142857142857, "precision": 0.9055912007332723, "recall": 0.9010487916096671, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9070541005543873, "precision": 0.8727192466156563, "recall": 0.9442012258218578, "support": 12563.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9072455769171802, "eval_loss": 0.6337578892707825, "eval_macro avg": { "f1-score": 0.8377360983666469, "precision": 0.8551314892926634, "recall": 0.8262777184318136, "support": 32613.0 }, "eval_runtime": 4.9902, "eval_samples_per_second": 16.232, "eval_steps_per_second": 2.204, "eval_weighted avg": { "f1-score": 0.9024627717609818, "precision": 0.902034384321647, "recall": 0.9072455769171802, "support": 32613.0 }, "step": 1134 }, { "epoch": 15.0, "eval_B-Claim": { "f1-score": 0.6715867158671587, "precision": 0.7368421052631579, "recall": 0.6169491525423729, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8902077151335311, "precision": 0.8287292817679558, "recall": 0.9615384615384616, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.896551724137931, "precision": 0.8816489361702128, "recall": 0.9119669876203577, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6393111798735368, "precision": 0.7221884498480243, "recall": 0.5734974656046343, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.898728813559322, "precision": 0.8393351800554016, "recall": 0.9671682626538988, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9101119213820535, "precision": 0.892125382262997, "recall": 0.9288386531879328, "support": 12563.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9105878024100819, "eval_loss": 0.6874175667762756, "eval_macro avg": { "f1-score": 0.8437854385647904, "precision": 0.8429813336239642, "recall": 0.8514227118782368, "support": 32613.0 }, "eval_runtime": 4.9796, "eval_samples_per_second": 16.266, "eval_steps_per_second": 2.209, "eval_weighted avg": { "f1-score": 0.9069419606556991, "precision": 0.9065117879543761, "recall": 0.9105878024100819, "support": 32613.0 }, "step": 1215 }, { "epoch": 16.0, "eval_B-Claim": { "f1-score": 0.676923076923077, "precision": 0.6827586206896552, "recall": 0.6711864406779661, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.8980891719745223, "precision": 0.8924050632911392, "recall": 0.9038461538461539, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8902606310013718, "precision": 0.8878248974008208, "recall": 0.8927097661623109, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6403442166540116, "precision": 0.6730513434424049, "recall": 0.6106685976345644, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9115737631520036, "precision": 0.8953386103781882, "recall": 0.9284085727314182, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.9030989460437312, "precision": 0.8924990283715507, "recall": 0.9139536734856324, "support": 12563.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 12536.0 }, "eval_accuracy": 0.9067549750099654, "eval_loss": 0.6374069452285767, "eval_macro avg": { "f1-score": 0.8457556865355311, "precision": 0.8462682233676798, "recall": 0.8458247435054351, "support": 32613.0 }, "eval_runtime": 5.004, "eval_samples_per_second": 16.187, "eval_steps_per_second": 2.198, "eval_weighted avg": { "f1-score": 0.905181158084066, "precision": 0.9041324706485238, "recall": 0.9067549750099654, "support": 32613.0 }, "step": 1296 }, { "epoch": 17.0, "eval_B-Claim": { "f1-score": 0.7001620745542951, "precision": 0.6708074534161491, "recall": 0.7322033898305085, "support": 295.0 }, "eval_B-MajorClaim": { "f1-score": 0.910828025477707, "precision": 0.9050632911392406, "recall": 0.9166666666666666, "support": 156.0 }, "eval_B-Premise": { "f1-score": 0.8870175438596491, "precision": 0.9054441260744985, "recall": 0.8693259972489684, "support": 727.0 }, "eval_I-Claim": { "f1-score": 0.6722241519045965, "precision": 0.6459724076546507, "recall": 0.7006999758629012, "support": 4143.0 }, "eval_I-MajorClaim": { "f1-score": 0.9159550561797752, "precision": 0.9029685423128047, "recall": 0.9293205654354765, "support": 2193.0 }, "eval_I-Premise": { "f1-score": 0.8987500505642975, "precision": 0.9137193617371279, "recall": 0.884263312902969, "support": 12563.0 }, "eval_O": { "f1-score": 0.999600989545926, "precision": 1.0, "recall": 0.9992022973835354, "support": 12536.0 }, "eval_accuracy": 0.9066016619139607, "eval_loss": 0.5605483055114746, "eval_macro avg": { "f1-score": 0.8549339845837495, "precision": 0.8491393117620675, "recall": 0.8616688864758607, "support": 32613.0 }, "eval_runtime": 5.0022, "eval_samples_per_second": 16.193, "eval_steps_per_second": 2.199, "eval_weighted avg": { "f1-score": 0.9078958614075368, "precision": 0.9097251400220888, "recall": 0.9066016619139607, "support": 32613.0 }, "step": 1377 } ], "logging_steps": 500, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "total_flos": 2436852375783000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }