{ "best_global_step": 471, "best_metric": 0.7473118279569892, "best_model_checkpoint": "/linkhome/rech/genini01/udd26kf/scratch/weborganizer/models/runs/answerdotai--ModernBERT-base_TopicAnnotations-Llama-3.1-8B_bsz512_lr1e-4_epochs5_warmup0.1_url1_TopicAnnotations-Llama-3.1-405B-FP8_bsz512_lr1e-4_epochs5_warmup0.1_url1/checkpoint-471", "epoch": 4.9728, "eval_steps": 500, "global_step": 780, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.64, "grad_norm": 8.25, "learning_rate": 9.686609686609687e-05, "loss": 2.1544, "num_input_tokens_seen": 249204064, "step": 100 }, { "epoch": 1.0, "eval_validation.parquet_accuracy": 0.8451, "eval_validation.parquet_accuracy__0": 0.9019607843137255, "eval_validation.parquet_accuracy__1": 0.7975460122699386, "eval_validation.parquet_accuracy__10": 0.9105882352941177, "eval_validation.parquet_accuracy__11": 0.8687258687258688, "eval_validation.parquet_accuracy__12": 0.8734622144112478, "eval_validation.parquet_accuracy__13": 0.6827956989247311, "eval_validation.parquet_accuracy__14": 0.8229461756373938, "eval_validation.parquet_accuracy__15": 0.8427947598253275, "eval_validation.parquet_accuracy__16": 0.8194842406876791, "eval_validation.parquet_accuracy__17": 0.8260135135135135, "eval_validation.parquet_accuracy__18": 0.8732876712328768, "eval_validation.parquet_accuracy__19": 0.861764705882353, "eval_validation.parquet_accuracy__2": 0.8159203980099502, "eval_validation.parquet_accuracy__20": 0.7183098591549296, "eval_validation.parquet_accuracy__21": 0.8975409836065574, "eval_validation.parquet_accuracy__22": 0.7981651376146789, "eval_validation.parquet_accuracy__23": 0.863481228668942, "eval_validation.parquet_accuracy__3": 0.9017857142857143, "eval_validation.parquet_accuracy__4": 0.8697829716193656, "eval_validation.parquet_accuracy__5": 0.8246268656716418, "eval_validation.parquet_accuracy__6": 0.8907168037602821, "eval_validation.parquet_accuracy__7": 0.7784946236559139, "eval_validation.parquet_accuracy__8": 0.8932584269662921, "eval_validation.parquet_accuracy__9": 0.8230403800475059, "eval_validation.parquet_accuracy_conf50": 0.8559460563955864, "eval_validation.parquet_accuracy_conf50__0": 0.9108910891089109, "eval_validation.parquet_accuracy_conf50__1": 0.8087774294670846, "eval_validation.parquet_accuracy_conf50__10": 0.9148936170212766, "eval_validation.parquet_accuracy_conf50__11": 0.8745098039215686, "eval_validation.parquet_accuracy_conf50__12": 0.8857142857142857, "eval_validation.parquet_accuracy_conf50__13": 0.7085714285714285, "eval_validation.parquet_accuracy_conf50__14": 0.8338150289017341, "eval_validation.parquet_accuracy_conf50__15": 0.8609865470852018, "eval_validation.parquet_accuracy_conf50__16": 0.8240469208211144, "eval_validation.parquet_accuracy_conf50__17": 0.8319039451114922, "eval_validation.parquet_accuracy_conf50__18": 0.8788927335640139, "eval_validation.parquet_accuracy_conf50__19": 0.8761329305135952, "eval_validation.parquet_accuracy_conf50__2": 0.8350515463917526, "eval_validation.parquet_accuracy_conf50__20": 0.7323529411764705, "eval_validation.parquet_accuracy_conf50__21": 0.9009628610729024, "eval_validation.parquet_accuracy_conf50__22": 0.8113207547169812, "eval_validation.parquet_accuracy_conf50__23": 0.865979381443299, "eval_validation.parquet_accuracy_conf50__3": 0.9195046439628483, "eval_validation.parquet_accuracy_conf50__4": 0.8807495741056218, "eval_validation.parquet_accuracy_conf50__5": 0.8358778625954199, "eval_validation.parquet_accuracy_conf50__6": 0.9020310633213859, "eval_validation.parquet_accuracy_conf50__7": 0.7986425339366516, "eval_validation.parquet_accuracy_conf50__8": 0.8951841359773371, "eval_validation.parquet_accuracy_conf50__9": 0.8345498783454988, "eval_validation.parquet_accuracy_conf75": 0.9065606361829026, "eval_validation.parquet_accuracy_conf75__0": 0.967391304347826, "eval_validation.parquet_accuracy_conf75__1": 0.8828125, "eval_validation.parquet_accuracy_conf75__10": 0.9493670886075949, "eval_validation.parquet_accuracy_conf75__11": 0.9037656903765691, "eval_validation.parquet_accuracy_conf75__12": 0.9176470588235294, "eval_validation.parquet_accuracy_conf75__13": 0.7906976744186046, "eval_validation.parquet_accuracy_conf75__14": 0.9027303754266212, "eval_validation.parquet_accuracy_conf75__15": 0.9281767955801105, "eval_validation.parquet_accuracy_conf75__16": 0.886986301369863, "eval_validation.parquet_accuracy_conf75__17": 0.8814229249011858, "eval_validation.parquet_accuracy_conf75__18": 0.9176029962546817, "eval_validation.parquet_accuracy_conf75__19": 0.9185185185185185, "eval_validation.parquet_accuracy_conf75__2": 0.9047619047619048, "eval_validation.parquet_accuracy_conf75__20": 0.8022813688212928, "eval_validation.parquet_accuracy_conf75__21": 0.9327485380116959, "eval_validation.parquet_accuracy_conf75__22": 0.8741007194244604, "eval_validation.parquet_accuracy_conf75__23": 0.9246031746031746, "eval_validation.parquet_accuracy_conf75__3": 0.9473684210526315, "eval_validation.parquet_accuracy_conf75__4": 0.9242718446601942, "eval_validation.parquet_accuracy_conf75__5": 0.8879310344827587, "eval_validation.parquet_accuracy_conf75__6": 0.9322033898305084, "eval_validation.parquet_accuracy_conf75__7": 0.848, "eval_validation.parquet_accuracy_conf75__8": 0.9339622641509434, "eval_validation.parquet_accuracy_conf75__9": 0.9001447178002895, "eval_validation.parquet_accuracy_label_average": 0.8398538864075228, "eval_validation.parquet_accuracy_label_average_conf50": 0.8508892890353281, "eval_validation.parquet_accuracy_label_average_conf75": 0.9024790252593734, "eval_validation.parquet_accuracy_label_min": 0.6827956989247311, "eval_validation.parquet_accuracy_label_min_conf50": 0.7085714285714285, "eval_validation.parquet_accuracy_label_min_conf75": 0.7906976744186046, "eval_validation.parquet_loss": 0.5004527568817139, "eval_validation.parquet_proportion_conf50": 0.9788, "eval_validation.parquet_proportion_conf75": 0.8551, "eval_validation.parquet_runtime": 10.52, "eval_validation.parquet_samples_per_second": 950.571, "eval_validation.parquet_steps_per_second": 29.753, "num_input_tokens_seen": 390215936, "step": 157 }, { "epoch": 1.2752, "grad_norm": 9.875, "learning_rate": 8.262108262108262e-05, "loss": 1.8475, "num_input_tokens_seen": 499147424, "step": 200 }, { "epoch": 1.9152, "grad_norm": 7.53125, "learning_rate": 6.837606837606838e-05, "loss": 1.7317, "num_input_tokens_seen": 751160992, "step": 300 }, { "epoch": 2.0, "eval_validation.parquet_accuracy": 0.8526, "eval_validation.parquet_accuracy__0": 0.8725490196078431, "eval_validation.parquet_accuracy__1": 0.8128834355828221, "eval_validation.parquet_accuracy__10": 0.9176470588235294, "eval_validation.parquet_accuracy__11": 0.9073359073359073, "eval_validation.parquet_accuracy__12": 0.9138840070298769, "eval_validation.parquet_accuracy__13": 0.7419354838709677, "eval_validation.parquet_accuracy__14": 0.7818696883852692, "eval_validation.parquet_accuracy__15": 0.8427947598253275, "eval_validation.parquet_accuracy__16": 0.8481375358166189, "eval_validation.parquet_accuracy__17": 0.8733108108108109, "eval_validation.parquet_accuracy__18": 0.8732876712328768, "eval_validation.parquet_accuracy__19": 0.8205882352941176, "eval_validation.parquet_accuracy__2": 0.7860696517412935, "eval_validation.parquet_accuracy__20": 0.7830985915492957, "eval_validation.parquet_accuracy__21": 0.9344262295081968, "eval_validation.parquet_accuracy__22": 0.8562691131498471, "eval_validation.parquet_accuracy__23": 0.9078498293515358, "eval_validation.parquet_accuracy__3": 0.8541666666666666, "eval_validation.parquet_accuracy__4": 0.8414023372287145, "eval_validation.parquet_accuracy__5": 0.8208955223880597, "eval_validation.parquet_accuracy__6": 0.8883666274970623, "eval_validation.parquet_accuracy__7": 0.7784946236559139, "eval_validation.parquet_accuracy__8": 0.8960674157303371, "eval_validation.parquet_accuracy__9": 0.8111638954869359, "eval_validation.parquet_accuracy_conf50": 0.8627911728647323, "eval_validation.parquet_accuracy_conf50__0": 0.8811881188118812, "eval_validation.parquet_accuracy_conf50__1": 0.8244514106583072, "eval_validation.parquet_accuracy_conf50__10": 0.9219858156028369, "eval_validation.parquet_accuracy_conf50__11": 0.9137254901960784, "eval_validation.parquet_accuracy_conf50__12": 0.9214285714285714, "eval_validation.parquet_accuracy_conf50__13": 0.7771428571428571, "eval_validation.parquet_accuracy_conf50__14": 0.7947976878612717, "eval_validation.parquet_accuracy_conf50__15": 0.8565022421524664, "eval_validation.parquet_accuracy_conf50__16": 0.8533724340175953, "eval_validation.parquet_accuracy_conf50__17": 0.8782161234991424, "eval_validation.parquet_accuracy_conf50__18": 0.8788927335640139, "eval_validation.parquet_accuracy_conf50__19": 0.8368580060422961, "eval_validation.parquet_accuracy_conf50__2": 0.8041237113402062, "eval_validation.parquet_accuracy_conf50__20": 0.8, "eval_validation.parquet_accuracy_conf50__21": 0.936726272352132, "eval_validation.parquet_accuracy_conf50__22": 0.8679245283018868, "eval_validation.parquet_accuracy_conf50__23": 0.9072164948453608, "eval_validation.parquet_accuracy_conf50__3": 0.8761609907120743, "eval_validation.parquet_accuracy_conf50__4": 0.8534923339011925, "eval_validation.parquet_accuracy_conf50__5": 0.8320610687022901, "eval_validation.parquet_accuracy_conf50__6": 0.8984468339307049, "eval_validation.parquet_accuracy_conf50__7": 0.7873303167420814, "eval_validation.parquet_accuracy_conf50__8": 0.8980169971671388, "eval_validation.parquet_accuracy_conf50__9": 0.8211678832116789, "eval_validation.parquet_accuracy_conf75": 0.9124079055081277, "eval_validation.parquet_accuracy_conf75__0": 0.9347826086956522, "eval_validation.parquet_accuracy_conf75__1": 0.8984375, "eval_validation.parquet_accuracy_conf75__10": 0.9620253164556962, "eval_validation.parquet_accuracy_conf75__11": 0.9372384937238494, "eval_validation.parquet_accuracy_conf75__12": 0.9529411764705882, "eval_validation.parquet_accuracy_conf75__13": 0.8294573643410853, "eval_validation.parquet_accuracy_conf75__14": 0.8686006825938567, "eval_validation.parquet_accuracy_conf75__15": 0.9226519337016574, "eval_validation.parquet_accuracy_conf75__16": 0.9143835616438356, "eval_validation.parquet_accuracy_conf75__17": 0.9209486166007905, "eval_validation.parquet_accuracy_conf75__18": 0.9250936329588015, "eval_validation.parquet_accuracy_conf75__19": 0.8962962962962963, "eval_validation.parquet_accuracy_conf75__2": 0.8809523809523809, "eval_validation.parquet_accuracy_conf75__20": 0.8593155893536122, "eval_validation.parquet_accuracy_conf75__21": 0.9576023391812866, "eval_validation.parquet_accuracy_conf75__22": 0.8992805755395683, "eval_validation.parquet_accuracy_conf75__23": 0.9603174603174603, "eval_validation.parquet_accuracy_conf75__3": 0.9192982456140351, "eval_validation.parquet_accuracy_conf75__4": 0.8990291262135922, "eval_validation.parquet_accuracy_conf75__5": 0.8879310344827587, "eval_validation.parquet_accuracy_conf75__6": 0.9282920469361148, "eval_validation.parquet_accuracy_conf75__7": 0.84, "eval_validation.parquet_accuracy_conf75__8": 0.9339622641509434, "eval_validation.parquet_accuracy_conf75__9": 0.8900144717800289, "eval_validation.parquet_accuracy_label_average": 0.8485205882320762, "eval_validation.parquet_accuracy_label_average_conf50": 0.8592178717576693, "eval_validation.parquet_accuracy_label_average_conf75": 0.909118863250162, "eval_validation.parquet_accuracy_label_min": 0.7419354838709677, "eval_validation.parquet_accuracy_label_min_conf50": 0.7771428571428571, "eval_validation.parquet_accuracy_label_min_conf75": 0.8294573643410853, "eval_validation.parquet_loss": 0.4816047251224518, "eval_validation.parquet_proportion_conf50": 0.9788, "eval_validation.parquet_proportion_conf75": 0.8551, "eval_validation.parquet_runtime": 8.307, "eval_validation.parquet_samples_per_second": 1203.799, "eval_validation.parquet_steps_per_second": 37.679, "num_input_tokens_seen": 783399104, "step": 314 }, { "epoch": 2.5504, "grad_norm": 7.59375, "learning_rate": 5.413105413105414e-05, "loss": 1.5837, "num_input_tokens_seen": 999700736, "step": 400 }, { "epoch": 3.0, "eval_validation.parquet_accuracy": 0.8558, "eval_validation.parquet_accuracy__0": 0.8627450980392157, "eval_validation.parquet_accuracy__1": 0.7607361963190185, "eval_validation.parquet_accuracy__10": 0.9035294117647059, "eval_validation.parquet_accuracy__11": 0.8764478764478765, "eval_validation.parquet_accuracy__12": 0.8980667838312829, "eval_validation.parquet_accuracy__13": 0.7473118279569892, "eval_validation.parquet_accuracy__14": 0.839943342776204, "eval_validation.parquet_accuracy__15": 0.8427947598253275, "eval_validation.parquet_accuracy__16": 0.830945558739255, "eval_validation.parquet_accuracy__17": 0.839527027027027, "eval_validation.parquet_accuracy__18": 0.8801369863013698, "eval_validation.parquet_accuracy__19": 0.8117647058823529, "eval_validation.parquet_accuracy__2": 0.7860696517412935, "eval_validation.parquet_accuracy__20": 0.7633802816901408, "eval_validation.parquet_accuracy__21": 0.9289617486338798, "eval_validation.parquet_accuracy__22": 0.8562691131498471, "eval_validation.parquet_accuracy__23": 0.8805460750853242, "eval_validation.parquet_accuracy__3": 0.8660714285714286, "eval_validation.parquet_accuracy__4": 0.8530884808013356, "eval_validation.parquet_accuracy__5": 0.8694029850746269, "eval_validation.parquet_accuracy__6": 0.900117508813161, "eval_validation.parquet_accuracy__7": 0.7741935483870968, "eval_validation.parquet_accuracy__8": 0.8904494382022472, "eval_validation.parquet_accuracy__9": 0.8669833729216152, "eval_validation.parquet_accuracy_conf50": 0.8663669799754802, "eval_validation.parquet_accuracy_conf50__0": 0.8712871287128713, "eval_validation.parquet_accuracy_conf50__1": 0.7711598746081505, "eval_validation.parquet_accuracy_conf50__10": 0.9078014184397163, "eval_validation.parquet_accuracy_conf50__11": 0.8823529411764706, "eval_validation.parquet_accuracy_conf50__12": 0.9089285714285714, "eval_validation.parquet_accuracy_conf50__13": 0.7828571428571428, "eval_validation.parquet_accuracy_conf50__14": 0.8511560693641619, "eval_validation.parquet_accuracy_conf50__15": 0.8565022421524664, "eval_validation.parquet_accuracy_conf50__16": 0.8357771260997068, "eval_validation.parquet_accuracy_conf50__17": 0.8456260720411664, "eval_validation.parquet_accuracy_conf50__18": 0.8858131487889274, "eval_validation.parquet_accuracy_conf50__19": 0.8277945619335347, "eval_validation.parquet_accuracy_conf50__2": 0.8041237113402062, "eval_validation.parquet_accuracy_conf50__20": 0.7794117647058824, "eval_validation.parquet_accuracy_conf50__21": 0.9312242090784044, "eval_validation.parquet_accuracy_conf50__22": 0.8710691823899371, "eval_validation.parquet_accuracy_conf50__23": 0.8797250859106529, "eval_validation.parquet_accuracy_conf50__3": 0.8885448916408669, "eval_validation.parquet_accuracy_conf50__4": 0.8637137989778535, "eval_validation.parquet_accuracy_conf50__5": 0.8816793893129771, "eval_validation.parquet_accuracy_conf50__6": 0.9115890083632019, "eval_validation.parquet_accuracy_conf50__7": 0.7850678733031674, "eval_validation.parquet_accuracy_conf50__8": 0.8923512747875354, "eval_validation.parquet_accuracy_conf50__9": 0.878345498783455, "eval_validation.parquet_accuracy_conf75": 0.9145129224652088, "eval_validation.parquet_accuracy_conf75__0": 0.9239130434782609, "eval_validation.parquet_accuracy_conf75__1": 0.85546875, "eval_validation.parquet_accuracy_conf75__10": 0.9493670886075949, "eval_validation.parquet_accuracy_conf75__11": 0.9121338912133892, "eval_validation.parquet_accuracy_conf75__12": 0.9450980392156862, "eval_validation.parquet_accuracy_conf75__13": 0.8294573643410853, "eval_validation.parquet_accuracy_conf75__14": 0.9129692832764505, "eval_validation.parquet_accuracy_conf75__15": 0.9392265193370166, "eval_validation.parquet_accuracy_conf75__16": 0.8972602739726028, "eval_validation.parquet_accuracy_conf75__17": 0.8893280632411067, "eval_validation.parquet_accuracy_conf75__18": 0.9288389513108615, "eval_validation.parquet_accuracy_conf75__19": 0.8925925925925926, "eval_validation.parquet_accuracy_conf75__2": 0.8809523809523809, "eval_validation.parquet_accuracy_conf75__20": 0.8479087452471483, "eval_validation.parquet_accuracy_conf75__21": 0.9502923976608187, "eval_validation.parquet_accuracy_conf75__22": 0.9064748201438849, "eval_validation.parquet_accuracy_conf75__23": 0.9404761904761905, "eval_validation.parquet_accuracy_conf75__3": 0.9228070175438596, "eval_validation.parquet_accuracy_conf75__4": 0.9067961165048544, "eval_validation.parquet_accuracy_conf75__5": 0.9224137931034483, "eval_validation.parquet_accuracy_conf75__6": 0.9374185136897001, "eval_validation.parquet_accuracy_conf75__7": 0.8453333333333334, "eval_validation.parquet_accuracy_conf75__8": 0.9276729559748428, "eval_validation.parquet_accuracy_conf75__9": 0.9305354558610709, "eval_validation.parquet_accuracy_label_average": 0.8470618003326092, "eval_validation.parquet_accuracy_label_average_conf50": 0.8580792494248763, "eval_validation.parquet_accuracy_label_average_conf75": 0.9081139825449239, "eval_validation.parquet_accuracy_label_min": 0.7473118279569892, "eval_validation.parquet_accuracy_label_min_conf50": 0.7711598746081505, "eval_validation.parquet_accuracy_label_min_conf75": 0.8294573643410853, "eval_validation.parquet_loss": 0.4807276427745819, "eval_validation.parquet_proportion_conf50": 0.9788, "eval_validation.parquet_proportion_conf75": 0.8551, "eval_validation.parquet_runtime": 8.2886, "eval_validation.parquet_samples_per_second": 1206.475, "eval_validation.parquet_steps_per_second": 37.763, "num_input_tokens_seen": 1176307328, "step": 471 }, { "epoch": 3.1856, "grad_norm": 6.53125, "learning_rate": 3.988603988603989e-05, "loss": 1.5392, "num_input_tokens_seen": 1250925472, "step": 500 }, { "epoch": 3.8256, "grad_norm": 7.0625, "learning_rate": 2.564102564102564e-05, "loss": 1.4928, "num_input_tokens_seen": 1499507040, "step": 600 }, { "epoch": 4.0, "eval_validation.parquet_accuracy": 0.8567, "eval_validation.parquet_accuracy__0": 0.8725490196078431, "eval_validation.parquet_accuracy__1": 0.8006134969325154, "eval_validation.parquet_accuracy__10": 0.9105882352941177, "eval_validation.parquet_accuracy__11": 0.888030888030888, "eval_validation.parquet_accuracy__12": 0.9086115992970123, "eval_validation.parquet_accuracy__13": 0.7419354838709677, "eval_validation.parquet_accuracy__14": 0.8271954674220963, "eval_validation.parquet_accuracy__15": 0.851528384279476, "eval_validation.parquet_accuracy__16": 0.8510028653295129, "eval_validation.parquet_accuracy__17": 0.8817567567567568, "eval_validation.parquet_accuracy__18": 0.8664383561643836, "eval_validation.parquet_accuracy__19": 0.8088235294117647, "eval_validation.parquet_accuracy__2": 0.8059701492537313, "eval_validation.parquet_accuracy__20": 0.7492957746478873, "eval_validation.parquet_accuracy__21": 0.924863387978142, "eval_validation.parquet_accuracy__22": 0.8379204892966361, "eval_validation.parquet_accuracy__23": 0.863481228668942, "eval_validation.parquet_accuracy__3": 0.8779761904761905, "eval_validation.parquet_accuracy__4": 0.8464106844741235, "eval_validation.parquet_accuracy__5": 0.8731343283582089, "eval_validation.parquet_accuracy__6": 0.881316098707403, "eval_validation.parquet_accuracy__7": 0.810752688172043, "eval_validation.parquet_accuracy__8": 0.8904494382022472, "eval_validation.parquet_accuracy__9": 0.8396674584323041, "eval_validation.parquet_accuracy_conf50": 0.8674908050674295, "eval_validation.parquet_accuracy_conf50__0": 0.8811881188118812, "eval_validation.parquet_accuracy_conf50__1": 0.8119122257053292, "eval_validation.parquet_accuracy_conf50__10": 0.9148936170212766, "eval_validation.parquet_accuracy_conf50__11": 0.8980392156862745, "eval_validation.parquet_accuracy_conf50__12": 0.9178571428571428, "eval_validation.parquet_accuracy_conf50__13": 0.7771428571428571, "eval_validation.parquet_accuracy_conf50__14": 0.838150289017341, "eval_validation.parquet_accuracy_conf50__15": 0.8654708520179372, "eval_validation.parquet_accuracy_conf50__16": 0.8563049853372434, "eval_validation.parquet_accuracy_conf50__17": 0.8867924528301887, "eval_validation.parquet_accuracy_conf50__18": 0.8719723183391004, "eval_validation.parquet_accuracy_conf50__19": 0.824773413897281, "eval_validation.parquet_accuracy_conf50__2": 0.8247422680412371, "eval_validation.parquet_accuracy_conf50__20": 0.7676470588235295, "eval_validation.parquet_accuracy_conf50__21": 0.9270976616231087, "eval_validation.parquet_accuracy_conf50__22": 0.8522012578616353, "eval_validation.parquet_accuracy_conf50__23": 0.8625429553264605, "eval_validation.parquet_accuracy_conf50__3": 0.9009287925696594, "eval_validation.parquet_accuracy_conf50__4": 0.858603066439523, "eval_validation.parquet_accuracy_conf50__5": 0.8854961832061069, "eval_validation.parquet_accuracy_conf50__6": 0.8936678614097969, "eval_validation.parquet_accuracy_conf50__7": 0.8235294117647058, "eval_validation.parquet_accuracy_conf50__8": 0.8923512747875354, "eval_validation.parquet_accuracy_conf50__9": 0.8503649635036497, "eval_validation.parquet_accuracy_conf75": 0.9156823763302537, "eval_validation.parquet_accuracy_conf75__0": 0.9347826086956522, "eval_validation.parquet_accuracy_conf75__1": 0.88671875, "eval_validation.parquet_accuracy_conf75__10": 0.9544303797468354, "eval_validation.parquet_accuracy_conf75__11": 0.9288702928870293, "eval_validation.parquet_accuracy_conf75__12": 0.9509803921568627, "eval_validation.parquet_accuracy_conf75__13": 0.8217054263565892, "eval_validation.parquet_accuracy_conf75__14": 0.9027303754266212, "eval_validation.parquet_accuracy_conf75__15": 0.9392265193370166, "eval_validation.parquet_accuracy_conf75__16": 0.910958904109589, "eval_validation.parquet_accuracy_conf75__17": 0.9308300395256917, "eval_validation.parquet_accuracy_conf75__18": 0.9138576779026217, "eval_validation.parquet_accuracy_conf75__19": 0.8851851851851852, "eval_validation.parquet_accuracy_conf75__2": 0.9047619047619048, "eval_validation.parquet_accuracy_conf75__20": 0.8365019011406845, "eval_validation.parquet_accuracy_conf75__21": 0.9488304093567251, "eval_validation.parquet_accuracy_conf75__22": 0.89568345323741, "eval_validation.parquet_accuracy_conf75__23": 0.9365079365079365, "eval_validation.parquet_accuracy_conf75__3": 0.9333333333333333, "eval_validation.parquet_accuracy_conf75__4": 0.8932038834951457, "eval_validation.parquet_accuracy_conf75__5": 0.9224137931034483, "eval_validation.parquet_accuracy_conf75__6": 0.9230769230769231, "eval_validation.parquet_accuracy_conf75__7": 0.8746666666666667, "eval_validation.parquet_accuracy_conf75__8": 0.9339622641509434, "eval_validation.parquet_accuracy_conf75__9": 0.91027496382055, "eval_validation.parquet_accuracy_label_average": 0.8504296666277162, "eval_validation.parquet_accuracy_label_average_conf50": 0.8618195935008668, "eval_validation.parquet_accuracy_label_average_conf75": 0.9113955826658905, "eval_validation.parquet_accuracy_label_min": 0.7419354838709677, "eval_validation.parquet_accuracy_label_min_conf50": 0.7676470588235295, "eval_validation.parquet_accuracy_label_min_conf75": 0.8217054263565892, "eval_validation.parquet_loss": 0.47853514552116394, "eval_validation.parquet_proportion_conf50": 0.9788, "eval_validation.parquet_proportion_conf75": 0.8551, "eval_validation.parquet_runtime": 8.3896, "eval_validation.parquet_samples_per_second": 1191.949, "eval_validation.parquet_steps_per_second": 37.308, "num_input_tokens_seen": 1566401088, "step": 628 }, { "epoch": 4.4608, "grad_norm": 7.5625, "learning_rate": 1.1396011396011397e-05, "loss": 1.4653, "num_input_tokens_seen": 1745927840, "step": 700 }, { "epoch": 4.9728, "eval_validation.parquet_accuracy": 0.8571, "eval_validation.parquet_accuracy__0": 0.8725490196078431, "eval_validation.parquet_accuracy__1": 0.7914110429447853, "eval_validation.parquet_accuracy__10": 0.9105882352941177, "eval_validation.parquet_accuracy__11": 0.8918918918918919, "eval_validation.parquet_accuracy__12": 0.9033391915641477, "eval_validation.parquet_accuracy__13": 0.7419354838709677, "eval_validation.parquet_accuracy__14": 0.8314447592067988, "eval_validation.parquet_accuracy__15": 0.8558951965065502, "eval_validation.parquet_accuracy__16": 0.8481375358166189, "eval_validation.parquet_accuracy__17": 0.875, "eval_validation.parquet_accuracy__18": 0.8595890410958904, "eval_validation.parquet_accuracy__19": 0.8117647058823529, "eval_validation.parquet_accuracy__2": 0.8109452736318408, "eval_validation.parquet_accuracy__20": 0.7436619718309859, "eval_validation.parquet_accuracy__21": 0.9262295081967213, "eval_validation.parquet_accuracy__22": 0.8440366972477065, "eval_validation.parquet_accuracy__23": 0.863481228668942, "eval_validation.parquet_accuracy__3": 0.8809523809523809, "eval_validation.parquet_accuracy__4": 0.8497495826377296, "eval_validation.parquet_accuracy__5": 0.8731343283582089, "eval_validation.parquet_accuracy__6": 0.8883666274970623, "eval_validation.parquet_accuracy__7": 0.7956989247311828, "eval_validation.parquet_accuracy__8": 0.9044943820224719, "eval_validation.parquet_accuracy__9": 0.8420427553444181, "eval_validation.parquet_accuracy_conf50": 0.8678994687372292, "eval_validation.parquet_accuracy_conf50__0": 0.8811881188118812, "eval_validation.parquet_accuracy_conf50__1": 0.8025078369905956, "eval_validation.parquet_accuracy_conf50__10": 0.9148936170212766, "eval_validation.parquet_accuracy_conf50__11": 0.8980392156862745, "eval_validation.parquet_accuracy_conf50__12": 0.9125, "eval_validation.parquet_accuracy_conf50__13": 0.7771428571428571, "eval_validation.parquet_accuracy_conf50__14": 0.8424855491329479, "eval_validation.parquet_accuracy_conf50__15": 0.8699551569506726, "eval_validation.parquet_accuracy_conf50__16": 0.8533724340175953, "eval_validation.parquet_accuracy_conf50__17": 0.8799313893653516, "eval_validation.parquet_accuracy_conf50__18": 0.8650519031141869, "eval_validation.parquet_accuracy_conf50__19": 0.8277945619335347, "eval_validation.parquet_accuracy_conf50__2": 0.8298969072164949, "eval_validation.parquet_accuracy_conf50__20": 0.7647058823529411, "eval_validation.parquet_accuracy_conf50__21": 0.9284731774415406, "eval_validation.parquet_accuracy_conf50__22": 0.8584905660377359, "eval_validation.parquet_accuracy_conf50__23": 0.8625429553264605, "eval_validation.parquet_accuracy_conf50__3": 0.9040247678018576, "eval_validation.parquet_accuracy_conf50__4": 0.8620102214650767, "eval_validation.parquet_accuracy_conf50__5": 0.8854961832061069, "eval_validation.parquet_accuracy_conf50__6": 0.9008363201911589, "eval_validation.parquet_accuracy_conf50__7": 0.8076923076923077, "eval_validation.parquet_accuracy_conf50__8": 0.9065155807365439, "eval_validation.parquet_accuracy_conf50__9": 0.8527980535279805, "eval_validation.parquet_accuracy_conf75": 0.9163840486492808, "eval_validation.parquet_accuracy_conf75__0": 0.9347826086956522, "eval_validation.parquet_accuracy_conf75__1": 0.87890625, "eval_validation.parquet_accuracy_conf75__10": 0.9518987341772152, "eval_validation.parquet_accuracy_conf75__11": 0.9288702928870293, "eval_validation.parquet_accuracy_conf75__12": 0.9470588235294117, "eval_validation.parquet_accuracy_conf75__13": 0.8217054263565892, "eval_validation.parquet_accuracy_conf75__14": 0.9061433447098977, "eval_validation.parquet_accuracy_conf75__15": 0.9447513812154696, "eval_validation.parquet_accuracy_conf75__16": 0.910958904109589, "eval_validation.parquet_accuracy_conf75__17": 0.924901185770751, "eval_validation.parquet_accuracy_conf75__18": 0.9101123595505618, "eval_validation.parquet_accuracy_conf75__19": 0.8888888888888888, "eval_validation.parquet_accuracy_conf75__2": 0.9047619047619048, "eval_validation.parquet_accuracy_conf75__20": 0.8326996197718631, "eval_validation.parquet_accuracy_conf75__21": 0.9502923976608187, "eval_validation.parquet_accuracy_conf75__22": 0.9028776978417267, "eval_validation.parquet_accuracy_conf75__23": 0.9325396825396826, "eval_validation.parquet_accuracy_conf75__3": 0.9368421052631579, "eval_validation.parquet_accuracy_conf75__4": 0.9009708737864077, "eval_validation.parquet_accuracy_conf75__5": 0.9224137931034483, "eval_validation.parquet_accuracy_conf75__6": 0.9308996088657105, "eval_validation.parquet_accuracy_conf75__7": 0.864, "eval_validation.parquet_accuracy_conf75__8": 0.940251572327044, "eval_validation.parquet_accuracy_conf75__9": 0.9117221418234442, "eval_validation.parquet_accuracy_label_average": 0.8506808235334006, "eval_validation.parquet_accuracy_label_average_conf50": 0.8620143984651407, "eval_validation.parquet_accuracy_label_average_conf75": 0.9116353999015111, "eval_validation.parquet_accuracy_label_min": 0.7419354838709677, "eval_validation.parquet_accuracy_label_min_conf50": 0.7647058823529411, "eval_validation.parquet_accuracy_label_min_conf75": 0.8217054263565892, "eval_validation.parquet_loss": 0.47900858521461487, "eval_validation.parquet_proportion_conf50": 0.9788, "eval_validation.parquet_proportion_conf75": 0.8551, "eval_validation.parquet_runtime": 8.446, "eval_validation.parquet_samples_per_second": 1183.995, "eval_validation.parquet_steps_per_second": 37.059, "num_input_tokens_seen": 1949274656, "step": 780 }, { "epoch": 4.9728, "num_input_tokens_seen": 1949274656, "step": 780, "total_flos": 1.297523316772307e+18, "train_loss": 1.6634563641670423, "train_runtime": 573.9155, "train_samples_per_second": 696.967, "train_steps_per_second": 1.359 } ], "logging_steps": 100, "max_steps": 780, "num_input_tokens_seen": 1949274656, "num_train_epochs": 5, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.297523316772307e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }