Datasets:
GPUMODE
/
backendbench_tests

Dataset card Data Studio Files
xet
Community
6
operator name
stringclasses
180 values
used in model
stringclasses
155 values
args
stringlengths
19
5.24k
aten._softmax.default
HuggingFace/GPTNeoForCausalLM
((T([1, 16, 128, 128], f32), -1, False), {})
aten._softmax.default
HuggingFace/GPTNeoForSequenceClassification
((T([1, 16, 128, 128], f32), -1, False), {})
aten._softmax.default
HuggingFace/DebertaV2ForMaskedLM
((T([1, 24, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/DebertaV2ForQuestionAnswering
((T([1, 24, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/ElectraForCausalLM
((T([1, 4, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/YituTechConvBert
((T([1, 6, 512, 512], f16), -1, False), {})
aten._softmax.default
TIMM/eca_halonext26ts
((T([1024, 1, 64, 144], f16), -1, False), {})
aten._softmax.default
HuggingFace/BlenderbotSmallForCausalLM
((T([1024, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/BlenderbotSmallForConditionalGeneration
((T([1024, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/eca_halonext26ts
((T([1024, 4, 16, 144], f16), -1, False), {})
aten._softmax.default
TIMM/eca_halonext26ts
((T([1024, 4, 64, 144], f16), -1, False), {})
aten._softmax.default
TIMM/swin_base_patch4_window7_224
((T([1024, 8, 49, 49], f16), -1, False), {})
aten._softmax.default
TIMM/tnt_s_patch16_224
((T([12544, 4, 16, 16], f16), -1, False), {})
aten._softmax.default
TIMM/levit_128
((T([128, 12, 16, 16], f16), -1, False), {})
aten._softmax.default
HuggingFace/MBartForConditionalGeneration
((T([128, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/PegasusForCausalLM
((T([128, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/TrOCRForCausalLM
((T([128, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/levit_128
((T([128, 16, 16, 49], f16), -1, False), {})
aten._softmax.default
TIMM/levit_128
((T([128, 4, 196, 196], f16), -1, False), {})
aten._softmax.default
TIMM/visformer_small
((T([128, 6, 196, 196], f16), -1, False), {})
aten._softmax.default
TIMM/visformer_small
((T([128, 6, 49, 49], f16), -1, False), {})
aten._softmax.default
TIMM/coat_lite_mini
((T([128, 8, 197, 40], f16, stride=(189120, 40, 960, 1)), 2, False), {})
aten._softmax.default
TIMM/coat_lite_mini
((T([128, 8, 3137, 8], f16, stride=(602304, 8, 192, 1)), 2, False), {})
aten._softmax.default
TIMM/levit_128
((T([128, 8, 49, 196], f16), -1, False), {})
aten._softmax.default
TIMM/levit_128
((T([128, 8, 49, 49], f16), -1, False), {})
aten._softmax.default
TIMM/coat_lite_mini
((T([128, 8, 50, 64], f16, stride=(76800, 64, 1536, 1)), 2, False), {})
aten._softmax.default
TIMM/coat_lite_mini
((T([128, 8, 785, 16], f16, stride=(301440, 16, 384, 1)), 2, False), {})
aten._softmax.default
HuggingFace/DistilBertForMaskedLM
((T([16, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
TorchBench/BERT_pytorch
((T([16, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/MobileBertForMaskedLM
((T([16, 4, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/PLBartForCausalLM
((T([192, 128, 128], f16), -1, False), {})
aten._softmax.default
TorchBench/hf_Longformer
((T([2, 1024, 12, 513], f16, stride=(6303744, 513, 525312, 1)), -1, True), {})
aten._softmax.default
TorchBench/hf_BigBird
((T([2, 12, 12, 64, 512], f16), -1, False), {})
aten._softmax.default
TorchBench/hf_BigBird
((T([2, 12, 64, 1024], f16), -1, False), {})
aten._softmax.default
TorchBench/hf_BigBird
((T([2, 12, 64, 448], f16), -1, False), {})
aten._softmax.default
TIMM/cait_m36_384
((T([2, 16, 1, 577], f16), -1, False), {})
aten._softmax.default
HuggingFace/MegatronBertForCausalLM
((T([2, 16, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/cait_m36_384
((T([2, 16, 576, 576], f16, stride=(5308416, 1, 9216, 16)), -1, False), {})
aten._softmax.default
HuggingFace/AlbertForMaskedLM
((T([2, 64, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/AlbertForQuestionAnswering
((T([2, 64, 512, 512], f16), -1, False), {})
aten._softmax.default
TIMM/sebotnet33ts_256
((T([256, 1024, 1024], f16), -1, False), {})
aten._softmax.default
HuggingFace/MBartForCausalLM
((T([256, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/Speech2Text2ForCausalLM
((T([256, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/swin_base_patch4_window7_224
((T([256, 16, 49, 49], f16), -1, False), {})
aten._softmax.default
TIMM/sebotnet33ts_256
((T([256, 256, 256], f16), -1, False), {})
aten._softmax.default
TIMM/mobilevit_s
((T([256, 4, 16, 16], f16), -1, False), {})
aten._softmax.default
TIMM/mobilevit_s
((T([256, 4, 256, 256], f16), -1, False), {})
aten._softmax.default
TIMM/mobilevit_s
((T([256, 4, 64, 64], f16), -1, False), {})
aten._softmax.default
TIMM/sebotnet33ts_256
((T([256, 64, 64], f16), -1, False), {})
aten._softmax.default
TorchBench/attention_is_all_you_need_pytorch
((T([256, 8, 31, 31], f16), -1, False), {})
aten._softmax.default
TorchBench/attention_is_all_you_need_pytorch
((T([256, 8, 31, 33], f16), -1, False), {})
aten._softmax.default
TorchBench/attention_is_all_you_need_pytorch
((T([256, 8, 33, 33], f16), -1, False), {})
aten._softmax.default
HuggingFace/YituTechConvBert
((T([3072, 9, 1], f16), 1, False), {})
aten._softmax.default
TIMM/twins_pcpvt_base
((T([32, 1, 3136, 49], f16), -1, False), {})
aten._softmax.default
HuggingFace/BartForConditionalGeneration
((T([32, 1024, 1024], f16), -1, False), {})
aten._softmax.default
HuggingFace/DistilBertForQuestionAnswering
((T([32, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/M2M100ForConditionalGeneration
((T([32, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/XGLMForCausalLM
((T([32, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/resnest101e
((T([32, 2, 1, 128], f16), 1, False), {})
aten._softmax.default
TorchBench/timm_resnest
((T([32, 2, 1, 128], f16), 1, False), {})
aten._softmax.default
TIMM/resnest101e
((T([32, 2, 1, 256], f16), 1, False), {})
aten._softmax.default
TorchBench/timm_resnest
((T([32, 2, 1, 256], f16), 1, False), {})
aten._softmax.default
TIMM/resnest101e
((T([32, 2, 1, 512], f16), 1, False), {})
aten._softmax.default
TorchBench/timm_resnest
((T([32, 2, 1, 512], f16), 1, False), {})
aten._softmax.default
TIMM/resnest101e
((T([32, 2, 1, 64], f16), 1, False), {})
aten._softmax.default
TorchBench/timm_resnest
((T([32, 2, 1, 64], f16), 1, False), {})
aten._softmax.default
TIMM/twins_pcpvt_base
((T([32, 2, 784, 49], f16), -1, False), {})
aten._softmax.default
HuggingFace/MobileBertForQuestionAnswering
((T([32, 4, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/twins_pcpvt_base
((T([32, 5, 196, 49], f16), -1, False), {})
aten._softmax.default
TIMM/twins_pcpvt_base
((T([32, 8, 49, 49], f16), -1, False), {})
aten._softmax.default
HuggingFace/GPT2ForSequenceClassification
((T([4, 12, 1024, 1024], f16), -1, False), {})
aten._softmax.default
HuggingFace/RobertaForCausalLM
((T([4, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/DebertaForMaskedLM
((T([4, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/DebertaForQuestionAnswering
((T([4, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
TorchBench/hf_Bert
((T([4, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
TorchBench/hf_GPT2
((T([4, 12, 512, 512], f16), -1, False), {})
aten._softmax.default
HuggingFace/XLNetLMHeadModel
((T([4, 16, 512, 512], f16), 3, False), {})
aten._softmax.default
TIMM/swin_base_patch4_window7_224
((T([4096, 4, 49, 49], f16), -1, False), {})
aten._softmax.default
HuggingFace/OPTForCausalLM
((T([48, 128, 128], f16), -1, True), {})
aten._softmax.default
TorchBench/hf_Bart
((T([48, 512, 512], f16), -1, False), {})
aten._softmax.default
TIMM/botnet26t_256
((T([512, 256, 256], f16), -1, False), {})
aten._softmax.default
TIMM/eca_botnext26ts_256
((T([512, 256, 256], f16), -1, False), {})
aten._softmax.default
TIMM/botnet26t_256
((T([512, 64, 64], f16), -1, False), {})
aten._softmax.default
TIMM/eca_botnext26ts_256
((T([512, 64, 64], f16), -1, False), {})
aten._softmax.default
TorchBench/fastNLP_Bert
((T([6, 12, 476, 476], f16), -1, False), {})
aten._softmax.default
HuggingFace/BartForCausalLM
((T([64, 1024, 1024], f16), -1, False), {})
aten._softmax.default
TIMM/volo_d1_224
((T([64, 12, 1, 197], f16), -1, False), {})
aten._softmax.default
HuggingFace/BertForMaskedLM
((T([64, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/BertForQuestionAnswering
((T([64, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
HuggingFace/RobertaForQuestionAnswering
((T([64, 12, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/volo_d1_224
((T([64, 12, 196, 196], f16), -1, False), {})
aten._softmax.default
TIMM/beit_base_patch16_224
((T([64, 12, 197, 197], f16), -1, False), {})
aten._softmax.default
TIMM/vit_base_patch16_224
((T([64, 12, 197, 197], f16), -1, False), {})
aten._softmax.default
TIMM/deit_base_distilled_patch16_224
((T([64, 12, 198, 198], f16), -1, False), {})
aten._softmax.default
HuggingFace/PegasusForConditionalGeneration
((T([64, 128, 128], f16), -1, False), {})
aten._softmax.default
TIMM/jx_nest_base
((T([64, 16, 1, 196, 196], f16), -1, False), {})
aten._softmax.default
TIMM/pit_b_224
((T([64, 16, 65, 65], f16), -1, False), {})
aten._softmax.default
TIMM/swin_base_patch4_window7_224
((T([64, 32, 49, 49], f16), -1, False), {})