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.embedding.default
HuggingFace/MobileBertForMaskedLM
((T([512, 512], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/MobileBertForQuestionAnswering
((T([512, 512], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/BlenderbotSmallForCausalLM
((T([512, 512], f16), T([128], i64)), {})
aten.embedding.default
HuggingFace/BlenderbotSmallForConditionalGeneration
((T([512, 512], f16), T([128], i64)), {})
aten.embedding.default
HuggingFace/BertForMaskedLM
((T([512, 768], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/BertForQuestionAnswering
((T([512, 768], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/DistilBertForMaskedLM
((T([512, 768], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/DistilBertForQuestionAnswering
((T([512, 768], f16), T([1, 128], i64)), {})
aten.embedding.default
HuggingFace/DebertaForMaskedLM
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/DebertaForQuestionAnswering
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/GoogleFnet
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/LayoutLMForMaskedLM
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/LayoutLMForSequenceClassification
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/YituTechConvBert
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
TorchBench/hf_Bert
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
TorchBench/hf_DistilBert
((T([512, 768], f16), T([1, 512], i64)), {})
aten.embedding.default
HuggingFace/RobertaForCausalLM
((T([512, 768], f16), T([4, 128], i64), 0), {})
aten.embedding.default
TorchBench/fastNLP_Bert
((T([512, 768], f16), T([6, 476], i64, stride=(0, 1))), {})
aten.embedding.default
HuggingFace/RobertaForQuestionAnswering
((T([512, 768], f16), T([64, 128], i64), 0), {})
aten.embedding.default
HuggingFace/TrOCRForCausalLM
((T([514, 1024], f16), T([8, 128], i64)), {})
aten.embedding.default
HuggingFace/CamemBert
((T([514, 768], f16), T([1, 512], i64), 1), {})
aten.embedding.default
TorchBench/attention_is_all_you_need_pytorch
((T([9521, 512], f16), T([256, 31], i64, stride=(1, 256)), 1), {})
aten.embedding.default
TorchBench/attention_is_all_you_need_pytorch
((T([9521, 512], f16), T([256, 33], i64, stride=(1, 256)), 1), {})
aten.embedding_dense_backward.default
HuggingFace/AllenaiLongformerBase
((T([1, 1024, 768], f16), T([1, 1024], i64), 1, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GPT2ForSequenceClassification
((T([1, 1024, 768], f16), T([1, 1024], i64), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BigBird
((T([1, 1024, 768], f16), T([1, 1024], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BigBird
((T([1, 1024, 768], f16), T([1, 1024], i64), 4096, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_BigBird
((T([1, 1024, 768], f16), T([1, 1024], i64), 4096, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/AllenaiLongformerBase
((T([1, 1024, 768], f16), T([1, 1024], i64), 4098, 1, False), {})
aten.embedding_dense_backward.default
HuggingFace/AllenaiLongformerBase
((T([1, 1024, 768], f16), T([1, 1024], i64), 50265, 1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BigBird
((T([1, 1024, 768], f16), T([1, 1024], i64), 50358, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/MegatronBertForCausalLM
((T([1, 128, 1024], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MegatronBertForQuestionAnswering
((T([1, 128, 1024], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GPTNeoForCausalLM
((T([1, 128, 2048], f16), T([1, 128], i64), 2048, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GPTNeoForSequenceClassification
((T([1, 128, 2048], f16), T([1, 128], i64), 2048, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GPTNeoForCausalLM
((T([1, 128, 2048], f16), T([1, 128], i64), 50257, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GPTNeoForSequenceClassification
((T([1, 128, 2048], f16), T([1, 128], i64), 50257, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MobileBertForMaskedLM
((T([1, 128, 512], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MobileBertForQuestionAnswering
((T([1, 128, 512], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BertForMaskedLM
((T([1, 128, 768], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BertForQuestionAnswering
((T([1, 128, 768], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DistilBertForMaskedLM
((T([1, 128, 768], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DistilBertForQuestionAnswering
((T([1, 128, 768], f16), T([1, 128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/ElectraForCausalLM
((T([1, 512, 128], f16), T([1, 512], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/ElectraForCausalLM
((T([1, 512, 128], f16), T([1, 512], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/AlbertForMaskedLM
((T([1, 512, 128], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/AlbertForQuestionAnswering
((T([1, 512, 128], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/ElectraForCausalLM
((T([1, 512, 128], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/ElectraForQuestionAnswering
((T([1, 512, 128], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_Albert
((T([1, 512, 128], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaV2ForMaskedLM
((T([1, 512, 1536], f16), T([1, 512], i64), 128100, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaV2ForQuestionAnswering
((T([1, 512, 1536], f16), T([1, 512], i64), 128100, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaV2ForMaskedLM
((T([1, 512, 1536], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaV2ForQuestionAnswering
((T([1, 512, 1536], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/CamemBert
((T([1, 512, 768], f16), T([1, 512], i64), 1, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DistillGPT2
((T([1, 512, 768], f16), T([1, 512], i64), 1024, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_GPT2
((T([1, 512, 768], f16), T([1, 512], i64), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/YituTechConvBert
((T([1, 512, 768], f16), T([1, 512], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/YituTechConvBert
((T([1, 512, 768], f16), T([1, 512], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/GoogleFnet
((T([1, 512, 768], f16), T([1, 512], i64), 32000, 3, False), {})
aten.embedding_dense_backward.default
HuggingFace/CamemBert
((T([1, 512, 768], f16), T([1, 512], i64), 32005, 1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GoogleFnet
((T([1, 512, 768], f16), T([1, 512], i64), 4, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DistillGPT2
((T([1, 512, 768], f16), T([1, 512], i64), 50257, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaForMaskedLM
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/DebertaForQuestionAnswering
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/GoogleFnet
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForMaskedLM
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForSequenceClassification
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/YituTechConvBert
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_Bert
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_DistilBert
((T([1, 512, 768], f16), T([1, 512], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/CamemBert
((T([1, 512, 768], f16), T([1, 512], i64), 514, 1, False), {})
aten.embedding_dense_backward.default
TorchBench/speech_transformer
((T([10, 22, 512], f16), T([10, 22], i64), 1014, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BlenderbotSmallForCausalLM
((T([128, 512], f16), T([128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BlenderbotSmallForConditionalGeneration
((T([128, 512], f16), T([128], i64), 512, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MBartForCausalLM
((T([16, 128, 1024], f16), T([16, 128], i64), 1026, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MBartForCausalLM
((T([16, 128, 1024], f16), T([16, 128], i64), 50265, 1, False), {})
aten.embedding_dense_backward.default
HuggingFace/MobileBertForMaskedLM
((T([16, 128, 128], f16), T([16, 128], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/MobileBertForMaskedLM
((T([16, 128, 512], f16), T([16, 128], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/PLBartForCausalLM
((T([16, 128, 768], f16), T([16, 128], i64), 1026, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/BERT_pytorch
((T([16, 128, 768], f16), T([16, 128], i64), 20005, 0, False), {})
aten.embedding_dense_backward.default
TorchBench/BERT_pytorch
((T([16, 128, 768], f16), T([16, 128], i64), 3, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/DistilBertForMaskedLM
((T([16, 128, 768], f16), T([16, 128], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/PLBartForCausalLM
((T([16, 128, 768], f16), T([16, 128], i64), 50005, 1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 512, 768], f16), T([16, 512], i64), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 512, 768], f16), T([16, 512], i64), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 512, 768], f16), T([16, 512], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 512, 768], f16), T([16, 512], i64), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 512, 768], f16), T([16, 512], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 512, 768], f16), T([16, 512], i64), 30522, 0, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForMaskedLM
((T([16, 512, 768], f16), T([16, 512], i64, stride=(2048, 4)), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/LayoutLMForSequenceClassification
((T([16, 512, 768], f16), T([16, 512], i64, stride=(2048, 4)), 1024, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BartForConditionalGeneration
((T([2, 1024, 1024], f16), T([2, 1024], i64), 1026, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/BartForConditionalGeneration
((T([2, 1024, 1024], f16), T([2, 1024], i64), 50265, 1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_Longformer
((T([2, 1024, 768], f16), T([2, 1024], i64), 1, -1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_Longformer
((T([2, 1024, 768], f16), T([2, 1024], i64), 4098, 1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_Longformer
((T([2, 1024, 768], f16), T([2, 1024], i64), 50265, 1, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_BigBird
((T([2, 1024, 768], f16), T([2, 1024], i64), 50358, 0, False), {})
aten.embedding_dense_backward.default
TorchBench/hf_BigBird
((T([2, 1024, 768], f16), T([2, 1024], i64, stride=(0, 1)), 2, -1, False), {})
aten.embedding_dense_backward.default
HuggingFace/M2M100ForConditionalGeneration
((T([2, 128, 1024], f16), T([2, 128], i64), 128112, 1, False), {})